PDF hosted at the Radboud Repository of the Radboud University
Nijmegen
The following full text is a publisher's version.
For additional information about this publication click this link.
http://repository.ubn.ru.nl/handle/2066/127229
Please be advised that this information was generated on 2015-01-25 and may be subject to
change.
Children from low-income and
immigrant backgrounds:
to what extent are they at risk?
© 2014 Marije Janssen
Layout: Gert Jan Bosgra
Production: Ipskamp
ISBN: 978-94-6259-192-9
Children from low-income and
immigrant backgrounds:
to what extent are they at risk?
Proefschrift
ter verkrijging van de graad van doctor
aan de Radboud Universiteit Nijmegen
op gezag van de rector magnificus prof. mr. S.C.J.J. Kortmann,
volgens besluit van het college van decanen
in het openbaar te verdedigen op dinsdag 24 juni 2014
om 10.30 uur precies
door
Marije Janssen
geboren op 24 oktober 1981
te Nijmegen
Promotoren
prof. dr. A.M.T. Bosman
prof. dr. P.P.M. Leseman (Universiteit Utrecht)
Copromotor
dr. J.T.A. Bakker
Manuscriptcommissie
prof. dr. J.M.A.M. Janssens
prof. dr. E.P.J.M. Elbers (Universiteit Utrecht)
prof. dr. A.E.M.G. Minnaert (Rijksuniversiteit Groningen)
Opgedragen aan Laila, Reda en Driss
Paranimfen
Janneke Janssen
Dorothee Schenkelaars
Contents
CHAPTER 1
General introduction
9
CHAPTER 2
Differential relationships between language skills and working memory
in Turkish-Dutch and native-Dutch first-graders from low-income families
35
CHAPTER 3
Phoneme awareness, vocabulary, and word decoding in monolingual and
bilingual Dutch children
59
CHAPTER 4
Limited language proficiency is not necessarily detrimental for the
wellbeing of children from low-income and/or immigrant backgrounds
77
CHAPTER 5
Differential trust between parents and teachers of children from lowincome and immigrant backgrounds
91
CHAPTER 6
General discussion
111
CHAPTER 7
Nederlandse samenvatting
127
Dankwoord
135
Curriculum vitae
137
7
8
chapter 1
general introduction
9
chapter 1
10
general introduction
The years before a child goes to school are important for a successful academic career,
which in turn provides the conditions necessary for achieving a fulfilling position in our
society. Children from lower socio-economic classes often lack the cognitive and social
skills that are required to successfully participate in formal education (Leseman & Cordus,
1994). Preschool education may increase chances for disadvantaged children when they
enter primary school. To reduce differences between children in academic performance,
school career, and eventually social success, preschool education received a prominent
place in educating children in the Netherlands. Since the nineties, the pre-school policy
of the Netherlands aimed at offering preschool education to disadvantaged children in
order to stimulate their cognitive, language, and, social-emotional development (Reezigt,
2003). If these skills were to improve before they enter primary school, they should be
better prepared for the more formal way of learning at school (Blok & Leseman, 1996).
Understanding the advantages of preschool education is closely related to
understanding that the pre-school period is a critical one for learning fundamental
language, cognitive and social-emotional skills (Leseman, 2001). Shonkoff and
Phillips (2000) conclude in their study that the period of early childhood is eminently
characterized as one - neurobiological - sensitive period.
The Educational Priority Policy in the Netherlands aimed at increasing
opportunities for disadvantaged children in primary education due to social,
economic, or cultural conditions. The language policy of the schools and preschool
institutions was directly aimed at improving the proficiency of the Dutch language.
Professional approaches to teaching, instruction, and attention for general learning
skills of students were central concerns. The preschool period was strongly linked to
the first period of primary education.
In order to prepare young children from lower socio-economic classes for formal
learning in primary schools, a number of different projects have been set up during
the past few decades in both the Netherlands and abroad. These projects aimed at
developing preschool skills (for an overview, see Leseman, Otter, Blok, & Deckers, 1998;
van der Wolf, 1999). The preschool programs offered as part of these projects are diverse
in nature. Some programs focus on providing wide support in terms of family and child
rearing, others tend to focus on the individual child’s education. Furthermore, some
programs are primarily concerned with providing in-home support (i.e., ‘home-based
programs’), whereas other programs provide support in day-care centres, playgroups and
elementary schools (i.e., ‘centre-based programs’).
The so-called ‘Taalstimuleringsprogramma’ (Language Stimulation Program, or ‘LSP’),
which was conducted in the city of Nijmegen in the Netherlands, is one of the programs
11
chapter 1
that fell within the category of educational pre-school programs, aimed at improving
the linguistic and cognitive development as well as the socio-emotional development of
children from lower socio-economic classes, regardless of their ethnicity. In view of its
setup (see below), this program can be considered a combination of a home-based and a
centre-based program.
Certain groups of both ethnically Dutch children and children of immigrant descent
have accumulated an educational deficit that is large enough to considerably impair their
chances of academic and social success (Leseman & Cordus, 1994). The goal of our
research is to contribute to the effort of solving this problem. It provides insight into
the role of some specific factors of academic success, namely, cognitive development
(i.e., phonological awareness and working memory), socio-emotional development, and
trust between parents and teachers. This thesis concerns aspects that are associated
with educational delays in children from lower socio-economic backgrounds, of which
minorities constitute a large proportion.
Insight into the extent to which children’s socio-emotional development and
behaviour may contribute to social and academic success could inspire policy makers
to assign more time and attention in schools to this particular aspect of development.
Insight into children’s cognitive development as well as parental educational theories in
terms of their children’s academic career and into the extent to which these contribute to
the effectiveness of the child’s academic career could improve communication between
teachers and parents.
This chapter provides the background to the studies presented in Chapters 2 to 5.
It starts with a description of the background of this study. Then, each of the major
aspects of this thesis will be discussed: Language and cognitive development and
parental involvement. The final part of the chapter presents the outline of the remainder
of the thesis.
Historical background
Toddler-language program developed by Kion and HAN
The LSP was developed and applied between August 1991 and May 1998, by the
Speech Therapy Department of the HAN (´Hogeschool van Arnhem en Nijmegen´ or
University of Applied Sciences of Arnhem and Nijmegen), in close collaboration with and
commissioned by the Kion foundation (´Kinderopvang Nijmegen´ or Day-Care Nijmegen).
In 1996, the LSP was honoured by the Stichting Logopedie Fonds België-Nederland (DutchBelgian Foundation for Speech Therapy) with the S.L.F. Award (Van den Heuvel, 1996).
Prior to 1998, the program was conducted and evaluated on a somewhat limited scale
12
general introduction
(De Beer & Van den Heuvel, 1995; Scholten & Snelder, 1997) in a number of playgroups
in some of Nijmegen’s less affluent neighbourhoods (known as intensive playgroups).
As of August ’98, the program is conducted in every one of the city’s 20 intensive
playgroups, and was financed by the GOAP (‘Gemeentelijk Onderwijs Achterstanden
Plan’ or Municipal Plan on Educational Gaps). Ever since the VVE-policy (‘Voor- en
Vroegschoolse Educatie’ or Pre- and Early-school education) was implemented in January
of 2002, ten regular playgroups and a day-care centre have also been added, with a grand
total of 67 groups of children. Cooperation and coordination with elementary schools
have been structured in the context of the VVE-policy and were entrusted to 13 alliances
between playgroups and elementary schools.
Short description of the LSP
The LSP had two main objectives. First, it was intended to improve the speech and
language development of toddlers from lower socio-economic classes in a way that would
lead to better opportunities throughout their academic careers. Second, the program
focused on improving parental involvement in the speech and language development of
their toddler, by providing them with information and insight in terms of the role that
the parents could play in this development.
The LSP was implemented in two ways. The first of which focused on the language
deficit in children of immigrant descent, whereas the other focused on second language
acquisition in children of immigrant descent. Both contained 20 different themes,
including ‘winter’, ‘playground’ and ‘stuffed animals’, which were dealt within 40
language hours (themes are covered twice). Each language hour was embedded within
a series of recurring activities, namely the welcoming song, food and beverages, free
playtime, and the goodbye song. The language of the program did not increase with
respect to complexity during the program, neither in the ethnically Dutch group, nor
in the group with children from immigrant descent. This setup ensured that a toddler
could enrol in the program at any time. The language hours were held weekly in groups
of up to six toddlers. Two adults supervised them: A teacher and an assistant.
The language exercises in the program for Dutch toddlers focused on basic
preconditions, including concentration/listening, eye contact, oral motor skills, socioemotional and communicative development, as well as language preconditions, including
active and passive vocabulary, syntax, audibility, word types, and understanding. The
language exercises in the program for children of immigrant descent were, more than
those aimed at the ethnically Dutch children, geared to materializing the language
in a sensory way, making the language visible through gestures, sounds, materials,
touching, etcetera. Furthermore, the program for children of immigrant descent used
a vocabulary list to register the words that were offered to the children. During the
13
chapter 1
hours in which the toddlers attended the playgroup, the themes that were dealt with
during the language hours were revisited, repeated and elaborated by way of activities.
The VVE-policy started in 2002 and it provided the extra manpower that was needed in
order to pay more attention to the smaller groups and individual children during regular
playgroup hours.
The role of the parents was an important aspect of the LSP. Prior to and during
the program, parent-teacher conferences were held. The parents were provided with
information on language development, language stimulation, and language acquisition.
Parents of children who were enrolled in the program attended the meetings aimed at
the activities they should do with their child at home and explained the role of language
and second language acquisition in the toddler’s development. Also, a contact register
was kept for the duration of the LSP. This contact register identified certain exercises and
elements of the language hours that parents were asked to do at home with their toddler.
The contact register for Dutch children also contained advice from the teacher, aimed
at the child’s specific language problem. Parents described their own experiences with
this advice. The contact register for parents of children of immigrant toddlers employed
visual aids, allowing parents to see which subjects were covered during the language
hours enabling to discuss them with their child in their own language.
The LSP’s short-term effects
In 1994, a study was conducted to evaluate the short-term effects of the LSP on
immigrant low-income children (de Beer & van den Heuvel, 1995). A comparison
was made of the language development of two groups, an experimental group and a
control group. The toddlers in the experimental group participated in the LSP for six
months. The toddlers in the control group did not participate in the LSP. However, the
parents of the latter group were provided with information on the study and exercises to
stimulate their child’s language development, as were the parents of the toddlers in the
experimental group.
Results from the Language Assessment Remediation and Screening Procedure
(LARSP, Schlichting, 1988) indicated that toddlers from the control group, when
compared to the experimental group, had fallen behind in their morphosyntactic and
pragmatic skills by an average of two months. Furthermore, the toddlers from the
experimental group displayed a two-month advantage in terms of their general language
understanding, determined by the language understanding section of the Reynell test
on concentration, readiness to listen, memory, and vocabulary (van Eldik, Schlichting,
Lutje Spielberg, van der Meulen, & van der Meulen, 1997). Note that this is a two-month
advantage compared to regular models of development. Toddlers from the control group
did not display similar progress.
14
general introduction
Kion-RU project
In 2002, the Radboud University Nijmegen and childcare centre Nijmegen (KION)
started to discuss the possibilities of a longitudinal study to analyse the effects of the
Language Stimulation program used by KION at that time. Because this preschool
program met the requirements for a successful early childhood program and participation
was guaranteed by Kion, some important conditions were met to conduct an evaluative
study of this preschool program. At the same time, however, a number of important
conditions for a valid effect study were lacking. The intervention was of limited duration
and intensity, only 40 language hours. More importantly, a purely experimental/control
group design was impossible to achieve, because almost all children participated in the
intervention and the parents of those who did not, were unwilling to give permission.
An additional problem emerged over the years in the group who had received the
intervention: Children and their parents dropped out of the study for various reasons,
which lead to incompleteness of the cohort data.
Although the original plan of this study had to be abolished, it was nevertheless
possible to study factors related to language delay of low-income native-Dutch and
minority children and their effects over the years. Therefore, the focus of the present
study moved to factors of language development, socio-emotional development, the
impact of bilingualism on phonological awareness and working memory, and the
influence of trust between parents and teachers on language development. Each of these
aspects will be introduced below.
The role of language development in education
Language development in children starts from the moment they are born. Through
interaction with the environment, the child learns the language it hears. During the
years before they start attending school, children learn the basic rules of the language
that is spoken in their environment. Disadvantaged children in the Netherlands, usually
children from low-income and/or immigrant backgrounds, often have an insufficient
command of the Dutch language when they go to primary school. In the following
paragraphs we will discuss the impact of a delayed language development for bilingual
children and children from low-income backgrounds.
Bilingualism
In the first years of life, children learn a culturally appropriate way of using language by
interacting with others. Not only the interaction with the parents is important, also being
subjected to conversations with others, such as (pre)school teachers and peers, advances
15
chapter 1
language development (Bornstein, Haynes, & Painter, 1998; Hoff, 2006; Hoff & Naigles,
2002). For example, Huttenlocher, Vasilyeva, Cymerman, and Levine (2002) investigated
individual differences in the acquisition of syntactic skills of 4-year-old monolingual
children related to language input. They distinguished between multi-clause (complex)
sentences and simple sentences in both language production and comprehension. They
found a high correlation between the number of multi-clause sentences produced and
comprehended by the children and the number of multi-clause sentences used by the
parents as well as the teachers of the preschools attended by the children.
In present society, the majority of children grow up in a bilingual or even in a
multilingual environment (Bialystok, 2007; Grosjean, 1982). Grosjean considers a
monolingual person as someone using only one language on a regular basis. For instance,
in the Netherlands, many native-Dutch usually have a reasonable command of English
and German, and in some cases also of French. Thus, according to this definition, the
majority of the Dutch are bilingual. However, the proficiency in the second language does
not attain the proficiency of a native speaker of that language. Exposure to and use of a
second language of a bilingual is usually limited to the school environment. People who are
well-balanced bilinguals usually learned to speak both languages at home simultaneously.
Simultaneous bilinguals are exposed to two languages in about the same degree from
birth. When a child is exposed to a second language after acquiring the first language this
is called successive bilingualism (Bhatia, 2006; Yip & Matthews, 2007).
Language acquisition in monolinguals and bilinguals is found to follow the same
developmental path (Gutiérrez-Clellen, Simon-Cereijido, & Wagner, 2008; Li, 2005;
Paradis & Genesee, 1996). Both mono- and bilingual children start with babbling, than
become gradually more competent in word learning through the one-word stage, twoword stage, multiword stage and finally acquire more complex language in the multiclause stage (de Houwer, 2006; Li, 2005; Pearson, Fernandez, Lewedeg, & Oller, 1997).
Language input is of utmost importance for language acquisition. Language input by the
caregiver is gradually adjusted to the attained level of language proficiency of the child
(Bates & Goodman, 1997; Schaerlaekens & Gillis, 1987). Particularly the quantity and
quality of language input is important for language outcomes. Because language input is
known to vary vastly between families, social classes, and ethnic-cultural communities
this could explain the early arising of individual and group differences (Hoff, 2006).
The (strong) relation between input and outcome in language development is a critical
issue in the case of bilingualism. In order to become a balanced and proficient bilingual,
children must be exposed to both languages in a roughly equal and, for each language
separately, sufficient degree (Oller & Eilers, 2002; Pearson, 2007).
These conditions, however, are rarely fulfilled in the case of bilingual immigrant
communities. Not only the quantity of exposure is likely to differ between the first
16
general introduction
and the second language, also the quality may reveal profound differences in terms
of vocabulary, grammar, and pragmatics (Scheele, Leseman,& Mayo, 2010). The first
language is mostly used for informal and interpersonal communication at home,
whereas the second language is usually the language of schooling and public media.
Thus, a bilingual, even if he or she has been exposed to the two languages from birth,
does not necessarily acquire the two languages equally well. This will depend on the
quality and quantity of the input the child receives in each language. Although the
language development of a bilingual child follows the same developmental stages as
of monolinguals in either language, there are also important differences. In an early
stage, young bilingual children are often found to mix the two languages as if both
languages are represented in a single system in the brain. In a later stage, however,
language mixing, or code-switching as it is called, seems more purposefully, depending
on the communicative context and the functions of language use, suggesting that both
languages are now represented separately (Bialystok, 2007; Petitto et al., 2001).
Language development in children from lower socio-economic and minority
backgrounds
Research has shown that the quality and the quantity of language input, especially
the quality, determines language outcome to a large degree, and this may explain the
differences that have been observed among individuals and between groups (Hoff,
2006).The acquisition of language emerges from social interactions and intentions,
which in fact is mainly studied for Western middle-class mothers. In other cultures
and/or socio-economic strata, many children may not be as actively stimulated and
engaged in interaction starting from a young age. Children from other cultures and
lower socio-economic strata (SES) mostly overhear speech from adults instead of actively
participating in a joint-attention framework (Tomasello, 2005). Less educated mothers
were found to talk less with their children, have a more directive style of communicating,
and use a less varied vocabulary and shorter utterances (Hoff, 2006; Hoff, Laursen,
&Tardif, 2002; Lieven, 1994). These differences in language input between higher and
lower SES mothers are related to children’s language development. Differences were
found for expressive as well as for comprehensive language outcomes (Huttenlocher
et al., 2002). Hoff and Tian (2005) found similar results for SES-related influences on
language development for Western and Asian culture. For example, immigrant families
in the Netherlands show a more disciplined and authoritarian parenting style than
native Dutch families (Pels, 2000).
Children from lower SES groups and children from ethnic minority backgrounds
appear to have a much harder time to successfully attend mainstream education. At
present, it is believed that the major problem to be solved is the development of the
17
chapter 1
language skills of these groups. Although these children may show a language delay,
this does not necessarily mean they have a language problem. A language delay may be
caused by cognitive problems like a low IQ or a medical cause like hearing problems.
When there are no clear factors causing the language delay a language delay is most
likely the result of insufficient exposure to the language at hand (van der Ploeg, Lanting,
Galindo Garre, & Verkerk, 2007). A delay caused by insufficient exposure occurs in
immigrant children when the home-language does not correspond with the school
language, or in native-Dutch children when they are exposed to the Dutch language in
an insufficient way.
Causes of a language delay
In general terms, immigrant children accumulate a larger educational deficit than
native-Dutch children from lower social economical background. This often appears
to be the result of a low socio-economic background combined with a language deficit,
which exists before these children even start elementary school (Ledoux, 1996). This
language deficit may be explained by the linguistic-, cultural-, and socio-economical
aspects within the family. The language delay that immigrant children face is the fact
that, within their families, Dutch is spoken only on rare occasions or not at all. This
means that the first time these children come into contact with the Dutch language is
when they enrol in elementary school or kindergarten, in contrast to their native-Dutch
peers, who generally speak no other language than Dutch. If the immigrant family does
in fact speak Dutch at home, usually the same problems arise as those that occur in
Dutch families from lower socio-economic backgrounds. The question then presents
itself whether the risk might still possibly lie within the cultural differences. A cultural
difference could be, for example, that some cultures put less emphasis on activities like
talking to children or reading to them.
The language problems of immigrant children’s parents might be an important
cause of the delay in education. When parents of immigrant children do not speak Dutch
with their children, their vocabulary might be small when they enrol in elementary
school or kindergarten. These children are not able to understand instruction properly
or may have problems understanding materials used for education. As a result, they
may have difficulties processing information. The final conclusion of Aarts, de Ruiter,
and Verhoeven (1996) was that the language delay of parents, family culture, socialcultural orientation of the child, and social-economic status cause language delay of
immigrant children. Social-economical status appears to be more predictive for level
of education than language skills. However, Kervezee (2005) revealed that the delay of
immigrant children is considerable larger than the delay of native-Dutch children from
low-income families. Child-, family-, and school characteristics determine the level of
18
general introduction
academic achievement. Social en cultural aspects also play an important role. Children
and their parents who have a positive attitude towards the Dutch language and culture,
and who speak Dutch and use more language-oriented interactions affect the children’s
academic achievement positively. In addition, the degree in which parents support their
child and the extent in which they use the Dutch language at home also affects academic
achievements (Aarts et al., 1996; Driessen & van der Slik, 2004; Leseman, Sijsling, JapA-Joe, & Sahin, 1995).
Many studies show that socio-economical status is an important determinant in
language acquisition. Chapman, Chapman, Kwapil, Eckblad, and Zinser (1994) and
Leseman (2007) mentioned risk factors for language development associated with a
lower SES including malnutrition, parental psychopathology, adverse neighbourhoods,
large number of children, authoritarian parenting style, illness, or unemployment within
the family. They found that for each risk factor the language score of a child went down.
Risk factors for educational disadvantage are factors within the child and/or factors
within their environment. Level of education of parents is said to be the best predictor
of school success of children (Meijnen, 2003). For disadvantaged immigrant children
the fact that the school language is their second language might be an additional risk
factor. Although, school success seemed more related to SES than to ethnicity. Higher
SES immigrant parents seemed, just like higher SES native-Dutch parents able to
stimulate their children and use childrearing principles in order to stimulate cognitive
development of their children. These children were better prepared for primary school
(Meijnen, 2003).
Although a recent improvement in the educational position of primarily immigrant
children can be discerned, a persistent deficit still remains (Reezigt, 2003). The
disappointing results of the Dutch educational deficit policy have ultimately
demonstrated in the nineties that both immigrant and native-Dutch children from low
socio-economic backgrounds have accumulated considerable educational deficits by the
time they start elementary school, despite the efforts and funds that have been invested
in educational deficit policies throughout the years. This is partly the reason why more
and more attention is directed towards preventing deficits from emerging in the years
before children start elementary school, as well as during their first years in school.
Consequences of a poor or inappropriate development of language
The initial educational deficit of immigrant children from low socio-economic
backgrounds, especially in terms of language, could impair their ability to fully take
advantage of primary school education, which may adversely affect their academic
performance throughout their educational career. At the end of primary school,
Turkish and Moroccan children have an average delay of 2 to 2,5 years, whereas Dutch
19
chapter 1
children from low-income families show a delay of 1 year (Kervezee, 2005). Immigrant
children appear to dropout more often (5.2% of the immigrant children and 2.7% of the
Dutch children). Eleven percent of the non-western immigrant children and seventeen
percent of the Dutch children eventually obtained their basic educational qualification
(Centraal Bureau voor de Statistiek [CBS], 2011). Tesser and Iedema (2001) showed that
children from low-income minority families go to primary school with a delay in their
cognitive- and (Dutch) language development. These children first have to acquire the
language used in school in order to profit from lessons (Elzer, 2005; van Elten, 2003).
When children from minority backgrounds show a delay in their language development
and sensory, cognitive, neurological, or emotional problems cannot explain this delay,
then this could be the result of a lack of language input (Goorhuis & Schaerlaekens,
2000). Particularly children growing up in a disadvantaged situation may profit from
pre-school education. As a consequence pre-school programs were designed in order to
reduce educational disadvantages.
Language acquisition in relation to cognitive and socio-emotional development
Based on the above, it is clear that language development is strongly dependent on
the opportunities for learning and language input provided by the environment of the
child. Also child characteristics for language processing, like memory or the abilities for
hearing, and discrimination sounds are needed. Two cognitive skills affected by language
development are phonological awareness and working memory (D’Angiulli, Siegel, &
Serra, 2001; de Jong & van der Leij, 2003; Scheltinga, van der Leij, & van Beinum, 2003;
Snowling, 2000). To learn a newly encountered word, a phonological representation
must be created in working memory. There can be phonological problems in vocabulary
learning, phonological deficits, or problems with working memory that can cause
problems in learning new verbal information. Phonological problems or problems with
working memory may impair the acquisition of new phonological sequences (Aguiar &
Brady, 1991; Service, 1992).
Working memory. Language acquisition strongly depends on the capacity to construct
and temporarily maintain a phonological representation of speech input in short term
memory (Gathercole, 2006). Phonological short-term memory is part of the working
memory system and functions to keep phonological information available for further
processing, integration, and understanding (Gazzaniga & Heatherton, 2003). For a
proper vocabulary development, accurate perception, storage, and retrieval of words
are of great interest. Verbal memory capacity has been found to affect vocabulary and
reading acquisition (de Jong, 1998; Elbro, 1996; Gathercole & Baddeley, 1993; Leather &
Henry, 1994).
20
general introduction
According to a recent reformulation of the original model of Baddeley and Hitch
(1974), working memory consists of four components: Phonological loop, visuospatial
sketchpad, central executive, and the episodic buffer. Both the phonological loop and the
central executive were found to be good predictors of language acquisition. The central
executive is expected to have a more general contribution to language comprehension
(Bialystok, 2005; Gathercole & Baddeley, 1993; Papagno & Vallar, 1995). When complex
syntactical and textual structures must be processed, controlled processing and optimal
use of different memory subsystems are required (cf. Kintsch, 2004). A number of
studies have shown that individual differences in the capacity of the phonological loop
predict language acquisition (e.g., Baddeley, Gathercole, & Papagno, 1998; Gathercole,
2006; Majerus, Poncelet, Greffe, & Van der Linden, 2006). Measures of verbal shortterm memory span like digit recall have been found to strongly correlate with different
aspects of language development, such as receptive vocabulary knowledge, vocabulary
specificity, mean length of utterances, and syntactic diversity (Adams & Gathercole,
2000). Short-term memory is, according to these studies, of particular importance for
the acquisition of language.
Phonological awareness. “An explicit awareness of the phonological structure of
the words in one’s language” is called phonological awareness (Torgesen, Wagner, &
Rashotte, 1994, p. 276), whereas phoneme awareness is “…an awareness of sounds in
spoken (not written) words that is revealed by such abilities as rhyming, matching initial
consonants, and counting the number of phonemes in spoken words” (Stahl & Murray,
1994, p. 221). Phoneme awareness is closely related to vocabulary and is one of the best
predictors of word-decoding abilities in both monolingual (see Blachman, 2000 for a
review) as well as bilingual children (e.g., Stuart, 1999, 2004). Note that, research has
also shown that when children start to learn to read and write, phonological skills, and
in particular phoneme-awareness skills develop as a result of the orthographic input
of written language, that is when they are learning to read and spell (Liberman &
Liberman, 1990; Morais, 1991).
Socio-emotional development. Language and communicative skills appear to be
closely linked to a child’s socio-emotional development and behaviour (Schaerlaekens,
2002; van Hell, 2002; van Lieshout & Haselager, 1993). A healthy socio-emotional
development has shown to improve future academic performances (e.g., Green,
Forehand, Beck, & Vosk, 1980). Children who are behind in terms of their language
and communicative skills are more likely to develop behavioural problems (Coster,
Goorhuis-Brouwer, Nakken, & Lutje Spelberg, 2002; Eleveld, Nakken, & GoorhuisBrouwer, 1994; Rice, Sell, & Hadley, 1991), which in turn may lead to children leaving
21
chapter 1
school prematurely and engaging in delinquent behaviour (Parker & Asher, 1987).
Insufficient mastery of the dominant, school language may not just be related to
an educational delay, it may also cause the development of social-emotional problems.
A number of studies have revealed that language development and social-emotional
development mutually affect each other and language difficulties may affect children’s
social-emotional development negatively (Coplan & Armer, 2005; Stanton-Chapman,
Justice, Skibbe, & Grant 2007).The relationship between language development and
socio-emotional development appears to be bi-directional a language delay may influence
communication and in turn affect social-emotional and/or behavioural development
(Redmond & Rice, 1998), and at the same time social-emotional development is essential
for a healthy language development (Goorhuis & Schaerlaekens, 2000).
The relationship between language delays and social-emotional problems has been
studied thoroughly in clinical samples (i.e., a speech or language impairment) with
mainly children from middle to high socio-economic backgrounds who are mostly from
Caucasian origin (Beitchman, Wilson, Brownlie, Walters, & Lancee, 1996; StantonChapman et al., 2007). Robust knowledge about the relationship between language delays
and social-emotional problems in children who merely lag behind in the development
of the dominant language of the society they live in is, however, lacking. Although
children from low-income and minority backgrounds are at risk for developing socialemotional or behaviour problems (Leseman, 2002; Raver, 2002), there is no evidence
that a language delay in this group is related to the development of social-emotional
problems per se.
Parental involvement
Parents play an important, if not the most important, role in the cognitive development
of their children. Their involvement has always been found to be determinative for the
success of all kinds of (language) stimulation programs that were aimed at ensuring
proper coordination between school and home environments (Leseman & de Jong, 1998).
The LSP also assigned a large role to the parents. During the child’s participation
in the LSP, the parents were expected to play an active part. During parent-teacher
conferences, they were informed on their role in the toddler’s language development and
they were provided with games and other materials with which they can stimulate their
child’s language skills. Furthermore, the contact register, which was kept by both parents
and LSP teachers, guaranteed continuity between a child’s home and its playgroup.
Parental involvement, however, can only be controlled up to a certain level. Fantuzzo,
Davis, and Ginsburg (1995) argue that there is too wide a divergence in parental behaviour
to facilitate full control. The fact that it matters is proven, but why it matters is a question
that, according to them, is more difficult to answer. The question becomes more
22
general introduction
pressing given that most parents have their child’s best interests at heart. Subsequently,
most parents, regardless of social background and ethnicity, will do whatever they can
to ensure a successful academic career for their child. In this light, the fact that some
parents succeed in their intentions while others fail should not be seen as result of their
level of commitment or involvement.
Rather, different views on the way children learn and the way children ought to learn
are to blame for the fact that the school’s agenda is frequently not in line with that of the
home (Serpell, 1997). In other words, parents often adhere to widely varying ‘theories on
education’, which appear to vary greatly in terms of their ability to successfully prepare
a child for school. The literature on early literacy represents a striking illustration of
this phenomenon. While some parents consider reading to be a skill that is difficult to
master, requiring a step-by-step approach, others are able to teach their children how to
read in a practical and playful manner (Baker, Sonnenschein, Serpell, Fernandez-Fein,
& Scher, 1994).
More generally, in certain social circles the line between playing and learning is less
rigid and, from an early age, playing is considered to be a part of the learning process,
and vice versa. Thus, it might simply be due to a particular type of child-rearing that some
children are considerably better equipped to start their academic careers (Sonnenschein,
Brody, & Munsterman, 1996). Generally, these differences in child rearing are linked to
socio-economic status.
During the past 40 years, the role of parents in the educational system has changed
drastically from relatively separate to shared responsibilities (Adams & Christenson,
2000). Parental involvement is considered to be an important part of development,
because parents can present their children with opportunities for development, such
as reading aloud and talking to their child about the story, support the child to read, the
presence of adult and children books at home, going to the library with the child and
emphasizing positive feelings towards reading (Baker, Afflerbach, & Reinking, 1996;
McCarthey, 2000; Sonnenschein et al., 1996).
Academic performance of children is a dynamic and complex process of child
characteristics, home variables, school variables and home-school partnership especially
for low income and minority children. It is assumed that when home, school, and
community, accomplish a strong relationship and cooperate with one another this will
foster parental involvement, irrespective of social economical class or ethnic background,
and as a consequence may improve children’s development. Therefore parents and
teachers must be willing to trust each other, because “Partnership is based on mutual
trust” (Deslandes, 2001, p. 2).
An important aspect of the home-school relationship is found to be teachers’
trust. Teachers’ trust has also been found to predict academic performance (Goddard,
23
chapter 1
Tschannen-Moran, & Hoy, 2001; Sui-Chu & Douglas, 1996). Furthermore, research
shows that trust is usually based on cultural norms and obligations about educating
children. When parents are from a different socio-economic stratum or another ethnic
background than their child’s teacher, this may easily lead to misunderstanding,
stereotypes, and different expectations, which in turn may affect the trust-relationship
negatively or amplify distrust (Bakker & Denessen, 2007). Because disadvantaged
children are more vulnerable for school failure, trust in these groups is even more
important to help them making school progress (Goddard et al., 2001).
Outline of the thesis
The second chapter of this thesis concerns the relationship between working
memory skills and language development of monolingual and bilingual, especially
Turkish-Dutch children from similar low socio-economic background. Chapter 3
concerns the relationship between phonological awareness and language development
of monolingual and bilingual children, specifically young Turkish-Dutch children from
similar low socio-economic background. The fourth chapter addresses the question:
What is the relationship between a language delay in preschool and the development of
socio-emotional and behaviour problems in low-income Dutch and immigrant children.
The fifth chapter will be about the influence of trust between parents and teachers on
academic performances of children. Finally the last chapter summarizes the findings
and discussions and will give some recommendations.
24
general introduction
References
Aarts, R., de Ruiter, J. J., & Verhoeven, L. (1996). Eigen-taalvaardigheid en schoolsucces
van Turkse en Marokkaanse leerlingen [First language skills and academic
achievement of Turkish and Moroccan students]. Pedagogische Studiën, 73, 275-290.
Adams, A., & Gathercole, S. (2000). Limitations in working memory: Implications
for language development. International Journal of Language and Communication
Disorders, 35, 95-116. doi: 10.1080/136828200247278
Adams, K. S., & Christenson, S. L. (2000). Trust and the family-school relationship
examination of parent-teacher differences in elementary and secondary grades.
