Family language policies, reported language use and proficiency in Russian – Hebrew bilingual children in Israel

This article was downloaded by: [University College London]On: 23 December 2011, At: 03:38Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Multilingual andMulticultural DevelopmentPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/rmmm20

Parents' assessment of their preschoolchildren's bilingual development in thecontext of family language policyMila Schwartz a b & Victor Moin ca Oranim Academic College of Education, Israelb Edmond J. Safra Brain Research Center for the Study of LearningDisabilities, University of Haifa, Israelc The Center for Research and Study of the Family, Schoolof Social Work, Faculty of Social Welfare & Health Sciences,University of Haifa, Israel

Available online: 19 Dec 2011

To cite this article: Mila Schwartz & Victor Moin (2011): Parents' assessment of their preschoolchildren's bilingual development in the context of family language policy, Journal of Multilingualand Multicultural Development, DOI:10.1080/01434632.2011.638078

To link to this article: http://dx.doi.org/10.1080/01434632.2011.638078

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae, and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,

demand, or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

Parents’ assessment of their preschool children’s bilingual development inthe context of family language policy

Mila Schwartza,b* and Victor Moinc

aOranim Academic College of Education, Israel; bEdmond J. Safra Brain Research Center forthe Study of Learning Disabilities, University of Haifa, Israel; cThe Center for Research andStudy of the Family, School of Social Work, Faculty of Social Welfare & Health Sciences,University of Haifa, Israel

Parents’ assessment of children’s development in the first and the second languageis an essential part of their family language policy (FLP) and an importantcomponent of parent�child communication. This paper presents a pilot studyfocused on Russian-speaking immigrant parents’ assessment of their children’slanguage knowledge in Russian as a first language and Hebrew as the secondlanguage in the context of their FLP. The research questions were as follows: Howis parents’ assessment of their children’s bilingual language development linked totheir choice of bilingual versus monolingual preschool education? To what degreeare parents’ reports of their children’s language knowledge similar or different totheir children’s actual language knowledge? Which domains of language knowl-edge do parents relate to or ignore when assessing their children’s languagedevelopment? The sample consisted of 27 children (14 from bilingual and 13 frommonolingual kindergartens), and their parents. Two sets of measurements wereused, one to obtain parents’ reports on child’s knowledge of Russian and Hebrewand the other to assess children. The finding points out parents’ insensitiveness tothe length of the children’s utterances and their tendency to rationalise FLP byoverestimating their children’s general language knowledge.

Keywords: family language policy; parents’ assessment; bilingual children’slanguage development; bilingual education

Introduction

Parents’ assessment of their children’s language development is a significant

component of parent�child communication, and an integral part of family language

policy (FLP). Clearly, parents do not use special scientific methods, such as

the Communicative Development Inventories, to assess their children’s language

development. As lay people, they tend to apply more general subjective assessments

in their daily parent�child communication such as ‘good-bad’ or ‘better-worse’ to

estimate their children’s language knowledge. These general assessments serve as

genuine instruments in their FLP.

In this paper, the results of our pilot study of immigrant parents’ general

assessments of their children’s bilingual development as a part of FLP are presented.

More specifically, we aimed to examine to what degree are bilingual parents’ reports

of their children’s language knowledge similar or different to their children’s actual

language knowledge in Russian and Hebrew. In addition, our focus was on

*Corresponding author. Email: [email protected]

Journal of Multilingual and Multicultural Development

2011, 1�21, iFirst article

ISSN 0143-4632 print/ISSN 1747-7557 online

# 2011 Taylor & Francis

http://dx.doi.org/10.1080/01434632.2011.638078

http://www.tandfonline.com

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

investigating domains of language knowledge which parents relate to or ignore when

assessing their children’s language development. Finally, we were interested to

understand how parents’ assessment of their children’s bilingual language develop-

ment is linked to the family’s language policy, and in particular, to their choice of

bilingual versus monolingual preschool education.

The article is a third stage of the broad research project entitled ‘Family language

policy of Russian-Hebrew speaking immigrants in Israel concerning the preschool

bilingual development of their children’ (as detailed in the Research Design section). Inthe first stage of this project, we examine immigrant parents’ language policy towards

their children’s bilingual development and education (Schwartz et al. 2010; Schwartz,

Moin, and Leikin 2011). In the second stage of this project, our focus was on the

language knowledge among immigrant parents’ children. The present study as a third

stage of the project aimed to show relations between immigrant parents’ FLP (parents’

assessments of preschool children’s bilingual knowledge) and bilingual development of

their children (children’s language knowledge in their L1, Russian, and L2, Hebrew,

measured in accordance with their performance on different tests.

Family language policy in immigrant context

Family Language Policy is a relatively new area of research, bringing together two

fields of study: language policy and child language acquisition (King and Fogle 2006;

King, Fogle, and Logan-Terry 2008; Schwartz 2010). Spolsky (2004) distinguished

between three components in the language policy of a community: its language

practices, language beliefs or ideology, and language management, which constitutesefforts to modify or influence that practice.

The FLP plays a particularly important role in language development of children

in immigrant families (Schwartz 2010; Spolsky 2007). A growing number of studies

are focusing on language policy of different groups of immigrant families in specific

sociocultural conditions: e.g. Latinos in the USA (Worthy and Rodriguez-Galindo

2006; Zentella 2005), Afrikaans-speaking parents in New Zealand (Barkhuizen

2006), Iranian families in Washington (Felling 2006) and Chinese-speaking parents

in California (Amaral 2001). Many of these researchers stressed that immigrationmobilises lay people’s awareness of language (e.g. Barkhuizen 2006; Yelenevskaya

and Fialkove 2003).

Parents’ assessment of children’s language is considered a part of their language

policy. Previous studies have shown that immigrant parents tend to evaluate their

children’s bilingual development (Barkhuizen and Knoch 2006; Okita 2002;

Schwartz, Moin, and Leikin 2011). The parents’ assessment might lead them to

rethink their language practice and implement a clear intervention plan to support

the child’s bilingual development. For example, to improve L1 knowledge, parentsmay decide to send their children to the country of origin during the summer or enrol

them in an L1-speaking summer camp, as described by Caldas and Caron-Caldas

(2000, 2002). In a study of South African immigrants in Australia, Barkhuizen (2006)

found that parents who were dissatisfied with their children’s L1 knowledge searched

for an external supporting sociolinguistic environment. They strove to provide the

children with maximum exposure to L1, and often chose residential suburbs with a

high concentration of South African immigrants. The similar tendency is notable

among Russian-Hebrew speaking immigrants in Israel who are the target populationof our study. Thus, this community tends to appreciate their original culture and

2 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

language (Ben-Rafael et al. 2006), and in order to maintain them, an organization of

immigrant Russian teachers has established a wide network of preschool Russian-

Hebrew education in Israel.

The immigrant family context typically provides L1 input at home. The reports of

immigrant parents might differ from reports of parents practicing dual language

input at home from the onset of the child’s language acquisition (e.g. as in the case of

one-parent-one-language context). Immigrant parents such as any other adult L2

learners have personal experience with adult second language (L2) acquisitionpatterns when they are cognitively and linguistically mature. As a result, they might

develop a high level of language awareness. Maturity and personal experience with

L2 acquisition seem to help immigrant parents to provide relatively reliable lay

assessments of their children’s language knowledge. Nevertheless, these parents are

inexperienced in monolingual L2 acquisition, and might therefore provide mistaken

reports about unusual or irregular trends in their children’s L2 acquisition. Thus, the

present study focused on the link between immigrant parents’ assessment of

children’s language knowledge in the context their FLP and with the children’sactual knowledge in heritage and host languages.

