Content uploaded by Rahat Zaidi
Author content
All content in this area was uploaded by Rahat Zaidi on Sep 11, 2015
Content may be subject to copyright.
http://ecr.sagepub.com/
Journal of Early Childhood Research
http://ecr.sagepub.com/content/11/1/3
The online version of this article can be found at:
DOI: 10.1177/1476718X12449453
2013 11: 3 originally published online 12 October 2012Journal of Early Childhood Research
Rahat Naqvi, Keoma J Thorne, Christina M Pfitscher, David W Nordstokke and Anne McKeough
diverse classrooms
Reading dual language books: Improving early literacy skills in linguistically
Published by:
http://www.sagepublications.com
can be found at:Journal of Early Childhood ResearchAdditional services and information for
http://ecr.sagepub.com/cgi/alertsEmail Alerts:
http://ecr.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://ecr.sagepub.com/content/11/1/3.refs.htmlCitations:
What is This?
- Oct 12, 2012OnlineFirst Version of Record
- Jan 29, 2013Version of Record >>
by guest on January 29, 2013ecr.sagepub.comDownloaded from
Journal of Early Childhood Research
11(1) 3 –15
© The Author(s) 2012
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1476718X12449453
ecr.sagepub.com
Reading dual language books:
Improving early literacy skills in
linguistically diverse classrooms
Rahat Naqvi, Keoma J Thorne, Christina M Pfitscher,
David W Nordstokke and Anne McKeough
University of Calgary, Canada
Abstract
Research has determined that dual language books have a positive effect on literacy achievement,
motivation, and family involvement in children’s schooling. In this study we used quantitative methods
to complement the largely qualitative extant research. We analyzed the early literacy skills of 105
kindergarten children (45 comparison, 60 treatment) with diverse language backgrounds (35% English,
31% Punjabi, 16% Urdu, 18% other languages) from eight kindergarten classes in four suburban Canadian
schools. Statistical analyses indicated that children who were read to using dual language books, written
in French, Punjabi, and Urdu, demonstrated significantly greater gains in graphophonemic knowledge
than children who were read to in English only. This gain occurred specifically in children who spoke
the targeted languages at home; children who did not speak the targeted languages were not negatively
affected. Findings are discussed in terms of developing metalinguistic awareness and directions for
practice and research are discussed.
Keywords
dual language books, early childhood education, early literacy, linguistic diversity
North America is highly linguistically diverse. In 1990, one in 20 public school students in
kindergarten to grade 12 residing in the United States was an English language learner (ELL)
(Capps et al., 2005), that is, a student who speaks English either not at all or with enough limita-
tions that he or she cannot fully participate in mainstream English instruction (Ministry of
Education Alberta, 2009). Today the figure is one in nine (Capps et al., 2005). Demographers
estimate that in 20 years it might be one in four. The ELL population has grown from two mil-
lion to five million since 1990, a period when the overall school population increased only 20
percent (Capps et al., 2005). Similarly, in 2006 there were over six million immigrants living in
Canada (Statistics Canada, 2010). Approximately 58 percent of these immigrants have come
Corresponding author:
Rahat Naqvi, Faculty of Education EDT 1118, University of Calgary, 2500 University Drive, Calgary, Alberta, Canada
T2N 1N4.
Email: rnaqvi@ucalgary.ca
11110.1177/1476718X12449453Naqvi et al.Journal of Early Childhood Research
2012
Article
by guest on January 29, 2013ecr.sagepub.comDownloaded from
4 Journal of Early Childhood Research 11(1)
from Asia and the Middle East (Liang, 2010), with the result that the linguistic composition of
Canada has been significantly impacted. As an example, the top immigrant home languages in
a large urban centre in western Canada are: English, Chinese languages, Punjabi, Tagalog,
Spanish, German, Vietnamese, Arabic, Polish, and Urdu (Liang, 2010). Data from the 2006
Canadian Census showed that between 2001 and 2006, 80 percent of Canadian immigrants had
a mother tongue other than English or French (Statistics Canada, 2010). These immigration
patterns have resulted in a steady increase in the number of school children whose home lan-
guage is not English.