Journal of School Psychology, 38, 477-497. doi:10.1016/S0022-4405(00)00048-0
Aguiar, L., & Brady, S. (1991). Vocabulary acquisition and reading ability. Reading and
Writing: An Interdisciplinary Journal, 3, 413-425. doi: 10.1007/BF00354971
Baddeley, A., Gathercole, S., & Papagno, C. (1998). The phonological loop as a language
learning device. Psychological Review, 105, 158-173. doi: 10.1037//0033-295X.105.1.158
Baddeley, A. D., & Hitch, G. J. (1974). Working Memory. In G. A. Bower (Ed.), The
psychology of learning and motivation (pp. 47-90). New York, NY: Academic Press.
Baker, L., Afflerbach, P., & Reinking, D. (1996). Developing engaged readers in school
and home communities: An overview. In L. Baker, P. Afflerbach, & D. Reinking
(Eds.), Developing engaged readers in school and home communities (pp. xiii). Mahwah,
NJ: Erlbaum.
Baker, L., Sonnenschein, S., Serpell, R., Fernandez-Fein, S., & Scher, D. (1994).
Opportunities for literacy learning in the homes of urban preschoolers. In L. M.
Morrow (Ed.), Family literacy: Connections in schools and communities (pp. 236-252).
Newark, DE: International Reading Association.
Bakker, J. T. A., & Denessen, E. (2007). The concept of parent involvement. Some theoretical
and empirical considerations. In H. Phtiaka & S. Symeonidou (Eds.), Schools and
families in partnership: Looking in the future (pp. 238-252). Nicosia, Cyprus: ERNAPE.
Bates, E., & Goodman, J. (1997). On the inseparability of grammar and the lexicon:
Evidence from acquisition, aphasia and real-time processing. Language and Cognitive
Processing, 12, 507-586. doi: 10.1080/016909697386628
Beitchman, J. H., Wilson, B., Brownlie, E. B., Walters, H., & Lancee, W. (1996). Longterm consistency in speech/language profiles: I. Developmental and academic
outcomes. Journal of the American Academy of Child and Adolescent Psychiatry, 35,
804-814. doi: 10.1097/00004583-199606000-00021
Bhatia, T. K. (2006). Bilingualism and second language learning. In K. Brown (Ed.),
Encyclopedia of languages and linguistics (pp. 16-22). Amsterdam, the Netherlands:
Elsevier. doi: 10.1016/B0-08-044854-2/00620-9
25
chapter 1
Bialystok, E. (2005). Consequences of bilingualism for cognitive development. In
J. F. Kroll & A. M. B de Groot (Eds.), Handbook of bilingualism: Psycholinguistic
approaches (pp. 417-432). New York, NY: Oxford University Press.
Bialystok, E. (2007). Acquisition of literacy in bilingual children: A framework for
research. Language Acquisition, 57, 45-77. doi: 10.1111/1467-9922.00180
Blachman, B. (2000). Phonological awareness. In M. Kamil, P. Mosenthal, P. D.
Pearson, & R. Barr (Eds.), Handbook of reading research (Vol. 3; pp. 483–502).
Mahwah, NJ: Erlbaum.
Blok, H., & Leseman, P. P. M. (1996). Effecten van voorschoolse stimuleringsprogramma’s:
een review van reviews [Effects of preschool programs: A review of reviews].
Pedagogische Studiën, 73, 184-197.
Bornstein, M. H., Haynes, O. M., & Painter, K. M. (1998). Sources of child vocabulary
competence: A multivariate model. Journal of Child Language, 25, 367-394. doi:
10.1017/S0305000998003456
Centraal Bureau voor de Statistiek (2011). Jaarboek onderwijs in cijfers 2011 [Yearbook
education in numbers 2011]. Den Haag, the Netherlands: Centraal Bureau voor de
Statistiek.
Chapman, L. J., Chapman, J. P., Kwapil, T. R., Eckblad, M., & Zinser, M. C. (1994).
Putatively psychosis-prone subjects 10 years later. Journal of Abnormal Psychology,
103, 171-183. doi: 10.1037/0021-843X.103.2.171
Coplan, R. J., & Armer, M. (2005). Talking yourself out of being shy: Shyness,
expressive vocabulary, and socioemotional adjustment in preschool. Merrill-Palmer
Quarterly, 51, 20-41. doi: 10.1353/mpq.2005.0004
Coster, F. W., Goorhuis-Brouwer, S. M., Nakken, H., & Lutje Spelberg, H. C. (2002).
Pedagogische aspecten van taalontwikkelingsstoornissen [Educational aspects
of language developmental disorders]. In J. G. van Hell, A. de Klerk, D. Straus,
& T. Torremans (Eds.), Taalontwikkeling en taalstoornissen: theorie, diagnostiek en
behandeling (pp. 125-135). Leuven, Belgium: Garant.
D’Angiulli, A., Siegel, L. S., & Serra, E. (2001). The development of reading in Italian
and English in bilingual children. Applied Psycholinguistics, 22, 479-507. doi:
10.1017/S0142716401004015
de Beer, J., & van den Heuvel, I. (1995). Taalstimulering en peuterspeelzalen. Verslag van
een effectmeting [Language stimulation and play groups. Report of an effect study].
Internal report Hogeschool Nijmegen.
de Houwer, A. (2006). The acquisition of two languages from birth: A case study. Cambridge,
United Kingdom: Cambridge University Press. doi: 10.1017/CBO9780511519789
de Jong, P. F. (1998). Working memory deficits of reading disabled children. Journal of
Experimental Child Psychology, 70, 75-96. doi: 10.1006/jecp.1998.2451
26
general introduction
de Jong, P. F., & van der Leij, A. (2003). Developmental changes in the manifestation of
a phonological deficit in dyslexic children learning to read a regular orthography.
Journal of Educational Psychology, 95, 22-40. doi: 10.1037//0022-0663.95.1.22
Deslandes, R. (2001). A vision of home-school partnership: Three complementary
conceptual frameworks. In F. Smit, K. van der Wolf, & P. Sleegers (Eds.), A Bridge
to the Future. Collaboration between parents, schools, and communities (pp. 11-23).
Nijmegen, the Netherlands: Institute for Applied Social Sciences (ITS).
Driessen, G., & van der Slik, F. (2004). Sociaal milieu, etnische herkomst, opvoeding
en competenties van kleuters [Background, etnical origin, education, and infant
competencies]. Tijdschrift voor Orthopedagogiek, 43,525-537.
Elbro, C. (1996). Early linguistic abilities and reading development: A review and a
hypothesis. Reading and Writing, 8, 1-33. doi: 10.1007/BF00577023
Eleveld, J., Nakken, H., & Goorhuis-Brouwer, S. M. (1994). Taal en sociaalemotionele problemen [Language and socio-emotional problems]. Tijdschrift voor
Orthopedagogiek, 33, 550-556.
Elzer, M. J. (2005). Talen naar een moedertaal [Care for mother tongue]. Tijdschrift
voor Orthopedagogiek, 3, 130-135.
Fantuzzo, J., Davis, G., & Ginsburg, M. (1995). Effects of parent involvement
in isolation or in combination with peer tutoring on student self-concept and
mathematics achievement. Journal of Educational Psychology, 87, 272-281. doi:
10.1037//0022-0663.87.2.272
Gathercole, S. (2006). Nonword repetition and word learning: The nature
of the relationship. Applied Psycholinguistics, 27, 513-543. doi: 10.1017/
S0142716406060383
Gathercole, S. E., & Baddeley, A. D. (1993). Working memory and language. Hillsdale,
NJ: Erlbaum.
Gazzaniga, S., & Heatherton, T. F. (2003). Psychological science. Mind, brain, and
behaviour. New York, NY: W.W. Norton & Company.
Goddard, R. D., Tschannen-Moran, M., & Hoy, W. (2001). A multilevel examination
of the distribution and effects of teacher trust in students and parents in urban
elementary schools. Elementary School Journal, 102, 3-17. doi: 10.1086/499690
Goorhuis, S. M., & Schaerlaekens, A. M. (2000). Handboek taalontwikkeling,
taalpathologie en taaltherapie bij Nederlandssprekende kinderen [Handbook
language development, language pathology, and language therapy in Dutch
speaking children]. Utrecht, the Netherlands: De Tijdstroom.
Green, K. D., Forehand, R., Beck, S. J., & Vosk, B. (1980). An assessment of the
relationship among measures of children’s social competence and children’s
academic achievement. Child Development, 51, 1149-1156. doi: 10.2307/1129556
27
chapter 1
Grosjean, F. (1982). Life with two languages: An introduction to bilingualism. Cambridge,
MA: Harvard University Press.
Gutiérrez-Clellen, V. F., Simon-Cereijido, G., & Wagner, C. (2008). Bilingual children
with language impairment: A comparison with monolinguals and second language
learners. Applied Psycholinguistics, 29, 3-19. doi: 10.1177/1367006909103530
Hoff, E. (2006). How social contexts support and shape language development.
Developmental Review, 26, 55-88. doi: 10.1016/j.dr.2005.11.002
Hoff, E., Laursen, B., & Tardif, T. (2002). Socio-economic status and parenting. In M. H.
Bornstein (Ed.), Handbook of parenting (pp. 231-252). Mahwah, NJ: Erlbaum.
Hoff, E., & Naigles, L. (2002). How children use input to acquire a lexicon. Cognition and
Language, 73, 418-433. doi: 10.1111/1467-8624.00415
Hoff, E., & Tian, C. (2005). Socio-economic status and cultural influences on language.
Journal of Communication Disorders, 38, 271-278. doi: 10.1016/j.jcomdis.2005.02.003
Huttenlocher, J., Vasilyeva, M., Cymerman, E., & Levine, S. (2002). Language input and
child syntax. Cognitive Psychology, 45, 337-374. doi: 10.1016/S0010-0285(02)00500-5
Kervezee, C. (2005). Onderwijsverslag 2003/2004 [Education report 2003/2004]. Utrecht,
the Netherlands: Inspectie van Onderwijs.
Kintsch, W. (2004). The construction-integration model of text comprehension and its
implications for instruction. In R. B. Ruddell & N. J. Unrau (Eds.), Theoretical models
and processes of reading (5th ed.; pp.1270-1328). Newark, DE: International Reading
Association.
Leather, C. V., & Henry, L. A. (1994). Working memory span and phonological awareness
tasks as predictor of early reading ability. Journal of Experimental Child Psychology, 58,
88-111. doi: 10.1006/jecp.1994.1027
Ledoux, G. (1996). School- en klaskenmerken basisonderwijs en speciaal onderwijs: beschrijvende
rapportage op basis van het PRIMA-cohortonderzoek 1994/1995. [School and classroom
characteristics in primary and special education: Descriptive study based on PRIMA
cohort 1994/1995]. Amsterdam, the Netherlands: SCO-Kohnstamm Instituut.
Leseman, P. (2007). Achterstandenbeleid: voorbij de voor- en vroegschoolse periode
[Delay policies: Past the preschool period]. In P. A. H. van Lieshout, M. S. S. van der
Meij, & J. C. I. De Pree (Eds.), Bouwstenen voor betrokken jeugdbeleid (pp. 113-128).
Amsterdam, the Netherlands: Amsterdam University Press.
Leseman, P. P. M. (2001). Aanzetten voor onderzoeksprogrammering in de voor- en
vroegschoolse periode [Preludes for research programs in the preschool period].
Amsterdam, the Netherlands: SCO-KI.
Leseman, P. P. M. (2002). Early childhood education and care for children from low income
or minority backgrounds. A paper for discussion at the OECD Oslo Workshop, June
6-7, 2002. Amsterdam, the Netherlands: Universiteit van Amsterdam.
28
general introduction
Leseman, P. P. M., & Cordus, J. (1994). (Allochtone) kleuters meer aandacht. Adviesrapport
van de commissie (voor)schoolse educatie in opdracht van de minister van Welzijn,
Volksgezondheid en Cultuur [(Immigrant) children more attention. The advisory
committee for childhood (preschool) education on behalf of the Minister of
Welfare, Health and Culture]. Rijswijk, the Netherlands: Ministerie van Welzijn,
Volksgezondheid en Cultuur/ Ministerie van Onderwijs en Wetenschappen.
Leseman, P. P. M., & de Jong, P. F. (1998). Home literacy: Opportunity, instruction,
cooperation, and social-emotional quality predicting early reading achievement.
Reading Research Quarterly, 33, 294-319. doi: 10.1598/RRQ.33.3.3
Leseman, P. P. M., Otter, M. E., Blok, H., & Deckers, P. (1998). Effecten van voor- en
vroegschoolse educatieve centrumprogramma’s: een meta-analyse van studies
gepubliceerd tussen 1985 en 1996 [Effects of preschool centre-based programs: A
meta-analysis of studies published between 1985 and 1996]. Nederlands Tijdschrift
voor Opvoeding, Vorming en Onderwijs, 14, 134-154.
Leseman, P. P. M., Sijsling, F. F., Jap-A-Joe, S. R., & Sahin, S. (1995). Gezinsdeterminanten
van de cognitieve ontwikkeling van vierjarige Nederlandse, Surinaamse en Turkse
kleuters [Family determinants of cognitive development for four-year-old Dutch,
Surinamese and Turkish children]. Pedagogische Studiën, 72, 186-205.
Li, R. (January 2005). Childhood bilingualism: Current status and future directions,
Washington DC, April 22-23, 2004. Retrieved from http://www.nichd.nih.gov/
publications/pubs/upload/Childhood-Bilingualism_2005.pdf
Liberman, I. Y., & Liberman, A. M. (1990). Whole language vs. code emphasis:
Underlying assumptions and their implications for reading instruction. Annals of
Dyslexia, 40, 51-75. doi: 10.1007/BF02648140
Lieven, E. V. M. (1994). Cross-linguistic and cross-cultural aspects of language addressed
to children. In C. Galloway & B. J. Richards (Eds.), Input and interaction in language
acquisition (pp. 56-74). Cambridge, UK: Cambridge University Press.
Majerus, S., Poncelet, M., Greffe, C., & Van der Linden, M. (2006). Relations between
vocabulary development and verbal short-term memory: The relative importance of
short-term memory for serial order and item information. Journal of Experimental
Child Psychology, 93, 95-119. doi: 10.1016/j.jecp.2005.07.005
McCarthey, S. J. (2000). Home-school connections: A review of the literature. Journal of
Educational Research, 99, 145-167. doi: 10.1080/00220670009598703
Meijnen, G. W. (2003). Onderwijsachterstanden in basisscholen [Educational disadvantages
in primary schools]. Leuven, Belgium: Garant.
Morais, J. (1991). Phonological awareness: A bridge between language & literacy. In D.
Sawyer & B. Fox (Eds.), Phonological awareness in reading: The evolution of current
perspectives (pp. 31-71). New York, NY: Springer.
29
chapter 1
Oller, D. K., & Eilers, R. E. (2002). Language and literacy in bilingual children. Clevedon,
UK: Multilingual Matters.
Papagno, C., & Vallar, G. (1995). Verbal short-term memory and vocabulary learning
in polyglots. The Quarterly Journal of Experimental Psychology, 48, 98-107. doi:
10.1080/14640749508401378
Paradis, J., & Genesee, F. (1996). Syntactic acquisition in bilingual children: Autonomous
or independent? Studies in Second Language Acquisition, 18, 1-25. doi: 10.1017/
S0272263100014662
Parker, J. G., & Asher, S. R. (1987). Peer relations and later adjustment: Are lower-accepted
children “at risk”? Psychological Bulletin, 102, 357-389. doi: 10.1037/0033-2909.102.3.357
Pearson, B. Z. (2007). Social factors in childhood bilingualism in the United States.
Applied Psycholinguistics, 28, 399-410. doi: 10.1017/S014271640707021X
Pearson, B. Z., Fernandez, S. C., Lewedeg, V., & Oller, D. K. (1997). The relation of input
factors to lexical learning by bilingual infants. Applied Psycholinguistics, 18, 41-58. doi:
10.1017/S0142716400009863
Pels, T. (2000). Opvoeding en integratie. Een vergelijkende studie van recente onderzoeken
naar gezinsopvoeding en de pedagogische afstemming tussen gezin en school. [Education
and integration. A comparative study of recent studies on child raringstyle and homeschool alignment]. Assen, the Netherlands: Van Gorcum.
Petitto, L. A., Katerelos, M., Levy, B. G., Gauna, K., Tetreault, K., & Ferraro, V. (2001).
Bilingual signed and spoken language acquisition from birth: Implications for the
mechanism underlying early bilingual language acquisition. Journal of Child Language,
28, 453-496. doi: 10.1017/S0305000901004718
Raver, C. C. (2002). Emotions matter: Making the case for the role of young children’s
emotional development for early school readiness. Social Policy Report, 16, 3-18.
Redmond, S. M., & Rice, M. L. (1998). The socioemotional behaviours of children with
SLI: Social adaptation or social deviance? Journal of Speech, Language, and Hearing
Research, 41, 688-700. PMID: 9638932
Reezigt, G. (2003). Voor- en vroegschoolse educatie [Pre- and early childhood education]. In W.
Meijnen (Ed.), Onderwijsachterstanden in basisscholen (pp. 79-97). Leuven, Belgium: Garant.
Rice, M. L., Sell, M. A., & Hadley, P. A. (1991). Social interactions of speech- and languageimpaired children. Journal of Speech and Hearing Research, 34, 1299-1307. doi: 10.1044/
jshr.3406.1299
Schaerlaekens, A. M. (2002). Taalverwerving, een multifactorieel proces: pleidooi voor
vroege detectie en hulp [Language acquisition, a multifactorial proces: Plea for early
detection and help]. In J. G. van Hell, A. de Klerk, D. E. M. Straus, & T. Torremans
(Eds.), Taalontwikkeling en taalstoornissen: theorie, diagnostiek en behandeling (pp. 37-49).
Leuven, Belgium: Garant.
30
general introduction
Schaerlaekens, A. M., & Gillis, S. (1987). De taalverwerving van het kind [Language acquisition
of the child]. Groningen, the Netherlands: Wolters-Noordhoff.
Scheele, A. F., Leseman, P. P. M., & Mayo, A. Y. (2010). The home language environment of
mono- and bilingual children and their language proficiency. Applied Psycholinguistics,
31, 117-140. doi: 10.1017/S0142716409990191
Scheltinga, F., van der Leij, A., & van Beinum, F. (2003). Importance of phonological skill
and underlying processes to reading achievement. Proceedings of the Institute of Phonetic
Sciences, 25, 21-30.
Schlichting, L. (1988). TARSP. Taalontwikkelingsschaal van Nederlandse kinderen van
1-4 jaar [Language development scale of Dutch children 1-4 years]. Amsterdam, the
Netherlands: Swets en Zeitlinger.
Scholten, S., & Snelder, A. (1997). Taalstimuleringsproject voor allochtone peuters. Leidt
vroegtijdige interventie tot uitbreiding van de woordenschat van allochtone peuters?
[Language stimulation project for immigrant children. Does early intervention lead to
vocabulary development?]. Logopedie en Foniatrie, 7/8, 179-185.
Serpell, R. (1997). Literacy connections between school and home: How should we evaluate
them? Journal of Literacy Research, 29, 587-616. doi: 10.1080/10862969709547975
Service, E. (1992). Phonology, working memory, and foreign language learning. The
Quarterly Journal of Experimental Psychology, 45, 21-50. doi: 10.1080/14640749208401314
Shonkoff, J. P., & Phillips, D. A. (Eds.). (2000). From neurons to neighborhoods: The science
of early childhood development. Washington, DC: National Academy Press.
Snowling, M. J. (2000). Language and literacy skills: Who is at risk and why? In D. V.
M. Bishop & L. B. Leonard (Eds.), Speech and language impairment in children: Causes,
characteristics, interventions, and outcome (pp. 245–260). Hove, United Kingdom:
Psychology Press.
Sonnenschein, S., Brody, G., & Munsterman, K. (1996). The influence of family beliefs
and practices on children’s early reading development. In L. Baker, P. Afflerbach, & D.
Reinking (Eds.), Developing engaged readers in school and home communities (pp. 3-20).
Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Stahl, S., & Murray, B. (1994). Defining phonological awareness and its relationship
to early reading. Journal of Educational Psychology, 86, 221-234. doi: 10.1037/00220663.86.2.221
Stanton-Chapman, T. L., Justice, L. M, Skibbe, L. E., & Grant, S. L. (2007). Social and
behavioural characteristics of preschoolers with specific language impairment. Topics
in Early Childhood Special Education, 27, 98-109. doi: 10.1177/02711214070270020501
Stuart, M. (1999). Getting ready for reading: Early phoneme awareness and phonics
teaching improves reading and spelling in inner-city second language learners. British
Journal of Educational Psychology, 69, 587-605. doi: 10.1348/000709999157914
31
chapter 1
Stuart, M. (2004). Getting ready for reading: A follow-up study of inner city second
language learners at the end of Key Stage I. British Journal of Educational Psychology,
74, 15-36. doi: 10.1348/000709904322848806
Sui-Chu, E. H., & Douglas, W. J. (1996). Effects of parental involvement on grade
achievement. Sociology of Education, 69, 126-141. doi: 10.2307/2112802
Tesser, P. T. M., & Iedema, J. (2001). Rapportage minderheden 2001. Deel I Vorderingen
op school [Report on minorities in 2001. Part 1 Progress in school]. Den Haag, the
Netherlands: Sociaal en Cultureel Planbureau.
Tomasello, M. (2005). Constructing a language. A usage-based theory of language
acquisition. Cambridge, MA: Harvard University Press.
Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (1994). Longitudinal studies of
phonological processing and reading. Journal of Learning Disabilities, 27, 276-286.
doi: 10.1177/002221949402700503
van den Heuvel, I. (1996). Taalstimulering in peuterspeelzalen, een rol voor logopedie
[Language stimulation in preschool, a role for speech therapy]. Logopedie en
Foniatrie, 12, 322-327.
van der Ploeg, C. P. B., Lanting, C. I., Galindo Garre, F., & Verkerk, P. H. (2007).
Screening op taalachterstanden en spraakstoornissen bij kinderen van 1 tot 6
jaar door de jeugdgezondheidszorg. Deelrapport 1: inventarisatie instrumenten
[Screening for language delays and speech disorders in 1-6 year old children
by youth health service. Report part 1: Inventory instruments]. Leiden, the
Netherlands: TNO.
van der Wolf, J. C. (1999). Voor- en buitenschoolse educatieve programma’s: beschrijving
en effectiviteit [Preschool education programmes: Description and effectiveness].
In P. P. M. Leseman (Ed.), Onderwijskundig Lexicon III. Achtergronden van leerlingen
(pp. 75-95). Alphen aan de Rijn, the Netherlands: Samsom H. D. Tjeenk Willink.
van Eldik, M. C. M., Schlichting, J. E. P. T., Lutje Spielberg, H. C., van der Meulen,
B. F., & van der Meulen, S. (1997). Reynell test voor taalbegrip, handleiding.
[Reynell test for language comprehension, manual]. Nijmegen, the Netherlands:
Berkhout.
van Elten, M. (2003). Schoolprestaties van allochtone kinderen [Academic achievement
of immigrant children]. Onderwijsmagazine Inzicht. Retrieved June, 2009, from
http://www.voo.nl/inzicht.html
van Hell, J. G. (2002). Taalontwikkeling en taalstoornissen: een introductie
[Language development and language disorders: An introduction]. In J. G. van
Hell, A. de Klerk, D. E. M. Straus, & T. Torremans (Eds.), Taalontwikkeling en
taalstoornissen: theorie, diagnostiek en behandeling (pp. 9-36). Leuven, Belgium:
Garant.
32
general introduction
van Lieshout, C. F. M., & Haselager, G. J. T. (1993). De structuur en organisatie van de
begeleiding van de sociaal-emotionele ontwikkeling van leerlingen [The structure
and organization of the guidance of the socio-emotional development of children].
Tijdschrift voor Orthopedagogiek, 32, 87-102.
Yip, V., & Matthews, S. (2007). Relative clauses in Cantonese-English bilingual
children: Typological challenges and processing motivations. Studies in Second
Language Acquisition, 29, 277-300. doi: 10.1017/S0272263107070143
33
34
Chapter 2
differential relationships between
language skills and working memory in
turkish-dutch and native-dutch firstgraders from low-income families
Accepted, pending revisions, for publication in Journal of Research in Reading.
35
chapter 2
Abstract
In the Netherlands, Turkish-Dutch children constitute a substantial group of
children who learn to speak Dutch at the age of three after they acquired their home
language. These children are generally academically less successful, because of their
limited proficiency in the Dutch language. Research also suggests a role for working
memory as an indicator of academic success. This study investigated the relationship
between language skills and working memory in a group of Turkish-Dutch and nativeDutch children both from low-income families. The findings revealed reduced Dutch
language and Dutch working-memory skills for Turkish-Dutch children compared to
native-Dutch children. Working memory in the native-Dutch children was unrelated to
their language skills, whereas in the Turkish-Dutch children strong correlations were
found both between Turkish language skills and Turkish working-memory performance
and between Dutch language skills and Dutch working-memory performance. Reduced
language proficiencies and reduced working-memory skills appear to manifest itself in
strong relationships between working memory and language skills in the respective
languages of the Turkish-Dutch children.
36
differential relationships between language skills and working memory
The majority of first- and second-generation immigrants in the Netherlands
is bilingual. The largest group is from Turkish origin (384.000; the entire Dutch
population constitutes 16.5 million people; Centraal Bureau voor de Statistiek [CBS],
2008). The first language that children in this community learn is the language of their
parents, which is predominantly Turkish. Substantial and prolonged exposure to Dutch
as a second language usually begins when the Turkish-Dutch children enter pre-school.
This often results in a large group of children from minority backgrounds entering
preschool with insufficient knowledge of the Dutch language. A large national-cohort
study revealed that children from low-income minority families start primary school
with a delay in their cognitive and Dutch language development of about one standard
deviation relative to the average of middle to high income native-Dutch children (Tesser
& Iedema, 2001). The effect is that they cannot benefit optimally from formal education
in reading, spelling, and mathematics (Elzer, 2005; van Elten, 2003), shown by the fact
that Turkish-Dutch children repeat grades twice as often as native-Dutch children from
a similar socio-economic background (Aarts, de Ruiter, & Verhoeven, 1996).
Being a non-native speaker may thus be a disadvantage as shown by the, on average,
smaller vocabularies of bilinguals in each language compared to monolinguals
(Bialystok & Luk, 2012; Pearson, 2007; Scheele, Leseman, & Mayo, 2010). Differences
in vocabulary may appear early and may increase over time. Moreover, differences in
vocabulary are found to affect educational achievement in the long run (Baker, Simons,
& Kameenui, 1995; Oller & Eilers, 2002; Pearson, Fernandez, Lewedeg, & Oller, 1997).
The importance of language skills for academic success is revealed by a large number
of studies that provide substantial evidence for a positive relationship between language
skills and verbal working memory in both monolingual and bilingual populations.
Measures of verbal memory span strongly correlate with different aspects of language
development, such as receptive vocabulary knowledge, vocabulary specificity, mean
length of utterances, and syntactic diversity. Monolingual children and adults with good
working-memory skills tend to have better language skills and/or reveal better academic
performance (e.g., Adams, Bourke, & Willis, 1999; Adams & Gathercole, 2000; Baddeley,
Gathercole, & Papagno, 1998; Gathercole & Alloway, 2004; Gathercole, Alloway, Willis,
& Adams, 2006; Gathercole & Baddeley, 1993; Gathercole, Service, Hitch, Adams, &
Martin, 1999; Gathercole, Willis, Emslie, & Baddeley, 1992; Majerus, Poncelet, Greffe,
& Van der Linden, 2006; Pickering, 2006).
Examples of bilingual children and adults show that English word span and Englishsounding nonword span is positively related to English language skills in 12-year old
Chinese students learning English as a second language (Cheung, 1996). English
37
chapter 2
phonological-memory skills of English adults learning to speak Spanish predicted their
oral skills in Spanish (O’Brien, Segalowitz, Collentine, & Freed, 2006). Adult Italian
students who are polyglot speakers (fluent in three or more languages) revealed faster
acquisition of a set of new Russian words in a paired-associate learning test than students
in a bilingual control group (Papagno & Vallar, 1995). English-sounding nonwordrepetition skill is strongly related to English-language skills in Finnish-speaking
children, aged 9-10 learning English (Service, 1992). Thus, being highly proficient in a
native and/or second language appears to be an asset for academic development.
That bilingualism may also be an advantage above being monolingual with respect to
the development of cognitive skills, such as working memory, metalinguistic, cognitive,
and conceptual processing has also been suggested. The explanation runs roughly as
follows: Bilingual children need to learn two grammatical systems and must be able to
keep the two systems apart, because they have to decide which language, which words,
and which syntactic structure to use. These skills require highly developed executive
functions such as attention shifting and inhibition, functions that are responsible for
the control of cognitive processes in working memory (Bialystok, 2009).
Substantial evidence for this hypothesis comes from Bialystok and her colleagues
(e.g., Bialystok, 2002; Morales, Calvo, & Bialystok, 2013; Poulin-Dubois, Blaye, Coutya, &
Bialystok, 2011) as well as from other studies. For example, the polyglots of Papagno and
Vallar (1995) had superior Italian short-term memory skills than the bilingual control
group. Kormi-Nouri et al. (2008) studying Persian monolingual, Turkish-Persian
bilingual, and Kurdish-Persian bilingual children (aged 9-10 years, 13-14 years, and 1617 years) showed that the two groups of bilingual children performed better on various
types of Persian, episodic and semantic memory tasks than monolingual children. This
effect was stronger for older bilingual children than for younger ones.
However, working-memory skills of a group of 6-8 years old, middle to upper class
bilingual children, living in Luxembourg, with Luxembourgish as their second was
similar to that of monolingual Luxembourgish-speaking children when tested in their
second language (Engel de Abreu, 2011). Similarly for the comparison between 8-year
old bilingual children, from low-income backgrounds, living in Luxembourg whose
first language was Portuguese and their second Luxembourgish and monolingual
Portuguese-speaking children living in Portugal. When tested in their first language no
performance differences emerged on the working-memory tasks between the language
groups (Engel de Abreu, Cruz-Santos, Tourinho, Martin, & Bialystok, 2012). Finally,
English memory skills of lower-class Portuguese English-speaking bilingual children
in Canada did not differ from those of monolingual English-speaking children (Da
Fontoura & Siegel, 1995).
Why is it that some studies report a clear advantage of being bilingual, whereas
38
differential relationships between language skills and working memory
others do not. One possible explanation has been provided by Bialystok in 2001. She
argued that a bilingual advantage with respect to working memory is more likely to
occur when there is a high proficiency in the two languages (see also Cummins, 2000).
Other factors that may explain bilingual advantage are social class and exposure to both
languages. Almost all positive findings with respect to a bilingual advantage pertain to
middle- or upper-class children who had been exposed to both languages since their
birth and who also used both languages daily. The only exception is the study conducted
by Engel de Abreu (2011). Her sample of bilingual children, exposed to both languages
from birth and who came from middle to upper class background, did not outperform
a control group of monolingual children. Note that the monolingual children had better
scores on language measures pertaining to vocabulary and syntax.
All three studies (Da Fontoura & Siegel, 1995; Engel de Abreu, 2011; Engel de Abreu
et al., 2012) in which no bilingual advantage was found concerned Portuguese as first
or native language and in two of them the participating children came from low-income
backgrounds. To our knowledge no other study has been conducted in which working
memory was studied in disadvantaged children and related to their language skills.
The present study attempts to contribute to the knowledge concerning the role
of working memory in the development of children who are exposed to their second
language long after they started learning their native language; a common situation
in immigrant children from poor families. We focussed on working memory, because
it is a good predictor of academic success (e.g., de Jong, 1998; Siegel & Ryan, 1989; St
Clair-Thompson & Gathercole, 2006). After all, in education, verbal information and
instructions must be stored, manipulated, and processed, in order to keep up with
the academic demands of the curriculum; working memory is the capacity to keep
information active in order to use it for further processing. Unlike short-term memory,
it requires monitoring of the information in memory. The most widely used workingmemory model is that of Baddeley and Hitch (originally 1974; see Baddeley, 2000 for
an adaptation). They assume a three-component system with a so-called attentioncontrol system known as the ‘central executive’ with two subsidiary systems, namely,
the ‘phonological loop’ and the ‘visuospatial sketchpad’, holding verbal and acoustic
information, and visuospatial information, respectively, in a temporary store.
In line with earlier research, we focussed on the central executive, because of its
role in maintaining as well as manipulating information (this system is responsible for
the control of cognitive processes), and on the phonological loop, because of its role in
keeping information in store by rehearsing it. Measures that are assumed to assess the
phonological loop and the central executive, rather than the visuospatial sketchpad have
been shown to be good predictors of language skills (e.g., Gathercole & Baddeley, 1993;
Messer, Leseman, Boom, & Mayo, 2010; Papagno & Vallar, 1995).