Parents’ reports of language knowledge

Interest in parents’ reports as an instrument for assessing language development in

young children is mounting, in both the clinical and the research domains (Bates and

Carnevale 1993; Feldman et al. 2005; Fenson et al. 1994). Parent reports are an

efficient research tool for several reasons. They are cost effective, enable datacollection from a large number of children, and do not induce the distractibility,

fatigue and hesitancy that many children display under experimental conditions

(Bates and Carnevale 1993; Feldman et al. 2005; Fenson et al. 1994; Guiberson and

Rodriguez 2010; O’Toole and Fletcher 2010; Ring and Fenson 2000). One of the

most frequently used parent report instruments is the MacArthur-Bates Commu-

nicative Development Inventories (CDI) (Feldman et al. 2005). Concurrent and

predictive validity of parents’ reports in general has been widely tested regarding

children who speak different languages (e.g. O’Toole and Fletcher 2010), childrenwith development disability, children of different age groups, bilingual children, and

concerning different domains of language development. In general, the results were

positive showing significant correlations between parents’ reports and children’s

performance in many language domains.

However, some researchers were skeptical concerning the accuracy of the parent

reports (Feldman et al. 2000). Regarding language domains, for example, some

controversial data were found regarding the correlation of the parents’ reports for

vocabulary and mean length of utterance (MLU). In a study of children withlanguage delay, Thal et al. (1999) reported high correlations between parents’ reports

of children’s vocabulary and grammatical complexity, and of children’s vocabulary

(r�0.86) and MLU (0.69). At the same time, Thordardottir and Weismer (2010),

who aimed to adapt the CDI for Icelandic children, found no significant correlation

between direct assessment of MLU and parents’ reports of children’s sentence

complexity.

Although many studies examined parents’ reports of bilingual children’s language

knowledge, information about this issue is lacking in the literature. First, most studieswere focused on parents’ reports of L1 history and knowledge (Paradis, Emmerzael,

Journal of Multilingual and Multicultural Development 3

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

and Duncan 2001), and relatively few studies aimed to obtain parents’ reports of

bilingual children’s language knowledge in L1 and L2 (Gutierrez-Clellen and Kreiter

2003; Marchman and Martınez-Sussmann 2002; Pease-Alvarez 1993). Second, up to

now the immigrant parents’ reports on children’s language addressed mostly the

patterns of family language and literacy practice in L1 and L2 (Duursma et al. 2007;

Tabors and Snow 1994), however, no data exist about their lay assessments of

children’s language knowledge. Third, parents’ lay assessments were studied solely as

a research and evaluation tool (e.g. CDI). However, these assessments could beconsidered as an integral part of the parents’ daily childrearing experience within the

framework of their FLP. In this case, parents’ assessments might be influenced by

their language ideology (e.g. believe in importance of early L2 immersion for child’s

language development), practice (e.g. parent�child language and literacy games in L1

or L2) and management (e.g. choice of bilingual education or community Sunday

school and vice versa. The present study addressed these issues.

Different language knowledge domains

Research on parents’ reports mainly addressed young children’s expressive and

receptive vocabulary, sentence complexity and MLU. In our study, we proposed to

extend the scope of the language domain, distinguishing between depth and breadth

of vocabulary knowledge and between paradigmatic and syntagmatic sense relations,

which are the two fundamental meaning levels of a word (Cruse 1986; Jakobson

1971; Lyons 1968). Similar to the studies mentioned in the previous section, we

focused also on Mean Length of Utterance as a measure of children’s languageproductivity and especially of syntactic development.

Vocabulary knowledge or knowing a word implies knowing many things about

the word; its literal meaning and various connotations, the types of syntactic

construction into which it is incorporated (for details, see Bloom 2000; Nagy and

Scott 2000). This wealth of word representations refers to depth of lexical knowledge

(i.e. how well we know these words, or the qualitative aspects of word knowledge, e.g.

knowledge of word concept dog, including categorisation, animal, and common

characteristics, which are essential for this word concept, e.g. barking and wagging its

tail). In contrast, the breadth of the lexical knowledge refers to the number of words

we know. Research on L2 lexical knowledge has typically measured vocabulary

breadth rather than depth (Wesche and Paribakht 1996). At the same time,

measuring breadth has been criticised and considered of limited value because it

ignores the continuous nature of acquiring new words. Thus, examining vocabulary

breadth alone does not throw light on the quality of the children’s acquired

vocabulary and fails to provide important information about their level of mastery of

a word’s sense relations. In light of this, in the present study we examined bothdimensions of lexical knowledge, which complement one another.

Regarding paradigmatic and syntagmatic sense relations, paradigmatic relations

are hierarchical, such as superordination and subordination, and therefore reflect

hierarchical relations that exist between items. For example, the superordinate

category furniture includes such subordinate-level (basic-level) categories as chair,

table and bed. In addition, within the paradigmatic sense relations, the lexical

hierarchy can be partonomic, a part-whole hierarchy (meronomy), as in the case of

body parts (head/nose). By contrast, syntagmatic sense relations represent horizontalrelations between items. They provide information about the object’s appearance,

4 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

location or use, e.g. ‘a watermelon is sweet and tasty’ (descriptive characteristic) or ‘a

hammer is something to pound with’ (functional description).

Paradigmatic sense relations are considered to be strongly related to the

development of high order cognition skills, such as conceptualisation, categorisation,

classification and de-contextualisation of word concepts. These relations develop and

improve within the educational environment (Anglin 1985; Ordonez et al. 2002;

Verhallen and Schoonen 1993, 1998; Vygotsky 1962). In addition, it has been found

that the ability to produce the subordinate for basic concepts (paradigmaticinformation), such as a dog is an animal, develops later than the ability to provide

a description of these concepts (syntagmatic information) (Anglin 1985). More

specifically, Anglin (1985) showed that the onset of paradigmatic sense relations is

around age three and that three-year-olds can supply the names of subordinates for

concrete objects (e.g. food or animals) and can provide their part-whole hierarchy

(e.g. a dog has legs and a tail). To sum up, paradigmatic sense relations are associated

with more mature semantic organisation than the syntagmatic sense relations, which

reflect vocabulary richness by providing descriptive, associative and metaphoricinformation about a variety of distinctive object attributes.

Mean length of utterance

All children develop expressive language skills in the same sequential order. As they

mature, their utterances increase in length and can be correlated with their

chronological age. Brown (1973) was the first to use the MLU to measure young

children’s syntactic development. He found that MLU provided a more accurate

reflection of the young children’s language development than their chronological age.

This measure was found to be valid between ages 1.6 and 4 years old. Concerning the

usage of MLU as a tool for bilingual children’s language development has been

recently proposed for measuring of language dominance for each language atdifferent stages (Yip and Matthews 2007). Note, however, that this measure is

recognised to be useful for within-language comparisons, and may not be directly

comparable across languages, especially those of different morphological types

(Dopke 1998). In the present study, we used this tool to examine whether immigrant

parents’ assessment of children’s general language knowledge is related to the

children’s MLU in the target languages.