This demographic shift in the linguistic composition of North American classrooms has
generated unique challenges for educators. In response, the development and implementation
of strategies that help ELLs acquire skills in both their home language and English, have been
encouraged by educational researchers and policy makers (Gutiérrez et al., 2010; Lesaux and
Gava, 2006). Such stakeholders have noted that access to appropriate resource materials, along
with appropriate teaching strategies, can substantially improve the academic achievement of
ELLs (e.g. Brisk and Harrington, 2007). Despite these suggestions, mainstream primary school
teachers are largely ill equipped to address the needs of students from diverse ethnic and lin-
guistic backgrounds because they often do not have access to appropriate materials or home
language speakers who can assist in the classroom (Naqvi, 2009). Nor do early childhood edu-
cators typically have training in effective teaching strategies to appropriately address these
children’s academic needs with respect to home and English literacy acquisition and skill
development (Canadian Council on Learning, 2006). This is especially important in early child-
hood education, which focuses on literacy acquisition and early reading skills (e.g. phonologi-
cal awareness, letter and word recognition, comprehension), as these abilities are foundational
to later language and literacy development (Dickinson and Tabors, 2001; Paris, 2005; Whitehurst
et al., 1994).
Metalinguistic awareness is another essential part of literacy development (Laurent and
Martinot, 2009) and refers to the development of an explicit awareness of linguistic form and
structure (Cazden, 1974). ‘This ability to reflect upon and manipulate the structural features of
language – e.g. phonological, morphological and syntactic structure – has been defined as meta-
linguistic’ (Laurent and Martinot, 2009: 29). ‘Learning to read, as a formal linguistic task, requires
the learner to develop an explicit awareness of his/her language, which must be intentionally
monitored, an awareness of language such that it can become the object of discussion’ (Laurent
and Martinot, 2009: 30). Research has shown that learning two languages helps to develop meta-
linguistic awareness (Bialystok et al., 2005) as bilingual learners use their metalinguistic aware-
ness to compare and contrast two language systems to discover commonalities and differences
(Koda, 2008). Consequently, they tend to have better phonological awareness than monolingual
children if their home language is syllabically complex (Campbell and Sais, 1995).
Dual language book reading programs are one approach that may address the diverse ethnic and
linguistic composition of classrooms, as they target home language literacy and literacy develop-
ment in English. Dual language books offer the same narrative in two languages, typically English
and another target language, with illustrations to link visual and textual representations. Researchers
have argued that dual language books help children feel a part of a community (Ma, 2008; Mullis,
2007; Robertson, 2006), develop their personal and cultural identity (Fort and Stechuk, 2008; Ma,
2008; Robertson, 2006; Taylor et al., 2008), improve literacy in their home language (Ma, 2008;
Sneddon, 2009), increase their metalinguistic awareness (Robertson, 2006), and improve their
English literacy (Cummins, 2007). Most of this research, however, has focused on older children,
who read at least part of the stories themselves, and has overlooked emergent-literacy classrooms
by guest on January 29, 2013ecr.sagepub.comDownloaded from
Naqvi et al. 5
where dual language books must be read aloud. Additionally, many studies have examined the
impact of dual language books written in a child’s home language on the child’s home language
literacy and less is known about how dual language books written in English and another language
impact children’s English literacy. Further, research regarding the potential literary benefits of dual
language books for children who do not speak or read in the target dual language is not available.
Finally, the research on dual language books is based primarily on qualitative methods and case
studies and although this evidence is valuable, empirical research is required to advance our under-
standing (Sneddon, 2009).
As a result of these gaps in the corpus of dual language book research, the benefits of dual
language books cannot be confidently asserted (Tsow, 1986), warranting classroom-based empiri-
cal research (Sneddon, 2009). Thus, the current study sought to empirically investigate the effect
that reading dual language books aloud has on the foundational literacy skills of young children
from diverse ethnic and linguistic backgrounds.1 The study is premised on the belief that the use
of evidence-based methods and instructional materials known to develop early literacy skills is
critical for families and early childhood educators. We chose to read aloud to the children because
this practice has been demonstrated to be one of the primary ways that children gain alphabetic
knowledge (Scarborough and Dobrich, 1994), understand book and print conventions and pat-
terns (Neuman et al., 2000), and learn the relationship between words and meaning (Gold and
Gibson, 2001). Reading books aloud also offers children a good model of expert reading
(Mcquillan, 2009).