39
chapter 2
In this study, three types of tasks were used in order to assess verbal-working
memory: Digit recall, backward-digit recall, and listening recall. Digit recall only takes
into account the storage function, whereas the complex tasks (listening recall and
backward-digit recall) also account for the processing functions. In accordance with
the model of Baddeley and Hitch (1974), digit recall measures the short-term storage
capacity of the phonological loop, backward-digit recall and listening recall include also
the central executive.
A final issue that will be addressed in this paper is the language in which workingmemory will be tested. In most studies, working memory is either tested in the native
language or in the second language. Because of a unique situation in the Netherlands
with respect to the Turkish language, it was possible to test language skills as well as
assess working memory in both Dutch and Turkish.
The aim of the present study is twofold. The first goal is to compare working-memory
skills of bilingual Turkish-Dutch children with those of monolingual native-Dutch
children, from low-income families in the Netherlands. A second goal is to investigate
the relationship between language skills and verbal-working memory within the two
language groups. The following questions will guide our investigation:
1. Are Dutch language skills of bilingual Turkish-Dutch children from low-income
backgrounds indeed lower than that of native-Dutch children?
2. Is Turkish still better developed than Dutch in first-grade Turkish-Dutch
children, and what is the relationship between Dutch and Turkish language
skills in Turkish-Dutch children?
3. Is there a difference between Dutch working memory of Turkish-Dutch children
and native-Dutch children?
4. Do Turkish-Dutch children perform better on Turkish working-memory tasks
than on Dutch working-memory tasks?
5. To what extent are Dutch language skills related to Dutch working-memory skills,
and is this different for Turkish-Dutch children and native-Dutch children?
6. To what extent are Turkish language skills related to Turkish working-memory
skills in Turkish-Dutch children?
40
differential relationships between language skills and working memory
Method
Participants
In this study participated 38 Turkish-Dutch and 48 native-Dutch children who all
attended first grade during testing. Information on gender and age is presented in
Table 1. All children were recruited from the same poor inner-city neighbourhoods
with low-income and immigrant families (i.e., all families had a low socio-economic
status). The Turkish-Dutch children were born in the Netherlands, but 95% of their
parents were born in Turkey. All Turkish-Dutch children learned Turkish as their first
language and for the majority of them (65%) this language was still the best developed
by the age of 3. Almost all children (80%) were to some extent exposed to Dutch as a
second language before the age of three by watching Dutch television or playing with
Dutch speaking children, including older siblings who already attended Dutch primary
schools. Nonetheless, starting in pre-school meant for most of them a strong increase
in Dutch language input. The native-Dutch children were born in the Netherlands
and grew up in low-income families and only spoke Dutch, and thus are considered
monolingual. When they were three years old, all Turkish-Dutch as well as the nativeDutch children attended a pre-school program for disadvantaged children, because of
their limited Dutch language proficiency (a centre-based program to enhance language
skills and socio-emotional development).
To assess general language and cognitive differences between the two groups, a
language-comprehension test (i.e., Reynell test for language comprehension by van
Eldik, Schlichting, Lutje Spielberg, van der Meulen, & van der Meulen, 1997) and a
nonverbal intelligence test (i.e., the Standard Progressive Matrices or SPM, Raven, 1958)
Table 1 Descriptive Statistics of Participants Group. Mean Age in Years and Months and Raw
Scores on the Standard Progressive Matrices (SPM) and the Reynell
Gender
Language group
SPM
Reynell
7;2
25.0
71.0
0.6
7.5
8.9
Mean
7;4
27.7
78.9
SD
0;6
6.2
4.1
n
Boys
Girls
Mean
age
Turkish-Dutch
38
24
14
Mean
SD
native-Dutch
48
25
23
41
chapter 2
were administered. There were no intellectual differences between the two groups
(F< 1) as assessed by the SPM (see Table 1 for the means scores on this test). With
respect to language comprehension, however, it appeared that the native-Dutch children
outperformed the Turkish-Dutch children F(1, 51) = 10.89, p = 002.
Materials
To measure the language skills of the two language groups two subtests of the
Diagnostic Test of Bilingualism (i.e., vocabulary and sentence imitation) were used,
developed by the national institute of educational testing (Cito; Verhoeven, Narain,
Extra, Konak, & Zerrouk, 1995). All children were tested in Dutch, and the Turkishspeaking children were also tested in Turkish.
Vocabulary. The productive-vocabulary test consists of 40 pictures. Children were
presented with a picture book displaying one picture per page. The children had to
answer the question ‘What is that?’ or ‘What happens here?’ A correct answer was
rewarded with one point. When a child failed five consecutive items the test was ended.
The minimum score was 0 and the maximum score was 40. The Dutch and Turkish
versions of the active-vocabulary test are highly reliable with Cronbach’s alpha being .90
and 85, respectively.
Sentence-imitation task. This test measures syntactic knowledge. Children were orally
presented with 20 sentences, one by one, and asked to repeat each sentence as accurately
as possible. For each sentence, the correct reproduction of two distinct grammatical
structures was scored: function words and word order. The mean sentence length and
mean number of nominal and verbal phrases were the same in the two languages. The
minimum score was 0 and the maximum sore was 40. The Dutch and Turkish sentenceimitation tests are highly reliable with Cronbach’s alpha being .95 for both tests.
To measure verbal-working memory, three subtests of the Automated working
Memory Assessment Battery (AWMA; Alloway, 2007) were adapted for Dutch and
Turkish by Messer et al. (2010). The AWMA can be used to test children between the
ages of 4.5 and 11.5 year. Each test begins with a series of practice trials immediately
followed by the test trials. The test ends when three or more errors within a block of
trials were made. The score for that test reflects the number of correct responses up to
the point at which the test was ended.
Digit Recall. The child had to repeat a sequence of voiced digits (1 to 9) in the same
order as presented. The test started with a block of one digit and ended with a block of
nine digits. The test consisted of 42 items divided in nine blocks of six trials each that
42
differential relationships between language skills and working memory
increased in difficulty. Each correct trial was rewarded with one point. The minimum
score was 0 and the maximum score was 42. For the Turkish-Dutch children, parallel
digit span tests in both Dutch and Turkish, using the count words from 1 to 9 in Turkish,
were administered on two different occasions within a two months’ period.
Listening recall. The child listened to a series of sentences and had to judge whether a
sentence was true or false, for instance ‘lions have legs and tomatoes play football’. At the
same time the child had to memorize the last word of each sentence. After all sentences
were presented and evaluated, the child had to recall the last word of each sentence, in
the same order as presented. The sentences were presented in growing set sizes starting
with a one-sentence trial and ending with a six-sentence trial. The entire test consisted
of 36 items divided in six blocks of six trials that increased in difficulty. Each correct
trial was rewarded with one point. The minimum score was 0 and the maximum score
was 36. For the Turkish-Dutch group, again, parallel versions in Turkish and Dutch were
administered. In order to obtain parallel versions, the original test was slightly adapted.
The target words to be remembered were not the last words of the sentences, but the first
words (always a highly frequent noun) because in Turkish the last words of a sentence
often are grammatical morphemes with a grammatical function without a clear lexical
meaning.
Backward digit recall. The child had to repeat a sequence of spoken digits (1 to 9) in
reverse order. The test started with a block of two digits and ended with a block of seven
digits. The test consisted of 36 items divided in six blocks of six trials that increased in
difficulty. Each correct trial was rewarded with one point. The minimum score was 0
and the maximum score was 36. The Turkish-Dutch children were given Turkish and
Dutch parallel tests.
Psychometric characteristics of the English version of all three working-memory
tests were reported in Alloway, Gathercole, and Pickering (2006) and proved satisfactory.
Reliability measures of the Dutch and Turkish versions of the AWMA listening-recall
test were developed within the doctoral research project of Messer (2010). Measures
taken at ages 5 and 6 in a sample of Dutch and Turkish-Dutch children showed moderate
stability over a one-year interval (r = .45, p< .001) for the Dutch version (n =136) and
similarly for the Turkish version (r = .54, p< .001; n = 65), indicating sufficient testretest reliability. Concurrent correlations at age 6 of the Dutch and Turkish language
version of the listening recall test with concurrent working-memory measures were
overall satisfactory (r = .49, p< .001, and r = .31, p< .01) with backward digit recall (r = .36,
p< .001, and r = .14, not significant) with visuospatial odd-one-out, indicating sufficient
construct validity.
43
chapter 2
Procedure
All children were tested individually in a quiet room in their school between
February and May. Tasks were presented in a fixed order. First the children were tested
in Dutch by a non-Turkish speaking Dutch person, two months later a native-speaking
Turkish researcher tested the Turkish-Dutch children in their mother tongue on all
three working-memory tests as well as on vocabulary and sentence imitation.
Results
The results’ section is divided in three parts. The first and second sections present
the results of the language tests and working-memory tests, respectively, examining
the differences between Turkish-Dutch and native-Dutch children on these skills. In
the third section the relationships between the working memory tests and the language
tests are presented.
Language skills
First t tests for independent samples were used to compare performance on Dutch
productive vocabulary and Dutch sentence imitation between Turkish-Dutch and nativeDutch children. The first columns of Table 2 present the descriptive statistics of each
group on both tests in Dutch. The scores of Native-Dutch children were significantly
better than those of the Turkish-Dutch children on Dutch productive vocabulary,
t(66) = 6.51, p< .0001, as well as on Dutch sentence imitation, t(35) = 2.65, p = .01.
Next, performance of the Turkish-Dutch children on the Dutch language tests was
compared with their performance on the Turkish language tests by means of t tests for
Table 2 Mean and Standard Deviation on Vocabulary and Sentence Imitation in Dutch and
in Turkish (for Turkish-Dutch Children only)
Language group Vocabulary
Turkish-Dutch
Mean 19.5
SD 4.9
native-Dutch
Mean 27.5
SD 4.6
44
Dutch language
Sentence imitation
Vocabulary
Turkish language
Sentence imitation
28.5
6.3
15.3
6.8
17.9
7.8
33.9
5.9
-
-
differential relationships between language skills and working memory
paired samples. The last columns of Table 2 present the descriptive statistics of TurkishDutch children on the Turkish and Dutch versions of productive vocabulary and sentence
imitation. Turkish-Dutch children were better on Dutch productive vocabulary than on
Turkish productive vocabulary, t(17) = 2.41, p = .03. They were also better on Dutch
sentence imitation than on Turkish sentence imitation, t(15) = 5.05, p< .0001.
Pearson correlations were also computed to assess relationships among and between
language skills. Table 3 presents the correlations. The figures show that the language skills
of the native-Dutch children did not correlate significantly, whereas those of the TurkishDutch children revealed significant and high correlations between Dutch productive
vocabulary and Dutch sentence imitation and between Turkish productive vocabulary
and Turkish sentence imitation. There were no significant relationships between Dutch
and Turkish vocabulary or between Dutch and Turkish sentence imitation.
To summarise, these findings reveal that Dutch language skills of native-Dutch
children were better than those of Turkish-Dutch children. Turkish-Dutch children had
better Dutch language skills than Turkish language skills. Dutch productive vocabulary
and Dutch sentence imitation were unrelated skills in native-Dutch children, but
highly related in Turkish-Dutch children (Fisher Z = 2.11, p = .01). Turkish productive
vocabulary and Dutch sentence imitation were also highly related, but Turkish
Table 3 Pearson Correlations between and among Dutch and Turkish Language Tests
Language group
Turkish-Dutch
native-Dutch
Dutch vocabulary * Dutch sentence imitation
r .73
p .001
n 21
.16
.57
16
Turkish vocabulary * Turkish sentence imitation
r .53
p .01
n .22
Dutch vocabulary * Turkish vocabulary
r .24
p .34
n 18
Dutch sentence imitation * Turkish sentence imitation
r .01
p .99
n 16
45
chapter 2
vocabulary and Dutch vocabulary or Turkish sentence imitation and Dutch sentence
imitation were not.
Working memory
First the results of the working memory tests of the Turkish-Dutch and the nativeDutch children on the Dutch version of the working-memory tests will be presented.
Second, a comparison will be made for the Turkish-Dutch children only between
achievements in their first language, Turkish, and their second language, Dutch.
Monolingual vs. Bilingual children. A 2 (language group: Turkish-Dutch vs. nativeDutch) by 3 (memory task: digit recall vs. listening recall vs. backward digit recall)
multivariate analyses of variance (MANOVA) was performed on the percentages correct
answers on the Dutch working-memory tests. Language group was a between-subjects
variable and memory task a within-subjects variable. Table 4 presents the descriptive
statistics for each group regarding the mean number of correct items on all three Dutch
working-memory tests.
The main effect of language group was significant F(1, 84) = 15.21, p< .001, partial
η2 = .15. Native-Dutch children outperformed the Turkish-Dutch participants. The
main effect of test was also significant F(2, 82) = 274.67, p< .0001, partial η2 = .77.
The results showed no significant language group by memory task interaction, F< 1.
Bonferroni-adjusted post-hoc tests were then carried out to investigate the differences
on the working memory tests. Performance on digit recall was significantly better than
on listening recall and on backward digit recall, both p’s < .05; the difference between
listening recall and backward digit recall was not significant. To summarise, these
findings show that native-Dutch children were better at all three memory tasks than
Table 4 Mean and Standard Deviation of Percentage Correct Items of the Dutch Working
Memory Tests of the Two Participant Samples
Language group
Turkish-Dutch
Digit recall
Listening recall
Backward digit recall
Mean 43.2
SD 5.9
n 38
18.9
11.4
38
20.8
13.2
38
Mean 48.1
SD 6.3
n 48
26.5
9.6
48
25.8
6.8
48
native-Dutch
46
differential relationships between language skills and working memory
Turkish-Dutch children, and that both Turkish-Dutch and native-Dutch children were
better at digit recall than listening recall and backward digit recall.
Bilingual children test in both languages. These analyses pertain to the TurkishDutch children only who were tested in both Dutch and their mother tongue Turkish.
A 2 (language: Dutch vs. Turkish) by 3 (memory task: digit recall vs. listening recall vs.
backward digit recall) MANOVA was performed on the percentages correct answers of
the working memory tests. Language was a between-subjects variable and memory task
a within-subjects variable. Table 5 presents the mean and standard deviations of each
working-memory test in both languages.
The main effect of memory task was significant F(2, 46) = 109.14, p< .0001, partial
η2 = .83, whereas the main effect of language was not, F(1, 23) = 1.43, p = .24, partial
η2 = .06. The results also showed a significant language by memory task interaction,
F(2, 46) = 3.65, p = .03, partial η2 = .14. Paired-samples t tests were then carried out
to investigate the differences between the tested languages. A significant difference in
percentage correct answers was found for digit recall t(26) = -4.32, p< .0001. With respect
to listening recall and backward digit recall, however, no such difference emerged,
t(25) = .49, p = .63, and t(24) = -.82, p = .42, respectively.
Pearson correlations between the Turkish and Dutch versions of each of the three
working-memory tests were also conducted. Table 5 shows that Turkish digit recall and
Dutch digit recall did not correlate significantly; similarly for listening recall. Turkish
and Dutch backward digit recall did, however, correlate significantly. To summarise,
these findings revealed that Turkish-Dutch children performed better on digit recall
when tested in Turkish than in Dutch, but they obtained similar scores for listening recall
Table 5 Mean and Standard Deviation of Percentages Correct Items on Working Memory of
Turkish-Dutch Children Tested in both Dutch and Turkish (n = 24)
Language
Dutch
Digit recall
Listening recall
Backward digit recall
Mean 44.7
SD 5.1
20.4
9.4
23.5
12.0
Mean 50.3
SD 7.6
18.2
11.6
25.1
12.2
.33, p = .10
.53, p = .006
Turkish
Pearson correlation
.28, p = .17
Note. The means of the Dutch working-memory tests presented in this table deviate slightly from those in Table 4, because not all
Turkish-Dutch children who were administered the Dutch tests received the Turkish version.
47
chapter 2
and backward digit recall in Turkish and Dutch. Interestingly, a strong performance
association between the two skills was apparent in backward digit recall only.
The relationships between working memory and language skills
Pearson correlations were computed between the three Dutch tests for working
memory (i.e., digit recall, listening recall, and backward digit recall) and the two Dutch
language tests (i.e., productive vocabulary and sentence imitation) for the native-Dutch
children and for the Turkish-Dutch children. The findings are presented in Table 6.
None of the correlations of the native-Dutch sample reached significant levels, whereas
five out of six correlations reached substantial and significant levels in the Turkish-Dutch
sample. To substantiate the differences between the two language groups, Fisher-Z tests
were conducted on the six correlational comparisons. It appeared that the correlational
differences between the two language groups pertaining to the association between the
language test ‘vocabulary’ on the one hand and the three working-memory tasks on the
other were significantly larger in the Turkish-Dutch group than in the native-Dutch group.
Note also, that the correlations in the first group were significant, whereas those in the latter
group were not. With respect to the association between ‘sentence imitation’ and the three
working-memory tasks the correlations did not differ significantly between language groups.
Table 6 Pearson Correlations between all Three Tests of Dutch Working Memory and the Two
Dutch Language Tests for both Language Groups
Vocabulary
Working-memory test
Turkish-Dutch
n = 21
native-Dutch
n = 47
Sentence imitation
Turkish-Dutch
n = 21
native-Dutch
n = 16
Digit recall
r
p
Fisher Z
Listening recall
r
p
Fisher Z
Backward digit recall
r
p
Fisher Z
.72
.0001
.05
.75
3.07, p = .001
.63
.002
.49
.02
.07
.66
1.67, p = .04
.47
.03
.49
.06
-0.07, p = .47
.50
.02
-.02
.88
2.03, p = .02
.22
.34
.41
.11
-0.58, p = .28
Note. The figures in bold represent significant results.
48
.26
.34
1.31, p = .09
differential relationships between language skills and working memory
Finally, Pearson correlations were computed between the three Turkish tests for
working memory and the two Turkish language tests for the Turkish-Dutch children.
Turkish productive vocabulary is significantly related to Turkish digit recall and Turkish
listening recall, but not with Turkish backward digit recall. Turkish sentence imitation
correlated significantly with Turkish digit recall, Turkish listening recall, and Turkish
backward digit recall (see Table 7).
To summarise, these findings reveal that Dutch working-memory skills were
unrelated to Dutch language skills in the native-Dutch children, but highly related in
Turkish-Dutch children. Note that differences in correlational strength between the
language groups were only apparent with respect to vocabulary and not for sentence
imitation. Also Turkish working-memory skills were highly related to Turkish language
skills in the Turkish-Dutch children.
Table 7 Pearson Correlations between all Three Tests of Turkish Working Memory and the
Two Turkish Language Tests for the Turkish-Dutch Children only
Vocabulary
Sentence
imitation
Working-memory test
Digit recall
r .44
p .04
n 23
.58
.009
21
r .64
p .001
n 22
.54
.02
20
r .33
p .13
n 23
.44
.05
21
Listening recall
Backward digit recall
Note. The figures in bold represent significant results.
Discussion
The goal of this study was twofold. One, compare working memory skills between
bilingual Turkish-Dutch children and monolingual native-Dutch children from lowincome families in the Netherlands. Two, establish the relationships between language
49
chapter 2
performance and working memory within each language group. The answers to the
six questions stated in the introduction will provide the background for addressing the
main goals of the present study.
Language skills
Native-Dutch children had indeed better Dutch language skills than Turkish-Dutch
children, both with respect to vocabulary and sentence imitation. An important finding
was that the Turkish-Dutch children performed better on the Dutch language tests than
on the Turkish ones. At the age of 6 or 7, Turkish-Dutch children have certainly acquired
a great deal of knowledge regarding the Dutch language, but their skills are not yet
at the level of those of native-Dutch children from similar backgrounds. The fact that
performance on Turkish tests was below that of Dutch tests suggests that the development
of Turkish in this group is slowing down. The children from these Turkish background
only hear Turkish at home or in the family. Unlike in the past, these children are unable
to attend Turkish lessons, because financing of extracurricular language education for
children who are non-native speakers of Dutch was stopped some years ago.
An interesting result was the differential relationship between performance on Dutch
vocabulary and Dutch sentence imitation of the two language groups. Performance on
these tasks was unrelated in native-Dutch children, but highly related in Turkish-Dutch
children as was Turkish vocabulary and Turkish sentence imitation. Note that in the
group with the more limited language skills relationships between different language
tasks was strong. We return to this finding below.
Memory
Native-Dutch children outperformed the Turkish-Dutch children on all three Dutch
memory tasks. Thus children with the better language skills also had a better memory
performance. Both language groups scored better on digit recall, a task tapping in the
storage function of memory, than on listening recall and backward digit recall, tasks that
refers storage as well as processing or manipulating information in memory.
Turkish memory performance was only superior on digit recall; the Turkish-Dutch
children performed equally well on Dutch listening recall and Turkish listening recall
and on Dutch backward digit recall and Turkish backward digit recall. Storage of Turkish
numbers appears to be the only aspect of memory that it is still better developed than
Dutch numbers in Turkish-Dutch children. The manipulation of information in memory
is equally well developed in Dutch and Turkish. In a recent study Janssen, Bosman, and
Leseman (2013) showed that Dutch phoneme awareness of Turkish-Dutch children in a
comparable sample was better than in Turkish.
The work of Da Fontoura and Siegel (1995) presented an opportunity to assess
50
differential relationships between language skills and working memory
performance difference on working memory tasks in first and second language of a
group of Portuguese-English bilingual children. Irrespective of their reading level, all
children performed better on the English version of the working-memory task than on
the Portuguese one (see Note 1, for the statistical analyses). Because it is impossible to
compare the level of proficiency of the participants of the present study with that of Da
Fontoura and Siegel (1995), it may be worthwhile to conduct a comparative study that
will shed some more light on the development of memory and language in bilingual
children. Note also that a strong performance association emerged between the Dutch
and Turkish version of the backward digit recall task, no such correlations existed for
digit recall and listening recall. Performance on Backward digit recall was worse than on
digit recall but equally good on listening recall.
Relationship between memory and language skills
With respect to associations between language and memory skills, an interesting
pattern emerged: Dutch working memory and Dutch language are unrelated skills in
native-Dutch children, but highly related in Turkish-Dutch children. Turkish working
memory and Turkish language skills are also highly related skills in Turkish-Dutch
children. These findings combined with the inferior language skills of the Turkish-Dutch
children suggest that a minimal level of language development is required to strengthen
verbal-working memory skills. Stated differently, limited exposure to language input,
suggests that experience with a particular language (i.e., Dutch) determines, at least
partly, the capacity of verbal-working memory in that language.
Sufficient semantic and syntactic knowledge required to support the capacity of verbalworking memory in listening recall in the Turkish-Dutch group, may not have been
sufficiently developed yet, neither in Turkish, nor in Dutch. Turkish-Dutch children’s
performance on the Turkish sentence-imitation task indicated low syntactic sensitivity in
Turkish, which was even lower than their syntactic sensitivity in Dutch. Being exposed
to a second language that is the dominant language in society after development of
the first language has started offers a possible explanation. Listening recall requires
good language proficiency. Languages proficiency provides options for chunking
and integrating verbal (semantically and syntactically structured) information, as is
especially needed in performing listening recall tasks (Service, 1992). A language delay
may therefore limit verbal-working memory and may slow down language acquisition
(Messer et al., 2010; Thorn & Frankish, 2005).
Strong associations between skills seem to indicate that skills have not yet fully
developed. An example from the reading literature reveals that phonological skills are
strongly related with reading performance, but only at the onset of reading development.
After children gain reading experience the correlation between phonological skills and
51
chapter 2
reading drops and usually even disappears (e.g., de Jong & van der Leij, 2002; Furness
& Samuelsson, 2009; Patel, Snowling, & de Jong, 2004). Thus, after children become
more proficient in skills, the initial relationship between performances diminishes
over time.
Implications
The strong association between language skills and verbal-working memory in
children with limited language skills likely points to reciprocal effects in a developmental
process of mutually constituting abilities (see also Jones, Gobet, & Pine, 2008; Messer et
al., 2010). This has a number of implications. First of all, children, who as a consequence
of being raised bilingually and growing up in a language poor environment are
lagging behind in the language of the school, face a double problem. Not only are they
disadvantaged in school language as such, but probably also in the ability to learn school
language from the input provided at school. Moreover, this effect may easily spread
to several subject matter areas involving understanding instruction, learning verbally
stated knowledge, and reading comprehension. Several studies, indeed, indicated that
poor working memory is a predictor of persistent learning difficulties in several school
subjects. Gathercole, Alloway, Willis, and Adams (2006) conclude that “working memory
acts as a bottleneck for learning” (p. 17). Gathercole and Alloway (2004) observed
children with poor working memory to have more difficulties in following instructions,
keeping place in a complex task, coping with simultaneous storage and processing
demands, and longer-term remembering. Given the tendency in the literature to view
working memory as a domain-general ability which is hardly dependent on experience
and instruction (Swanson, 2001), the present study adds an important new perspective
to these analyses, namely that verbal working-memory problems, may – at least partly
– be caused by language deficiencies that could be remediated by supporting language
development at an early age.
Focusing on verbal-working memory, we were not able to find evidence for cognitive
advantages of bilingualism in this study. As a matter of fact, we found disadvantages
in language proficiency and verbal-working memory as a probable consequence of
being bilingual, noting that the mono- and bilingual groups in the present study did
not differ in nonverbal cognitive ability and socio-economic background. Therefore, the
disadvantages in the verbal working-memory tasks could be explained by differences in
language proficiency and language input, in full agreement with the idea that becoming
more familiar with a particular language increases verbal working memory capacity for
that language.
52
differential relationships between language skills and working memory
Note
1. Table 2 of Da Fontoura and Siegel (1995) presents the mean scores for a group
of normally achieving readers (n = 24) on Portuguese working memory (5.1,
SD = 1.7) and on English working memory (6.0, SD = 2.1). The authors did not
present the statistics, but a t test for dependent samples revealed a significant
difference, t(23) = -2.1, p< 0.05. A similar computation for a group of reading
disabled children also proved to be significant, t(11) = -2.3, p< 0.05; Portuguese
working memory (3.7, SD = 1.5) and English working memory (4.9, SD = 1.8).
53
chapter 2
References
Aarts, R., de Ruiter, J. J., & Verhoeven, L. (1996). Eigen-taalvaardigheid en schoolsucces
van Turkse en Marokkaanse leerlingen [First language skills and academic
achievement of Turkish and Moroccan students]. Pedagogische Studiën, 73, 275-290.
Adams, A., & Gathercole, S. (2000). Limitations in working memory: Implications
for language development. International Journal of Language and Communication
Disorders, 35, 95-116. doi: 10.1080/002075999399701
Adams, A. M., Bourke, L. J., & Willis, C. S. (1999). Working memory and spoken
language comprehension in young children. International Journal of Psychology, 34,
364-373. doi: 10.1080/002075999399701
Alloway, T. P. (2007). Automated Working Memory Assessment. Translated and reproduced
by permission of the Psychological Corporation, UK.
Alloway, T. P., Gathercole, S. E., & Pickering, S. J. (2006). Verbal and visuospatial shortterm and working memory in children: Are they separable? Child Development, 77,
1698-1716. doi: 10.1111/j.1467-8624.2006.00968.x
Baddeley, A. (2000) The episodic buffer: A new component of working memory? Trends
in Cognitive Sciences, 4, 417-423. doi: 10.1016/S1364-6613(00)01538-2
Baddeley, A. D., Gathercole, S., & Papagno, C. (1998). The phonological loop as a language
learning device. Psychological Review, 105, 158-173. doi: 10.1037//0033-295X.105.1.158
Baddeley, A. D., & Hitch, G. J. (1974). Working memory. In G. A. Bower (Ed.), The
psychology of learning and motivation (pp. 47-90). New York, NY: Academic Press.
Baker, S. K., Simmons, D. C., & Kameenui, E. J. (1995). Vocabulary acquisition: Synthesis
of the research. Eugene, OR: University of Oregon, National Center to Improve the
Tools of Educators.
Bialystok, E. (2001). Bilingualism in development: Language, literacy and cognition. New
York, NY: Cambridge University Press.
Bialystok, E. (2002). Acquisition of literacy in bilingual children: A framework for
research. Language Acquisition, 57, 45-77. doi: 10.1111/1467-9922.00180
Bialystok, E. (2009). Bilingualism: The good, the bad, and the indifferent. Bilingualism:
Language and Cognition, 12, 3-11. doi: 10.1017/S1366728908003477
Bialystok, E., & Luk, G. (2012). Receptive vocabulary differences in monolingual and
bilingual adults. Bilingualism: Language and Cognition, 15, 397-401. doi: 10.1017/
S136672891100040X
Centraal Bureau voor de Statistiek (2008). Jaarrapport integratie 2008 [Year report
integration 2008]. Den Haag, the Netherlands: Centraal Bureau voor de Statistiek.
Cheung, H. (1996). Nonword span as a unique predictor of second-language vocabulary
learning? Developmental Psychology, 32, 867-873. doi: 10.1037/0012-1649.32.5.867
54
differential relationships between language skills and working memory
Cummins, J. (2000). Language, power and pedagogy: Bilingual children in the crossfire.
Buffalo, NY: Multilingual Matters. Ltd.
Da Fontoura, H. A. & Siegel, L. S. (1995). Reading, syntactic, and working memory skills
of bilingual Portuguese–English Canadian children. Reading and Writing, 7, 139-153.
doi: 10.1007/BF01026951
de Jong, P. F. (1998). Working memory deficits of reading disabled children. Journal of
Experimental Child Psychology, 70, 75-96. doi: 10.1006/jecp.1998.2451
de Jong, P. F. & van der Leij, A. (2002). Effects of phonological abilities and linguistic
comprehension on the development of reading. Scientific Studies of Reading, 6, 51-77.
doi: 10.1207/S1532799XSSR0601_03
Elzer, M. J. (2005). Talen naar een moedertaal [Care for mother tongue]. Tijdschrift voor
Orthopedagogiek, 3, 130-135.
Engel de Abreu, P. M. J. (2011). Working memory in multilingual children: Is there a
bilingual effect? Memory, 19, 529-537. doi: 10.1080/09658211.2011.590504
Engel de Abreu, P. M. J., Cruz-Santos, A., Tourinho, C. J., Martin, R., & Bialystok, E. (2012).
Bilingualism enriches the poor: Enhanced cognitive control in low-income minority
children. Psychological Science, 23, 1364-1371. doi: 10.1177/0956797612443836
Furness, B., & Samuelsson, S. (2009). Preschool cognitive and language skills predicting
Kindergarten and Grade 1 reading and spelling: A cross-linguistic comparison.
Journal of Research in Reading, 32, 275-292. doi: 10.1111/j.1467-9817.2009.01393.x
Gathercole, S. E., & Alloway, T. P. (2004). Working memory and classroom learning.
Journal of the Professional Association for Teachers of Students with Specific Learning
Difficulties, 17, 2-12.
Gathercole, S. E., Alloway, T. P., Willis, C., & Adams, A. (2006). Working memory in
children with reading disabilities. Journal of Experimental Child Psychology, 93, 265281. doi: 10.1016/j.jecp.2005.08.003
Gathercole, S. E., & Baddeley, A. D. (1993). Working memory and language. Hillsdale, NJ:
Erlbaum.
Gathercole, S. E., Service, E., Hitch, G. J., Adams, A., & Martin, A. J. (1999). Phonological
short term memory and vocabulary development: Further evidence on the nature
of the relationship. Applied Cognitive Psychology, 13, 65-77. doi: 10.1002/(SICI)10990720
Gathercole, S. E., Willis, C., Emslie, H., & Baddeley, A. (1992). Phonological memory
and vocabulary development during the early school years: A longitudinal study.
Developmental psychology, 28, 887-898. doi: 10.1037//0012-1649.28.5.887
Janssen, M., Bosman, A. M. T., & Leseman, P. P. M. (2013). Phoneme awareness,
vocabulary, and word decoding in monolingual and bilingual Dutch children.
Journal of Research in Reading, 36, 1-13. doi: 10.1111/j.1467-9817.2011.01480.x
55
chapter 2
Jones, G., Gobet, F., & Pine, J. M. (2008). Computer simulations of developmental
change: The contributions of working memory capacity and long-term knowledge.
Cognitive Science, 32, 1148-1176. doi: 10.1080/03640210802073689
Kormi-Nouri, R., Shojaei, R., Moniri, S., Gholami, A., Moradi, A. Akbari-Zardkhaneh,
S., & Nilsson, L. (2008). The effect of childhood bilingualism on episodic and
semantic memory tasks. Scandinavian Journal of Psychology, 49, 93-109. doi: 10.1111/
sjop.2008.49.issue-2
Majerus, S., Poncelet, M., Greffe, C., & Van der Linden, M. (2006). Relations between
vocabulary development and verbal short-term memory: The relative importance of
short-term memory for serial order and item information. Journal of Experimental
Child Psychology, 93, 95-119. doi: 10.1016/j.jecp.2005.07.005
Messer, M. H. (2010). Verbal short-term memory and vocabulary development. Utrecht, the
Netherlands: Doctoral dissertation Utrecht University .