Socio-cultural context of the study

The paper focuses on Russian-Hebrew speaking immigrant parents’ general

assessments of their children’s bilingual development in the context of FLP. It is a

well-known fact that Israel is populated largely by successive waves of immigrants.Between 1989 and 2008, 992,236 immigrants arrived in Israel from the former Soviet

Union (FSU), making up approximately one-sixth of the total Israeli population

(Statistical Abstract of Israel 2010). This huge influx of newcomers has resulted in a

consolidated community structure at both the formal (e.g. political) and informal

levels. In addition, Russian has attained the status of Israel’s third language (Olstein

1995), after Hebrew, the official language, and English. This status includes a state-

sponsored Russian-language radio network and TV channel, and the publication of

some 50 Russian-language newspapers and magazines. The Russian community’smain acculturation strategy is integration, combining maintenance of the original

Journal of Multilingual and Multicultural Development 5

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

culture and adaptation to the host culture with relatively rapid and successful

acquisition of Hebrew (Horenczyk and Ben-Shalom 2006; Horowitz and Leshem

1998; Lissitsa 2007; Olshtain and Kotik 2000). This strategy was in accord with their

relatively high sociocultural profile. Approximately 60% of the FSU immigrants in

the workforce work in professions that require an academic education (Leshem and

Lissak 1999). Thus, this unique sociocultural context allows focus on a research

population with a favourable background for providing controls for influencing

factors, such as parents’ education level (Thordardottir and Weismer 2010) andknowledge of L1 and L2 (Gutierrez-Clellen and Kreiter 2003; Marchman and

Martınez-Sussmann 2002).

The present study

The research questions:

(1) To what degree are parents’ reports of their children’s language knowledge

similar or different to their actual language knowledge in Russian (L1) and

Hebrew (L2)?

(2) Which domains of language knowledge do parents relate to or ignore when

assessing their children’s language development?

(3) How is the parents’ assessment of their children’s bilingual language

development linked to the family’s language policy and, in particular, to

their choice of bilingual versus monolingual preschool education?

Method

The research design

The article presents the results of the third stage of an extensive research project

focusing on the language development of bilingual children entering bilingual versus

monolingual kindergartens. As noted earlier, the first stage investigated how

immigrant parents described and explained their FLP concerning their preschoolchildren’s bilingual development. In addition, we explored the factors linked to the

parents’ choice of bilingual or monolingual kindergarten for their child. The stage

was divided into two parts: the first part was quantitative and used questionnaires

and the second part was qualitative and used in-depth semi-structured interviews. In

the quantitative part, the study design was based on a comparison of two groups of

parents: those who chose bilingual Russian-Hebrew-speaking kindergartens versus

monolingual Hebrew-speaking kindergartens for their children (Schwartz et al.

2010). The research included immigrants from the FSU in Israel, who had Israeli-born children aged three and four years old. We used multistage, random, purposeful

sampling. In the first stage, we randomly chose three bilingual and five monolingual

kindergartens in similar neighbourhoods in the north of Israel, in a region with a

large Russian-speaking immigrant community. In the second stage, we chose 112

parents, whose children attended these kindergartens and who met the following

inclusion criteria: (1) the child was born in Israel; (2) they were 3�4 years old; (3)

Russian was their first and dominant language at home (Schwartz, Moin, and Leikin

2011). We distributed 112 questionnaires in Russian (L1) to the parents and askedboth parents to complete the questionnaires separately. One hundred and one

6 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

parents from 55 families agreed to participate in the research and returned the

questionnaires. The qualitative part of the study provided insight into the parents’

FLP and the experience of raising bilingual children, by using the life stories told in

the separate semi-structured interviews with mothers and fathers who had chosen

bilingual versus monolingual kindergartens.

In the second stage of the project, we compared the bilingual lexical knowledge of

the children from the bilingual kindergartens with that of the children from the

monolingual kindergartens. This comparison aimed to examine the role of earlybilingual education in L1 maintenance and L2 acquisition among children coming

from immigrant families. Twenty-seven children, whose parents were participants of

the first stage of the project, were randomly selected: 14 from the bilingual

kindergartens and 13 from the monolingual kindergartens.

In the present study (the third stage of the project), to compare the parents’

reports with the children’s actual bilingual knowledge, new databases were created by

selecting the parents of the children who were tested in the second stage of the project

(n�27) from the general parents’ sample (n�101). Second, ordinal scales, whichwere used in the first and second stages of the project, were reconstructed to 100-

point scales in a monotonic way. Finally, new variables were created to characterise

the children’s general language knowledge in accordance with the parents’ report and

test results (see ‘Data analysis’).

Participants

The sample of the secondary comparative analysis consisted of 27 children (14 fromthe bilingual kindergartens, hereafter the bilingual kindergarten, and 13 from the

monolingual kindergartens, hereafter the monolingual kindergarten) and their 27

sets of parents. The mean age of the bilingual group was 37.4 months and 37.9

months in the monolingual group (the difference was not significant, t�0.72). The

only significant difference between these groups was in the number of years in

kindergarten (M �1.1 years for bilingual group versus M �1.7 years for mono-

lingual group, t�4.5**).

The parent sample included only mothers, because the response rate for thissubsample was higher among the mothers (100%) than among the fathers (79%). In

addition, no significant differences were found in the general parent sample between

mothers’ and fathers’ reports of their children’s bilingual knowledge in both types of

kindergarten (t�0.15 for knowledge of Hebrew and t�0.25 for knowledge of

Russian). Thus, this paper presents the parent report data based only on the mothers’

reports and we used the term ‘parents’ report’ rather than ‘mothers’ report’

throughout the paper.

The mothers’ mean age was 32 years (SD �2.8) in the bilingual group and 29years (SD �2.7) in the monolingual group (the difference was statistically

significant, t�4.5, p�.007). No significant differences were found between these

groups for mothers’ education level (15 years’ education in the bilingual group and

16 years in the monolingual group, t�1.3). All mothers obtained some kind of

educational experience in Israel, their duration of residence in the host country was

12 and 11 years, respectively, t�0.8.

All participants were minority-language mothers. The reported L1 of all mothers

and their children was Russian, which was the dominant communication languagebetween family members, including the child. In all families, both parents were

Journal of Multilingual and Multicultural Development 7

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

bilingual. In accordance with t-test no differences were found between the two

groups in the mothers’ knowledge of Hebrew and Russian. The parents’ mastery in

both target languages was relatively high, according to self-assessment on five-point

scale (for Russian: M�4.6, SD �0.58 for the bilingual group; and M �4.7,

SD�0.56 for the monolingual group; for Hebrew: M�4.3, SD �0.73 for the

bilingual group; and M�4.2, SD �0.94 for the monolingual group). In both

groups, Russian language knowledge was higher than Hebrew language knowledge

according to the one-paired t-test (t�2.70, p�.008 for the bilingual group andt�3.25, p�.002 for the monolingual group).

Instruments

In this study, two sets of measurements were used, one to obtain parents’ reports ontheir child’s knowledge of Russian and Hebrew and the other to assess children. The

former was measured by instruments such as parents’ reports and general

assessments of their children’s language knowledge in Hebrew and Russian; and

the latter was measured using children’s language knowledge in Hebrew and Russian

in accordance with their performance on tests examining different language

knowledge domains.