Purpose of the research
In this article we present academic outcome findings from the first year of a two-year longitudinal
study on the effects of dual language books across academic and cultural outcomes. The primary
target language was English and the dual languages were French, Urdu, or Punjabi. Dual language
books written in English/French were used because of the bilingual nature of Canada, and books
written in English/Punjabi and English/Urdu were used given the high proportion of native Punjabi
and Urdu speakers in western Canada. Our intent was two-fold. First we aimed to determine if the
early reading skills (i.e. knowledge of alphabet, conventions, and meaning) of children who were
read to using dual language books (treatment group) would be different to children who were read
to in English only (comparison group). Second, we sought to establish if the early reading skills
differed among English, Punjabi, and Urdu speaking children who received the dual language
instruction to investigate if dual language book reading may have more of an impact for children
who speak the dual language at home than for children who do not.
Methods
Participants
Child participants were drawn from four schools that were located within a 3-km radius of each
other in a highly ethnically and linguistically diverse suburban area in a large western Canadian
city. Each school had two kindergarten classes.2 Approximately 160 kindergarten students were
eligible for participation in the study and the parents of 115 of these students gave permission for
their children to participate in the study. Attendance was recorded throughout the study and data
from children who missed 40 percent of the dual language book reading program were eliminated.
Arguably, students who missed almost half of the lessons would not accrue the potential benefit of
the DLB program and, if included in the statistical analysis, might result in a Type 2 error (i.e. a
by guest on January 29, 2013ecr.sagepub.comDownloaded from
6 Journal of Early Childhood Research 11(1)
false negative or finding no significant effect when one existed). With these students eliminated
and due to attrition over the course of the program, results are based on 105 participants: 45
students in the comparison group (27 boys, 18 girls) and 60 students in the treatment group (32
boys, 28 girls). The mean age at the time of pre-testing was five years for both the comparison
and the treatment group. Mean ages did not differ significantly between the treatment and com-
parison groups (t(103) = .369, p > .05). The mother tongues of participants in the comparison
group were 29 percent English, 36 percent Punjabi, 16 percent Urdu, and 19 percent other.
Similarly, the mother tongues of children in the treatment group were 40 percent English, 27
percent Punjabi, 17 percent Urdu, and 16 percent other. There were no native French speakers in
either group.
Teachers and readers also participated in the study by delivering the program. Teachers included
eight females who held university degrees in teaching and whose teaching experience ranged from
recently graduated to veteran. The readers included seven females and one male, each of whom
spoke English and either French, Punjabi or Urdu fluently. All guest readers were immigrants. All
were associated with the school in some fashion. More specifically, five readers had a child who
attended one of the schools, one reader had a grandchild who attended one of the schools, and one
reader was a support staff employee at one of the schools.
Measures
Test of Early Reading Ability, 3rd edn (TERA-3; Reid et al., 2001), a standardized test of early
reading abilities for children ages 3–6 to 8–6 years, measured the participants’ early English read-
ing skills across three areas: a) alphabet: graphophonemic knowledge and the recognition of
printed letters and words; b) conventions: understanding of the conventions of written language
and reading; and c) meaning: comprehension that printed language conveys meaning and infor-
mation. While the TERA-3 yields several scores, only standard scores (SSs) for the three subtests
(alphabet, conventions, and meaning) were of interest in this study as they allowed us to target
specific skill development. SSs have a mean of 10 and a standard deviation of 3, and are classified
as falling within the very poor range (1–3), poor range (4–5), below average range (6–7), average
range (8–12), above average range (13–14), superior range (15–16), or very superior range (17–
20). The TERA-3 has two equivalent forms (Form A and Form B), allowing for re-testing within
a short time period.