Messer, M. H., Leseman, P. P. M., Boom, J., & Mayo, A. Y. (2010). Phonotactic probability
effect in nonword recall and its relationship with vocabulary in monolingual and
bilingual preschoolers. Journal of Experimental Child Psychology, 105, 306-323. doi:
10.1016/j.jecp.2009.12.006
Morales, J., Calvo, A., & Bialystok, E. (2013). Working memory development in
monolingual and bilingual children. Journal of Experimental Child Psychology,
114, 187-202. doi: 10.1016/j.jecp.2012.09.002
O’Brien, I., Segalowitz, N., Collentine, J., & Freed, B. (2006). Phonological memory and
lexical, narrative, and grammatical skills in second language oral production by adult
learners. Applied Psycholinguistics, 27, 377-402. doi: 10.1017/S0142716406060322
Oller, D. K., & Eilers, R. E. (2002). Language and literacy in bilingual children. Clevedon,
United Kingdom: Multilingual Matters.
Papagno, C., & Vallar, G. (1995). Verbal short-term memory and vocabulary learning
in polyglots. The Quarterly Journal of Experimental Psychology, 48, 98-107. doi:
10.1080/14640749508401378
Patel, T. K., Snowling, M. J., & de Jong, P. F. (2004). A cross-linguistic comparison of
children learning to read in English and Dutch. Journal of Educational Psychology, 96,
785-797. doi: 10.1037/0022-0663.96.4.785
Pearson, B. Z. (2007). Social factors in childhood bilingualism in the United States.
Applied Psycholinguistics, 28, 399-410. doi: 10.1017/S014271640707021X
Pearson, B. Z., Fernandez, S. C., Lewedeg, V., & Oller, D. K. (1997). The relation of input
factors to lexical learning by bilingual infants. Applied Psycholinguistics, 18, 41-58. doi:
10.1017/S0142716400009863
Pickering, S. J. (2006). Working memory and Education. Burlington, NJ: Academic
press.
56
differential relationships between language skills and working memory
Poulin-Dubois, D., Blaye, A., Coutya, J., & Bialystok, E. (2011). The effects of bilingualism
on toddlers’ executive functioning. Journal of Experimental Child Psychology, 108,
567-579. doi: 10.1016/j.jecp.2010.10.009
Raven, J. C. (1958). Standard Progressive Matrices. London, United Kingdom: H. K. Lewis.
Scheele, A. F., Leseman, P. P. M., & Mayo, A. Y. (2010). The home language environment
of mono- and bilingual children and their language proficiency. Applied
Psycholinguistics, 31, 117-140. doi: 10.1017/S0142716409990191
Service, E. (1992). Phonology, working memory, and foreign language
learning. The Quarterly Journal of Experimental Psychology, 45, 21-50. doi:
10.1080/14640749208401314
Siegel, L. S., & Ryan, E. B. (1989). The development of working memory in normally
achieving and subtypes of learning disabled children. Child Development, 60, 973980. doi: 10.2307/1131037
St Clair-Thompson, H. L., & Gathercole, S. E. (2006). Executive functions and achievements
in school: Shifting, updating, inhibition and working memory. The Quarterly Journal
of Experimental Psychology, 59, 745-759. doi: 10.1080/17470210500162854
Swanson, H. L. (2001). A subgroup analysis of working memory in children
with reading disabilities. Journal of Learning Disabilities, 34, 249-263. doi:
10.1177/002221940103400305
Tesser, P. T. M., & Iedema, J. (2001). Rapportage minderheden 2001. Deel I Vorderingen
op school [Report Minorities 2001. Part 1 Achievements in School]. Den Haag, the
Netherlands: Sociaal en Cultureel Planbureau.
Thorn, A. S. C., & Frankish, C. R. (2005). Long-term knowledge effects on serial recall
of nonwords are not exclusively lexical. Journal of Experimental Child Psychology, 31,
729-735. doi: 10.1037/0278-7393.31.4.729
van Eldik, M. C. M., Schlichting, J. E. P. T., Lutje Spielberg, H. C., van der Meulen, B. F.,
& van der Meulen, S. (1997). Reynell test voor taalbegrip, handleiding. [Reynell test for
language comprehension, manual]. Nijmegen, the Netherlands: Berkhout.
van Elten, M. (2003). Schoolprestaties van allochtone kinderen [Academic achievement
of immigrant children]. Onderwijsmagazine Inzicht. Retrieved June, 2009, from
http://www.voo.nl/inzicht.html
Verhoeven, L., Narain, G., Extra, G., Konak, O. A., & Zerrouk, R. (1995). Toets Tweetaligheid
Handleiding [Manual Test for Bilingualism]. Arnhem, the Netherlands: Cito.
57
58
Chapter 3
phoneme awareness, vocabulary,
and word decoding in monolingual
and bilingual dutch children
Published as:
Janssen, M., Bosman, A. M. T., & Leseman, P. P. M. (2013). Phoneme awareness,
vocabulary, and word decoding in monolingual and bilingual Dutch children. Journal of
Research in Reading, 36, 1-13. doi: 10.1111/j.1467-9817.2011.01480.x
59
chapter 3
Abstract
The aim of this study was to investigate whether bilingually raised children in the
Netherlands, who receive literacy instruction in their second language only, show an
advantage on Dutch phoneme-awareness tasks compared to monolingual Dutch-speaking
children. Language performance of a group of 47 immigrant first-grade children with
various different cultural backgrounds and a subsample of 29 Turkish-Dutch bilingual
immigrant children was compared with those of 15 first-grade monolingual native
Dutch children from similar low socio-economic backgrounds. All children were tested
on Dutch phoneme awareness, vocabulary, and word decoding. The Turkish-Dutch
children were also tested on Turkish phoneme awareness and Turkish vocabulary.
Dutch vocabulary scores of the bilingual children were below that of the monolingual
Dutch children. Neither the entire group of bilingual children nor the subsample of
Turkish-Dutch children were better or worse on phoneme-awareness than monolingual
Dutch children. However, Turkish-Dutch children scored better on the Dutch tasks for
phoneme awareness and vocabulary than on the Turkish tasks. Language proficiency in
the adopted language of bilingual children appears to quickly exceed that of their native
language, when no instruction in the first language is provided.
60
phoneme awareness, vocabulary, and word decoding
In the Netherlands, the majority of first and second, and even part of the thirdgeneration of immigrants is bilingual (3.2 million on a population of 16.5 million
people). The largest group is from Turkish origin (384.000) and the second largest is
from Moroccan descent (349.000; Centraal Bureau voor de Statistiek [CBS], 2008).
The children in this community usually learn to speak the language of their parents,
their first language, at home. Substantial and prolonged exposure to Dutch as a second
language begins for most of them not before they start kindergarten at the age of 4.
Being exposed to the second language after having acquired the first language is a
called successive or consecutive bilingualism as opposed to simultaneous bilingualism.
Simultaneous bilinguals are exposed to two languages in about the same degree from
birth (Bhatia, 2006; Yip & Matthews, 2007).
Language input is known to be strongly related to language proficiency for both
monolingual and bilingual speakers (e.g., Huttenlocher, Haight, Bryk, Seltzer,
& Lyons, 1991) and it varies vastly between families, social classes, and ethnic-cultural
communities (Hoff, 2006). Because language input is critical in bilingual development,
children must be exposed to both languages in a roughly equal and, for each language
separately, sufficient degree to become balanced and proficient speakers (Oller & Eilers,
2002). Unfortunately, however, in the case of bilingual immigrant communities, at least
in the Netherlands, these conditions are rarely fulfilled.
Not only the quantity of exposure is likely to differ between the first and the second
language, also the quality may reveal profound differences in terms of vocabulary,
grammar, and pragmatics (Duursma, Romero-Contreras, Szuber, Proctor, & Snow,
2007; Scheele, Leseman & Mayo, 2010). The first language is often used mostly for
informal and interpersonal communication at home, whereas the second language is
usually the language at school and in public media. Thus, even if a bilingual has been
exposed to the two languages from birth, he or she does not necessarily acquire the two
languages equally well. This will depend on the quality and quantity of the input the
child receives in each language. Although language development of a bilingual child
usually follows the same developmental stages as of monolinguals in either language,
there are also important differences. In an early stage, young bilingual children are often
found to mix the two languages. In a later stage, however, language mixing, or codeswitching as it is called, seems to be more purposeful, depending on the communicative
context and the functions of language use (Bialystok, 2007).
Being raised a bilingual may have disadvantages as well as advantages (Bialystok,
2007) with corresponding hypotheses with respect to cross-linguistic effects.
Disadvantages of bilingualism are manifest in for example the on average smaller
61
chapter 3
vocabularies in each of their languages compared to monolinguals (Scheele et al., 2010).
Verhoeven (2000) presented data on the comparison between monolingual Dutch
speaking children and Dutch second language learners from Turkish, Moroccan, and
children from previous Dutch colonies. His study showed that Dutch bilingual children
had a significantly smaller vocabulary in their second (i.e., Dutch) language compared
with monolingual Dutch children. Oller (2005) maintains that bilingual children have
smaller vocabularies in both their first and second language, but adds that “lexicalized
concepts of the bilingual are differentially ‘distributed’ across the two languages such
that some concepts are lexicalized in one language but not in the other and vice versa”
(p. 1744). Note, however, that this difference in vocabulary may affect educational
achievement negatively, especially with regard to reading comprehension (see also
Jiménez, Garcia, & Pearson, 1995). These findings led to the ‘competition hypothesis’,
stating that competition between the first and second language causes negative effects
in the development of bilingual language acquisition.
Advantages of bilingualism have been shown in the domain of meta-linguistic,
cognitive, and conceptual processing, as well as with respect to executive, attentional,
and control skills. Because bilingual children need to be able to keep the two systems
apart, they have to decide when to use which language, which words, and which syntactic
structure. These skills require highly developed executive functions, functions that are
responsible for the control of cognitive processes, including working memory, attention
shifting, and inhibition. Positive transfer of knowledge and skills in the first language to
the acquisition of the second language is referred to as ‘the linguistic interdependence
hypothesis’ (cf., Cummins, 1981).
According to the linguistic interdependency hypothesis, bilinguals with sufficient
proficiency in their first language should be able to use conceptual-semantic knowledge
and higher-level ‘academic’ language skills acquired in the first language to learn the
lexicon, use rules, and academic uses of the second language, which should be reflected
in positive correlations between first and second language test scores. Evidence for the
linguistic interdependency hypothesis is provided by numerous studies (e.g., Bialystok,
2005; Bialystok, Luk, & Kwan, 2005; D’Angiulli, Siegel, & Serra, 2001; Durgunoglu &
Hancin, 1992; Scheele et al., 2010).
The present study aims at investigating the role of bilingualism in the development
of an important meta-linguistic skill, namely phonological or more precisely, phoneme
awareness. Phonological awareness is “an explicit awareness of the phonological
structure of the words in one’s language” (Torgesen, Wagner, & Rashotte, 1994,
p. 276), whereas phoneme awareness is “…an awareness of sounds in spoken (not
written) words that is revealed by such abilities as rhyming, matching initial consonants,
and counting the number of phonemes in spoken words” (Stahl & Murray, 1994, p. 221).
62
phoneme awareness, vocabulary, and word decoding
Phoneme awareness is closely related to vocabulary and is one of the best predictors of
word-decoding abilities in both monolingual (e.g., Bradley & Bryant, 1983; Lundberg,
Frost, & Peterson, 1988; see Blachman, 2000 for a review) as well as bilingual children
(e.g., Stuart, 1999, 2004). Note that, research has also shown that when children start to
learn to read and write, phonological skills, and in particular phoneme-awareness skills
develop as a result of the orthographic input of written language, that is when they are
learning to read and spell (Liberman & Liberman, 1990; Morais, 1991).
There is ample evidence for cross-language transfer of (instruction) of phoneme
awareness in the first language on literacy skills in the second language (e.g., Chen,
Xu, Nguyen, Hong, & Wang, 2010; D’Angiulli et al., 2001; Durgunoglu & Hancin,
1992; Kim, 2009; Leikin, Schwartz, & Share, 2010). In the majority of these studies,
bilingual children received instruction in both their first and in their second language.
In the present study, however, children only received literacy instruction in their second
language. The question, then, is whether children from immigrant families in the
Netherlands, who are successive bilinguals, still have an advantage over monolingual
children from similar low socio-economic backgrounds with respect to phoneme
awareness.
As said earlier, the Dutch population from immigrant descent in the Netherlands
is substantial (20%). The majority is from non-Western countries and has a low socioeconomic background. These children usually start acquiring their second language
(i.e., Dutch) when they enter kindergarten, which is around the age of four. Although
the largest group is from Turkish origin, the vast majority is from a huge range of
different countries. These particular circumstances gave us the opportunity to look into
the issue of linguistic transfer in children with a variety of different first languages
but also presented the possibility to study a subgroup of Turkish-Dutch children more
specifically.
We chose to compare phoneme-awareness skills and vocabulary development in the
second language of a group of bilingual children and in the first language of monolingual
children in the Netherlands. Because of the existence of a Turkish standardized
vocabulary test and the possibility of developing Turkish phoneme-awareness tests as
well, it was feasible to compare phoneme awareness and vocabulary in Turkish and
Dutch in a subsample of the entire group of immigrants.
Durgunoglu and Oney (1999) established an interesting finding with respect to
the goal of the present study, namely that phoneme awareness in monolingual Turkish
children appears to develop more easily than in monolingual English children. They
used three different phoneme-awareness tasks (phoneme segmentation, initialphoneme deletion, and final-phoneme deletion) and in accordance with their hypothesis,
the Turkish children could manipulate syllables more easily and performed better on
63
chapter 3
the phoneme-segmentation and phoneme-deletion task than the English children. For
initial-phoneme deletion the children scored at similar levels, albeit the Turkish children
were significantly better at final-phoneme deletion than at initial-phoneme deletion.
They explained the more advanced skills of the Turkish-speaking children in terms of
language characteristics. In general, Turkish is a much more transparent language than
English. It has a more consistently-defined syllable structure, stronger vowel harmony,
and a rather transparent morphological structure.
The Turkish writing system is one of the most transparent alphabetic languages.
English, on the other hand is a rather opaque. The differential consistency between
phonology and orthography in various languages is referred to as orthographic depth
(Ziegler et al., 2010). Spanish, Finnish, Turkish are transparent orthographies, because
graphemes correspond fairly consistently with phonemes, whereas English and Hebrew
are opaque, because the consistency between graphemes and phonemes is low. Although
Dutch has a more transparent writing system than English (see Bosman, Vonk, & van
Zwam, 2006) it is still less transparent than Turkish (see for a description of the Turkish
language, Durgunoglu, 2006, and for the Dutch, Bosman, de Graaff, & Gijsel, 2006).
To summarise, the main goal of the present study is to investigate whether
bilingually raised children in the Netherlands, who receive literacy instruction in their
second language only, still show an advantage on Dutch phoneme-awareness tasks
compared to monolingual Dutch-speaking children. Note that, all former studies
pertained to bilingual students who received instruction in both their first and their
second language. Additionally, we will study the difference in phoneme awareness
of Turkish-Dutch children, a subsample of the entire bilingual group, in their native
language Turkish and their second language Dutch. To obtain insight in the acquisition
of Dutch vocabulary and Dutch word decoding in the experimental groups, the scores
on these tests will be compared also. Moreover, Turkish-Dutch children will be tested on
Turkish vocabulary as well, which enables us to compare their Dutch vocabulary with
their Turkish vocabulary.
Method
Participants
The study involved 47 immigrant Dutch and 15 native Dutch first grade children
(mean age in May was 7 year and 3 months). All children had participated in a special,
half-day pre-school program for socio-economically disadvantaged children when they
were three years old. The program lasted about a year. The selection to participate in
this program was based on the observations of teachers who were experienced with
64
phoneme awareness, vocabulary, and word decoding
language development of young children. Children who were clearly lagging behind in
their language development were eligible for participation.
All children lived in poor inner-city neighbourhoods with a majority of lowincome and immigrant families. Background information was gathered by a parent
questionnaire. The native-Dutch children were born in the Netherlands and grew up in
low-income families in which the predominant language was Dutch; all were considered
monolingual. All immigrant children were bilingual, representing five different first
languages: Turkish (n = 29), Moroccan (Berber and Arabic; n = 15), Bosnian (n = 1),
Dari (n = 1), and Somalia (n = 1). Although all immigrant children were born in the
Netherlands, their fathers and 95% of their mothers were born in the country of origin.
For all immigrant children, the language of the parents was their first language and
for the majority of them (65%) this was still the best-developed language by the age
of three. Almost all children (80%) were exposed to Dutch as a second language well
before they were three years old, by watching Dutch television or playing with Dutch
speaking children, including older siblings who already attended Dutch primary schools.
Nonetheless, starting in pre-school meant for most of them a strong increase in Dutch
language input. Table 1 provides an overview of the participants and the mean score on
the Standard Progressive Matrices (SPM, Raven, 1958), a nonverbal intelligence test.
The mean scores of the Dutch and immigrant children on the SPM did not differ
significantly, F< 1. For the largest subgroup within the immigrant group, the TurkishDutch subgroup, measurement instruments in the first language Turkish were available,
allowing us to examine differences and similarities in the relationships of phoneme
awareness with first and second language. Therefore, in the remainder we will report
the results of the comparison between the native-Dutch children and the entire group
of immigrant children (including the Turkish-Dutch children) and between the nativeDutch children and the subgroup of Turkish-Dutch children.
Table 1 Participants and their Mean Score (SD in parentheses) on the Standard Progressive
Matrices
n
Participant Group
Native-Dutch
Immigrant
Turkish-Dutcha
a
15
47
29
Boys
9
27
19
Gender
Girls
6
20
10
Standard Progressive Matrices
27.5 (6.6)
24.5 (8.2)
25 (7.5)
The Turkish-Dutch group constitutes a subsample of the total group of immigrant children.
65
chapter 3
Materials
To measure phoneme awareness three tests were used: phoneme segmentation, initialphoneme deletion and final-phoneme deletion (Durgunoglu & Oney, 1999). The original
words in the tests used in the study of Durgunoglu and Oney were adapted to the purpose
of this research, in order to be administered in both Dutch and Turkish.
Phoneme segmentation. This task consisted of the same eight pseudowords that were
used by Durgunoglu and Oney (1999): e, a, mo, ep, le, ir, fim, mul. They varied between
one and three phonemes. All eight words were pronounceable in both Dutch and Turkish.
Children had to repeat the word and tell the experimenter how many phonemes they
heard. A correct answer was rewarded with one point. The minimum score was 0 and the
maximum score was eight.
Initial- and final-phoneme deletion. This task consisted of 12 pseudowords of three
phonemes each. Six of the original pseudowords used by Durgunoglu and Oney (1999)
could be used for the present purpose, that is, hez, niz, fid, ped, nug, and div. The other six
pseudowords used by Durgunoglu and Oney had to be changed to meet required criteria
and resulted in the following six pseudowords, that is, len, mek, gam, jul, tis, and san. The
original pseudowords had to be changed because four of them, when pronounced were
actual Dutch words. Another reason was that the original pseudowords were selected with
respect to rhyming familiarity in both English and Turkish. For our purpose a pseudoword
either had to rhyme with many words (high-familiar) or with few words (low-familiar) in
both Dutch and Turkish. The pseudowords len, gam, mek, jul, tis and san rhymed with
high-familiar Dutch words. They had a mean number of 13.2 Dutch rhyming neighbour
words. The pseudowords hez, niz, fid, ped, nug, div were low-familiar Dutch pseudowords.
They had a mean number of 2.2 rhyming neighbours in Dutch. The pseudowords: len, tis,
jam, hez, niz, and san were high-familiar Turkish words. They had a mean number of 7.0
rhyming neighbours in Turkish. The words fid, ped, nug, jul, div, and mek were low-familiar
Turkish words. They had a mean number of 1.8 rhyming neighbours in Turkish. The
children were asked to repeat the word with the initial phoneme deleted. Then they were
asked to repeat the words with the final phoneme deleted. The same pseudowords were
used in both phoneme-deletion tasks. A correct answer was rewarded with one point. The
minimum score was 0 and the maximum score was 12.
Vocabulary. Children’s productive vocabulary was assessed with the Diagnostic Test of
Bilingualism, developed by the national institute of educational testing, CITO (Verhoeven,
Narain, Extra, Konak, & Zerrouk, 1995). The test consisted of 40 pictures. All children were
tested in Dutch, the Turkish-Dutch children were also tested in Turkish. Children were
66
phoneme awareness, vocabulary, and word decoding
presented with a picture book displaying one picture per page. The children had to answer
the question ‘What is that?’ or ‘What happens here?’ A correct answer was rewarded with
one point. When a child failed five consecutive items the test was ended. The minimum
score was 0 and the maximum score was 40. The task of the commission of test affairs
(COTAN) of the Dutch Institute of Psychologists (NIP) is to evaluate psychological and
educational tests. Only test that are evaluated positively, that is, reliability and criterion
validity need to be at least sufficient (the current test was rated ‘good’) can be used validly
for individual assessment.
Word decoding. Children’s word decoding skills were assessed by the Drie-minutentest or DMT [Three-Minutes-Test] a standardised and norm-referenced reading test
(Verhoeven, 1995). The DMT is a single-word reading-decoding test consisting of three
cards with 150 words divided over 5 columns. The cards increase in difficulty of the
orthographic structure. The first card consist of VC, CV and CVC words, the second card
of monosyllabic word with consonant clusters and the third card consisted of multisyllabic
words. The score is the number of words read correctly in one minute. For this study only
Card 1 was used, because this corresponded with first-grade level. The reliability of this test
is high, Cronbach’s alpha > .90 (Moelands, Kamphuis, & Verhoeven, 2003).
Procedure
All children were tested individually in a quiet room in their school between February
and May. The DMT was administered in May. Tasks were presented in a fixed order, that
is, in increasing order of complexity. First the children were tested in Dutch, two months
later a Turkish researcher tested the Turkish-Dutch children in their mother tongue for
vocabulary and phoneme awareness.
Results
The results of the phoneme-awareness tasks will be discussed first, followed by
those of the language tests (i.e., vocabulary and word decoding). Performance of the
native-Dutch group will be compared with that of the entire immigrant group first and
then with a subsample of Turkish-Dutch children only. Subsequently, performance of
Turkish-Dutch children on both Turkish and Dutch phoneme awareness and vocabulary
will be compared.
Phoneme awareness
Dutch vs. Immigrant children. A 2 (bilingual background: Dutch vs. immigrant)
67
chapter 3
by 3 (task: phoneme segmentation vs. initial-phoneme deletion vs. final-phoneme
deletion) multivariate analyses of variance (MANOVA) was performed on the percentage
correct on the Dutch phoneme-awareness tasks. Bilingual background was a betweensubjects variable and task a within-subjects variable. The upper and middle panel of
Table 2 presents the mean scores on the phoneme-awareness tasks of native-Dutch and
immigrant children.
The analysis revealed a non-significant bilingual background by task interaction,
F < 1. The main effect of bilingual background was also not significant, F< 1, but the main
effect of task was F(2, 59) = 3.74, p = .03, partial η2 = .11, observed power = .66. Bonferroniadjusted post-hoc tests revealed that phoneme segmentation was significantly better
than initial-phoneme deletion (p< .01; Cohen’s d = .48; effect size r = .24), and better
than final-phoneme deletion (p< .05; Cohen’s d = .53; effect size r = .26); the difference
between initial- and final-phoneme deletion was not significant.
Dutch vs. Turkish-Dutch children. The same analysis was performed for the Dutch
and Turkish-Dutch participants. The lower panel of Table 2 presents the mean scores on
the phoneme-awareness tasks of the Turkish-Dutch children.
Neither the bilingual background by task interaction nor the main effect of bilingual
background reached significance (both F’s<1). The main effect of task, however, was
significant F(2, 41) = 3.50, p = .04, partial η2 = .15, observed power = .62. Bonferroniadjusted post-hoc tests revealed that phoneme-segmentation performance was
Table 2 Mean and Standard Deviation of Percentages Correct Items of the PhonemeAwareness Tasks of the Native-Dutch Children, the Entire Immigrant Group, and the TurkishDutch Children
Group
Native Dutch
Phoneme segmentation
Initial-phoneme deletion Final-phoneme deletion
Mean 90.8
SD 15.3
80.6
32.4
87.8
25.8
Mean 93.6
SD 11.3
82.6
28.3
79.8
27.6
Mean 93.9
SD 12.3
81.9
29.1
81.9
27.4
Immigrant
Turkish-Dutcha
a
The Turkish-Dutch group constitutes a subsample of the total group of immigrant children.
68
phoneme awareness, vocabulary, and word decoding
significantly better than initial-phoneme deletion (p< .05, Cohen’s d = .50; effect size
r = .24) and marginally so than final-phoneme deletion (p< .10; Cohen’s d = .43; effect size
r = .21); the difference between initial- and final-phoneme deletion was not significant.
Turkish-Dutch children. These analyses pertained to the Turkish-Dutch children only
who were tested in both Dutch and their mother tongue Turkish. A 2 (tested language:
Dutch vs. Turkish) by 3 (task: phoneme segmentation vs. initial-phoneme deletion vs. finalphoneme deletion) MANOVA was performed on the percentages correct answers of the
phoneme-awareness tasks. Tested language was a between-subjects variable and task a
within-subjects variable. Table 3 presents the mean scores of the Turkish-Dutch children.
The interaction effect between tested language and task was not significant,
F(2, 46) = 1.94, p = .15, partial η2 = .08, observed power = .38. The main effect of tested
language was significant, F(1, 23) = 11.39, p = .003, partial η2 = .33, observed power = .90.
Performance on the Dutch phoneme-awareness tasks (84.5%) was better than on the
Turkish tasks (68.4%). The main effect of task was also significant, F(2, 46) = 12.51,
p = .001, partial η2 = .35, observed power = .99. Bonferroni-adjusted post-hoc tests
revealed that performance on phoneme segmentation (87.5%) was significantly better
than on initial-phoneme deletion (75.0%; Cohen’s d = .54; effect size r = .26), which in
turn was significantly better than on final-phoneme deletion (66.8%; Cohen’s d = .29;
effect size r = .14), all p’s < .05. Caveat, inspection of the means revealed that this effect
was particular visible in the Turkish task, the Dutch task showed the pattern observed in
the bilingual immigrant group and the monolingual Dutch group.
Table 3 Mean and Standard Deviation of Percentages Correct of the Turkish and Dutch
Phoneme-Awareness Tasks of the Turkish-Dutch Children (n =24)
Language
Dutch
Phoneme segmentation
Initial-phoneme deletion Final-phoneme deletion
Mean 93.8
SD 12.7
80.6
31.4
79.1
29.3
Mean 81.3
SD 20.2
69.4
33.9
54.5
39.8
Turkish
Vocabulary and word decoding
Native-Dutch children scored significantly higher on the Dutch vocabulary test
than the entire group of immigrant children (t(47) = 3.68, p< .001; Cohen’s d = 1.15;
69
chapter 3
effect size r = .50), and higher than the Turkish-Dutch children (t(34) = 4.58, p< .001;
Cohen’s d = 1.53; effect size r = .61). A paired-samples t test revealed that the TurkishDutch children performed better on the Dutch vocabulary test than on the Turkish one,
t(17) = 2.41, p = .03; Cohen’s d = 0.67; effect size r = .32. Native-Dutch children did not
perform better on the word-decoding test than the entire group of immigrant children
or the Turkish-Dutch children (both t’s < 1).Table 4 presents the descriptive statistics of
each group.
Table 4 Mean and Standard Deviation on Vocabulary Tests in Dutch and in Turkish (for
Turkish-Dutch Children only) and Word Decoding in Dutch
Participant Group
Native Dutch
Vocabulary
Dutch
Turkish
Word decoding
Dutch
Mean 27.3
SD 5.3
34.92
18.9
Mean 21.1
SD 5.5
36.76
17.82
Immigrant
Turkish-Dutcha
Mean 19.5
SD 4.9
a
15.5
6.8
40.7
20.29
The Turkish-Dutch group constitutes a subsample of the total group of immigrant children.
Discussion
The central issue of this study was whether bilingually raised children in the
Netherlands, who receive literacy instruction in their second language only, still show
an advantage on Dutch phoneme-awareness tasks compared to monolingual Dutchspeaking children. The results are clear: This is not the case. Neither in the entire group
of bilingual children nor in the subsample of Turkish-Dutch children was phonemeawareness better than that of monolingual children. Note that performance on the
phoneme-awareness tasks of the bilingual children was not worse either.
The three phoneme-awareness tasks revealed differences in difficulty. For the
Dutch and the entire immigrant group, phoneme segmentation was easier than initial
and final phoneme deletion. This means that the children found it easier to count the
number of phonemes in the pseudowords than deleting an initial or a final phoneme by
pronouncing the remaining part of the pseudoword.
70
phoneme awareness, vocabulary, and word decoding
In the Turkish-Dutch group a somewhat different pattern was obtained. Like the
entire immigrant group and the monolingual Dutch group, they were better on
phoneme segmentation than on the two phoneme-deletion tasks. However, with
respect to the phoneme-deletion tasks, they showed better performance on initialphoneme deletion than on final-phoneme deletion. This is interesting, because this
effect is the opposite from what was found in the Durgunoglu and Oney (1999) study
with monolingual Turkish-speaking children, but coincides with work by de Graaff,
Hasselman, Bosman, and Verhoeven (2008) who studied monolingual Dutch-speaking
children. When comparing phoneme-awareness performance of the Turkish-Dutch in
Dutch and Turkish it appeared that they performed better on the Dutch administration
of the phoneme-awareness tasks than on the Turkish ones; mean correct was 84.5%
and 68.4%, respectively. It seems that our bilingual Turkish-Dutch children behaved
more like monolingual Dutch-speaking children than monolingual Turkish-speaking
children. This finding requires some thought.
The Dutch phoneme-awareness, administered two months prior to the Turkish
tasks, did not cause a learning effect or a natural maturation effect due to on-going
instruction. This is all the more remarkable, because the stimuli were exactly the same.
The only difference was that the second administration of the tasks was done in Turkish,
by a Turkish native speaker. A likely explanation is provided by earlier findings in the
literature about the development of phoneme awareness and literacy instruction. There
is ample evidence showing that phoneme awareness develops in response to reading
instruction. This fact has been established in a large variety of languages (see for Czech,
Caravolas & Bruck, 1993; for Portuguese, Cardoso-Martins, 1995; for Italian, Cossu,
Shankweiler, Liberman, Katz, & Tola, 1988; for Turkish, Durgunoglu & Oney, 1999;
for Dutch, Patel, Snowling, & de Jong, 2004; for English, Stahl & Murray, 1994; for
German, Wimmer, Landerl, Linortner, & Hummer, 1991). Thus, phoneme awareness
in Dutch of the Turkish-Dutch children may be better than in Turkish, simply because
they only learned to read and spell in Dutch, no instruction in the reading and spelling
of the Turkish language was provided (cf., van der Leij, Bekebrede, & Kotterink, 2010).
Another indication of the Turkish-Dutch superior proficiency in Dutch than in
Turkish was provided by the vocabulary tests. Turkish-Dutch children also obtained a
higher score on the Dutch than on the Turkish vocabulary test, albeit the mean score
on the Dutch vocabulary test was significantly lower than those of the monolingual
Dutch children from similar socio-economic backgrounds. Together with their superior
phoneme awareness in Dutch, these findings suggest that the Turkish-Dutch children’s
proficiency in Dutch is growing and may in the long run surpass their proficiency in
Turkish. To what extent this is due to a lack of formal instruction in their native language
is unclear, but this finding extends those of Leseman (2000), who found that Dutch
71
chapter 3
vocabulary of Turkish children increases rapidly from the age of four onwards, and is
corroborated by work of Secada (1991).
Before finishing this paper, we would like to discuss an important limitation with
respect to generalization. The results presented here are based on a relatively small
sample, and the comparison between first- and second language performance was
limited to Turkish-Dutch children. The question remains whether similar effects are
to be found in children from other ethnic groups. Our second largest subsample of
immigrant children was from Moroccan descent. It is however, not possible to compare
their first and second language performance on phoneme awareness or vocabulary,
because these children speak different languages, namely, Arabic and Berber. The
majority of Dutch Moroccan children speak Berber. Berber is a spoken language only,
which makes it hard to develop proper stimuli for conducting an experiment that
assesses phoneme-awareness.
Nevertheless, our study is the first attempt to test whether bilingually raised
children who do not receive language instruction in their first language still have an
advantage over children who are monolingual with respect to a metalinguistic skill such
as phoneme awareness. The general conclusion of this study is that these children do
not profit or suffer from knowing two languages when it concerns the acquisition of
phoneme awareness. Not only was their performance on phoneme-awareness similar
to that of bilingual children, their word-decoding performance was too. Although the
Dutch vocabulary of the Turkish-Dutch subsample was poorer than those of the native
Dutch speakers, their language proficiency in their adopted language was better than in
their native language Turkish.