Parents’ reports

The child’s knowledge of Russian and Hebrew. The parents were asked to rate their

child’s knowledge of Russian and Hebrew using three ‘can-do’ speaking and

comprehension items, and to estimate the child’s language knowledge compared to

other children of the same age (‘Please, try to assess your child’s level of Russian in

relation to his/her friends’). All responses were given on a 5-point scale (‘not at all’,‘bad’, ‘not good enough’, ‘good’, and ‘very good’). The general assessments of

knowledge of the child’s Russian and Hebrew were calculated as the mean score of

the parents’ responses to these three questions. Cronbach’s alpha was 0.83 for

knowledge of Russian and 0.93 for Hebrew. To run a comparative analysis of the

parents’ assessment and children’s language knowledge, which were measured on

100-point scale, the 5-point scale of parents’ assessment was reconstructed to 100-

point in a monotonic way. A balance between children’s language knowledge in L1

and L2 presents an essential characteristic of their bilingual development (Oller andPearson 2002). This balance was calculated as the discrepancy between the parents’

assessments of their child’s knowledge in Russian and Hebrew languages. If the delta

score is positive and high in accordance with parents’ assessments, the child is more

dominant in the Russian language. If the delta score is negative, it points out the

Hebrew language dominance, and if it is around 0, it means balanced Hebrew-

Russian language knowledge.

Children’s language knowledge

Children’s knowledge of Hebrew and Russian was measured in accordance with their

performance on tests examining different domains of language knowledge: (1)

Breadth of vocabulary knowledge (receptive vocabulary test, productive vocabulary

test); (2) Depth of vocabulary knowledge (paradigmatic and syntagmatic senserelations); and (3) MLU.

8 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

(1) Breadth of vocabulary knowledge

Receptive vocabulary�Peabody Picture Vocabulary Test (PPVT-R) (Nevo 1979,

Hebrew version; after Dunn 1965). This test required the child to indicate which of

four pictures matched a spoken word. Test-retest reliability for the Hebrew and

Russian versions was reported at 0.70 and 0.84, respectively (Aram and Biron 2004).

Productive vocabulary (Schwartz et al. 2007). The test included 10 semantic

categories (animals, transportation, electrical devices, musical instruments, clothes,

toys, furniture, fruit and vegetables, body parts and cooking utensils), which

contained 34 items (e.g. pair of pants, train) taken from the CDI (Fenson et al.

1993) and were adapted to Hebrew and Russian. The items were presented to the

participants in the form of coloured pictures, and the child was asked to name them.

Each picture was scored in the following way: 2 points for the correct answer, 1 pointfor an answer from the same semantic category (e.g. coat instead of shirt) and 0

points for an incorrect answer. The maximum score was 68. Internal consistencies for

the Hebrew and Russian versions (alpha) were 0.92 and 0.90.

(2) Depth of vocabulary knowledge: paradigmatic sense relations

Categorical identification test (Schwartz et al. 2007). This test tapped receptive

vocabulary skills. Children were shown eight pages, each including four pictures �two pictures belonging to the same semantic category (e.g. chair and cupboard which

belong to the category of furniture) and two pictures belonging to other semantic

categories (e.g. dog which belongs to the category of animals and apple which

belongs to the category of fruits). All pictures presented words which were graded by

the preschool teachers and speech therapists as words that the child often uses. The

pictures were presented in random order. First, the child was asked to name thepresented pictures, then to point to two pictures belonging to the target semantic

category, e.g. furniture (chair and cupboard). Scores were calculated in the following

way: 2 points for the correct answer (identification of two pictures), 1 point for

pointing to only one or two correct pictures and one incorrect picture and 0 points

for an incorrect answer. The maximum score was 16. Internal consistencies for the

Hebrew and Russian versions (alpha) were 0.82 and 0.65.

Semantic fluency test (Hebrew version adapted from Spreen and Strauss 1998). In

this task, tapping productive vocabulary skills, the child was required to name as

many category members as possible in 60 seconds. The semantic categories were

animals and food. The scores were calculated by allocating one point for each

correctly retrieved item.

Word description (productive vocabulary skill). This task, also tapping productive

vocabulary skills, was adopted from Rom and Moreg (1999), and Verhallen and

Schoonen (1993) (after Richard and Hanner 1985). Children were presented with

three stimulus words, high-frequency concrete nouns (watermelon, dog and

refrigerator) and asked to give as many dimensions of meaning as they could think

of in answer to an open-ended prompt (e.g. ‘Tell me everything you know about the

watermelon’). The analysis was based on the model proposed by Verhallen andSchoonen (1993). The two subcategories of word description, which expressed two

Journal of Multilingual and Multicultural Development 9

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

types of vocabulary knowledge, were measured: paradigmatic and syntagmatic sense

relations.

Paradigmatic sense relations: (1) Hierarchical taxonomical (superordinate) and (2)

Hierarchical partonomic (part-whole constituents). For example, a child defined a

dog as ‘an animal that has legs, a back and a head’ (paradigmatic sense relations:

superordinate and a part-whole hierarchy).

Syntagmatic sense relations: Descriptive information about the object (size, shape,

colour, taste, location, function, use). For example, a child described a dog as

‘barking’, ‘can be found at home’ (syntagmatic sense relations: descriptive informa-

tion about function and location).

Mean Length of Utterance as a measure of language development was calculated by

collecting utterances spoken by a child during the script description task and

dividing the number of words by the number of utterances. We used the open

encouragement method which is built on verbal encouragement only, to produce

stories about one-time personal experiences or familiar scripted events (Berman and

Slobin 1994). The present study chose ‘a birthday party’ script since it is familiar and

common among all children in day care. The child was asked to describe his or her

birthday party by applying elicitation questions (e.g. ‘What happens at the end?’).

Data analysis

The results of children’s performance on the tasks in different language knowledge

domains were analysed in the following way:

(1) First, all scales were recomputed as a percentage of the successfully

performed items relative to the maximum scores for each of the following

tests: receptive vocabulary, productive vocabulary and categorical identifica-

tion test. For other tests (paradigmatic and syntagmatic sub-scales; semanticfluency, and MLU) percentages were computed relative to the best

performance on each of these tests among the participants. For example,

the best MLU result among the participants of this study was seven units, and

therefore the percentage of the other participants’ successful performance on

MLU was calculated relative to this number (7). Following this, the

percentages that indicated the participants’ performance on the test items

were interpreted and analysed as pseudo-metric 100-point ordinal scales.

(2) Three composite language knowledge variables were constructed as follows:(1) General language knowledge in Russian, as the mean score of the

children’s performance on all the Russian language tasks; five tests and two

subtests (Cronbach’s alpha �0.77), (2) General language knowledge in

Hebrew, as the mean score of the children’s performance on all language

tasks in Hebrew (Cronbach’s alpha �0.70) and (3) A balance between

children’s language knowledge in L1 and L2 is an important characteristic of

their bilingual development. This balance was calculated as the discrepancy

between the children’s knowledge of the Russian and Hebrew languages. Ifthe delta score is positive and high, the child is more dominant/proficient in

10 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

Russian; if the delta score is negative, it points out Hebrew dominance, and if

it is about 0, it means balanced Hebrew-Russian language knowledge.

The research variables

Based on the theoretical background presented earlier, the research included the

following main groups of variables (see Figure 1):

(1) FLP towards bilingual development of their children. FLP was operationalised in

this study by parents’ choice of bilingual (Hebrew�Russian speaking) versus

monolingual (Hebrew-speaking) kindergarten; (2) Parents’ general assessments of

the children’s language knowledge in Hebrew and Russian. These assessments were

combined based on parents’ assessments of children’s perceptive (speaking) and

receptive (comprehension) skills in each language and also assessments of their

language knowledge compared to other children of the same age; and (3) Children’s

actual language knowledge in Hebrew and Russian.