Procedures
First, participation was sought from schools that have a large and diverse immigrant population,
particularly Urdu and Punjabi speaking people. School board administrators circulated a brief
description of the proposed project to all elementary schools within one quadrant of the city in
which the study took place, due to its high immigrant population. Four elementary schools (two
kindergarten classes in each school) agreed to participate. The study’s purpose and procedures
were explained fully to the schools’ administrators and kindergarten teachers. Second, when
school participation was granted, parents of potential child participants were contacted by school
administrators and teachers through written information prepared by the researchers, which was
available in English, Urdu and Punjabi. As well, the research team provided information sessions
at the schools. These oral presentations and question and answer sessions were also available in
English, Urdu and Punjabi. Third, a final list of dual language books was selected, based on age-
appropriate level and interest, and on availability of the books (see the Appendix Table for a list
by guest on January 29, 2013ecr.sagepub.comDownloaded from
Naqvi et al. 7
of the dual language books used). Fourth, readers who were fluent in the various languages were
sought to help with instruction through discussion with school personnel, the schools’ newslet-
ters, and word of mouth. Prior to instructing, the readers attended a one-hour workshop on the
principles and functions of dual language book reading prior to delivering and conducting the
readings. The aim was to help establish consistency in reading and teaching methods across
classes and schools. Finally, informed consent was received from the parents of kindergarten
participants, teachers and administrators, and readers. This initial portion of the research required
approximately six weeks to complete.
School administrators randomly assigned one class in each school to the treatment group and
the other class to the comparison group. The work with the children took place over 15 weeks.
First, children were pre-tested with the TERA-3 Form A over a two-week period within six-weeks
of the start of the school year. Research assistants, who were trained in assessment, individually
administered the TERA-3 to the participants during school hours in the school building. Test
administration was done in a standardized fashion as stipulated in the TERA-3 manual. Second,
the 11-week reading program began immediately after the pre-testing. Classroom teachers and
readers delivered the dual language book reading program and each reading session was video-
recorded. Classroom teachers read the English text and sometimes the French text, whereas read-
ers read in Punjabi, Urdu, and sometimes French. Readings occurred in the classroom and each
week a new story was read (i.e. 11 stories in total). Both groups received the reading program
three times per week and each reading took between 15 and 25 minutes. For each reading, the
children in the treatment group were concurrently read to (i.e. page-by-page) in English and the
dual language (i.e. either French, Punjabi, or Urdu), whereas children in the comparison group
were read to in English only. In both groups, teachers encouraged the children to ask questions,
make predictions, summarize, and share their thoughts and feelings related to the story content.
Third, at the end of the 11-week program, post-testing with the TERA-3 Form B was completed
over a two-week period, again in standardized fashion and under the same conditions as the pre-
test. During the length of the study, all children continued to receive their regular classroom lit-
eracy instruction.
Data analysis
Independent sample t-tests were conducted on pre-test data to compare for differences between the
baseline reading abilities of a) the treatment and control groups and b) the speakers of the target
dual languages (Urdu and Punjabi) to speakers of all other languages. To determine the effective-
ness of dual language books in promoting early reading skills, simple change scores (i.e. TERA-3
pre-test SS minus TERA-3 post-test SS) of the comparison and treatment groups were compared.
Simple change scores of Urdu and Punjabi speaking children (referred to as the UP group) and
children speaking all other languages (referred to as the non-UP group) were also compared
within and across the treatment and comparison groups. Due to the presence of several outliers
that likely skewed the sample distributions, a non-parametric statistical approach (i.e. Mann-
Whitney U test) was used to investigate group differences. A significance level of .05 was set for
all comparisons.