72
phoneme awareness, vocabulary, and word decoding
References
Bhatia, T. K. (2006). Bilingualism and second language learning. In K. Brown (Ed.),
Encyclopedia of languages and linguistics (pp. 16-22). Amsterdam, the Netherlands:
Elsevier. doi: 10.1016/B0-08-044854-2/00620-9
Bialystok, E. (2005). Consequences of bilingualism for cognitive development. In J. F.
Kroll & A. M. B de Groot (Eds.), Handbook of bilingualism: Psycholinguistic approaches
(pp. 417-432). New York, NY: Oxford University Press.
Bialystok, E. (2007). Language acquisition and bilingualism: Consequences for
a multilingual society. Applied Psycholinguistics, 28, 393-397. doi: 10.1017/
S0142716407070208
Bialystok, E., Luk, G., & Kwan, E. (2005). Bilingualism, biliteracy, and learning to read:
Interactions among languages and writing systems. Scientific Studies of Reading, 9,
43-61. doi: 10.1207/s1532799xssr0901_4
Blachman, B. (2000). Phonological awareness. In M. Kamil, P. Mosenthal, P. D. Pearson,
& R. Barr (Eds.), Handbook of reading research (Vol. 3; pp. 483–502). Mahwah, NJ:
Erlbaum.
Bosman, A. M. T., de Graaff, S., & Gijsel, M. A. R. (2006). Double Dutch: The Dutch
spelling system and learning to spell in Dutch. In R. M. Joshi & P. G. Aron (Eds.),
Handbook of orthography and literacy (pp. 135-150). Mahwah, NJ: Lawrence Erlbaum
Associates.
Bosman, A. M. T., Vonk, W., & van Zwam, M. (2006). Spelling consistency affects
reading in students with and without dyslexia. Annals of Dyslexia, 56, 271-300. doi:
10.1007/s11881-006-0012-4
Bradley, L., & Bryant, P. (1983). Categorizing sounds and learning to read: A causal
connection. Nature, 301, 419-421. doi: 10.1038/301419a0
Caravolas, M., & Bruck, M. (1993). The effect of oral and written language input on
children’s phonological awareness: A cross-linguistic study. Journal of Experimental
Child Psychology, 55, 1-30. doi: 10.1006/jecp.1993.1001
Cardosa-Martins, C. (1995). Sensitivity to rhymes, syllables, and phonemes in
literacy acquisition in Portuguese. Reading Research Quarterly, 30, 808-828. doi:
10.2307/748199
Centraal Bureau voor de Statistiek (2008). Jaarrapport integratie 2008 [Year report
integration 2008]. Den Haag, the Netherlands: Centraal Bureau voor de Statistiek.
Chen, X., Xu, F., Nguyen, T-K., Hong, G., & Wang, Y. (2010). Effects of cross-language
transfer on first-language phonological awareness and literacy skills in Chinese
children receiving English instruction. Journal of Educational Psychology, 102, 712728. doi: 10.1037/a0018802
73
chapter 3
Cossu, G., Shankweiler, D., Liberman, I. S., Katz, L., & Tola, G. (1988). Awareness of
phonological segments and reading ability in Italian children. Applied Psycholinguistics,
9, 1-16. doi: 10.1017/S0142716400000424
Cummins, J. (1981). The role of primary language development in promoting educational
success for language minority students. In California State Department of Education
(Ed.), Schooling and minority students: A theoretical framework (pp. 3-49). Los Angeles,
CA: National Dissemination and Assessment Center.
D’Angiulli, A., Siegel, L. S., & Serra, E. (2001). The development of reading in Italian
and English in bilingual children. Applied Psycholinguistics, 22, 479-507. doi: 10.1017/
S0142716401004015
de Graaff, S., Hasselman, F., Bosman, A. M. T., & Verhoeven, L. (2008). Cognitive and
linguistic constraints on phoneme isolation in Dutch kindergartners. Learning and
Instruction, 18, 391-403. doi: 10.1016/j.learninstruc.2007.08.001
Durgunoglu, A. Y. (2006). How language characteristics influence Turkish literacy
development. In R. M. Joshi & P. G. Aaron (Eds.), Handbook of orthography and
literacy (pp. 219-229). Mahwah, NJ: Erlbaum.
Durgunoglu, A. Y., & Hancin, B. J. (1992). An overview of cross-language transfer in
bilingual reading. In R. J. Harris (Ed.), Cognitive processing in bilinguals (pp. 391411). Amsterdam, the Netherlands: Elsevier Science Publishers. doi: 10.1016/S01664115(08)61507-1
Durgunoglu, A. Y., & Oney, B. (1999). A cross-linguistic comparison of phonological
awareness and word recognition. Reading and Writing: An Interdisciplinary Journal,
11, 281-299. doi: 10.1080/09669760220114836
Duursma, E., Romero-Contreras, S., Szuber, A., Proctor, P., & Snow, C. E. (2007).
The role of home literacy and language environment on bilinguals’ English and
Spanish vocabulary development. Applied Psycholinguistics, 28, 171-190. doi: 10.1017/
S0142716406070093
Hoff, E. (2006). How social contexts support and shape language development.
Developmental Review, 26, 55-88. doi: 10.1016/j.dr.2005.11.002
Huttenlocher, J., Haight, W., Bryk, A., Seltzer, M., & Lyons, T. (1991). Early vocabulary
growth: Relation to language input and gender. Developmental Psychology, 27, 236248. doi: 10.1037//0012-1649.27.2.236
Jiménez, R. T., Garcia, E. G., & Pearson, P. D. (1995). Three children, two languages,
and strategic reading: Case studies in bilingual/monolingual reading. American
Educational Research Journal, 32, 67-97. doi: 10.3102/00028312032001067
Kim, Y-S. (2009). Crosslinguistic influence on phonological awareness for Korean–
English bilingual children. Reading & Writing, 22, 843-861. doi: 10.1007/s11145-0089132-z
74
phoneme awareness, vocabulary, and word decoding
Leikin, M., Schwartz, M., & Share, D. (2010). General and specific benefits of bi-literate
bilingualism: A Russian–Hebrew study of beginning literacy. Reading & Writing, 23,
269-292. doi: 10.1007/s11145-009-9210-x
Leseman, P. P. M. (2000). Bilingual vocabulary development of Turkish preschoolers
in the Netherlands. Journal of Multilingual and Multicultural Development, 21, 93-112.
doi: 10.1080/01434630008666396
Liberman, I. Y., & Liberman, A. M. (1990). Whole language vs. code emphasis:
Underlying assumptions and their implications for reading instruction. Annals of
Dyslexia, 40, 51-75. doi: 10.1007/BF02648140
Lundberg, I., Frost, J., & Peterson, O. (1988). Effects of an extensive program for
stimulating phonological awareness in preschool children. Reading Research
Quarterly, 23, 263-284. doi: 10.1598/PRQ.23.3.1
Moelands, F., Kamphuis, F., & Verhoeven, L. (2003). Verantwoording Drie-Minuten-Toets
[Justification of the Three-Minutes-Test]. Arnhem, the Netherlands: Cito.
Morais, J. (1991). Phonological awareness: A bridge between language & literacy. In D.
Sawyer & B. Fox (Eds.), Phonological awareness in reading: The evolution of current
perspectives (pp. 31-71). New York, NY: Springer.
Oller, D. K. (2005). The distributed characteristic in bilingual learning. In J. Cohen,
K. T. McAlister, K. Rolstad, & J. MacSwan (Eds.), Proceedings of the 4th International
Symposium on Bilingualism (pp. 1744-1749). Somerville, MA: Cascadilla Press.
Oller, D. K., & Eilers, R. E. (2002). Language and literacy in bilingual children. Clevedon,
England: Multilingual Matters.
Patel, T. K., Snowling, M. J., & de Jong, P. F. (2004). A cross-linguistic comparison of
children learning to read in English and Dutch. Journal of Educational Psychology, 96,
785-797. doi: 10.1037/0022-0663.96.4.785
Raven, J. C. (1958). Standard Progressive Matrices. London, United Kingdom: H. K.
Lewis.
Scheele, A. F., Leseman, P. P. M., & Mayo, A. Y. (2010). The home language environment
of mono- and bilingual children and their language proficiency. Applied
Psycholinguistics, 31, 117-140. doi: 10.1017/S0142716409990191
Secada, W. G. (1991). Degree of bilingualism and arithmetic problem solving in Hispanic
first graders. The Elementary School Journal, 92, 213-231. doi: 10.1086/461689
Stahl, S., & Murray, B. (1994). Defining phonological awareness and its relationship
to early reading. Journal of Educational Psychology, 86, 221-234. doi: 10.1037/00220663.86.2.221
Stuart, M. (1999). Getting ready for reading: Early phoneme awareness and phonics
teaching improves reading and spelling in inner-city second language learners.
British Journal of Educational Psychology, 69, 587-605. doi:10.1348/000709999157914
75
Stuart, M. (2004). Getting ready for reading: A follow-up study of inner city second
language learners at the end of Key Stage I. British Journal of Educational Psychology,
74, 15-36. doi: 10.1348/000709904322848806
Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (1994). Longitudinal studies of
phonological processing and reading. Journal of Learning Disabilities,27, 276-286.
doi: 10.1177/002221949402700503
van der Leij, A., Bekebrede, J., & Kotterink, M. (2010). Acquiring reading and vocabulary
in Dutch and English: The effect of concurrent instruction. Reading and Writing: An
Interdisciplinary Journal, 23, 415-434. doi: 10.1007/s11145-009-9207-5
Verhoeven, L. (1995). Drie Minuten Toets. Handleiding [Three-Minute Test. Manual].
Arnhem, the Netherlands: Cito.
Verhoeven, L. (2000). Components in early second language reading and spelling.
Scientific Studies of Reading, 4, 313-330. doi: 10.1207/S1532799XSSR0404_4
Verhoeven, L., Narain, G., Extra, G., Konak, O. A., & Zerrouk, R. (1995). Toets Tweetaligheid
Handleiding [Test for Bilingualism Manual].Arnhem, the Netherlands: Cito.
Wimmer, H., Landerl, K., Linortner, R., & Hummer, P. (1991). The relationship of
phonemic awareness to reading acquisition: More consequence than precondition,
but still important. Cognition, 40, 218-249. doi: 10.1016/0010-0277(91)90026-Z
Yip, V., & Matthews, S. (2007). Relative clauses in Cantonese-English bilingual children:
Typological challenges and processing motivations. Studies in Second Language
Acquisition, 29, 277-300. doi: 10.1017/S0272263107070143
Ziegler, J. C., Bertrand, D., Tóth, D., Csépe, V., Reis, A., Faísca, L. et al. (2010).
Orthographic depth and its impact on universal predictors of reading: A cross-language
investigation. Psychological Science, 21, 551-559. doi: 10.1177/0956797610363406
76
Chapter 4
limited language proficiency is not
necessarily detrimental for the
wellbeing of children from low-income
and/or immigrant backgrounds
Submitted for publication
77
chapter 4
Abstract
This study was designed to investigate whether a limited language proficiency in
children from low-income families, consisting of immigrant-Dutch and native-Dutch
children in the Netherlands affects their social-emotional development over the course
of four years. Socio-emotional development was measured by teacher ratings and
language proficiency was measured by the Reynell test for language comprehension.
The children were followed from preschool to first grade. The general conclusion was
that children with a more limited language proficiency, immigrant-Dutch children do
not develop more socio-emotional or behaviour problems than native-Dutch children,
based on the teachers rating scales.
78
language proficiency and social-emotional development
To learn a second language, just like learning a first language takes time and
depends on several factors like motivation, attitude, personal characteristics, or age
(Goorhuis-Brouwer & Schaerlaekens, 2000). A number of studies have revealed that
language development and social-emotional development mutually affect each other
and language difficulties may affect children’s social-emotional development negatively
(Coplan & Armer, 2005; Stanton-Chapman, Justice, Skibbe, & Grant, 2007). Social
skills are needed for the development in many areas of life, such as school, home, and
peer contact. The interest in social-emotional development of children has increased
over the last decade. Pianta, Steinberg, and Rollings (1995) argue that social-emotional
development is even more important for school success in preschool children and
kindergartens than academic skills are (see also Raver, 2002).
The present study focused on the relationship between socio-emotional development
and language proficiency in children who are native Dutch speaking or native speaking
in their home language and learning Dutch as second language. The immigrantDutch children in these usually learn to speak the language of their parents at home.
Substantial and prolonged exposure to Dutch as second language does not start before
they enter kindergarten. Like Cummins (1981) uses the terms BICS (Basic Interpersonal
Communicative Skills) and CALP (Cognitive Academic Language Proficiency) in order
to differentiate between language usage at home and at school. This usually results
in a large group of children from minority backgrounds with insufficient knowledge
of the Dutch language, which means that they cannot optimally benefit from formal
education in reading, spelling, and mathematics (Elzer, 2005; van Elten, 2003). A large
national-cohort study revealed that children from low-income minority families begin
in primary school with limited proficiency of the Dutch language of about one standard
deviation relative to the average of middle to high income Dutch native children (Tesser
& Iedema, 2001) and Turkish and Moroccan children repeat grades twice as often as
Dutch children (Aarts, de Ruiter, & Verhoeven, 1996).
The relationship between language development and socio-emotional development
appears to be bi-directional, that is, they affect each other reciprocally. A language delay
may restrict social interactions and as a consequence this may lead to peer rejection
and frustrations (Gertner, Rice, & Hadley, 1994). Conversely, extraverted, highly social
young children are found to talk more, which in turn enhances language development.
Shy, or withdrawn, children are often quiet, and found to have a smaller vocabulary
and produce less words per utterance than non-shy children (Crozier & Perkins, 2002;
Rice, Sell, & Hadley, 1991).Thus, a language delay may influence communication and in
turn affect social-emotional and/or behavioural development (Redmond & Rice, 1998),
79
chapter 4
and at the same time social-emotional development is essential for a healthy language
development (Goorhuis-Brouwer & Schaerlaekens, 2000).
Knowledge about the relationship between language and social-emotional
development mainly comes from research conducted with children with speech and
language impairments. These children tend to develop externalizing problems,
such as, attention problems, conduct disorder, and delinquency (Baker & Cantwell,
1987; Cantwell & Baker, 1991; Moffitt & Silva, 1988; Schonfeld, Shaffer, O’Conner,
& Portnoy, 1988), as well as internalizing problems, like anxiety disorder and socialemotional adjustment problems (Irwin, Carter, & Briggs-Gowan, 2002). Longitudinal
studies show that children who had language problems at 5 years of age, showed more
behavioural, emotional, and social problems at age 12 (Beitchman, Wilson, Brownlie,
Walters, & Lancee, 1996). In a follow-up study, Beitchman et al. (2001) showed that even
at the age of 19 these children had more psychiatric problems than a comparable group.
Children identified as having both speech and language problems in preschool were
most vulnerable in developing problem behaviour (Benasich, Curtiss, & Tallal, 1993).
The relationship between language proficiency and social-emotional problems has
been studied thoroughly in clinical samples (i.e., a speech or language impairment)
with mainly children from middle to high socio-economic backgrounds who are mostly
from Caucasian origin (Beitchman et al, 1996; Benasich et al., 1993; Stanton-Chapman
et al., 2007). Robust knowledge about the relationship between language proficiency
and social-emotional problems in children who merely lag behind in the development of
the dominant language of the society they live in is, however, lacking. Although children
from low-income and minority backgrounds are at risk for developing social-emotional
or behaviour problems (Leseman, 2002; Raver, 2002), there is no evidence that a limited
language proficiency in the school language in this group is related to the development
of social-emotional problems per se.
The present study focused on the relationship between socio/emotional development
and language proficiency involving low-income immigrant-Dutch children in the
Netherlands who were compared to low-income native-Dutch children living in the
same poor neighbourhoods, over the course of four years. The children were followed
from preschool to first grade.
Method
Participants
This study involved four years of data collection. At the start of the study (Year 1) all
children attended preschool. Their mean age was 40 months (SD = 10). The next year all
80
language proficiency and social-emotional development
children attended the first year of kindergarten, followed by the second of kindergarten,
and Grade 1. The numbers of children and their teachers who participated in this study
are presented in Table 1. The information on children’s socio-emotional development
was provided by the teachers and as such was dependent upon the teachers being willing
to return questionnaires (see below).
All children had participated in a special, half-day pre-school program for socioeconomically disadvantaged children when they were three years old. The program
lasted about a year. The selection to participate in this program was based on the
observations of teachers who were experienced with language development of young
children. Children who were clearly lagging behind in their development of the Dutch
language were eligible for participation. All immigrant-Dutch children were born in the
Netherlands, but 100% of the fathers and 96% of the mothers were born in the country
of origin. All children lived in poor inner-city neighbourhoods with a majority of lowincome and immigrant families. Background information was gathered by a parent
questionnaire.
Table 1 Number of Teachers who filled in the Questionnaires in each Year and the Number of
Children Tested on the Reynell Language test
Year
N of questionnaires returned
1
2
3
4
native Dutch
immigrant Dutch
N of Reynell tests administered
native Dutch
immigrant Dutch
30
86
21
45
21
70
19
60
30
85
28
72
27
85
33
94
Materials and procedure
To measure socio-emotional development three test were used over 4 years: Child
Behaviour Check List (CBCL), Behaviour Check List for Toddlers and Children in
Kindergarten (GVPK) and the Teacher Report form (TRF). To measure language
development the Reynell test for language comprehension was used in all 4 years.
The original CBCL questionnaire was adapted in the first year to the purpose of this
study, in order to measure social-emotional and behaviour problems appropriate for the
age of 3 years. Therefore, a selection of 10 items of the Behaviour Check List for Toddlers
81
chapter 4
and Children in Kindergarten, was used for this study. These were not included in the
CBCL and relevant for toddlers and children in kindergarten, and for that reason added
to the CBCL. Examples are: ‘is difficult to comfort’, ‘is avoided by other children’, ‘teases
other children’ and ‘is difficult to understand’. This combination of the CBCL and the
GVPK was used in the first year, the CBCL was used in the second and third year, and
the TRF was used in the last year when the children were in first grade.
Child Behaviour Check List (Dutch adaptation, Verhulst, van der Ende, & Koot, 1996).
To measure problem behaviour the CBCL was be used. The CBCL consists of 112 items
that measure socio-emotional and behaviour problems in children from 4-16 years. It
takes about 15 minutes to fill in the CBCL. In this case teachers rated their child with
respect to how true each item is now or within the past 6 months using the following
scale: 0 = not true (as far as you know); 1 = somewhat or sometimes true; 2 = very
true or often true. The CBCL provides scores for Internalizing, Externalizing, and Total
Problems and for the following sub-scales: Aggressive Behaviour, Anxious/Depressed,
Attention Problems, Rule-Breaking Behaviour, Social Problems, Somatic Complaints,
Thought Problems, and Withdrawn/Depressed. The answers to the questions must
concern actual behaviour within the last two months. Each child received a general score
on each subscale; it was also checked whether the child scored within the ‘clinical range’
on each subscale. This means that a child shows more problem behaviour in that subscale than expected given his or her age. The raw score can be translated into a t-score.
A t-score above 70 means a score at or above the 98th percentage or ‘clinical range’.
The commission of test affairs (COTAN) of the Dutch Institute of Psychologists (NIP)
evaluated its reliability as sufficient and its validity as ‘good’.
Behaviour Check List for Toddlers and Children in Kindergarten (Gedragsvragenlijst
voor Peuters en Kleuters, Goossens, Bruinsma, Dekker, & de Ruyter, 2000). The GPVK
is, like the CBCL, a questionnaire that is to be completed by the teacher, used to measure
problem behaviour in children. The questionnaire is put together for toddlers and
children in kindergarten. The total tests consist of 40 items. It takes about 10 minutes to
fill out the GVPK. As described above, because many items were similar to items of the
CBCL a selection of 10 items was used for this study.
Teacher Report form (Dutch adaptation, Achenbach, 1991). The TRF is designed to
obtain teachers’ reports of children’s socio-emotional and behaviour problems. The TRF
consist of 112 items with the same rating scale and scores as the CBCL. Teachers of firstgrade completed the TRF. The commission of test affairs (COTAN) of the Dutch Institute
of Psychologists (NIP) evaluated its reliability as ‘good’ and its validity as ‘sufficient’.
82
language proficiency and social-emotional development
Reynell test for language comprehension (van Eldik, Schlichting, Lutje Spielberg, van
der Meulen, & van der Meulen, 1997). Language comprehension was measured with
the Reynell test for children between the age of one year and two months and six years
and three months. The test consists of 43 items divided in 12 sections that increase in
difficulty. Within each section, the number of objects is pictured in a fixed order. The
questions are about these objects; children could indicate their answer to the questions
by pointing at one of the depicted objects. The commission of test affairs (COTAN) of
the Dutch Institute of Psychologists (NIP) evaluated the reliability and construct validity
of the Reynell test as ‘good’.
Procedure
Data were gathered between February and June in the same month each year.
The questionnaire was completed by the teacher, during this period, and a researcher
administered the language test individually in a quiet room at school.
Results
Given the small sample size no structural equation modelling (SEM) could be
conducted. Moreover, a repeated measure analysis with year as dependent variable was
not possible either, because of the limited number of children who participated in all four
years (10 Dutch and 27 immigrant children). We therefore conducted analyses for each
year separately. The first analysis concerned the difference in language development of
the native-Dutch and the children from an immigrant family, followed by the analyses
of social-emotional development, and finalized by a correlational analysis between
language and social-emotional development.
Language development. For each year a separate analysis of variance (ANOVA) was
conducted with ethnic background (Dutch vs. Immigrant) as a between-subjects factor
and language score as the dependent measure. These analyses revealed significant
differences between the groups during all years: Year 1: F(1, 114) = 69.58, p = .0001; Year 2:
F(1, 99) = 23.01, p = .0001; Year 3: F(1, 111) = 21.98, p = .0001; Year 4: F(1, 126) = 33.65,
p = .0001. Dutch children had a higher score on the language comprehension test
administered during each year than the immigrant children (see Table 2). In the last year
(Grade 1) differences between the two ethnic groups were still visible.
Social-emotional differences. For each year a 2 (ethnic background: Dutch vs.
Immigrant) by 4 (socio-emotional scale: withdrawn vs. somatic vs. aggressive vs.
83
chapter 4
anxious/depressed) MANOVA with ethnic background as between-subjects variable
and socio-emotional scale as within-subjects variable. Descriptive statistics (means
and standard deviations) are presented in Table 2 for the different subscales of the
behaviour checklist.
Year 1. In the first year, the main effect of ethnic background was not significant,
F< 1. The main effect of the socio-emotional scales was significant. Pillai’s trace = .51,
F(3, 112) = 39.29, p =.0001, partial η2 = .51. All children showed more withdrawn
and aggressive behaviour than anxious/depressed behaviour or somatic complaints.
The results showed no significant scale by ethnicity interaction, Pillai’s trace = .06,
F(3, 112) = 2.44, p = .07, partial η2 = .06.
Year 2. In the second year, the main effect of ethnic background was again not
significant, F< 1. The main effect of the socio-emotional scales was significant Pillai’s
trace = .58 , F(3, 62) = 27.92, p = .0001, partial η2 = .57. All children showed more
withdrawn and aggressive behaviour than anxious/depressed behaviour or somatic
complaints. The results showed no significant scale by task ethnicity interaction,
Pillai’s trace = .04 , F(3, 62) = .80, p = .50, partial η2 = .04.
Table 2 Number, Mean, and Standard Deviation on the CBCL-Scales and Language Tests
CBCL-scale
Withdrawn Somatic
AnxiousDepressed
Aggressive
Language
Reynell
Mean
SD
86 Mean
SD
.35
.28
.38
.35
.03
.08
.03
.10
.21
.26
.20
.26
.44
.47
.27
.27
46.8
17.5
21.4
13.4
Year 2
Dutch
21
Immigrant
45
Mean
SD
Mean
SD
.22
.23
.30
.25
.02
.06
.02
.06
.21
.22
.21
.21
.32
.38
.29
.31
59.8
12.4
44.1
15.4
Year 3
Dutch
21
Immigrant
70
Mean
SD
Mean
SD
.19
.20
.31
.30
.04
.08
.08
.15
.21
.24
.19
.20
.47
.56
.23
.22
75.1
5.9
67.9
7.3
Year 4
Dutch
19
Mean
SD
60 Mean
SD
.25
.19
.20
.26
.09
.19
.07
.16
.30
.27
.12
.15
.48
.52
.19
.23
77.9
4.8
70.0
7.3
Group
n
Year 1
Dutch
30
Immigrant
Immigrant
84
language proficiency and social-emotional development
Year 3. In the third year, the main effect of ethnic background was again not
significant, F< 1. The main effect of the socio-emotional scales was significant, Pillai’s
trace = .45 , F(3, 87) = 23.86, p =.0001, partial η2 = .45. All children showed more
withdrawn and aggressive behaviour than anxious/depressed behaviour or somatic
complaints. The results showed a significant scale by ethnicity interaction, Pillai’s
trace = .14 , F(3, 87) = 4.81, p = .004, partial η2 = .14. Immigrant children showed
significantly more withdrawn behaviour, t(90) = 30.18, p = .0001, whereas native
Dutch children showed significantly more aggressive behaviour, t(90) = 22.68,
p = .0001.
Year 4. On the last year, the main effect of ethnic background was significant,
F(1, 77) = 9.74, p =.003, partial η2 = .11, as well as the main effect of the socio-emotional,
scales Pillai’s trace = .34 , F(3, 75) = 12.93, p =.0001, partial η2 = .34. The results also
showed a significant scale by ethnicity interaction, Pillai’s trace = .20 , F(3, 75) = 6.25,
p = .001, partial η2 = .20. Dutch children were found to significantly show more anxious
/depressed behaviour, t(78) = 26.23, p = .0001,and significantly more aggressive
behaviour than the immigrant children, t(78) = 20.88, p = .0001.
Relationship between socio-emotional behaviour and language comprehension.
Correlations among the four scales of the scales of the behaviour checklist and the
language test (see Table 3) for each year separately, revealed no significant correlations
between the total score on the Reynell and the score on the subscales of the behaviour
checklist.
Table 3 Correlations between the Language-Comprehension Test (Reynell) and the CBCLScales for All Children in Years 1 to 4
Withdrawn
CBCL-scale
Somatic
Anxious
depressed
Aggressive
Year 1
-.19
-.12
-.02
.10
Year 2
-.30
.07
.19
.03
Year 3
-.48
-.30
-.08
.20
Year 4
-.17
.04
.02
.15
Language comprehension
Note. Bonferroni corrected, p value is .003.
85
chapter 4
Discussion
The present study was designed to investigate the hypothesis that a limited language
proficiency of low-income native-Dutch and immigrant Dutch children was related to socialemotional and/or behaviour problems (as reported by the teacher). The results obtained
with the language-comprehension task showed that immigrant children, as expected, have
a more limited proficiency in Dutch compared to low-income Dutch children. This limited
proficiency in Dutch was, however, not related to socio-emotional problems.
According to their teachers, immigrant-Dutch children did not develop more
socio-emotional or behaviour problems than native-Dutch children from similar SESbackgrounds with a higher command of the Dutch language. The only exception was
in year 3, where immigrant-Dutch children seemed to display stronger withdrawn
behaviour than their native-Dutch peers, but the native-Dutch children displayed
more aggressive behaviour. Note also, that in year 4 native-Dutch children showed
more anxious depressed and aggressive behaviour than immigrant-Dutch children.
The correlational analysis substantiates the conclusion that the relationship between
language development and socio-emotional development is limited or even absent in
the sample studied here; none of the correlations (see Table 1) between the scores on the
language-comprehension test were related to the scores on the socio-emotional scales.
Our findings do not support studies by others, who found that children with a language
delay show more withdrawn behaviour (e.g., Fujiki, Brinton, Isaacson, & Summer, 2001;
Glover-Gagnon & Nagle, 2004; Irwin et al., 2002; Noterdaeme & Amorosa, 1999) and
studies that reveal that preschool children with a specific language impairment (SLI)
show more social problems (more internalizing behavioural problems), as reported by
their parents than from children without a language impairment (Stanton-Chapman
et al., 2007). Note that these children were from middle SES families. Our findings
suggest that the children from the present study do not have a language impairment
or a delay. It seems that their language proficiency is limited rather than delayed due to
circumstances, such as a relatively poor language input.
A critical note with respect to the current study is the fact that teachers’ ratings
were used to assess socio-emotional behaviour. Studies on the use of rating scales
have shown that different factors may influence the reliability, especially the inter-rater
reliability, of rating scales. Caution should be exercised concerning the interpretation
and use of different informant sources (Daradkeh, 1993; Embregts, 2000; Konold &
Pianta, 2007). Judgments about the intensity and frequency of behaviour is affected
by situational influences on children’s behaviour, different conceptions of abnormal
behaviour, informant characteristics, opportunities for observing a child’s behaviour,
and the informants effects on the child’s behaviour and their knowledge of the child.
86
language proficiency and social-emotional development
Riddle and Rapoport (1976), however, showed teacher ratings to be stable over a
period of two years, despite the fact that different teachers evaluated the child. Our study
ran over a period of four years. Each year a different teacher assessed the socio-emotional
behaviour of the children, and children also changed schools. Nevertheless, the patterns
obtained here are highly stable and yield promising results for low-income children with
a limited command of the language. The conclusion that a limited language proficiency
is not necessarily detrimental for the wellbeing of children from low-income and/or
immigrant backgrounds appears justified.
87
chapter 4
References
Aarts, R., de Ruiter, J. J., & Verhoeven, L. (1996). Eigen-taalvaardigheid en schoolsucces
in Turkse en Marokkaanse leerlingen [First-language skills and school success in
Turkish and Moroccan children]. Pedagogische Studiën, 73, 275-290.
Achenbach, T. M. (1991). Manual of the Teacher’s Report Form and 1991 profile. Burlington,
VT: University of Vermont, Department of Psychiatry.
Baker, I., & Cantwell, D. P. (1987). A prospective psychiatric follow-up of children with
speech/language disorders. Journal of the American Academy of Child and Adolescent
Psychiatry, 26, 193-196. doi: 10.1097/00004583-198707000-00015
Beitchman, J. H., Wilson, B., Brownlie, E. B., Walters, H., & Lancee, W. (1996). Longterm consistency in speech/language profiles: I. Developmental and academic
outcomes. Journal of the American Academy of Child and Adolescent Psychiatry, 35,
804-814. doi: 10.1097/00004583-199606000-00021
Beitchman, J. H., Wilson, B., Johnson, C. J., Atkinson, L., Young, A., Adlaf, E. et al.
(2001). Fourteen-year follow-up of speech/language-impaired and control children:
Psychiatric outcome. Journal of the American Academy of Child and Adolescent
Psychiatry, 40, 75-83. doi: 10.1097/00004583-200101000-00019
Benasich, A., Curtiss, S., & Tallal, P. (1993). Language, learning, and behavioral
disturbances in childhood: A longitudinal perspective. Journal of the American
Academy of Child and Adolescent Psychiatry, 32, 585-594. doi: 10.1097/00004583199305000-00015
Cantwell, D. P., & Baker, L. (1991). Psychiatric and developmental disorders in children with
communication disorders. Washington, DC: American Psychiatric Press.
Coplan, R. J., & Armer, M. (2005). Talking yourself out of being shy: Shyness, expressive
vocabulary, and socioemotional adjustment in preschool. Merrill-Palmer Quarterly,
51, 20-41. doi: 10.1353/mpq.2005.0004
Crozier, W. R., & Perkins, P. (2002). Shyness as a factor when assessing children.
Educational psychology in Practice, 19, 239-244. doi: 10.1080/0266736022000010267
Cummins, J. (1981). The role of primary language development in promoting educational
success for language minority students. In California State Department of Education
(Ed.), Schooling and minority students: A theoretical framework (pp. 3-49). Los Angeles,
CA: National Dissemination and Assessment Center.
Daradkeh, T. K. (1993). Parent-teacher reliability in rating children on the 10-items
Conners Rating Scale. Egypt Journal of Psychiatry, 16, 52-56.
Elzer, M. J. (2005). Talen naar een moedertaal [Care for mother tongue]. Tijdschrift voor
Orthopedagogiek, 3, 130-135.
88
language proficiency and social-emotional development
Embregts, P. J. C. M. (2000). Reliability of the Child Behavior Checklist for the assessment
of behavioral problems of children and youth with mild mental retardation. Research
in Developmental Disabilities, 21, 31-41. doi: 10.1016/S0891-4222(99)00028-1
Fujiki, M., Brinton, B., Isaacson, T., & Summer, C. (2001). Social behaviors of children
with language impairment on the playground: A pilot study. Language, Speech, and
Hearing Services in Schools, 32, 101-113. doi: 10.1044/0161-1461(2001/008)
Gertner, B. L., Rice, M. L., & Hadley, P. A. (1994). Influences of communication
competence on peer preferences in a preschool classroom. Journal of Speech, Language,
and Hearing Research, 37, 899-907.
Glover-Gagnon, S., & Nagle, R. J. (2004). Relationships between peer interactive play
and social competence in at-risk preschool children. Psychology in the Schools, 41, 173189. doi: 10.1002/pits.10120
Goorhuis-brouwer, S. M. & Schaerlaekens, A. M. (2000). Handboek taalontwikkeling,
taalpathologie en taaltherapie bij Nederlandssprekende kinderen. [Handbook language
development, language pathology and language therapy with Dutch children].