Results

Parents’ reports and children’s general language knowledge

To address the first research question, we distinguished between two ways of

measuring similarity between the parents’ reports and the children’s results: through

Children’s actual languageknowledge

Breath of vocabulary knowledge

Depth of vocabulary knowledge

Syntagmatic sense relations

Receptivevocabulary

Categoryidentification

Semantic fluency

Productivevocabulary

Paradigmatic sense relations

MLU

Parents’ general assessments of their children

Understanding Talk skill Generalcomparative assessment

Family LanguagePolicy:

Parents’ choice ofbilingual vs. monolingual

kindergartens

Language:L1, Russian L2, Hebrew

Figure 1. The research variables.

Journal of Multilingual and Multicultural Development 11

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

correlation (relative similarity) and mean absolute discrepancy (see details in Kenny

and Acitelli 1994).

Relative similarity between the parents’ report and the children’s general language

knowledge was examined by Pearson correlation. The analysis showed significant

correlations for both target languages and for measure of balance between the

languages. The higher the parents’ assessment, the better was the children’s general

language knowledge in Russian and Hebrew and vice versa. The parents’ report was

found to be more similar to the general language knowledge level in Russian (r�.61,

pB.001) than in Hebrew (r�.48, pB.01). The highest relative similarity was found

for the balance between Russian/Hebrew language knowledge (r�.77, pB.001).

Absolute similarity was examined according to the discrepancy between the

parents’ report and children’s scores on general language knowledge in Russian and

Hebrew. The significance of absolute similarity was defined by paired sample t-test

(see Table 1). The differences were statistically significant on all measures. The

greatest difference was found between the parents’ report and children’s scores on

general language knowledge in Russian.

The results demonstrated that the parents assessed their children’s general

language knowledge to be significantly higher than their actual performance in both

Russian and Hebrew.

Criteria of the parents’ reports

To examine the second question, we calculated the correlation between the parents’

reports of their children’s language knowledge on the one hand, and the children’s

actual knowledge in different language domains on the other (through performance

on various language knowledge tasks) (see Table 2).

A clear-cut similarity was found in explicit representation of parents on the

significance of various domains of children’s language knowledge in both Russian

and Hebrew. We found that the parents related to the same language domains (i.e.

receptive and productive vocabulary, semantic fluency and syntagmatic sense

relations), and ignored the others in both languages. Hence, the parents’ reports

did not correlate at all with two out of three measures of the paradigmatic sense

relations (categorical identification and word definition), and with MLU in both

Russian and Hebrew. Second, the largest correlations were obtained between the

parents’ reports and the children’s productive vocabulary in Russian and syntagmatic

knowledge (word description task) in Hebrew.

Table 1. The absolute similarity between the parents’ reports and children’s general language

knowledge in Russian and Hebrew (100-point score).

MeasuresParents’reports

Children’sperformance Discrepancy t

General language knowledge inRussian

85 52 33 16.7*

General language knowledge inHebrew

61 36 25 6.5*

*p B.001.

12 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

The parents’ reports of the children’s general language knowledge in the bilingual versusmonolingual group

To investigate whether the parents’ reports differed according to their FLP regarding

the choice of bilingual versus monolingual education, we compared the parents’

reports and the children’s general language knowledge in L1 and L2 in two types of

preschool education settings. The significance of differences between these groups of

parents and children were analysed by t-test (see Table 3). Significant differences

were found between the parents’ reports and children’s general language knowledge

in two types of education settings: bilingual kindergarten versus monolingual

kindergarten.

The findings indicated great similarity between both groups of parents on their

reports about the children’s general language knowledge in Russian. However, the

children’s performance results showed that children in the bilingual group knew

Russian better than the children in the monolingual group did. As for Hebrew (L2),

it was found that the level of children’s’ knowledge in this language was similar in

both bilingual and monolingual kindergartens. At the same time, the parents from

the monolingual group assessed the children’s knowledge in Hebrew to be

significantly higher than the assessment by parents from the bilingual group.

Table 2. Correlation between the parents’ reports of children’s language knowledge in

Russian and Hebrew and the children’s different language knowledge domains.

Children’s knowledge in different language domainsParents’ report

Russian Hebrew

Breadth of vocabulary knowledgeReceptive vocabulary 0.51** 0.40*Productive vocabulary 0.59*** 0.41*

Depth of vocabulary knowledgeParadigmatic sense relationsCategorical identification 0.25 0.06Semantic fluency 0.38* 0.48**Paradigmatic sense relations 0.32 0.07Syntagmatic sense relations 0.51* 0.67***

Mean length of utterance 0.31 0.26

*p B.05; **pB.01; ***p B.001.

Table 3. The parents’ reports and the children’s language knowledge in two types of

preschool education settings: bilingual and monolingual kindergartens.

LanguageRussian Hebrew

Bilingualgroup

Monolingualgroup

Bilingualgroup

Monolingualgroup

Group M (SD) M (SD) t M (SD) M (SD) t

Children’s generallanguageknowledge

59 45 4.4* 31 41 2.8

Parents’ reports 87 83 0.89 46 76 4.8*

*p B.001.

Journal of Multilingual and Multicultural Development 13

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

The other way to study links between the educational setting, the parents’ report,

and the children’s language knowledge is by examining the balance between

children’s knowledge of Russian and Hebrew (see Figure 2). Balance between the

children’s general language knowledge in Russian and Hebrew was calculated in

accordance with (1) the parents’ reports of Russian and Hebrew and (2) the children’s

performance results on the general language knowledge measure. Most of the

parents’ assessments correctly reflected the real balance in Russian and Hebrew

language knowledge among children in bilingual and monolingual kindergartens. Ahigh significant correlation was found between the balance score in accordance with

the parents’ reports and the children’s language knowledge (r�.77, pB.01). The

parents’ report and the children’s performance demonstrated more balanced

bilingualism in the monolingual than in the bilingual kindergartens.

Discussion

This study focused on the role of the immigrant parents’ assessment of their bilingual

children’s language knowledge as an essential part of their FLP and in particular of

their family language management. Hence, the discussion of the results will be

presented in this context.

Relative and absolute similarity

Our results showed significant relative similarity between parents’ reports and their

children’s actual language knowledge in Russian and Hebrew. The higher the

parents’ assessment, the greater was the children’s general language knowledge inRussian and Hebrew and vice versa. These results are in line with the data obtained

from others’ studies, which reported significant correlations between parents’ reports

and children’s performance (for review, see: Feldman et al. 2005; O’Toole and

Fletcher 2010; Thordardottir and Weismer 2010).

Previous research also pointed out some sociocultural and linguistic factors, which

could influence the validity of parents’ reports, such as parents’ education, socio-

economic status or level of language mastery (e.g. Marchman and Martınez-Sussmann

4128

6 3

1

21

41

61

81

Bilingual kindergarten Monolingualkindergarten

Parent report Test examination

Figure 2. Balance between the children’s general language knowledge in Russian and Hebrew

in accordance with the parents’ reports and the children’s test examination. Note: The scale of

the balance ranged from �100 (Child knows only Hebrew) to 100 (Child knows only Russian);

the nearer the score to zero, the more balanced the child’s language knowledge.