Results
Median, interquartile range (IQR), and change score values for the comparison and treatment
groups are displayed in Table 1. Independent samples t-tests, which compared the means of two
by guest on January 29, 2013ecr.sagepub.comDownloaded from
8 Journal of Early Childhood Research 11(1)
independent samples, demonstrated that the early reading abilities of the comparison (N = 51) and
treatment (N = 64) groups across alphabet, conventions, and meaning did not differ prior to the
dual language book program (i.e. at pre-test). To examine if dual language books impacted early
literacy abilities, the reading skills change scores of the treatment and comparison groups were
compared using the Mann-Whitney test, which indicated that the treatment group experienced
significant greater change on the TERA-3 alphabet subtest than the comparison group (U = 918.5,
z = –2.8, p = .005). The medians demonstrate that children who received dual language book read-
ing experienced significantly greater gains in their recognition and knowledge of printed letters
and words than children who only heard stories in English. Change between the treatment and
comparison group did not differ significantly on the conventions or meaning subtests. However, it
is important to note that there was no decrement in performance on these subtests, which suggests
that provision of dual language book reading did not negatively impact students’ knowledge of the
conventions and meaning of print.
Due to a small sample size and the absence of French speaking children, to determine if the
reading skills of children in the treatment group differed depending on their understanding of the
target dual language (i.e. Punjabi or Urdu), treatment group data were separated into two. The UP
group was composed of 28 students (Urdu speaking = 10, Punjabi speaking = 18) and the non-UP
group was composed of 36 children (English speaking = 24, other language speaking = 12).
Median, IQR, and change score values for the UP and other groups are displayed in Table 2. Early
reading abilities as measured by the pre-test were first compared to establish a baseline. An inde-
pendent samples t-test demonstrated a significant difference between the UP and non-UP groups
on the TERA-3 meaning subtest (t(62) = –2.2, p = .03) and the medians show that the UP group
Table 1. Median TERA-3 reading standard scores (interquartile range) pre- and post-program and change
scores for comparison and treatment groups
TERA-3 Comparison group Treatment group
Pre Post Change Pre Post Change
Alphabet 8 (6–11) 10 (7.5–12) 0 (−1–4) 8 (6–10) 11 (9–12) 2 (1–4)
Conventions 8 (5–9) 9 (6.5–10.5) 1 (0–2.5) 7.5 (6–9) 9 (6–10) 1 (0–3)
Meaning 8 (7–10) 8 (5–11) 0 (−2.5–2) 8 (7–9) 9 (7–10.75) 1 (−1–2)
Note: Pre and post values are reported as standard scores with a mean of 10 and standard deviation of 3. TERA-3 = Test
of Early Reading Ability, 3rd edn.
Table 2. Median TERA-3 reading standard scores (interquartile range) pre- and post-program and change
scores for Urdu or Punjabi speakers and other language speakers in the treatment group
TERA-3 Treatment UP group Treatment other group
Pre Post Change Pre Post Change
Alphabet 8 (6–9) 11 (9–12) 3 (1.8–4.3) 8 (6–11) 11 (8.5–12.3) 1 (0–3.3)
Conventions 7.5 (5.3–9) 9 (6–11) 1 (−.3–3) 7.5 (6–9) 8.5 (6–10) 1 (0–2.3)
Meaning 8 (7–10) 8 (6–9.3) 1 (-1–2) 8 (7–10) 9.5 (7–11) 1 (−1–2)
Note: UP group = children’s whose home language is Urdu or Punjabi, Other Group = children’s whose home language
is neither Urdu nor Punjabi. Pre and post values are reported as standard scores with a mean of 10 and standard devia-
tion of 3. TERA-3 = Test of Early Reading Ability, 3rd edn.
by guest on January 29, 2013ecr.sagepub.comDownloaded from
Naqvi et al. 9
scored significantly lower on the TERA-3 meaning subtest than the non-UP group. There were no
significant differences between the groups on the alphabet or conventions subtests.
The Mann-Whitney test was then used to determine if there were differences between the read-
ing change scores of the UP and Other groups as a result of the dual language book reading (i.e.
within the treatment group). This analysis revealed a significant difference between the change
scores of the UP and non-UP groups on the TERA-3 alphabet subtest (U = 308.5, z = –2.0, p = .04).
As evident from the medians, the Urdu and Punjabi speaking children experienced significantly
greater gains in their alphabet knowledge, after the dual language book reading, than children who
spoke other languages. There were no differences between the change scores of the groups on the
conventions or meanings subtests.