Utrecht, the Netherlands: De Tijdstroom.
Goossens, S., Bruinsma, C., Dekker, P. H., & de Ruyter, P. A. (2000). Gedragsvragenlijst
voor peuters en kleuters: discriminante en convergente validiteit [Behavior questionnaire
for toddlers and kindergartners: Discriminant and convergent validity]. Internal
Report. Amsterdam, the Netherlands: Vrije Universiteit.
Irwin, J. R., Carter, A. S., & Briggs-Gowan, M. J. (2002). The social-emotional
development of ‘late talking’ toddlers. Journal of the American Academy of Child and
Adolescent Psychiatry, 41, 1324-1332. doi: 10.1097/00004583-200211000-00014
Konold, T. R., & Pianta, R. C. (2007). The influence of informants on ratings of children’s
behavioral functioning. A latent variable approach. Journal of Psychoeducational
Assessment, 25, 222-236. doi: 10.1177/0734282906297784
Leseman, P. P. M. (2002). Early childhood education and care for children from low income
or minority backgrounds. A paper for discussion at the OECD Oslo Workshop, June
6-7, 2002. Amsterdam: Universiteit van Amsterdam.
Moffitt, T. E., & Silva, P. A. (1988). Neuropsychological deficit and self-reported
delinquency in an unselected birth cohort. Journal of the American Academy of Child
and Adolescent Psychiatry, 27, 233-240. doi: 10.1097/00004583-198803000-00017
Noterdaeme, M., & Amorosa, H. (1999). Evaluation of emotional and behavioral problems
in language impaired children using the Child Behavior Checklist. European Child &
Adolescent Psychiatry, 8, 71-77. doi: 10.1007/s007870050087
Pianta, R. C., Steinberg, M. S., & Rollins, K. B. (1995). The first two years of school.
Teacher-child relationship and deflections in children’s classroom adjustment.
Development and Psychopathology, 7, 295-312. doi: 10.1017/S0954579400006519
89
Raver, C. C. (2002). Emotions matter: Making the case for the role of young children’s
emotional development for early school readiness. Social Policy Report, 16, 3-18.
Redmond, S. M., & Rice, M. L. (1998). The socioemotional behaviors of children with
SLI: Social adaptation or social deviance? Journal of Speech, Language, and Hearing
Research, 41, 688-700. doi: 10.1044/jslhr.4103.688
Rice, M. L., Sell, M. A., & Hadley, P. A. (1991). Social interactions of speech- and
language- impaired children. Journal of Speech and Hearing Research, 34, 1299-1307.
doi: 10.1044/jshr.3406.1299
Riddle, K. D., & Rapoport, J. L. (1976). A 2-year follow-up of 72 hyperactive boys.
Classroom behavior peer acceptance. The Journal of Nervous and Mental Diseases, 162,
126-134. doi: 10.1097/00005053-197602000-00007
Schonfeld, I. S., Shaffer, D., O’Conner, P., & Portnoy, S. (1988). Conduct disorder and
cognitive functioning: Testing three causal hypotheses. Child Development, 59, 9931007. doi: 10.2307/1130266
Stanton-Chapman, T. L., Justice, L. M, Skibbe, L. E., & Grant, S. L. (2007). Social and
behavioral characteristics of preschoolers with specific language impairment. Topics
in Early Childhood Special Education, 27, 98-109. doi: 10.1177/02711214070270020501
Tesser, P. T. M., & Iedema, J. (2001). Rapportage minderheden 2001. Deel I Vorderingen
op school [Report on minorities in 2001. Part 1 Progress at school]. Den Haag, The
Netherlands: Sociaal en Cultureel Planbureau.
van Eldik, M. C. M., Schlichting, J. E. P. T., Lutje Spielberg, H. C., van der Meulen, B. F.,
& van der Meulen, S. (1997). Reynell test voor taalbegrip, handleiding. [Reynell test for
language comprehension, manual]. Nijmegen, the Netherlands: Berkhout.
van Elten, M. (2003). Schoolprestaties van allochtone kinderen [Academic achievement
of immigrant children]. Onderwijsmagazine Inzicht. Retrieved June, 2003, from
http://www.voo.nl/inzicht.html
Verhulst, F. C., van der Ende, J., & Koot, H. M. (1996). Handleiding voor de CBCL/418 [Manual for the CBCL/4-18]. Rotterdam, the Netherlands: Erasmus Universiteit/
Sophia Kinderziekenhuis.
90
Chapter 5
differential trust between parents
and teachers of children from lowincome and immigrant backgrounds
Published as:
Janssen, M., Bakker, J. T. A., Bosman, A. M. T., Rosenberg, K., & Leseman, P. P. M.
(2012). Differential trust between teachers of children from low-income and immigrant
backgrounds. Educational Studies, 38, 383-396. doi: 10.1080/03055698.2011.643103
91
chapter 5
Abstract
This study was designed to investigate the trust relationship between parents and
teachers in first grade. Additional research questions were whether trust was related
to ethnicity and reading performance. The five facets of trust; benevolence, reliability,
competence, honesty, and openness, were measured on a 4-point Likert scale. Reading
performance was measured by the three-minute test. Parents were found to have more
trust in the reliability, competence, and honesty of teachers than teachers in parents.
Native-Dutch and immigrant parents have the same trust level towards their children’s
teacher. However, teachers reported lower trust levels on all five facets of trust for
immigrant parents. Our findings indicated that ethnicity has no influence on parents
trust in teachers, but ethnicity may explain teachers’ trust in parents. Some support
was found for the assumption that teachers’ trust plays a role in reading performance.
Children were found to have higher reading performance when teachers reported higher
trust in the benevolence and openness of parents.
92
differential trust between parents and teachers
Around 1960 specific groups in Dutch society showed severe educational
disadvantages (Mathijssen & Sonnemans, 1959; van Heek, 1968/2009). For example,
academic performance and school progress of children from lower socio-economic status
(SES) groups were considerably lower than that of children from higher-SES groups.
The hypothesis put forward by van Calcar, Soutenberg, and Tellegen (1968/2009) was
that education itself hampered the success of children from lower-SES groups. They
argued that lower-SES children were unable to show that they had sufficient skills to
attain higher forms of education. The major explanation was that family culture did not
correspond with middle-class school culture, and that a language delay was considered
to be an important aspect of their problem in first year of primary school.
In the early seventies attempts were made to change this situation. The notion that
children did not start primary school at the same developmental level, due to differences
in social and cultural backgrounds, became more and more accepted. Rupp (1971) found
evidence for the relation between ‘cultural-pedagogical aspects of upbringing’ and
academic performance. Cultural-pedagogical aspects expressed itself in involvement of
parents in the school, school attendance of the child, observation of positive changes
in child’s development, and stimulation of cognitive development. Thus, because lowincome children were mainly absent from higher forms of education, education for
disadvantaged children in the Netherlands obtained a prominent place on the national
Dutch political agenda.
One of the first initiatives in 1968 was the development of a preschool program. This
program included a family counselling program, a school preparation program, and
a language program. The most important assumption was that parental involvement
was of crucial importance to guarantee school success. That is why attempts were
made to improve the home-school relationship. The last decade preschool education
has become of particular importance in educating disadvantaged low-income and
immigrant children. A great many studies have shown the effectiveness of early
language interventions in the critical period before children are entering primary school
(see Leseman, 2009 for a review). The goal of these preschool programs was to prepare
children from socio-economically disadvantaged families entering formal schooling.
These programs covered various domains of development: Language, social-emotional
and cognitive development as well as parental (school) involvement and child-rearing
support. A number of studies showed that parental involvement predicts children’s
school performance in different domains (Blok et al., 2005; Epstein & Dauber, 1991;
Grolnick & Slowiaczek, 1994; Henderson & Mapp, 2002; Zellman & Waterman, 1998).
It has been known for some time that children from lower SES groups and children
93
chapter 5
from ethnic minority backgrounds have a much harder time to successfully participate
in mainstream education. At present, it is believed that the major problem to be solved is
the development of the language skills of these groups. Without a sufficient command
of the Dutch language they will be unable to maximally profit from education offered in
the school (Elzer, 2005; Stanton-Salazar, 1997; van Elten, 2003).
Children’s social contexts (family, school, community) are important predictors
of reading development. Particularly at a young age, parental involvement plays an
important role for providing opportunities to learn by reading aloud and talking to their
child about the story, support the child to read, the presence of adult and children books
at home, going to the library with the child and emphasizing positive feelings towards
reading (Baker, Afflerbach, & Reinking, 1996; McCarthey, 2000; Sonnenschein, Brody,
& Munsterman, 1996).
Bakker and Denessen (2007) argue that “parent involvement can vary from a narrow
perspective, defining parent involvement as parent involvement activities at school, to a
broad perspective, also including parenting behaviours at home and parents’ attitudes
towards their child’s school(ing)” (p. 240). They conclude “… that existing questionnaires
enable the measure of parent involvement as a multidimensional construct” (p. 241). A
distinction can be made between home-based and school-based involvement. Schoolbased behaviour is easier to observe for the teacher, like helping in the classroom,
engaging in school events, attending parent-teacher meetings or volunteer work. Homebased behaviour, less obvious for a teacher, refers to talking with the child about school,
helping with homework, or reading with the child. Particularly home-based involvement
is of crucial importance.
Bakker and Denessen (2007) also distinguish, apart from home-based and
school-based involvement, spontaneous parental involvement. This indirect parental
involvement could be teaching their child to follow rules and directives, encourage
their child’s self-esteem, helping their child understand his/her moral and ethnical
responsibilities, visit cultural activities, talk about television programs or being home
when the child returns from school. Teachers in their study, however, tended to consider
parental involvement as a one-dimensional construct. Teachers may disregard homebased involvement because it is literally withdrawn from their eyes. This may easily
lead to stereotypical images or prejudices, and as a consequence to lower expectations,
different teachers attitudes, and lower academic performance (Bakker, Denessen, &
Brus-Laeven, 2007; Bakker, Stoep, van den Heuvel, & Bouts, 2002). Mutual trust might
prevent this self-fulfilling prophecy.
Not only is the concept of parental involvement complex but establishing effective
partnership between low-income and/or immigrant parents and schools brings additional
complexities because of issues of prejudice, discrimination, language problems, low
94
differential trust between parents and teachers
level of education, or a lack of social network to reach goals that may reveal profound
differences in the quality and quantity of communication between parents and teachers
(Bakker et al., 2002; Denessen, Bakker, & Gierveld, 2007; Driessen & Jungbluth, 1994;
Ogbu, 1978). Low-income and immigrant parents were found to show more home-based
involvement like showing interest, encouraging the child or helping with homework
than school-based involvement (Denessen, Driessen, Smit, & Sleegers, 2001). The low
frequency of school-based involvement in low-income and immigrant parents may be
interpreted by teachers as a lack of involvement (Tett, 2004). However, although lowincome and immigrant parents are involved differently than white middle-class parents,
they do believe that education is of great importance.
It is assumed that when home, school, and community, accomplish a strong
relationship and cooperate with one another this will foster parental involvement,
irrespective of social economical class or ethnic background, and as a consequence may
improve children’s development. Therefore parents and teachers must be willing to trust
each other, because “Partnership is based on mutual trust” (Deslandes, 2001, p. 2).
Adams and Christenson (1998, p. 6) define trust as “confidence that another person
will act in a way to benefit or sustain the relationship, or the implicit or explicit goals of
the relationship, to achieve positive outcomes for students.” Hoy and Tschannen-Moran
(1999, p. 189) explicate the nature of the relationship in their definition of trust: “Trust
is an individual’s or group’s willingness to be vulnerable to another party based on the
confidence that the latter party is benevolent, reliable, competent, honest and open”.
Hoy and Tschannen-Moran (1999) distinguish five facets of trust. Each of these
facets represents a distinct aspect of trust. The first facet is benevolence or confidence
in the good will of the other party. The second facet, reliability, refers to predictability
of expectations. Competence, the third facet, means that teachers are expected to be
competent in teaching and parents are expected to be competent in child-rearing and
supporting their child’s learning. The fourth facet is honesty and refers to truthfulness,
integrity, and consistency. The fifth and last facet is openness. Openness and trust are
reciprocal, that is, when people trust each other they tend to be more open, and when
people are more open and share information a more trustful relationship will occur. All
facets of trust are needed to build a solid trusting relationship. In sum, a perfectly trustworthy person is, benevolent, reliable, competent, honest, and open (Hoy & TschannenMoran, 1999). An important aspect of the home-school relationship is found to be
teachers’ trust. Teachers’ trust has also been found to predict academic performance
(Goddard, Tschannen-Moran, & Hoy, 2001; Sui-Chu and Douglas, 1996; Swap, 1993;
Webb, 1992). Furthermore, research shows that trust is usually based on cultural norms
and obligations about educating children. When parents are from a different socioeconomic stratum or another ethnic background than their child’s teacher, this may
95
chapter 5
easily lead to misunderstanding, stereotypes, and different expectations, which in turn
may affect the trust-relationship negatively or amplify distrust (Allison & Messick, 1985;
Bakker et al., 2007).
Lee (2007) has put the levels of trust on a continuum; at the low end we find
distrust, and at the high end high trust. It appears that distrusting relationships may
lead to non-optimal student results, feelings of anxiety and insecurity, whereas hightrust relationships are generally beneficial for positive student achievements. Rempel,
Holmes, and Zanna (1985) described the lowest level of trust as a limited amount of
contact between parents and teacher, which remains only at the level of predictability.
A medium-trust level refers to a more defensive attitude in which positive behaviour
is seen as situation specific, and negative behaviour may be interpreted in a broader
context that refers to dependability. High trust refers to faith or emotional security.
Parents and teachers rely on each other and trust each other to maintain appointments
and take each other seriously.
It goes without saying that a perfect trust relationship does not exist. However, the
greater the contribution of each facet the more trust-worthy the relationship between
parents and school will be. Because it is a reciprocal phenomenon, it does not mean that
equal levels of trust are to be found between teachers and parents. In fact, Adams and
Christenson (1998) found that teachers’ trust mostly remains at the lowest level of trust,
that is, the predictability stage, whereas parents’ trust was found to be at a much higher
trust level. Parents must show initial trust by sending their child to school, whereas
teachers have little at stake personally.
Therefore, teachers are in search of trustworthy behaviour in parents, in this
view the amount of contact to display trustworthy behaviour is important. HoorensMaas and Naafs-Wilstra (1997) concluded that when teachers and parents build up a
positive relationship, feelings of safety and trust will develop, and they will become
more predictable with respect to each other and become partners in education. Because
disadvantaged children are more vulnerable for school failure, trust in these groups is
even more important to help them making school progress (Goddard et al., 2001).
Brewster and Railsback (2003) found poor communication and negative experiences
to be an important risk factor for the development of distrust. Distrust may occur when
parents doubt the effectiveness of school, the competence of the teacher, have conflicts
at school, or doubt the integrity of the school principle. When parents feel that teachers
or school principles do not take their problems seriously or are unable to solve problems,
this may lead to distrust. Another risk factor mentioned by Denessen et al. (2007) is
cultural differences between home and school. Smit, Driessen, and Doesborgh (2005)
found that traditional immigrant families in the Netherlands evaluated the teacherchild interactions as too informal. Traditional immigrant families prefer a more
96
differential trust between parents and teachers
authoritarian, formal teacher-child relation. Their ideas about parenting and education
differ in important ways from native Dutch (middle and high SES) parents. They
found that immigrant parents consider school to be responsible for the education of
their children and they, the parents, are responsible for teaching morals and values at
home. Nevertheless, to establish partnership Deslandes (2001) emphasized that parents
and teachers need to be open and willing to raise and educate the child mutually and
share responsibilities. In order to improve academic performance, motivation and,
development of the child, good parent-teacher interaction, or home-school partnership,
is of particular importance.
To summarize, several studies have shown that trust is of great interest for academic
performance. The majority of these studies addressed middle or high SES families.
None of these studies have provided information about trust between teachers and
parents from low income and immigrant backgrounds. In this study we try to build
upon the empirical evidence linking trust and academic performance. Research shows
that more trust may lead to better academic performance (Lee, 2007). The question
remains whether this relation is already present in first grade and whether trust may
be seen as a function of SES and/or ethnicity and whether trust affects academic
performance, especially reading. These questions are of practical importance for parents
and teachers who might want to improve the trust-relationship or look for ways to foster
and/or to repair trust. Empirical evidence for the relation between trust level, academic
performance and SES or ethnicity may help low-income and immigrant children in their
school career. This study contributes to evidence about a specific group of children who
are known to be at risk for academic failure or drop-out. We therefore, sought answers
to the following questions:
1. What facet of trust determines the relationship between teachers and parents?
2. Is trust related to ethnicity?
3. Is trust related to academic performance, that is, to reading?
Method
Participants
In this study participated 57 parents (28 were Native Dutch and 29 were immigrants)
and 23 teachers (n = 23). All parents completed a questionnaire (see below) regarding the
teacher of their child. The teachers completed the same questionnaire about the parents
97
chapter 5
of 38 native-Dutch and 70 immigrant children. The majority of these children and their
families lived in inner-city neighbourhoods from low-income and immigrant families
(n = 96), whereas a small sample of 20 children lived in neighbourhoods with a majority
of Caucasian high to middle-income families. The mean age of the children was 7.2 years.
Materials and procedure
To measure trust between parents and teachers, we developed two new instruments
that measured the five facets of trust (i.e., ‘benevolence’, ‘reliability’, ‘competence’, ‘honesty’,
and ‘openness’) described by Hoy and Tschannen-Moran (1999); one for parents and one
for teachers. The final list of items of the two questionnaires was analogous. The parents’
questionnaire consisted of all 17 items of the list developed by Adams and Christenson
(2000) and 17 new items. The teachers’ questionnaire consisted of all 19 items of Adams
and Christenson for teachers to which 15 new items were added. Both questionnaires
contained sentences that began with ‘I am confident that teachers...’ or ‘I am confident
that parents…’ followed by a statement about parent or teacher behaviour concerning the
child’s school performance. The answer options consisted of a 4-point Likert scale varied
from totally disagree to totally agree. The facet ‘Benevolence’ was measured with items
9, 12, 16, 17, 18, and 20; ‘Reliability’ with items 3, 6, 7, 10, 11, 30, and 34; ‘Competence’
with items 1, 2, 15, 19, 28 and 33; ‘Honesty’ with items 22, 25, 27, 29, 31, and 32; and
‘Openness’ with items 4, 5, 8, 13, 14, 23, 24, and 26. The final list is presented in the
Appendix. About 10 minutes were needed to complete the questionnaire.
Reliability of the questionnaire was assessed for the complete questionnaire. In order
to determine homogeneity of the items underlying one facet of trust, internal consistency
of the facets was evaluated. Cronbach’s alpha was computed in order to estimate how
consistent the subjects responded to the separate questions within each facet of trust.
Alpha of .70 or higher was considered to be sufficient. Internal consistencies for the
facets for parents were: benevolence .91, reliability .85, competence .81, honesty .92 and
openness .87. The internal consistencies for the five facets for teachers were: benevolence
.92, reliability .92, competence .90, honesty .90 and openness .91. This implies that
internal consistency was sufficient over the five facets of trust. The reliability of the
complete questionnaire for parents was .97 and for teachers .98.
Correlations between the facets are presented in Table 1. All correlations of the
five facets of trust from parents and teachers were significant. Because the internal
consistency of all facets and from the complete questionnaire was high this was
expected. This would indicate there is one underlying construct, trust. We choose to
maintain the five facets for further analyses, because these aspects were expected to lead
to better understanding of the concept trust between teachers and parents than a onedimensional analysis of trust.
98
differential trust between parents and teachers
Table 1 Pearson Correlations between Facets of Trust among Parents and among Teachers
Benevolence
Benevolence
Reliability
Competence
Honesty
Openness
.79
.82
.80
.75
Reliability
.88
.84
.86
.78
Competence
.91
.91
.79
.74
Honesty
.85
.85
.86
Openness
.85
.84
.87
.81
.82
Note. The right upper triangle contains the r’s of the teachers (n = 95); the left lower triangle those of the parents (n = 55). All
correlations are significant, p< .001.
Reading. Drie-minuten-test [Three-Minutes-Test] (DMT; Verhoeven, 1995) was used
to measure word reading. The DMT is a single-word reading test consisting of three
different cards. The first card consists of VC, CV and CVC words, the second card of
monosyllabic word with consonant clusters, and the third card consisted of multi-syllabic
words. Children are asked to read as quickly and accurately as they can the words on the
card (starting with the first one). The score is the number of words read correctly in one
minute. For this study only the first and second cards were used. The raw scores were
converted into norm scores therefore a lower score on the reading task indicates a high
reading level.
Data were gathered from first grade children in March. All children were tested by
their teacher at their school. The parents’ questionnaires were administrated between
March and June at home. The teachers questionnaires were send to the teachers and
they were asked to send them back.
Results
We will first analyse the data regarding the trust relationship between parents and
teachers. Then the relationship between trust and ethnicity is investigated, followed by the
question pertaining to parents’ trust in teachers and teachers’ trust in parents. Finally, we
will discuss the question whether trust is predictive for the acquisition of reading.
Trust relationship between teachers and parents
No significant correlations between the five facets of parents and teachers were found
(see Table 2) or between the view of parents and the view of teachers with respect to the
five facets of trust. Paired-sampled t tests were conducted in order to estimate whether
there were differences between the five facets of parents and teachers.
99
chapter 5
Table 2 T-tests and Correlations between Parents’ and Teachers’ Facets of Trust
Facet
Benevolence
Reliability
Competence
Honesty
Openness
Mean score
parents
3.37
3.45
3.44
3.50
3.33
Mean score
teachers
3.30
3.23
3.23
3.29
3.13
t-tests
Pearson correlations
t(41) = 0.75, p = .45
t(41) = 2.40, p = .02
t(41) = 2.43, p = .02
t(41) = 2.54, p = .02
t(41) = 1.81, p = .08
r = .05, p = .75
r = .20,p = .20
r = .20,p = .21
r = .24,p = .13
r = -.13,p = .42
Of 42 children both parents and teacher filled out the questionnaire. On the facets
‘benevolence’ and ‘openness’ no significant differences were found between parents
and teachers. The facets ‘reliability’, ‘competence’, and ‘honesty’ revealed significant
differences: Parents were found to have more trust in teachers to be reliable, competent
and honest than teachers in parents.
Ethnicity related to parents’ trust in teachers.
These analyses pertained to the relationship between ethnicity and trust. A 2
(ethnicity: native Dutch vs. Immigrant) by 5 (facet: benevolence vs. reliability vs.
competence vs. honest vs. openness) MANOVA was performed on the mean score of the
facets. Ethnicity was a between-subjects variable and facet a within-subjects variable.
The main effect of ethnicity was not significant F < 1. The trust of native Dutch parents
in teachers (M = 3.34, SD = .37) did not significantly differ from the trust of immigrant
parents in teachers (M = 3.39, SD = .43). The main effect of facet was significant F(4, 212)
= 5.56, p< .0001, partial η2 = .10. The results also showed a significant ethnicity by facet
interaction, F(4, 212) = 3.32, p = .01, partial η2 = .06. Thus, the differences between the
groups on the different facets of trust significantly differed from each other.
A repeated measure ANOVA was carried out for each group of ethnicity. For trust of
Dutch parents in the teacher, a significant main effect was found F(4, 108) = 10.99, p = .0001,
partial η2 =.29.Bonferroni-adjusted post-hoc tests were carried out to further investigate
the differences on the facets of trust. The mean score on openness was significantly lower
than reliability, competence and honesty (p< .05). The difference between benevolence
and openness was not significant. No other differences were significant. For trust of
immigrant parents in the teacher, no main effect for trust was found F< 1.The levels of all
five facets of trust were equal for immigrant parents (see Table 3).
Teachers’ trust in parents of different ethnicity.
These analyses pertained to the relationship between ethnicity of parents and
teachers’ trust. A 2 (ethnicity: native Dutch vs. Immigrant) by 5 (facet: benevolence vs.
100
differential trust between parents and teachers
Table 3 T-tests of Parents’ Trust in Teachers based on Parents’ Ethnic Backgrounds
Facet
Benevolence
Reliability
Competence
Honesty
Openness
Mean score of Dutch
parents (n = 28)
3.34
3.45
3.48
3.47
3.23
Mean score of immigrant
parents (n = 27)
3.28
3.34
3.31
3.39
3.32
t-tests
t(53) = .38, p= .70
t(53) = .99, p= .33
t(53) = 1.56,p= .13
t(53) = .68, p= .50
t(53) = -.75, p= .46
reliability vs. competence vs. honest vs. openness) MANOVA was performed on the
mean score of the facets. Ethnicity was a between-subjects variable and facet a withinsubjects variable.
The main effect of ethnicity was significant F(1, 93) = 11.91, p< .0001, partial η2 = .11.
Teachers’ trust in Dutch parents (M = 3.29, SD = 0.08) was significantly higher than
teachers’ trust in immigrant parents (M = 2.95, SD = .006). The main effect of facet
was significant F(4, 372) = 21.70, p< .0001, partial η2 = .19. The results also showed a
significant ethnicity by facet interaction, F(4, 372) = 4.20, p = .002, partial η2 = .04. A
repeated measure ANOVA was carried out for each facet of teachers’ trust in parents for
native Dutch and immigrant parents separately.
With respect to the trust of teachers in Dutch parents a significant main effect of
trust was found F(4, 136) = 5.22, p = .0001, partial η2 = .13. Bonferroni-adjusted posthoc tests were carried out to further investigate the differences on the facets of trust.
The mean score on openness was significantly lower than benevolence, reliability and
competence (p< .05). The difference between honesty and openness was not significant.
The trust of teachers in immigrant parents also revealed a significant main effect of
trust F(4, 236) = 24.57, p = .0001, partial η2 = 29. Bonferroni-adjusted post-hoc tests were
carried out to further investigate the differences on the facets of trust. The mean score
on openness was significantly lower than all other facets of trust (p< .05). Honesty was
found to be significantly higher than reliability, competence and openness.
A paired-sampled t test was then carried out in order to investigate the differences
between teachers trust in immigrant and Dutch parents. Significant differences were
found for all facets of teachers trust between immigrant and Dutch parents. Teachers
had significantly more trust in Dutch parents than in immigrant parents (see Table 4).
Trust and reading
To assess the strongest predictor of reading from trust, a stepwise multiple-regression
analysis was carried out. The results showed that the five facets of teachers’ trust were
101
chapter 5
Table 4 T-tests of Teachers’ Trust in Parents based on Parents’ Ethnic Backgrounds
Facet
Benevolence
Reliability
Competence
Honesty
Openness
Mean score of Dutch
parents (n = 35)
3.36
3.30
3.30
3.32
3.19
Mean score of immigrant
parents (n = 55)
3.01
2.96
2.96
3.09
2.73
t-tests
t(88) = 3.14, p= .0001
t(88) = 2.85, p= .0001
t(88) = 3.24, p= .0001
t(88) = 2.08,p= .04
t(88) = 3.84, p= .0001
related to reading Card 1 (F(5,93) = 2.32, p = 0.05) and Card 2 (F(5,91) = 2.32, p = .04).
Teachers’ trust in the benevolence of parents was found to be the best predictor for
reading Card 1 (Beta = -.55, t(5,88) = -2.06, p = 0.4) and Card 2 (Beta = -.53, t(5,86) = -1,96,
p = 0.53). The greater teachers’ trust in parents’ benevolence, the higher the reading level
of the child. A lower score on the reading task indicates a high reading level, explaining
the negative beta.
The second predictor openness that turned out to be a significant predictor for
reading Card 1 (Beta = 0.42, t(5.88) = 1.96, p = .05), was indicating that a higher level of
teachers’ trust in the openness of parents predicts a higher reading level of the child. No
other facet turned out to be significant.
Discussion
The present study was designed to investigate the trust relationship between parents
and teachers in first grade. The results obtained on the parents’ trust scale were found
to be different from the results obtained on the teachers’ trust scale. Parents were found
to have more trust in the reliability, competence, and honesty of teachers than teachers
have in parents. Our finding, that parents’ and teachers’ trust differ are in accordance with
the findings of Adams and Christenson (1998). These authors also found that parents’
trust in teachers was higher than teachers’ trust in parents for middle-school students.
Income and ethnicity did not appear to influence parents’ trust in teachers. They explained
these finding by stating that parents are more vulnerable, because they have at least some
confidence in the teacher to be a professional in teaching children. Parents must have some
basic trust in the teacher, because they leave their child with a stranger in order to teach
their child. Parents expect the teacher to be a professional. Teachers, on the other hand,
are not as vulnerable as parents, because they see themselves as the professional and judge
whether parents are trustworthy. Therefore trust may remain at the predictability stage.
102
differential trust between parents and teachers
The results of the second research question, pertaining to the effect of ethnicity on
trust, revealed that the facet ‘openness’ of teachers is less present than the other facets of
trust, judged by native-Dutch parents. This suggests that native-Dutch parents judge their
teachers as less open. For the immigrant parents no differences between the different
facets of trust were found. No differences were found between native-Dutch and immigrant
parents on mean trust level per facet except for openness. This indicates that native-Dutch
and immigrant parents have the same trust level towards their children’s teacher, expect
for openness. Conversely, teachers reported lower trust level on all five facets of trust in
immigrant parents than in native-Dutch parents. Our findings indicate that ethnicity
has no influence on parents’ trust in teachers, but ethnicity does affect teachers’ trust in
parents. The relationship between immigrant parents and teachers is more complicated,
than it is between native families and the teachers. The relationship may remain at the
lowest trust level, because many immigrant parents do not have a sufficient command of
the native (i.e., Dutch) language, which may limit the quantity and the quality of contact
and may lead to distrust in teachers. This discontinuity between home and school often
cause conflict situations and may interfere with a healthy trust relationship.
A second important variable affecting trust was socio-economic status (SES).
Goddard et al. (2001) found a strong association between SES and teachers’ trust. The
majority of the variance of teachers’ trust in their study was explained by SES and not
by ethnicity. Although this was not examined in the present study, our sample consisted
of low SES families, therefore this seem to be a possible explanation for low teachers
trust in parents. Hoff and Tian (2005) found similar results for SES-related influences
on language development for western and Asian culture. Cultural differences between
socio-economic classes may be harder to overcome than cultural differences based
on ethnicity. That is, people show higher-trust levels towards people more similar to
themselves, which is in educational context mainly based on social-economic status.
In line with this view, teachers might show equal trust levels in middle or high-class
immigrant and native-Dutch parents.
The results obtained on the reading task showed that teachers’ trust played an
important role in reading performance. Children were found to have higher reading
performance when teachers reported higher trust. Teachers’ trust in the benevolence
and openness of parents was found to be predictive of reading performance. A possible
explanation for these findings may be that when teachers judge parents to be more
willing and open, teachers will be more prepared to support parents and give them
specific advice about reading at home than to parents of which they think are less willing
and open. The need to build trust seems obvious. On the other hand, an alternative
explanation may be that when children are better readers this may lead to more teachers
trust in the benevolence and openness of parents.
103
chapter 5
Practical implications
Over the last decade much attention has been paid to educational disadvantages
of children from low-income and immigrant backgrounds. The effect being that the
educational priority policy was established in the Netherlands. Advancing parental
involvement and improving home-school partnership was one of the main goals.
Research had shown that the quality and quantity of home-school communication is
crucial for parental involvement (Slob, 1995; Smit et al., 2005; van Erp & Veen, 1990).
Our study proved that not only communication, but also teachers’ trust in parents plays
a crucial role in educating children at risk. Trust is of crucial importance for an effective
collaboration between parents and teachers. The importance of teacher’s trust became
even clearer when it was shown that it was associated with reading performance of the
children. Moreover, teachers were found to have lower trust-levels in immigrant parents
than in native-Dutch parents. To promote trust teachers need to have faith in all parents,
irrespective of their background. Educating children has to be a mutual goal for parents
and teachers. Therefore stimulation of educational involvement of parents within (pre)
school programs is best based on effective partnership. To establish this, it may help
to enhance teachers’ competence in communicating with different cultures (Booijink,
2007; Denessen et al., 2007; Smith et al., 2005). When teachers feel more competent
in communicating with immigrant parents, who might have limited knowledge of the
native (Dutch) language or who have different cultural manners to communicate, this
may facilitate mutual understanding and parents might feel less of a distance between
home and school which in turn might lead to more effective partnership.
What is probably needed is an open communication, mutual respect, listening to
each other, and honesty. When parents trust their teacher, they most likely believe that
the teacher is qualified, honest, reliable, and will act to achieve positive school-outcomes
for their child.
104
differential trust between parents and teachers
References
Adams, K. S., & Christenson, S. L. (1998). Differences in parent and teacher trust
levels: Implications for creating collaborative family-school relationships. Special
Services in the Schools, 14, 1-22. doi: 10.1300/J008v14n01_01
Adams, K. S., & Christenson, S. L. (2000). Trust and the family-school relationship
examination of parent-teacher differences in elementary and secondary grades.