14 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

2002; Thordardottir and Weismer 2010). In the same vein, the parents’ reports in our

study were significantly correlated with the children’s general language knowledge,

while our parents had relatively high education level, long duration of residence in the

host country and relatively high level of proficiency in L1 and L2.

The data concerning relative similarity between the immigrant parents’ assess-

ments and language knowledge of their children raise illusions that the immigrant

parents might assess their children’ language in an adequate way and as a result they

can built up their family language management respectively.However, the analysis of absolute similarity in our study demonstrated significant

differences between the parents’ reports and the children’s actual language knowl-

edge. The findings indicated the tendency of these immigrant parents to overestimate

their children’s knowledge in both languages, Russian and Hebrew. Several reasons

can explain this result. One of them is effect of experimental distractibility. Several

previous studies found that parents’ overestimation of children’s skills in general

might be attributed to the fact that young children might demonstrate some skills

inconsistently and only in familiar environments. Such inconsistency is a hallmark ofrecently learned skills requiring additional practice before they are generalisable to

new settings, namely, mastered (Diamond and Squires 1993). Glascoe and Dworkin

(1995) also noted that parental overestimates of language skills might be a function

of children’s greater willingness to verbalise at home. Another possible explanation is

parents’ general tendency to overestimate their children’s skills and abilities (Corder

et al. 2010; Pezdek, Berry, and Renno 2002).

The degree of discrepancy between the parents’ reports and children’s perfor-

mance can be related to subjective character of the parents’ assessments of thechildren’s language knowledge. In principle, lay people’s general assessments might

not be so accurate compared to the results of structured measuring system because of

possible cognitive and motivational bias (e.g. Kelley and Michela 1980; Kruglanski

and Ajzen 1983; Ross 1978). Moreover, it is noteworthy that the parents’ assessments

did not relate to the specific language domains, but evaluated their general language

development. As lay people, the parents might relate to some language domains

while ignoring others.

Finally, the degree of discrepancy between the parents’ reports and children’sperformance can be moderated by different factors, in particular, by the parents’

sociocultural characteristics and their FLP, as will be further discussed.

Criteria of parents’ assessments

Our study shed light on language domains which immigrant parents relate to or

ignore when assessing their children’s language development. More specifically,

correlations were found between parents’ general assessments of the children’slanguage development in both Russian and Hebrew in breadth of vocabulary

knowledge (receptive and productive vocabulary), which is in keeping with previous

findings (see Feldman et al. 2005). Relatively high correlations were also found with

syntagmatic sense relations in both languages.

Interestingly, at the same time, the parents’ reports in both L1 and L2 did not

relate to two out of three measures of the paradigmatic sense relations (categorical

identification and word definition). These data might be explained by fundamental

differences between paradigmatic and syntagmatic sense relations. The syntagmaticsense relations develop and are salient in early childhood. Paradigmatic sense

Journal of Multilingual and Multicultural Development 15

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

relations, however, develop only around age three and might be more salient for

educators than for parents. Thus, in their everyday practice, the parents might take

more notice of their children’s vocabulary richness (syntagmatic sense relations)

than their vocabulary reflecting high order cognition (paradigmatic sense relation),

which develops in a structural and planned way within educational settings (Snow

1990).

Concerning the length of the children’s utterance (MLU) as a measure of

children’s syntactic development, the parents’ general assessments of children’slanguage knowledge found to be unrelated to this essential domain. As our focus on

the parents’ assessments was as a part of their FLP, and not as an assessment tool,

the parents were not asked to report specifically about sentence complexity or length

as in CDI. Note, however, that this finding points out immigrant parents’

insensitiveness to the length of the children’s utterances versus vocabulary richness

in both L1 and L2. In addition, these data reinforces the results based on CDI in L1

and evidenced on inconsistent correlation between the parents’ reports of MLU and

children’s actual performance (Thordardottir and Weismer 2010). Thus, futureresearch on immigrant parents’ assessments of children’s syntactic development as a

part of general language knowledge is necessary.

Parents’ assessments of children’s language development as a rationalisation of theirFLP

The present study shows that immigrant parents tend to rationalise their FLP

through overestimation of children’s general language knowledge. More specifically,it seems that the level of their children’s language development could serve as direct

evidence for the correctness of our parents’ choice of bilingual or monolingual

kindergarten. However, why did the parents from the monolingual group have a

greater tendency to overestimate than the parents from the bilingual group? Because

this study was a part of the broad project on FLP (see ‘‘Method’’), we can answer

this question with regard to the findings obtained in the first stage of this project,

that is, from the in-depth, semi-structured interviews with the parents. The

importance of in-depth semi-structured interviews cannot be over-emphasisedbecause they provide a sensitive method for understanding the processes taking

place within the family (Okita 2002). Based on the data from the interviews with the

parents, we found that the parents in the monolingual group were more doubtful

about their choice, and experienced high internal conflict. In their reasoning about

children’s language development during in-depth interviews, all parents from the

bilingual and monolingual group believed that children’s immersion in the second

language (Hebrew) would not be easy, and possibly even traumatic. However, the

parents from the monolingual group believed that early immersion in L2 could lessenthe trauma. For example, one mother expressed this as follows: ‘Why didn’t I want

him to go to a Russian kindergarten? I think it’s better to overcome the trauma of the

language barrier as early as possible’ (Schwartz, Moin, and Leikin 2011, 13). Thus,

this mother’s perception of the bilingual setting seemed clear, as she believed that the

bilingual kindergarten would hamper L2 development. Accordingly, her misconcep-

tion led her to take the approach of ‘the younger the better’, assuming that younger

children learn L2 more quickly and easily than older children.

Ironically, the results of the present and previous studies provide evidence infavour of the opposite approach to early bilingual development and education: ‘first

16 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

language first’ and relatively late immersion in L2 (Baker 2000; Cummins 2000). This

was the approach adopted by the bilingual kindergartens in our study. The results

provided evidence that after less than one year of immersion in Hebrew, the children

in the bilingual group displayed a general language knowledge level that was similar

to that of the children in the monolingual group, who were enrolled in the

monolingual kindergartens significantly earlier.

Parents in the monolingual group expected their children to have a good

knowledge of Hebrew (L2) and this was one of the main reasons for their choice of

monolingual preschool education. Overestimation of the children’s actual knowledge

of Hebrew seemed to play a more important role in justifying their choice than for

parents in the bilingual group.

Our findings also showed that parents’ support of ‘the younger the better’

approach towards immersion in L2, as part of their FLP, might be misleading and

result in gradual L1 attrition (Wong Fillmore 1991). The parents in the

monolingual group overestimated their children’s knowledge of Russian to a greater

degree than the bilingual group’s parents. The data obtained from the in-depth

interviews with the parents during the qualitative part of the study (Schwartz,

Moin, and Leikin 2011) showed that this discrepancy might be rooted in the

conviction of the parents from the monolingual group that they could compensate

the lack of L1 support in kindergarten by regulating L1 development themselves.

This regulation could be carried out through enhancing L1 input at home (e.g.

reading and exposure to Russian media). In addition, although the parents from the

monolingual group believed in the tremendous power of the home language

environment, which could automatically guarantee L1 language acquisition, some

of them expressed the difficulty of coping with conflicting language demands. In

particular, they felt personally responsible for their child’s limited Russian input,

due to restricted time spent in ‘pure’ interaction with the child (Schwartz, Moin,

and Leikin 2011).