Journal of School Psychology, 38, 477-497. doi: 10.1016/S0022-4405(00)00048-0
Allison, S. T., & Messick, D. M. (1985). The group attribution error. Journal of
Experimental Social Psychology, 21, 563-579. doi: 10.1016/0022-1031(85)90025-3
Baker, L., Afflerbach, P., & Reinking, D. (1996). Developing engaged readers in school
and home communities: An overview. In L. Baker, P. Afflerbach, & D. Reinking
(Eds.), Developing engaged readers in school and home communities (pp. xiii). Mahwah,
NJ: Erlbaum.
Bakker, J., & Denessen, E. (2007). The concept of parent involvement. Some
theoretical and empirical considerations. In H. Phtiaka & S. Symeonidou (Eds.),
Schools and families in partnership: Looking in the future (pp. 238-252). Nicosia,
Cyprus: ERNAPE.
Bakker, J., Denessen, E., & Brus-Laeven, M. (2007). Socio-economic background,
parental involvement and teacher perceptions of these in relation to pupil
achievement. Educational Studies, 33, 175-190. doi: 10.1080/03055690601068345
Bakker, J., Stoep, J., van den Heuvel, W., & Bouts, L. (2002). Leerkrachtverwachtingen
en de oordeelsvorming over ouderlijke betrokkenheid [Teacher expectations and
parental-involvement judgements]. Pedagogische Studiën, 79, 376-388.
Blok, H., Fukkink, R. G., Gebhardt, E. C., & Leseman, P. P. M. (2005). The relevance
of delivery mode and other programme characteristics for the effectiveness of early
childhood intervention. International Journal of Behavioral Development, 29, 35-47.
doi: 10.1080/01650250444000315
Booijink, M. (2007). Terug naar de basis. Communicatie tussen leerkrachten en allochtone
ouders in het primair onderwijs. Onderzoeksrapport [Back to basics. Communication
between teachers and ethnic minority parents in primary education. Research
report]. Leiden, the Netherlands: Leiden University.
Brewster, C., & Railsback, J. (2003). Building trust with schools and diverse families: A
foundation for lasting partnerships. Portland, OR: Northwest Regional Educational
Laboratory. Retrievable from, http://educationnorthwest.org/webfm_send/453
Denessen, E., Bakker, J., & Gierveld, M. (2007). Parental involvement in multi ethnic
schools. The School Community Journal, 17, 27-44.
105
chapter 5
Denessen, E., Driessen, G., Smit, F., & Sleegers, P. (2001). Culture differences in
education: Implications for parental involvement and educational policies. In F.
Smit, K. van der Wolf, & P. Sleegers (Eds.), A bridge to the future. Collaboration
between parents, schools, and communities (pp. 55-66). Nijmegen, the Netherlands:
Institute for Applied Social Sciences (ITS).
Deslandes, R. (2001). A vision of home-school partnership: Three complementary
conceptual frameworks. In F. Smit, K. van der Wolf, & P. Sleegers (Eds.), A bridge
to the future. Collaboration between parents, schools, and communities (pp. 11-23).
Nijmegen, the Netherlands: Institute for Applied Social Sciences (ITS).
Driessen, G., & Jungbluth, P. (1994). Educational opportunities. Tackling ethnic,
class, and gender inequality through research. Munster, Germany/New York, NY:
Waxmann.
Elzer, M. J. (2005). Talen naar een moedertaal [Care for mother tongue]. Tijdschrift voor
Orthopedagogiek, 3, 130-135.
Epstein, J., & Dauber, S. (1991). School programs and teacher practices of parent
involvement in inner-city elementary and middle schools. The Elementary School
Journal, 91, 279-289. doi: 10.1086/461656
Goddard, R. D., Tschannen-Moran, M., & Hoy, W. K. (2001). A multilevel examination
of the distribution and effects of teacher trust in students and parents in urban
elementary schools. Elementary School Journal, 102, 3-17. doi: 10.1086/499690
Grolnick, W., & Slowiaczek, M. (1994). Parents’ involvement in children’s schooling:
A multidimensional conceptualization and motivational model. Child Development,
65, 237-252. doi: 10.2307/1131378
Henderson, A. T., & Mapp, K. L. (2002). A new wave of evidence: The impact of school,
family, and community connections on student achievement. Austin, TX: National
Center of Family & Community Connections with Schools, Southwest Educational
Development Laboratory. ED 474 521
Hoff, E., & Tian, C. (2005). Socio-economic status and cultural influences on language.
Journal of Communication Disorders, 38, 271-278. doi: 10.1016/j.jcomdis.2005.02.003
Hoorens-Maas, R., & Naafs-Wilstra, C. (1997). Een school om te vertrouwen [A school to
trust]. Houten, the Netherlands: EPN.
Hoy, A. W., & Tschannen-Moran, M. (1999). Implications of cognitive approaches to
peer learning for teacher education. In A. M. O’Donnell & A. King (Eds.), Cognitive
perspectives on peer learning (pp. 257-284). Mahwah, NJ: Lawrence Erlbaum
Associates Publishers.
Lee, S. (2007). The relationship between the student-teacher trust relationship and
school success in the case of Korean middle schools. Educational Studies, 33, 209216. doi: 10.1080/03055690601068477
106
differential trust between parents and teachers
Leseman, P .P. M. (2009). Integrated early childhood education and care: Combating
educational disadvantages of children from low income and immigrant families.
In Eurydice & EAC Executive Agency (Eds.), Early childhood education and care in
Europe: Tackling social and educational inequalities (pp.17-49). Brussel, Belgium:
European Commission.
Matthijssen, M. A. J. M., & Sonnemans, G. J. M. (1959). Schoolkeuze en schoolsucces bij
vhmo en ulo in Noord-Brabant [School choice and school success of vocational and
secondary education in Noord-Brabant]. Tilburg, the Netherlands: Zwijsen.
McCarthey, S. J. (2000). Home-school connections: A review of the literature. Journal of
Educational Research, 99, 145-167. doi: 10.1080/00220670009598703
Ogbu, J. U. (1978). Minority Education and Caste: The American System in Cross-Cultural
Perspective. New York, NY: Academic Press.
Rempel, J. K., Holmes, J. G., & Zanna, M. P. (1985). Trust in close relationships. Journal
of Personality and Social Psychology,49, 498-512. doi: 10.1037//0022-3514.49.1.95
Rupp, J. C. C. (1971). Opvoeding tot schoolweerbaarheid. Een Utrechtse kinderpsychologische
studie [Education for school fitness. An Utrecht child psychological study]. Groningen,
the Netherlands: Wolters-Noordhoff.
Slob, L. (1995). Onderwijspositie en ouderlijke betrokkenheid: over de positie van Turkse
leerlingen in het onderwijs en de betrokkenheid van Turkse ouders in Nijmegen
[Education and parental involvement: The position of Turkish children in education
and the involvement of Turkish parents in Nijmegen]. Nijmegen, the Netherlands:
Katholieke Universiteit Nijmegen,Wetenschapswinkel.
Smit, F., Driessen, G., & Doesborgh, J. (2005). Opvattingen van allochtone ouders over
onderwijs: tussen wens en realiteit [Opinions of immigrant parents about education:
Between wishes and reality]. Nijmegen, the Netherlands: ITS.
Sonnenschein, S., Brody, G., & Munsterman, K. (1996). The influence of family beliefs
and practices on children’s early reading development. In L. Baker, P. Afflerbach, &
D. Reinking (Eds.), Developing engaged readers in school and home communities (pp.
3-20). Mahwah, NJ: Erlbaum.
Stanton-Salazar, R. (1997). A social capital framework for understanding the socialization
of racial minority children and youths. Harvard Educational Review, 67, 1-40.
Sui-Chu, E. H., & Douglas, W. J. (1996). Effects of parental involvement on grade
achievement. Sociology of Education, 69, 126-141. doi: 10.2307/2112802
Swap, S. M. (1993). Developing home-school partnerships: From concepts to practice. New
York, NY: Teachers College Press.
Tett, L. (2004). Parents and school communities in Japan and Scotland: Contrasts in
policy and practice in primary schools. International Journal of Lifelong Education, 23,
259-273. doi: 10.1080/0260/37042000229228
107
chapter 5
van Calcar, C., Soutenberg, S., & Tellegen, B. (1968/2009). School, milieu en prestatie
[School, environment, and achievement]. In F. van Heek (Ed.), Het verborgen talent
(pp. 255-286). Meppel, the Netherlands: Boom.
van Elten, M. (2003). Schoolprestaties van allochtone kinderen [Academic achievement
of immigrant children]. Onderwijsmagazine Inzicht. Retrieved June, 2003, from
http://www.voo.nl/inzicht.html
van Erp, M., & Veen, A. M. (1990). De relatie met de ouders op Amsterdamse
onderwijsvoorrangsscholen: een onderzoek naar de stand van zaken op het thema schoolouders in het Amsterdamse OVB [The relation with the parents on education priority
schools in Amsterdam: An examination of the situation on the theme school-parents
in education priority policy]. Amsterdam, the Netherlands: Stichting Centrum voor
Onderwijsonderzoek van de Universiteit van Amsterdam.
van Heek, F. (1968/2009). Het verborgen talent: milieu, schoolkeuze en schoolgeschiktheid
[The hidden talent: Environment, school choice and school aptitude]. Meppel, the
Netherlands: Boom.
Verhoeven, L. (1995). Drie-Minuten-Toets. Handleiding [Three-Minute-Test. Manual].
Arnhem, the Netherlands: Cito.
Webb, M. O. (1992). The epistemology of trust and the politics of suspicion. Pacific
Philosophical Quarterly,73, 390-400.
Zellman, G. L., & Waterman, J. M. (1998). Understanding the impact of parent school
involvement on children’s educational outcomes. The Journal of Educational Research,
91, 370-380. doi: 10.1080/00220679809597566
108
differential trust between parents and teachers
Appendix
I am confident that the parents/teacher
1. is doing a good job teaching my child to read, write and calculate
2. is doing a good job teaching my child to follow rules and directions
3. is doing a good job helping my child resolve conflicts with peers
4. is doing a good job in keeping me well-informed of my child’s progress
5. is doing a good job encouraging my participation in my child’s education
6. is doing a good job disciplining my child
7. is easy to reach when I have a problem or question
8. keep me aware of all the information I need related to school
9. is doing a good job encouraging my child’s sense of self-esteem
10. is doing a good job encouraging my child to have a positive attitude toward learning
11. is doing a good job in helping my child understand his/her moral and ethnical
responsibilities
12. is friendly and approachable
13. is receptive to my input and suggestions
14. keeps me informed about cultural activities
15. respect me as a competent parent
16. is involved in the school progress of my child
17. care about my child
18. makes the most of what my child is able to do
19. are worthy my respect
20. will do what’s best for my child in the classroom
21. is willing to offer my child extra instruction
22. treat my child honest and right
23. involves me in making decisions
24. listens actively
25. takes me seriously when I ask a question
26. is open
27. have my child’s best interest at heart
28. pursues the same goals
29. supplies information that is correct and reliable
30. has a sufficient amount of time available when I want to discuss my concerns
31. keeps up appointments
32. takes responsibility for his/her own actions
33. is doing a good job in transmitting his/her skills and knowledge
34. is doing a good job in telling me how to help my child in the best manner
109
110
Chapter 6
general discussion
111
chapter 6
112
general discussion
The purpose of this chapter is to discuss the findings of this study and relate them to
recent developments in early and pre-school education. In 2002, the Radboud University
Nijmegen and the Childcare Centre Nijmegen (KION) in the Netherlands started a
longitudinal study to analyse the effects of their Language Stimulation program used
by KION at that time. Although the initial aim of this study was to investigate the effect
of their preschool program, it proved difficult to set up a proper experimental study. A
number of important conditions were lacking. An attempt to involve children who did
not participate in a preschool program failed, because no parents were willing to give
permission. Also the incompleteness of the cohort data over the years made it difficult to
determine the impact of the program in the long run. Moreover, except for 40 language
hours, many factors in the development of a child may influence (language) development.
However, important aspects of a limited language proficiency of low-income nativeDutch and minority children can be investigated. Therefore, the focus of this study was
on factors that may affect language development of children, including socio-emotional
development, the impact of cognitive factors and the effect of trust between parents and
teachers on language development.
While studying different aspects of the development of individual, educational,
and social factors in bilingual development of children from low socio-economical
backgrounds, one should remember that the development is the result of a complex
interaction of the child and its environment. Isolating single predictive factors is
therefore nearly impossible. This study has made an attempt to create an overview of a
number of different factors that may affect the development of a language delay, while
keeping our focus on cognitive factors (phonological awareness and working memory),
socio-emotional development, and trust between parents and teachers. This is how we
have attempted to create a comprehensive study of the different systems that interact
with one another. A developmental delay may originate from the co-existence of a
number of risk factors. This is what is known as risk-accumulation. This implies that
we should always observe a necessary level of nuance in phrasing our conclusions. First
we will summarize the findings of our study, followed by a number of implications and
recommendations.
Summary of results
Chapter 1 provided background information about Dutch children from low socioeconomical classes and discussed the assumption that pre-school education may increase
113
chapter 6
chances for disadvantaged children to enter primary school with sufficient background.
A short historical overview of Dutch pre-school policy is provided.
Furthermore, an overview is presented about the role of language development in
education, cognitive development, socio-emotional development and parent involvement.
Language development within this study is approached from an interactional
perspective, meaning that language develops through interaction, or language input.
The overview showed that language input is of great interest for vocabulary acquisition.
Language development in an interactional framework is not only a cognitive process
but also a highly social process, because language development arises within social
interactions. The overview showed that monolingual and bilingual children follow the
same developmental path for language acquisition, although there might be important
differences. It was emphasized that the quality and quantity of language input is of
importance too for language outcomes. Because language input is known to vary vastly
between families, social classes, and ethnic-cultural communities this could explain the
early rise of individual and group differences.
In the case of immigrant bilingual children not only the quantity of exposure is
likely to differ between the first and the second language, also the quality may reveal
profound differences in terms of vocabulary, grammar, and pragmatics. Studies about
language development in children from lower socio-economic backgrounds have shown
differences in language input and child-rearing styles between mothers from high and
from lower economic backgrounds. Differences in language input appear to be related
to their children’s language development. It was found that mainly the accumulation
of risk factors cause educational disadvantage. The fact that the school language is
their second language might be an additional risk factor for disadvantaged immigrant
children. Although school success seemed more related to SES than to ethnicity,
higher SES immigrant parents seemed, just like higher SES native Dutch parents, able
to stimulate their children and use those child rearing principles that stimulate their
children’s cognitive development. These children appeared to be better prepared for
primary school.
The overview also showed that immigrant children accumulate a larger educational
deficit than native-Dutch children from lower social economical background, which is
most likely the result of a language delay that existed before they started elementary
school. Child-, family-, and school characteristics seem to determine the level of
academic achievement. It is clear that language development is strongly dependent on
the opportunities for learning and language input provided by the environment of the
child. Also child characteristics for language processing, like memory or the hearing
abilities, and sound discrimination are needed. Two cognitive skills affected by language
development are phonological awareness and working memory.
114
general discussion
The overview raised important questions about pivotal factors for academic success
in children from low income and minority backgrounds, which were addressed in the
empirical studies of this thesis. The first question was about the role of verbal working
memory in learning Dutch as second language. Verbal-memory capacity has been found
to affect vocabulary and reading acquisition and is also important for language acquisition.
Second, the overview suggested that phoneme awareness is related to vocabulary
skills and reading (word decoding) in monolingual and bilingual children. With respect
to this study we were interested in the influence of phoneme awareness, because this is
one of the most important skills in learning to read an alphabetic orthography. Earlier
studies have shown that phoneme awareness is crucial for preliterate children learning
an alphabetic orthography.
The third question concerned social-emotional development, since social-emotional
development is one of the decisive factors for academic success. The overview showed that
language is acquired through social interaction or language input. Social development
in this view is of great importance for language development: A child has to learn to
communicate, to share, and direct intentions. Social and language development depend
on one another. Social skills are needed for adequate language development, because
language develops as a social process. The overview also showed that the quantity and
quality of language input is important for language outcomes. Because language input is
known to vary vastly between families, social classes, and ethnic-cultural communities
this could explain the early development of individual and group differences. Knowledge
about the relationship between limited language proficiency and social-emotional
problems in children who merely lag behind in the development of the dominant
language of the society they live in is lacking.
Finally, the importance of parent involvement was investigated. The overview suggested
that most parents, regardless of social background and ethnicity, do whatever they can to
ensure a successful academic career for their child. Parents often adhere to widely varying
‘theories on education’, which appear to vary greatly in terms of their ability to successfully
prepare a child for school. Parents may differ the way they rear their children. In families
form a high socio-economic background, the difference between playing and learning is
gradual; play is considered to be a part of the learning process.
During the past 40 years, the role of parents in the educational system has changed
drastically from relatively separate but shared responsibilities. It was found that children’s
social contexts/environments (family, school, community) are important predictors of
reading development. Especially at a young age parent involvement is expected to be of
particular importance. Academic performance of children is a dynamic and complex
process, of child characteristics, home variables, school variables, and home-school
partnership particularly for low income and minority children. The new millennium
115
chapter 6
has shown the emergence of partnership between parents and teachers. It is assumed
that when home, school, and community, accomplish a strong relationship and cooperate
with each other this may foster parental involvement. Therefore parents and teachers
must be willing to trust each other, because “partnership is based on mutual trust”.
Trust was found to be of critical importance for a strong home-school relationship.
Teachers’ trust also predicts academic performance. Because disadvantaged children
are more vulnerable to fail school, trust in these groups is even more important to help
them making school progress
Chapter 2 concerned the first empirical study of this thesis, examining the
relationship between verbal working memory and language skills in Turkish-Dutch
and native Dutch children from low-income families. We showed that language skills
and Dutch verbal-working memory were inferior for Turkish-Dutch children compared
to native-Dutch children. Verbal-working memory in the native-Dutch children was
unrelated to their language skills, whereas in the Turkish-Dutch children strong
correlations were found both between Turkish language skills and Turkish verbalworking memory and between Dutch language skills and Dutch verbal-working
memory. Limited exposure to a language appears to determine verbal-working memory
capacity in that language. The results revealed that language knowledge influences
performance on memory tasks. The Turkish-Dutch children show limited proficiency
in both languages, which may explain their low performance on the complex verbalworking memory tasks. This indicates that the relationship between language skill
and verbal-working memory is reciprocal. These complex tasks require good language
proficiency, like syntactic sensitivity and options for chunking and integrating verbal
information. Limited proficiency in a language may therefore limit verbal-working
memory and may slow down language acquisition. A second indication for the reciprocal
relationship between language and verbal-working memory was provided by the strong
correlations between Turkish language skills and Turkish verbal-working memory,
and between Dutch language skills and Dutch verbal-working memory in the TurkishDutch children. Although, in the native-Dutch children language skills appeared to be
unrelated to verbal-working memory skills. These findings suggest that a minimum
level of language development is required to strengthen verbal-working memory skills
and experience with a particular language determines, at least partly, the capacity of
verbal-working memory in that language.
Chapter 3 investigated the relationship between phoneme awareness, vocabulary
and reading (word decoding) in monolingual and bilingual Dutch children. There were
no significant differences on the phoneme awareness tasks between the Dutch children
116
general discussion
and the entire group of bilingual children or the subsample of Turkish-Dutch children.
The results indicated that there was neither an advantage nor a disadvantage for being
bilingual in phoneme awareness at grade one. Although the present study showed
that Turkish children lag behind their monolingual peers in vocabulary, note that the
two groups showed comparable scores on phoneme awareness and word decoding.
No significant correlations between the phoneme awareness tasks and productive
vocabulary or reading skills were obtained. The findings of our study also revealed
that the Turkish-Dutch children obtained a higher score on phoneme-segmentation
and final-phoneme deletion when tested in Dutch. Phoneme awareness develops in
response to reading instruction. Phoneme awareness of Turkish children in Dutch may
be better than in Turkish, simply because they have learnt to read in Dutch and that
Dutch was becoming their dominant language.
The general conclusion of this study is that these children do not profit or suffer
from knowing two languages when it concerns the acquisition of phoneme awareness.
Not only was their performance on phoneme-awareness similar to that of bilingual
children, their word-decoding performance was too. Although Dutch vocabulary of
the Turkish-Dutch subsample was poorer than those of the native Dutch speakers,
their language proficiency in their adopted language was better than in their native
language Turkish.
Chapter 4 addressed the influence of a limited language proficiency on the
development of social-emotional and/or behaviour problems of children from lowincome and minority backgrounds over the course of four years. The results of the
teacher ratings demonstrated that children with a more limited language proficiency,
that is, immigrant-Dutch children, do not develop more socio-emotional or behaviour
problems than native-Dutch children. Except for withdrawn behaviour in year 3, the
immigrant-Dutch children showed more withdrawn behaviour than their native-Dutch
peer, and the native-Dutch peers displayed more aggressive behaviour, according to their
teachers. In year 4 native-Dutch children showed more anxious depressed and aggressive
behaviour than immigrant-Dutch children. In our sample none of the correlations
between the scores on the language-comprehension test were related to the scores on
the socio-emotional scales. Limited language proficiency is not necessarily detrimental
for the wellbeing of children from low-income and/or immigrant backgrounds.
Chapter 5 aimed at investigating trust between parents and teachers of children
from low-income and minority backgrounds in first grade. The study on the relationship
of trust between parents and teachers has shown that the perception of parents in
relation to trust does not correspond to the teacher’s perception. The parents have
117
chapter 6
more trust in the reliability, competence, and honesty of the teacher than the teacher
has in the parents. A possible explanation is that parents must have some basic trust
level in teachers, because the teacher is a professional and they leave their child in
confidence with a stranger. Teachers’ trust remains longer at a predictability level,
because they first have to judge parents trustworthiness. The results also demonstrated
that teachers have, on average, a higher level of trust in native Dutch parents than they
do in immigrant parents. This warrants the conclusion that teacher’s trust in parents
is affected by parents’ ethnicity, whereas parents’ ethnicity does not depend on their
trust in the teacher. The results further indicated that the parents’ trust in the teacher
has no predictive value in terms of reading development, whereas, teacher’s trust in the
parents’ openness and their willingness to co-operate does appear to have a predictive
value in terms of the development of reading skills. Children were found to have higher
reading performance when teachers reported higher trust.
Implications and Recommendations for Practice
The implications and recommendations of the thesis are discussed with respect to the
different aspects studied in this thesis: Cognitive factors, socio-emotional development,
and trust between parents and teachers on language development.
Cognitive factors
The fact that the immigrant children who participated in the current study are
bilingual did not give them an advantage or disadvantage in terms of their phonological
awareness compared to their native Dutch peers at the end of first grade. To test language
transfer, the bilingual group should have received both reading instruction in their first
and in their second language, in order to measure whether first-language instruction
contributed to second-language learning. Future research on cross-lingual transfer
would be relevant and should include first-language reading proficiency, and compare
performance of Turkish children receiving first-language and second-language reading
instruction with Turkish children who only receive second-language instruction.
Results of cultural communities studied by Verhoeven (2007), Leseman (2000), and
Scheele, Leseman, and Mayo (2010) which were similar to the one in the present study,
support the view that there is cross-language interdependence. That is, transfer from the
first to the second language indicates that development in the first language advances
development in the second language.
These studies also report that the average level of the first and second language
proficiency of bilingual immigrant children (including Turkish-Dutch 4- to 6-year-
118
general discussion
olds) is (far) below that of native Dutch peers, based on a between-groups comparison.
Leseman (2000, p. 110) explains this difference for the Turkish children “because their
first language did not reach a sufficient mature level regarding the development of deep,
decontextualized lexical-conceptual knowledge, language analysis and metalinguistic
skills to have second language acquisition benefit from first language proficiency”.
According to Leseman (2001), the effectiveness of this immersion policy is questionable,
considering the persistent problems faced by bilingual students. Both Leseman (2000)
and Verhoeven (2007) encourage bilingual education for young immigrant children in
order to enhance their first language proficiency. It is believed that basic language skills
indirectly enhance language development in the second language. Some experts advocate
a strategy that further develops the first language (Leseman, 2000; Verhoeven, 2007).
According to them, the skills acquired by the child in the context of the first language are
important skills for learning a second language (transition-sequence model).
In contrast to this, we did not find an advantage or disadvantage in terms of our
bilingual children’s phonological awareness, but there was a relationship between
language skills and working memory. This relationship entails that children who were
raised bilingually in a linguistically poor environment, resulted in a language delay,
are in fact, faced with a double disadvantage. Not only do they have a language delay in
terms of the language that is used in their school, they are probably restricted in their
ability to learn that language as well. This could also result in difficulty understanding
(and following) instructions at school, if these instructions are given verbally.
A large number of studies have shown that poor working memory capacity is an
important precursor to learning difficulties. Children with limited working memory
capacity may have difficulty following instructions, keeping an eye on the overall
picture while performing complex tasks, dealing with simultaneous processing of
information and long-term information retention. Based on these problems, these
children appear to be in need of extended instruction and practice. Whether this need
will persist is unclear. It does seem important to be mindful of individual differences
in terms of learning, in order to allow the child to take optimal advantage of the
educational system, as it appears equally important to adapt instructions to this reality
as much as possible. It might also be effective to involve some kind of working-memory
training within a preschool program. The preschool period is after all eminently
characterized as one - neurobiological - sensitive period ideally suited for stimulating
development and guidance of children (Leseman, 2001; Shonkoff & Phillips, 2000).
In order to prepare children for formal learning instruction of elementary school,
most preschool programs aim at improving lingual and cognitive development. In this
view, working memory can be considered a cognitive skill and would fit easily into a
preschool program. The program ‘Tools of the Mind’ is a program that falls within the
119
chapter 6
category of educational pre-school programs, aimed at improving executive functions
(Diamond, Barnett, Thomas, & Munro, 2007). Diamond et al. found that the executive
functions of 4- to 5-year-old children from lower socio-economic backgrounds improved
by taking part in ‘Tools of the Mind’.
Socio-emotional development
The limited language proficiency in Dutch was not related to socio-emotional
problems. Our findings do not support studies by others, who found that children with
a language delay show more withdrawn behaviour (e.g., Fujiki, Brinton, Isaacson, &
Summer, 2001; Glover-Gagnon & Nagle, 2004; Irwin, Carter, & Briggs-Gowan, 2002;
Noterdaeme & Amorosa, 1999) and studies that reveal that preschool children with a
specific language impairment (SLI) show more social problems (more internalizing
behavioural problems), as reported by their parents than from children without a
language impairment (Stanton-Chapman, Justice, Skibbe, & Grant, 2007). Note that
these children were from middle SES families. Our findings indicated that the children
from the present study do not have a language impairment. Many factors determine
school success of children, including psychological, social and cultural conditions.
Also the quality and quantity of language input is of great importance for language
and cognitive development. As stated in the general introduction, language input is of
utmost importance for language acquisition. Although we refer in the introduction to
a language delay or impairment, studying the development of disadvantaged nativeDutch and immigrant-Dutch children we realized that that their language proficiency
is better described as limited rather than delayed or impaired, because circumstances,
such as a relatively poor language input, may be the reason why their language skills
are limited. A delay or even impairment suggests a developmental problem, whereas a
limitation refers to insufficient experience with the language. Between the ages of 3 to 6
it is hard to establish whether they have a developmental delay rather than just a limited
development due to limited exposure.
Trust
It has been shown that not only communication, but also trust plays an important
part in children’s success rate in school. Trust between parents and teachers appears
rather important in terms of constructing a productive partnership. Trust in teachers
has proven to be influential on the reading skills of students in the first grade. Also,
teachers were found to have more trust in native-Dutch parents than in immigrant
parents. In order to improve this relationship of trust, it is important that teachers
display a basic level of trust in all parents, regardless of their ethnicity, so that the child’s
education can become the common objective. Stimulating parental involvement in pre-
120
general discussion
school education should be based on partnership, and partnership is based on mutual
trust. In establishing an appropriate level of partnership, improving teachers’ abilities
to communicate with parents from different cultural backgrounds may prove useful
(Bakker & Denessen, 2007; Denessen, Bakker, & Gierveld, 2007; Smit, Driessen, &
Doesborgh, 2005). A situation in which teachers feel more confident in communicating
with immigrant parents, a process in which a language barrier or different cultural
values in terms of communicating might be inhibiting factors, may foster more mutual
understanding, which may decrease the distance parents perceive, which may in turn lead
to a more effective partnership. An open method of communication, mutual respect, the
willingness to listen and honesty seem to be necessary. Parents who trust their teacher
will be more easily convinced that the teacher is competent, honest and reliable, and
that he or she will do whatever it takes to make sure their child performs well in school.
Improving partnership between parents and schools is important. Engaging parents as
partners in educating children requires understanding of parental perceptions about
education, their aspirations and ambitions, their view on parenting and expectations
towards teachers and school, about raising children and responsibilities. The new
millennium has shown the emergence of partnership between parents and teachers of
which trust was found to be of critical importance for a strong home-school relationship
(Adams & Christenson, 1998; Goddard, Tschannen-Moran, & Hoy, 2001; Hoy & Tarter,
2004; Karoly & Gonzalez, 2011; Moore & Lasky, 1999).
General implications and recommendations
Based on theories about second language acquisition (see Cummins, 1991) as well
as research into bilingual education state that immersion in the second language
from preschool on is too early. According to Leseman (2001), the effectiveness of this
immersion policy is questionable, considering the persistent problems faced by bilingual
students. The immigrant-Dutch children in these studies usually learn to speak the
language of their parents at home. Substantial and prolonged exposure to Dutch as a
second language does not start before they enter kindergarten. This usually results in a
large group of children from minority backgrounds with insufficient knowledge of the
Dutch language, which means that they cannot optimally benefit from formal education
in reading, spelling, and mathematics (Elzer, 2005; van Elten, 2003). Both Leseman
(2000) and Verhoeven (2007) suggest bilingual education for young immigrant children
in order to enhance their first language proficiency. It is believed that basic language
skills indirectly enhance language development in the second language. According to
them, the skills acquired by the child in the context of the first language are important
skills for learning a second language (transition sequence model).
On the other hand, immigrant second language learners may be dependent on limited
121
chapter 6
input in an immersion environment, and may hear and use both languages in more
restricted settings. There may also be a negative transfer, or subtractive bilingualism,
referring to a situation in which the time spend on learning a second language leads
to a loss of language learning of the first language (Butler & Hakuta, 2004). Leseman
(2000) and Scheele et al. (2010) found weak or absent cross-lingual transfer in TurkishDutch children. They argued that the quality of the Turkish language environment was
overall low in the Turkish families and did not function as a buffer against subtractive
bilingualism. Their first language skills should be further developed before they should
learn the second language in order to attain cross-lingual transfer.
Many factors determine school success of children, including psychological, social
and cultural conditions. Also the quality and quantity of language input are of great
importance for language and cognitive development. Although bilingual pre-school
centre-based programs seem to be a solution against subtractive bilingualism, there
are some major practical concerns, like the diversity of children, the absence of native
speaking teachers. Also home-based bilingual programs are influenced by cultural
and socio-economic factors (Leseman & Van Tuijl, 2001). Despite these explanations of
second language learning, the question remains whether bilingual education for this
specific population is truly a solution against school failure. Notwithstanding the fact
that language is important, it remains only one of many factors influencing academic
performance of disadvantaged children.
This study does not justify generalized policy recommendations. The results are
based on a relatively small sample, and the studies concerning first- and second language
performance were limited to only Turkish-Dutch children. The question remains whether
similar effects are to be found in children from other ethnic groups. Nevertheless,
considering our results in relation to cognitive (language) tasks, and the importance of
effective partnership between schools and parents, offering pre-school education seems
necessary and needed. Offering additional support within the preschool programs, as
in the Dutch educational priority policy, seems to be in accordance with that concept.
Partnership might contribute to the decrease or prevention of educational disadvantage.
Within the current Dutch policy, intensive language education in elementary schools by
ways of transitional classes or extra language hours, all the while working on improving
partnership and trust between parents and schools, is a policy that appears to be supported
by the results of this study. This study is also supportive of the policy of considering
parents’ education level, rather than their country of origin, as a determinative factor
in attributing relative weight and risk of (development of) educational delays. After all,
when no other risk factors are present, the parents’ country of origin does not appear
to cause an educational delay. The emphasis on parental involvement and increasing
effective partnership seem to fit within the Dutch educational priority policy, because it
122
general discussion
is aimed at offering comprehensive pre-school programs, in which parental involvement
plays an important role as well.
Empirical research is needed to further clarify the results of this thesis. This research
highlights the fact that research assessing the effects of preschool programs or predictors
of language development within a specific group has generally adopted approaches that
do not reflect the contextual complexity and dynamics within which risk factors for
language delays, or limited language proficiency, emerge. That is, research should include
the children’s social contexts, family system, school and community, in which child risk
factors might emerge. It is possible that a child’s school performance not only reflects
child or family characteristics, but also the quality of the school, teacher’s interactions,
or social factors may influence academic development. Future research needs to adopt a
broader contextual approach in developing models of language development in children
from low-income and/or minority backgrounds. Specifically, in addition to child and
parent characteristics, information on parenting practices, the school environment, and
larger demographic and social factors influencing parenting or teachers practices needs
to be obtained and assessed in research to language development in children from low
income and minority backgrounds.
To improve participation of parents they deserve extra support, especially those who
tend to drop out in an early stage or participate only minimally, in order to minimize
the incompleteness of the cohort data over the years. The question remains why these
parents did not cooperate, did they understand the materials, did they understand the
larger concept of the study to positively affect the school carriers of children from lowincome and/or minority backgrounds? Understanding this process might increase
participation in the future. Possibly a more personal way of maintaining contact with the
families, with someone from within their own community, would in fact increase longterm participation. Especially during follow-up the help from within the community
might increase participation. Also a flexible attitude of the researchers is needed,
because appointments are not well complied with, telephone numbers or address change
or letters are not read. Future researchers will face a challenge in improving engagement
of low-income families in proper methodological longitudinal designs. Longitudinal
research is important to observe whether the findings of this study in cognitive and
social-emotional development and the influence of trust on language development affect
the school careers of these children.
Although the non-significant findings in this study for behaviour problems and the
finding that these children do not profit or suffer from knowing two languages when it
concerns the acquisition of phoneme awareness, are promising results for low-income
children. Of course this result is no guarantee that problems will not occur later in life.
In fact, it is unknown if and how working-memory problems develop when proficiency
123
chapter 6
of the Dutch language increases or whether working-memory deficits in first grade
might influence academic performance in long term. It is also not known whether low
levels of trust or even distrusting relationships in first grade affect the socio-emotional
development or predicts academic performance of these children. The children in our
study could still be at risk for developing socio-emotional or behavioural problems or
school dropout. Longitudinal research on this topic is needed to reveal predictors for
school dropout or other developmental problems of low-SES children in order to prevent
the emergence of problems later in life.
124
general discussion
References
Adams, K. S., & Christenson, S. L. (1998). Differences in parent and teacher trust
levels: Implications for creating collaborative family-school relationships. Special
Services in the Schools, 14, 1-22. doi: 10.1300/J008v14n01_01
Bakker, J., & Denessen, E. (2007). The concept of parent involvement. Some
theoretical and empirical considerations. In H. Phtiaka & S. Symeonidou (Eds.),
Schools and families in partnership: Looking in the future (pp. 238-252). Nicosia,
Cyprus: ERNAPE.
Butler, Y. G., & Hakuta, K. (2004). Bilingualism and second language acquisition.
In T. Bhatia & W. C. Ritchie (Eds.), The handbook of bilingualism (pp. 114–144).
Malden, MA: Blackwell.
Cummins, J. (1991). Conversational and academic language proficiency in bilingual
contexts. AILA Review, 8, 75-89.
Denessen, E., Bakker, J., & Gierveld, M. (2007). Parental involvement in multi ethnic
schools. The School Community Journal, 17, 27-44.
Diamond, A., Barnett, W. S., Thomas, J., & Munro, S. (2007). Preschool program
improves cognitive control. Science, 318, 1387-1388. doi: 10.1126/science.1151148
Elzer, M. J. (2005). Talen naar een moedertaal [Care for mother tongue]. Tijdschrift voor
Orthopedagogiek, 3, 130-135.
Fujiki, M., Brinton, B., Isaacson, T., & Summer, C. (2001). Social behaviors of children
with language impairment on the playground: A pilot study. Language, Speech, and
Hearing Services in Schools, 32, 101-113. doi: 10.1044/0161-1461(2001/008)
Glover-Gagnon, S., & Nagle, R. J. (2004). Relationships between peer interactive play
and social competence in at-risk preschool children. Psychology in the Schools, 41,
173-189. doi: 10.1002/pits.10120
Goddard, R. D., Tschannen-Moran, M., & Hoy, W. K. (2001). A multilevel examination
of the distribution and effects of teacher trust in students and parents in urban
elementary schools. The Elementary School Journal, 102, 3-17. doi: 10.1086/499690
Hoy, W. K., & Tarter, C. J. (2004). Organizational justice in schools: No justice
without trust. International Journal of Educational Management, 18, 250-259. doi:
10.1108/09513540410538831
Irwin, J. R., Carter, A. S., & Briggs-Gowan, M. J. (2002). The social-emotional
development of ‘late talking’ toddlers. Journal of the American Academy of Child
and Adolescent Psychiatry, 41, 1324-1332. doi: 10.1097/00004583-20021100000014
Karoly, L. A., & Gonzalez, G. C. (2011). Early care and education for children in
immigrant families. The future children, 21, 71-101. doi: 10.1353/foc.2011.0005
125
Leseman, P. P. M. (2000). Bilingual vocabulary development of Turkish preschoolers
in the Netherlands. Journal of Multilingual and Multicultural Development, 21, 93112. doi: 10.1080/01434630008666396
Leseman, P. P. M. (2001). Aanzetten voor onderzoeksprogrammering in de voor- en
vroegschoolse periode. [Sketches of research programs during preschool and early
learning]. Amsterdam, the Netherlands: SCO-KI.
Leseman, P. P. M., & Van Tuijl, C. (2001). Home support for bilingual development
of Turkish 4-6-year-old immigrant children in the Netherlands: Efficacy of a
home based educational programme. Journal of Multilingual and Multicultural
Development, 22, 309-324. doi: 10.1080/01434630108666438
Moore, S., & Lasky, S. (1999). Parent involvement in education: Models, strategies and
contexts. Commissioned by The Educational Quality and Accountability Office.
Toronto, Canada: Ontario Institute for Studies in Education, University of Toronto.
Noterdaeme, M., & Amorosa, H. (1999). Evaluation of emotional and behavioural
problems in language impaired children using the Child Behaviour Checklist.
European Child & Adolescent Psychiatry, 8, 71-77. doi: 10.1007/s007870050087
Scheele, A. F., Leseman, P. P. M., & Mayo, A. Y. (2010). The home language
environment of mono- and bilingual children and their language proficiency.
Applied Psycholinguistics, 31, 117-140. doi: 10.1017/S0142716409990191
Shonkoff, J. P., & Phillips, D. A. (2000). From neurons to neighborhoods: The science of
early childhood development. Washington, DC: National Academy Press.
Smit, F., Driessen, G., & Doesborgh, J. (2005). Opvattingen van allochtone ouders over
onderwijs: tussen wens en realiteit [Opinions of immigrant parents about education:
Between wishes and reality]. Nijmegen, the Netherlands: ITS.
Stanton-Chapman, T. L., Justice, L. M., Skibbe, L. E., & Grant, S. L. (2007).
Social and behavioural characteristics of preschoolers with specific language
impairment. Topics in Early Childhood Special Education, 27, 98-109. doi:
10.1177/02711214070270020501
van Elten, M. (2003). Schoolprestaties van allochtone kinderen [Academic achievement
of immigrant children]. Onderwijsmagazine Inzicht. Retrieved june, 2009, from
http://www.voo.nl/inzicht.html
Verhoeven, L. (2007). Early bilingualism, language transfer, and phonological
awareness. Applied Psycholinguistics, 28, 425-439. doi: 10.1017/S0142716407070233
126
Nederlandse samenvatting
(summary in dutch)
127
nederlandse samenvatting
128
summary in dutch
In 2002 startten de Radboud Universiteit Nijmegen en Kinderopvang Nijmegen (KION)
een longitudinaal onderzoek naar de resultaten van het taalstimuleringsprogramma dat op
dat moment door KION werd gebruikt. Het oorspronkelijke doel van deze studie was een
onderzoek naar het effect van hun voorschoolse programma. Het bleek echter moeilijk,
zo niet onmogelijk, om een goede experimentele studie op te zetten, omdat een aantal
belangrijke voorwaarden hiervoor ontbrak. Daarom is er voor gekozen om de aandacht
uit te laten gaan naar factoren die de ontwikkeling van de kinderen kunnen beïnvloeden,
waaronder de taal, de cognitieve en de sociaal-emotionele ontwikkeling, en de band tussen
ouders en leerkrachten.
Wie de verschillende aspecten van de taalontwikkeling van tweetalige kinderen uit
de lagere sociaaleconomische milieus wil bestuderen, weet zich geconfronteerd met een
ingewikkeld samenspel van factoren van individuele, opvoedkundige en sociale aard.
Taalontwikkeling vindt immers plaats onder de voortdurende invloed van een complex
samenspel van factoren die niet alleen het kind zelf betreffen, maar ook diens directe
sociale omgeving en alle mogelijke interacties tussen die beide. Het isoleren van een
enkele voorspellende factor is dan ook vrijwel onmogelijk.
In deze studie is een poging gedaan een aantal factoren die van invloed kunnen zijn
op de taalontwikkeling nader te beschouwen en te onderzoeken. Hierbij zijn cognitieve
factoren zoals het fonologisch bewustzijn en het werkgeheugen in ogenschouw genomen,
maar ook de socio-emotionele ontwikkeling en de vertrouwensband tussen ouders en
leerkrachten. Een beperkte (taal)ontwikkeling kan ontstaan vanuit een aantal risicofactoren
die gemeenschappelijk voorkomen en een optimale ontwikkeling in de weg kunnen staan.
Dat samen voorkomen van risicofactoren wordt ook wel risico-accumulatie genoemd. Dit
betekent dat bij het trekken van conclusies op het gebied van één enkele factor, altijd een
bepaalde mate van voorzichtigheid in acht moet worden genomen.
In het eerste hoofdstuk van dit proefschrift worden de achtergronden geschetst van
Nederlandse kinderen uit lagere sociaaleconomische milieus en wordt stilgestaan bij de
veronderstelling dat voor- en vroegschoolse educatie (VVE) de onderwijskansen van deze
kinderen in het basisonderwijs doet vergroten. Het biedt een kort historisch overzicht
van het Nederlandse VVE-beleid en verschaft inzicht in de oorsprong van onderhavig
onderzoek. In dat verband wordt stilgestaan bij de rol die de taalontwikkeling in het
onderwijs speelt, bij de cognitieve en sociaal-emotionele ontwikkeling, alsmede bij de
onderwijsbetrokkenheid van de ouders. Taalontwikkeling binnen deze studie wordt
benaderd vanuit een interactioneel perspectief, wat inhoudt dat taal zich ontwikkelt
dankzij interactie en taalaanbod. Uit het overzicht blijkt dat taalaanbod van groot belang is
voor de ontwikkeling van de woordenschat.
129
nederlandse samenvatting
Vanuit interactioneel perspectief bezien is taalontwikkeling niet slechts een cognitief,
maar ook een sociaal proces aangezien deze zich in interactie met de sociale omgeving
voordoet. Uit het overzicht blijkt dat eentalige en tweetalige kinderen op het gebied van
taalverwerving in beginsel eenzelfde ontwikkeling doormaken. Desondanks zijn er
belangrijke verschillen te constateren. Benadrukt wordt dat zowel de kwantiteit als de
kwaliteit van het taalaanbod van belang zijn voor de taalontwikkeling. Taalaanbod kan
sterk variëren tussen sociale klassen en etnisch-culturele gemeenschappen, waardoor
individuele en groepsverschillen verklaard kunnen worden. In het geval van allochtone
tweetalige kinderen kan niet alleen de hoeveelheid blootstelling aan de eerste en tweede
taal zorgen voor verschillen, maar kan ook de kwaliteit van het taalaanbod diepgaande
verschillen veroorzaken in termen van woordenschat, grammatica en pragmatiek.
Uit onderzoek is gebleken dat vooral de accumulatie van risicofactoren een
onderwijsachterstand veroorzaakt. Het feit dat de schooltaal veelal hun tweede taal
is, zou voor kansarme allochtone kinderen een bijkomende risicofactor kunnen zijn.
Hoewel schoolsucces meer gerelateerd leek aan sociaaleconomische status (SES) dan
aan etniciteit, bleken allochtone ouders meteen hogere SES net als autochtone ouders
met een hogere SES-achtergrond in staat om hun kinderen optimaal te stimuleren
en gebruik te maken van opvoedingsprincipes die de cognitieve ontwikkeling van
hun kinderen bevorderen. Deze kinderen bleken adequater voorbereid op het formele
leren op de basisschool dan de kinderen met een migrantenachtergrond uit de lagere
milieus. Uit het overzicht blijkt eveneens dat allochtone kinderen in het algemeen
een grotere onderwijsachterstand vertonen dan autochtone Nederlandse kinderen uit
vergelijkbare milieus, dat waarschijnlijk het gevolg is van hun taalachterstand die reeds
bestond voordat zij naar de basisschool gingen. Kind-, gezins- en schoolkenmerken
lijken bepalend te zijn voor de schoolprestaties van kinderen. Taalontwikkeling is sterk
afhankelijk van de mogelijkheden die een kind heeft om te leren en de kwaliteit van het
taalaanbod. Daarnaast zijn kenmerken van het kind zoals geheugen en het vermogen tot
klankdiscriminatie onontbeerlijk. Twee cognitieve vaardigheden die taalontwikkeling
beïnvloeden zijn fonologisch bewustzijn en werkgeheugen. Vanuit het overzicht kwamen
belangrijke vragen naar voren aangaande de bepalende factoren voor schoolsucces van
kinderen uit lage SES- en minderheidsgroepen, die in de empirische studies van dit
proefschrift nader werden onderzocht.
De eerste vraag betrof de rol van het verbaal-werkgeheugen bij het leren van
Nederlands als tweede taal. Uit onderzoek is gebleken dat het verbaal werkgeheugen
invloed heeft op de woordenschatontwikkeling, het leren lezen en tevens van belang is
voor de taalverwerving.
Ten tweede komt naar voren dat het fonemisch bewustzijn gerelateerd is aan de
ontwikkeling van de woordenschat en het leren lezen (woord decoderen). Dit gaat voor
130
summary in dutch
zowel de één- als tweetalige kinderen op. Met betrekking tot dit onderzoek waren we met
name geïnteresseerd in de invloed van het fonemisch bewustzijn, omdat dit, zoals eerder
onderzoek aantoonde, een van de belangrijkste voorwaarden voor het leren lezen is.
De derde vraag betreft de vraag naar de sociaal-emotionele ontwikkeling, omdat
gebleken is dat sociaal-emotionele ontwikkeling een doorslaggevende factor kan
zijn voor het latere schoolsucces. Uit het overzicht blijkt dat taal wordt verworven in
interactie met de sociale omgeving en dankzij taalaanbod. Vanuit dit perspectief bezien
is de sociale ontwikkeling van eminent belang voor de ontwikkeling van taal; een kind
moet leren om te communiceren. Taalontwikkeling en sociale ontwikkeling van elkaar
afhankelijk. Tot dusver ontbreekt het echter aan inzicht in de mogelijke relatie tussen de
beperkte taalvaardigheid van kinderen uit de lagere SES- en migrantenmilieus en hun
sociaal-emotionele ontwikkeling.
Tot slot wordt het belang van ouderlijke betrokkenheid bij de taalontwikkeling aan
nader onderzoek onderworpen. Uit de overzichtsstudie blijkt dat alle ouders, ongeacht
hun sociale achtergrond en etniciteit, alles in het werk stellen om de schoolloopbaan
van hun kind zo voorspoedig mogelijk te laten verlopen. Dat sommige ouders daarin
succesvoller dan andere zijn lijkt erg afhankelijk van de opvoedingscultuur thuis. De
één lijkt adequater op de schoolloopbaan voor te bereiden dan de ander. Doorgaans
worden dergelijke verschillen in opvoedingscultuur in verband gebracht met verschillen
in sociaal-economische status, maar verschillen in cultureel-etnische achtergrond doen
zich minstens zo sterk gevoelen.
De afgelopen 40 jaar is de rol van ouders binnen het onderwijs drastisch veranderd, van
een relatief gescheiden naar een gedeelde verantwoordelijkheid. De betrokkenheid van
ouders bij het onderwijs wordt meer en meer gezien als een belangrijke voorwaarde voor
een succesvolle schoolloopbaan. Vrij algemeen wordt aangenomen dat wanneer ouders,
school en de gemeenschap een sterke relatie opbouwen en met elkaar samenwerken, dit
de betrokkenheid van ouders vergroot en de leerprestaties van kinderen doet bevorderen.
Een kritische voorwaarde voor samenwerking is vertrouwen; vertrouwen tussen ouders
en leerkracht. Aangenomen mag dan ook worden dat wanneer ouders en leerkrachten
elkaar vertrouwen, dit de schoolprestaties van de kinderen ten goede komt.
Hoofdstuk 2 betreft de eerste empirische studie van dit proefschrift waarin de
relatie tussen het verbaal-werkgeheugen en taalvaardigheid van Turks-Nederlandse
en autochtone Nederlandse kinderen uit de lagere sociaaleconomische milieus
wordt onderzocht. We toonden aan dat de taalvaardigheid en de capaciteit van het
Nederlands verbaal-werkgeheugen van Turks-Nederlandse kinderen zwakker
waren dan die van autochtoon-Nederlandse kinderen. Verbaal-werkgeheugen bij
de autochtone Nederlandse kinderen was niet gerelateerd aan hun taalvaardigheid,
131
nederlandse samenvatting
terwijl bij de Turks-Nederlandse kinderen sterke correlaties gevonden werden
tussen zowel Turkse taalvaardigheid en de verbaal- werkgeheugenprestaties in het
Turks enerzijds en anderzijds tussen de Nederlandse taalvaardigheid en de verbaalwerkgeheugenprestaties in het Nederlands. Een beperkte blootstelling aan een
specifieke taal lijkt de capaciteit van het verbaal-werkgeheugen in die taal te bepalen.
Uit de resultaten bleek dat taalkennis de prestaties op geheugentaken beïnvloedt. De
Turks-Nederlandse kinderen laten zowel in het Turks als in het Nederlands een beperkte
vaardigheid zien, die hun lage prestaties op de complexe verbaal-werkgeheugen taken
kan verklaren. Daarmee wordt gesuggereerd dat de relatie tussen taalvaardigheid
en verbaal-werkgeheugen wederkerig is. De complexe taken vereisen een goede
taalvaardigheid, zoals syntactische gevoeligheid en het integreren van verbale
informatie. Beperkte beheersing van een taal kan dus het verbaal-werkgeheugen
beïnvloeden wat vervolgens de taalverwerving kan vertragen.
Een tweede indicatie voor de wederkerige relatie tussen taal en verbaal-werkgeheugen
zijn de sterke correlaties bij Turks-Nederlandse kinderen tussen hun Turkse
taalvaardigheid en hun verbaal-werkgeheugenprestaties in het Turks aan de ene kant
en tussen hun Nederlandse taalvaardigheid en hun verbaal- werkgeheugenprestaties
in het Nederlands aan de andere kant. Bij de autochtoon-Nederlandse kinderen bleek
de taalvaardigheid niet gerelateerd aan het verbaal-werkgeheugen. Dit suggereert dat
een minimumniveau van taalontwikkeling vereist is om het verbaal-werkgeheugen te
versterken. Klaarblijkelijk bepaalt ervaring met een bepaalde taal, althans gedeeltelijk,
de capaciteit van het verbaal-werkgeheugen in die taal.
In Hoofdstuk 3 wordt de relatie tussen fonemisch bewustzijn, woordenschat en lezen
(woord decoderen) bij eentalige en tweetalige Nederlandse kinderen nader onderzocht.
Bij de fonemisch-bewustzijnstaken konden geen significante verschillen tussen de
autochtoon-Nederlandse kinderen en de hele groep van tweetalige kinderen noch tussen
hen en de subgroep van Turks-Nederlandse kinderen vastgesteld worden. Dat neemt
overigens niet weg dat Turks-Nederlandse kinderen bij hun eentalige leeftijdsgenoten
op het gebied van woordenschat achterblijven, er rekening mee houdend dat de twee
groepen vergelijkbare scores vertoonden op de fonemisch-bewustzijnstaken en woorddecoderen (lezen). Er zijn geen significante correlaties gevonden tussen de fonemischbewustzijnstaken en productieve woordenschat of leesvaardigheden. Ook is gebleken
dat de Turkse kinderen een hogere score behaalden op foneem-segmentatie en foneemdeletie bij testen in het Nederlands. Foneembewustzijn ontwikkelt zich in reactie op het
leesonderwijs. Foneembewustzijn van Turkse kinderen in het Nederlands kan beter zijn
ontwikkeld dan in het Turks, simpelweg omdat ze hebben leren lezen in het Nederlands
en Nederlands steeds meer hun dominante taal wordt. De algemene conclusie van dit
132
summary in dutch
onderzoek is dat deze kinderen geen voor- of nadeel ervaren van tweetaligheid als het
gaat om het verwerven van een fonemisch bewustzijn. Niet alleen waren hun prestaties
op fonemisch-bewustzijnstaken vergelijkbaar met die van eentalige kinderen, ook
hun leesprestaties waren vergelijkbaar. Hoewel de Nederlandse woordenschat van de
Turks-Nederlandse kinderen geringer was dan die van de eentalige kinderen, bleek hun
taalvaardigheid in het Nederlands beter dan die in het Turks.
In Hoofdstuk 4 wordt ingegaan op de invloed van een beperkte taalvaardigheid op de
ontwikkeling van eventuele sociaal-emotionele en / of gedragsproblemen in een periode
van vier jaar. Uit de resultaten bleek dat naar de inschatting van de leerkrachten de
kinderen met een meer beperkte taalvaardigheid, de allochtoon-Nederlandse kinderen,
niet meer sociaal-emotionele of gedragsproblemen vertonen dan de autochtone
kinderen. Alleen bleken de allochtone kinderen in jaar 3 meer teruggetrokken gedrag
te vertonen dan hun autochtone leeftijdgenoten, en vertoonden de autochtone kinderen
naar het oordeel van hun leerkracht meer agressief gedrag. In jaar 4 vertoonden de
autochtone kinderen meer angstig depressief en agressief gedrag dan allochtone
kinderen. Binnen onze steekproef waren geen van de correlaties tussen de scores op de
taaltoets gerelateerd aan de scores op de schalen voor sociaal-emotioneel gedrag. Een
beperkte taalvaardigheid is klaarblijkelijk niet per definitie van invloed op het welzijn
van kinderen uit lage SES en / of van allochtone afkomst.
In Hoofdstuk 5 staat de vertrouwensband tussen ouders en leerkrachten van kinderen
uit groep 3 centraal. Belangrijkste uitkomst van ons onderzoek is nog wel dat het
vertrouwen bepaald niet wederzijds is. De ouders stellen beduidend meer vertrouwen in
de betrouwbaarheid, competentie en eerlijkheid van de leerkracht dan dat leerkrachten
in deze opzichten de ouders vertrouwen. Een mogelijke verklaring hiervoor is dat de
ouders een basisniveau van vertrouwen moeten hebben in een leerkracht. De leerkracht
heet immers de professional te zijn bij wie de ouders hun kind in vertrouwen achterlaten.
De resultaten toonden ook aan dat leerkrachten gemiddeld genomen meer vertrouwen
hebben in de autochtone dan in allochtone ouders. Het vertrouwen van de leerkracht
in de ouders wordt beïnvloed door de etniciteit van ouders, terwijl hun etniciteit niet
bepalend is voor het vertrouwen dat ze in de leerkracht stellen. De resultaten wijzen verder
uit dat het vertrouwen van de ouders in de leerkracht geen voorspellende waarde heeft
voor de leesontwikkeling van de kinderen, terwijl het vertrouwen van de leerkrachten
in de openheid van de ouders en hun bereidheid om met hen samen te werken wel
een voorspellende waarde heeft voor de ontwikkeling van de leesvaardigheid van de
kinderen. Kinderen bleken betere leesprestaties te vertonen wanneer de leerkrachten
meer vertrouwen in de ouders stelden.
133
nederlandse samenvatting
Veel factoren bepalen het schoolsucces van kinderen, waaronder psychologische,
sociale en culturele omstandigheden. Ook de kwaliteit en kwantiteit van het taalaanbod
zijn van groot belang voor de taal- en cognitieve ontwikkeling. Tweetalige centre-based
programma’s lijken een oplossing te bieden tegen subtractieve tweetaligheid (wanneer
het leren van een tweede taal interfereert met het leren van de eerste taal en de tweede taal
gaat domineren). Toch zijn er enkele belangrijke praktische bezwaren, zoals de diversiteit
van de kinderen en het ontbreken van leerkrachten die de eerste taal van kinderen uit
cultureel-etnische minderheidsgroepen als moedertaal beheersen. Ook de home-based
tweetalige programma’s worden beïnvloed door culturele en sociaaleconomische
factoren. De vraag blijft of tweetalig onderwijs voor deze specifieke populatie echt een
oplossing is om schools falen tegen te gaan. Want taal en taalontwikkeling mogen dan
uitermate belangrijk zijn, het blijven slechts enkele van de vele factoren die van invloed
zijn op de schoolprestaties van kinderen.
Onze resultaten op de cognitieve (taal)taken en het belang van een effectieve
samenwerking tussen scholen en ouders lijkt het aanbieden van voorschoolse educatie
noodzakelijk te maken. Extra ondersteuning binnen voorschoolse programma’s zoals dat
in het Nederlandse onderwijsvoorrangsbeleid gestalte krijgt, lijkt in overeenstemming
met dit concept. Partnerschap van ouders en school zou kunnen bijdragen aan het
verminderen of zelfs voorkomen van onderwijsachterstanden. Deze studie is eveneens
bevestigend voor het beleid waarbij niet het land van herkomst, maar het opleidingsniveau
van de ouders de bepalende factor is voor het toekennen extra ondersteuning aan een
school om onderwijsachterstanden weg te werken. Immers, bij afwezigheid van andere
risicofactoren, is de etnische achtergrond van de ouders niet de primaire oorzaak van een
onderwijsachterstand. Daarnaast lijkt het vergroten van de ouderlijke betrokkenheid en
het bevorderen van een effectieve partnerschap tussen school en ouders te passen binnen
het Nederlandse onderwijsvoorrangsbeleid. In veel voor- en vroegschoolse programma’s
is voor de ouders immers een belangrijke rol weggelegd.
134
135
Dankwoord
Het onderzoek dat aan dit proefschrift ten grondslag ligt, is voortgekomen uit een
project waar veel mensen bij betrokken zijn geweest. Een aantal van hen wil ik op deze
plek bedanken. Daarnaast zijn er een aantal mensen die ik in het bijzonder wil bedanken
voor de steun en hulp bij de totstandkoming van dit proefschrift.
Allereerst gaat mijn dank uit naar de ouders, de kinderen, de peuterspeelzaalleidsters
en leerkrachten voor hun bereidwilligheid om deel te nemen aan het project. Bijzondere
dank gaat uit naar Anna-Titia Goutbeek. Het eerste idee om onderzoek te doen naar
de kansen van Nijmeegse kinderen in achterstandsituaties was van haar afkomstig.
De eerste jaren van het onderzoek heeft zij met enorme inzet en betrokkenheid als
coördinator ervoor gezorgd dat alle kinderen getest konden worden. Helaas werd zij als
gevolg van een chronische ziekte gedwongen haar taak halverwege het project te staken.
Naast de coördinator ben ik ook zeer veel dank verschuldigd aan een grote groep
studenten (meer dan 100) die hielp bij de afname van de jaarlijkse toetsbatterij. Hiervan
zijn er 14 afgestudeerd als orthopedagoog op een scriptie binnen dit project. Om ze
bij naam en toenaam te noemen: Marloes Agten, Eline Arends, Esther Blom, Esther
Cornelissen, Moniek van Dijck, Loes Ketelaars, Marije Janssen, Kirsten Rosenberg,
Yvonne Schneider, Janneke van der Steen, Roelie Stellaard, Ammerens van der Touw,
Annemiek Verkoulen en Marleen Vermeulen. Janet van Hell die velen van hen begeleidde,
en bovendien behulpzaam was bij de ontwikkeling van het onderzoeksinstrumentarium,
bedank ik voor haar betrokkenheid tijdens de eerste fase van het onderzoek.
De Turkse en Marokkaanse kinderen hadden niet in hun eigen taal getest kunnen
worden zonder de toegewijde hulp van onderzoeksassistenten Emine Yildirim en Latifa
Karrouch. Ook hen ben ik bijzonder dankbaar.
Uiteraard had het project nooit uitgevoerd kunnen worden zonder de financiering
van de Gemeente Nijmegen en de inzet van de Stichting Kion. Ik dank beide voor het in
mij gestelde vertrouwen.
Dit proefschrift zou niet tot een goed einde gekomen zijn zonder de hulp van
twee van mijn promotoren, Anna Bosman en Joep Bakker. Anna en Joep, ontzettend
bedankt voor jullie inzet, tijd en enthousiasme tijdens de praktische uitvoering van het
onderzoek. Dit was een gigantische klus, maar mede dankzij jullie betrokkenheid en
expertise is het me gelukt. Uiteraard ook tijdens het schrijven was jullie theoretische
kennis, maar zeker ook de gezelligheid tijdens ons regelmatig overleg, onmisbaar. Joep,
bedankt voor alle tijd, inzet en enthousiasme bij zowel het project als het proefschrift.
136
Jouw expertise op het gebied van ouderlijke betrokkenheid en je zeer gewaardeerde,
kritische opmerkingen bij de onderwerpen waar jij wat verder vanaf stond, waren altijd
waardevol. Anna, zonder jouw humor, goede samenwerking, motiverende woorden en
jouw vertrouwen in mij, zou het proefschrift nu nog altijd niet zijn afgerond. Ik heb op
veel gebieden ontzettend veel van jullie beiden geleerd. Tot slot wil ik Paul bedanken
voor zijn bijdrage aan de totstandkoming van mijn proefschrift.
De leden van de manuscriptcommissie, professoren Janssens, Minnaert en Elbers,
bedank ik voor hun bereidheid om dit proefschrift te beoordelen.
Mijn paranimfen, Janneke en Dorothee bedank ik voor de welkome afleiding en
jullie luisterend oor voor al mijn verhalen, plezierige en minder plezierige, gedurende
een reeks van jaren. Ik vind het ontzettend leuk dat jullie mijn paranimfen willen zijn.
Mijn schoonzus Jamila en haar dochters, Sara en Lamyae: jullie deur stond altijd
voor ons open. Bedankt voor alle goede zorgen, speciaal voor Laila en Reda.
De steun en het vertrouwen van mijn ouders en echtgenoot waren vooral in de
laatste periode van mijn proefschrift onmisbaar. Pap en mam, bedankt voor jullie goede
adviezen, motiverende woorden en warme betrokkenheid. En speciaal voor mijn lieve
mam: bedankt voor de goede zorg voor Laila en Reda op de momenten dat ik aan mijn
proefschrift moest werken. Je bent een superoma! Het combineren van mijn werk op
school, de zorg voor de kinderen en het schrijven van mijn proefschrift was nooit gelukt
zonder de steun van mijn ouders en mijn lieve partner Souliman. Ontzettend bedankt!
Souliman, het heeft een lange tijd geduurd en ook voor jou was dit promotietraject, op
zijn zachts gezegd, niet altijd even gezellig. Menig avond, weekend en vakantie is in dit
proefschrift gaan zitten. Ik waardeer het zeer dat jij mij altijd hebt gesteund en altijd het
vertrouwen hebt gehad dat ik het af zou ronden.
Lieve Laila, Reda en Driss: mama is nu klaar. Nes tagseg kenieuw!
137
Curriculum Vitae
Marije Janssen is geboren op 24 oktober 1981 te Nijmegen. Zij behaalde in
2000 haar atheneumdiploma op de Nijmeegse Scholengemeenschap Groenewoud.
Vervolgens studeerde zij Pedagogische Wetenschappen en Onderwijskunde (PWO) aan
de Katholieke Universiteit Nijmegen, tegenwoordig Radboud Universiteit (RU), waar
zij in 2004 afstudeerde op haar scriptie over taalontwikkeling en taalstimulering bij
een- en tweetalige kinderen. Vanaf 2005 werkte zij als onderzoeksassistent en junior
docent bij Pedagogische Wetenschappen en Onderwijskunde, waar zij gedurende het
onderzoek dit proefschrift, als buitenpromovenda, is gaan schrijven. Vanaf 2008 is zij
als orthopedagoog verbonden aan het Flexcollege Nijmegen, een school voor overbelaste
jongeren die zijn uitgevallen in het reguliere onderwijs. Daarnaast heeft zij tijdens de
laatste drie jaar van haar proefschrift een drietal kinderen gekregen. Alle drie worden
tweetalig opgevoed.
138

Presentatie Jeroen Kortmann - Vereniging voor Mededingingsrecht

Download report as PDF - HCSS Centre for Strategic Studies

EFL Academic writing: What should Dutch business communication

NISL NEWS 20 - The Netherlands International School Lagos

Seminar FIDIN 2008

Title page_Contents_Acknowledgements_Map

ON THE ROAD FOR PEACE

Onboard assortiment Arke Shop

Welcome to - Van Duuren Districenters