Conclusion

In conclusion, the present study has shown that these immigrant parents tend to

rationalise their FLP (choice of preschool educational setting) by overestimating

their children’s general language knowledge. In addition, the parents’ expectations

about different educational settings (monolingual versus bilingual kindergarten) and

their presentation of children’s bilingual development, as a part of their family

language ideology, might influence their assessments of children’s language knowl-

edge. We assume that exploring parents’ lay assessments of children’s language

knowledge and studying rational and irrational mythical components of these

assessments may extend the understanding of FLP and bilingual development in

early childhood.

We acknowledge that this study is exploratory, and as a result, some of the data

patterns that emerged may be suggestive and require further investigation. In this

context, it is notable that the relatively small size of the sample, specific social and

cultural context, and characteristics of the research population (adult immigrants

from the FSU in Israel, bilingual parents of three- to four-year-old children) might

restrict to some extent the generalisability of our results.

Journal of Multilingual and Multicultural Development 17

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

References

Amaral, O.M. 2001. Parents’ decisions about bilingual program models. Bilingual ResearchJournal 25, no. 1�2: 1�23.

Anglin, J.M. 1985. The child’s expressible knowledge of word concepts. In Children’s language,Vol. 5, ed. K.E. Nelson, 77�127. Hillsdale, NJ: Erlbaum.

Aram, D., and S. Biron. 2004. Joint storybook reading and joint writing interventions amonglow SES preschoolers: Differential contributions to early literacy. Early ChildhoodResearch Quarterly 19, no. 4: 588�610.

Baker, C. 2000. A parents’ and teachers’ guide to bilingualism. Clevedon: Multilingual Matters.Barkhuizen, G. 2006. Immigrant parents’ perceptions of their children’s language practices:

Afrikaans speakers living in New Zealand. Language Awareness 15: 63�77.Barkhuizen, G., and U. Knoch. 2006. Macro-level policy and micro-level planning: Afrikaans-

speaking immigrants in New Zealand. Australian Review of Applied Linguistics 29, no. 1:310�8.

Bates, E., and G.F. Carnevale. 1993. New directions in research on language development.Developmental Review 13: 436�70.

Ben-Rafael, E., M. Lyubansky, O. Glockner, P. Harris, Y. Israel, W. Jasper, and J. Schoeps.2006. Building a Diaspora: Russian Jews in Israel, Germany and the USA. Leiden: Brill.

Berman, R.A., and D.I. Slobin. 1994. Relating events in narrative: A crosslinguisticdevelopment study. Hillsdale, NJ: Erlbaum.

Bloom, P. 2000. How children learn the meaning of words. Cambridge: MIT press.Brown, R. 1973. A first language: The early stages. Cambridge, MA: Harvard University Press.Caldas, S., and S. Caron-Caldas. 2000. The influence of family, school, and community on

bilingual preference: Results from a Louisiana/Quebec case study. Journal of AppliedPsycholinguistics 21, no. 3: 365�81.

Caldas, S., and S. Caron-Caldas. 2002. A sociolinguistic analysis of the language preferencesof adolescent bilinguals: Shifting allegiances and developing identities. Applied Linguistics23, no. 4: 490�514.

Corder, K., E. van Sluijs, A. McMinn, U. Ekelund, A. Cassidy, and S. Griffin. 2010.Perception versus reality: Awareness of physical activity levels of British children.American Journal of Preventive Medicine 38: 1�8.

Cruse, D.A. 1986. Lexical semantics. Cambridge: University Press.Cummins, J. 2000. Language, power, and pedagogy: Bilingual children in the crossfire. Clevedon:

Multilingual Matters.Diamond, K.E., and J. Squires. 1993. The role of parental report in the screening and

assessment of young children. Journal of Early Intervention 17: 107�15.Dopke, S. 1998. Competing language structures: The acquisition of verb placement by

bilingual German-English children. Journal of Child Language 25: 555�84.Dunn, L.M. 1965. Peabody picture vocabulary test. Circle Pines, MN: American Guidance

Service.Duursma, E., S. Romero-Contreras, A. Szuber, P. Proctor, and C.E. Snow. 2007. The role of

home literacy and language environment on bilinguals’ English and Spanish vocabularydevelopment. Applied Psycholinguistics 28: 171�90.

Feldman, H., P. Dale, T. Campbell, D. Colborn, M. Kurs-Lasky, H. Rockette, and J. Paradise.2005. Concurrent and predictive validity of parent reports of child language at ages 2 and3 years. Child Development 76: 856�68.

Feldman, H.M., C.A. Dollaghan, T.F. Campbell, M. Kurs-Lasky, J.E. Janosky, and J.L.Paradise. 2000. Measurement properties of the MacArthur Communicative DevelopmentInventories at ages one and two years. Child Development 71: 310�22.

Felling, S. 2006. Fading Farsi: Language Policy, Ideology, and Shift in the Iranian AmericanFamily. Unpublished doctoral thesis. Washington, DC: Georgetown University. Abstractno. DA3247344.

Fenson, L., P.S. Dale, J.S. Reznick, D. Thal, E. Bates, J. Hartung, S. Pethick, and J. Reilly.1993. The MacArthur Communicative Development Inventories: User’s guide and technicalmanual. San Diego: Singular Publishing Group.

Fenson, L., P. Dale, J.S. Reznick, E. Bates, D. Thal, and S. Pethick. 1994. Variability in earlycommunicative development. Monographs of the Society for Research in ChildDevelopment 59 (5, Serial No. 242).

18 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

Glascoe, F., and P. Dworkin. 1995. The role of parents in the detection of developmental andbehavioral problems. Pediatrics 95: 829�36.

Guiberson, M., and B. Rodriguez. 2010. Measurement properties and classification accuracyof two Spanish parent surveys of language development for preschool-age children.American Journal of Language Psychology 19: 225�37.

Gutierrez-Clellen, V.F., and J. Kreiter. 2003. Understanding child bilingual acquisition usingparent and teachers reports. Applied Psycholinguistics 24, no. 2: 267�88.

Horenczyk, G., and U. Ben-Shalom. 2006. Acculturation in Israel. In A Cambridge handbookof acculturation psychology, ed. D. Sam and J. Berry, 294�310. Cambridge: CambridgeUniversity Press.

Horowitz, T., and E. Leshem. 1998. The immigrants from the FSU in the Israeli culturalsphere. In Profile of an immigration wave, ed. M Sicron and T. Leshem, 291�333.Jerusalem, Israel: The Magnes Press, Hebrew University (in Hebrew).

Jakobson, R. 1971. Selected writings, Vol 3. The Hague: Mouton.Kelley, H., and J. Michela. 1980. Attribution theory and research. Annual Review Psychology

no. 31: 457�501.Kenny, D., and L. Acitelli. 1994. Measuring similarity in couples. Journal of Family Psychology

8: 417�31.King, K., and L. Fogle. 2006. Bilingual parenting as good parenting: Parents’ perspectives on

family language policy for additive bilingualism. The International Journal of BilingualEducation and Bilingualism 9, no. 6: 695�712.

King, K., L. Fogle, and A. Logan-Terry. 2008. Family language policy. Language andLinguistics Compass 2, no. 5: 907�22.

Kruglanski, A.W., and I. Ajzen. 1983. Bias and error in human judgment. European Journal ofSocial Psychology 13: 1�44.

Leshem, E., and M. Lissak. 1999. Development and consolidation of the Russian communityin Israel. In Roots and routes: Ethnicity and migration in global perspective, ed. S. Weil,135�71. Jerusalem: The Magnes Press, Hebrew University.

Lissitsa, S. 2007. Index of immigrants’ integration in Israel. In ‘Russian’ face of Israel: Thefeatures of social portrait, ed. M. Kenigshtein, 141�64. Jerusalem�Moscow: Gesharim. (InRussian).

Lyons, J. 1968. Introduction to theoretical linguistic. Cambridge: University Press.Marchman, V., and C. Martınez-Sussmann. 2002. Concurrent validity of caregiver/parent

report measures of language for children who are learning both English and Spanish.Journal of Language, and Hearing Research 45: 983�97.

Nagy, W.E., and J.A. Scott. 2000. Vocabulary processes. In Handbook of reading research, Vol.3, ed. M.L. Kamil, P. Mosenthal, P.D. Pearson, and R. Barr Vol. 3, 269�84. Mahwah, NJ:Erlbaum.

Nevo, B. 1979. [Peabody Picture Vocabulary Test]. Haifa:University of Haifa.

Okita, T. 2002. Invisible work: Bilingualism, language choice and childrearing in intermarriedfamilies. Amsterdam: John Benjamins.

Oller, D.K., and B.Z. Pearson. 2002. The effects of bilingualism: A background. In Languageand literacy in bilingual children, ed. D.K. Oller and R.E. Eilers, 3�22. Clevedon:Multilingual Matters.

Olshtain, E., and B. Kotik. 2000. The development of bilingualism in an immigrantcommunity. In Language, identity, and immigration, ed. E. Olshtein and G. Hornczyk,210�7. Jerusalem: The Magnes Press, Hebrew University.

Olstein, A. 1995. Paper presented at a conference on Ingathering of the exiles? The impact ofthe 1990s’ wave of immigration on central areas of Israeli life. Sociological SocietyImmigrant absorption section, Tel-Aviv University. (in Hebrew)

Ordonez, C.L., M.S. Carlo, C.E. Snow, and B. McLaughlin. 2002. Depth and breadth ofvocabulary in two languages: Which vocabulary skills transfer? Journal of EducationalPsychology 94, no. 4: 719�28.

O’Toole, C., and P. Fletcher. 2010. Validity of a parent report instrument for Irish-speakingtoddlers. First language 30: 199�217.

Journal of Multilingual and Multicultural Development 19

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

Paradis, J., K. Emmerzael, and T.S. Duncan. 2010. Assessment of English language learners:Using parent report on first language development. Journal of Communication Disorders25: 474�97.

Pease-Alvarez, C. 1993. Moving in and out of bilingualism: Investigating native languagemaintenance and shift in Mexican-descent children. Santa Cruz, CA and Washington, DC:National Center for Research on Cultural Diversity and Second Language Learning.

Pezdek, K., T. Berry, and P. Renno. 2002. Children’s mathematics achievement: The role ofparents’ perceptions and their involvement in homework. Journal of Education Psychology94: 771�7.

Richard, G.I., and M.A. Hanner. 1985. Language processing test. Moline, II: Linguisystems.Ring, E., and L. Fenson. 2000. The correspondence between parent report and child

performance for receptive and expressive vocabulary beyond infancy. First language 20:141�59.

Rom, A., and L. Moreg. 1999. [Assessment of oral language processing]. Tel-Aviv:Kibbutzim College of Education.

Ross, L. 1978. The intuitive psychologist and his shortcomings: Distortions in the attributionprocess. In Cognitive theories in social psychology, ed. L. Berkowitz, 337�84. New York:Academic Press.

Schwartz, M. 2010. Family language policy: Core issues of an emerging field. AppliedLinguistics Review 1, no. 1: 171�92.

Schwartz, M., M. Leikin, Y. Shaul, R. Fuhrman-Engel, and O. Skarbovsky. 2007. Languagetests [in Hebrew]. Unpublished tests. Haifa: University of Haifa.

Schwartz, M., V. Moin, and M. Leikin. 2011. Parents’ discourses about language strategies forthe child’s preschool bilingual development. Diaspora, Indigenous, and Minority Educa-tion: An International Journal 5: 149�66.

Schwartz, M., V. Moin, M. Leikin, and A. Breitkopf. 2010. Immigrants’ family languagepolicy toward children’s preschool bilingual education: Parents’ perspective. InternationalMultilingual Research Journal 4: 107�24.

Snow, C.E. 1990. The development of definition skill. Journal of Child Language 17: 697�710.Spolsky, B. 2004. Language policy. Cambridge: Cambridge University Press.Spolsky, B. 2007. Family language management: Some preliminaries. In Studies in language

and language education: Essays in honor of Elite Olshtain, ed. A. Stavans andI. Kupferberg, 429�49. Jerusalem: The Magnes Press, Hebrew University.

Spreen, O., and E. Strauss. 1998. A compendium of neuropsychological tests: Administration,norms, and commentary. New York: Oxford University Press.

Statistical Abstract of Israel. 2010. Central Bureau of Statistics, Vol. 61. Jerusalem:Government Publishing House.

Tabors, P., and C. Snow. 1994. English as a second language in preschools. In Educating secondlanguage children: The whole child, the whole curriculum, the whole community, ed.F. Genesee, 103�25. New York: Cambridge University Press.

Thal, D.J., L. O’Hanlon, M. Clemmons, and L. Fralin. 1999. Validity of a parent reportmeasure of vocabulary and syntax for preschool children with language impairment.Journal of Language, and Hearing Research 42, no. 2: 482�96.

Thordardottir, E.T., and S.E. Weismer. 2010. Language assessment via parent report:Development of a screening instrument for Icelandic children. First Language 16: 265�85.

Verhallen, M., and R. Schoonen. 1993. Lexical knowledge of monolingual and bilingualchildren. Applied Linguistics 14, no. 4: 344�63.

Verhallen, M., and R. Schoonen. 1998. Lexical knowledge in L1 and L2 third and fifthgraders. Applied Linguistics 19, no. 4: 452�70.

Vygotsky, L. S. 1962. Thought and language. Cambridge, MA: Harvard University Press.Wesche, M., and T.M. Paribakht. 1996. Assessing vocabulary knowledge: Depth versus

breadth. Canadian Modern Language Review 53: 13�40.Wong Fillmore, L. 1991. When learning a second language means losing the first. Early

Childhood Research Quarterly 6: 323�46.Worthy, J., and A. Rodriguez-Galindo. 2006. Mi hija vale dos personas’’: Latino immigrant

parents’ perspectives about their children’s bilingualism. Bilingual Research Journal 30:579�601.

20 M. Schwartz and V. Moin

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011

Yelenevskaya, M., and L. Fialkova. 2003. From ‘muteness’ to eloquence: Immigrants’narratives about languages. Language Awareness 12: 30�48.

Yip, V., and S. Matthews. 2007. The bilingual child: Early development and language contact.Cambridge: Cambridge University Press.

Zentella, A. ed. 2005. Building on strength: Language and literacy in Latino families andcommunities. New York: Teachers College Press.

Journal of Multilingual and Multicultural Development 21

Dow

nloa

ded

by [

Uni

vers

ity C

olle

ge L

ondo

n] a

t 03:

38 2

3 D

ecem

ber

2011