Based on the findings that children in the treatment group who spoke Urdu and Punjabi experi-
enced greater gains in their alphabet knowledge compared to children who spoke other languages,
further analyses were conducted. First, we investigated if differences in alphabet change scores were
evident between children who spoke Urdu and Punjabi (UP group) in the comparison (N = 27) and
treatment (N = 28) groups. The Mann-Whitney test revealed that there were significant differences
between the change of the comparison and treatment UP groups on the alphabet subtest (U = 155.0,
z = –2.9, p = .004), with the treatment UP group (Mdn change = 0, IQR = −1 – 4) demonstrating
greater alphabet knowledge gains than the comparison UP group (Mdn change = 3, IQR = 1.8 – 4.3).
A final Mann-Whitney test was used to examine if there were differences on the alphabet subtest
between children whose home language was not targeted in the dual language books (non-UP) in
the comparison (N = 24) and treatment (N = 36) groups. There were no significant differences
between the comparison (non-UP) group’s median change (0.5, IQR = 0 – 4) and the treatment
(non-UP) group’s median change (1, IQR = 0 – 3.3).
Discussion
The primary purpose of this study was to investigate the efficacy of using dual language books as
instructional materials to promote early literacy skills in linguistically and ethnically diverse young
children. Additionally, we explored whether the impact of dual language books varied according to
students’ home language relative to the languages contained in the dual language books. The main
finding from the study was that children who spoke Urdu or Punjabi gained significantly more
graphophonemic knowledge and knowledge printed letters and words, as measured by the alphabet
subtest of the TERA-3, when read to using dual language books in those languages than did their
peers who were read the same books in English only. This finding lends empirical support for the
claim that dual language books are a useful tool for developing these foundational literacy skills in
kindergarten classrooms and confirms research that has repeatedly demonstrated that pedagogical
practices and teaching materials impact children’s academic achievement (Conteh, 2007; Martin
et al., 2006; Robertson, 2006).
A second finding was that children in the comparison and treatment groups, who spoke neither
Urdu nor Punjabi, did not differ significantly in terms of graphophonemic knowledge and knowl-
edge printed letters and words when they were read English only books or dual language books in
those languages. This finding suggests that the use of dual language books does not hinder the
development of these foundational literacy skills in children who do not speak these languages.
This conclusion is further supported by the finding that there were no significant differences
between the comparison and treatment groups on the conventions and meaning subtests.
A third finding was that children in the treatment group, who spoke Urdu and Punjabi at home,
functioned significantly lower on the meaning subtest at pretest than children in the treatment group
by guest on January 29, 2013ecr.sagepub.comDownloaded from
10 Journal of Early Childhood Research 11(1)
who did not speak Urdu and Punjabi (i.e. non-UP group). Because the latter group was composed of
twice as many native English-speaking children as children who had another home language, it is
reasonable to assume that this difference in meaning knowledge was language related. Indeed, evi-
dence has suggested that ELLs demonstrate lower academic achievement test scores compared to their
English speaking peers (Ministry of Education Alberta, 2006). This trend has long been recognized in
previous studies of bilingual children learning to read in English (e.g. Paez et al., 2007). It is notewor-
thy that there was, however, no significant difference in meaning change scores between these two
groups, which suggests that the Urdu and Punjabi speakers did not remain behind their non-UP peers.
Beyond offering support for dual language books in linguistically diverse kindergarten class-
rooms, the current findings raise several intriguing questions. First and we believe most interest-
ing, what happens during dual language book reading that might impact alphabet knowledge? We
propose that exposing the UP group (i.e. native Urdu and Punjabi speakers) to different languages
and scripts alongside their home language supported the development of these children’s concepts
of words, letters, and graphophonic representation. This conceptual development, we argue, is
under-girded by metalinguistic awareness (i.e. knowledge of the form of language as distinct from
its content), which has been shown to be important for early reading development, particularly for
students who may be learning to read in a language they do not speak at home (Bialystok et al.,
2005). Observations of interaction between UP group members (pseudonyms: Jai, Ashprit, and
Aanisah) and readers in the current study demonstrate children’s metalinguistic awareness, as
shown in the following lesson transcript segment in which What Shall We Do with the Boo Hoo
Baby? was read in English and Punjabi.3
Teacher [reading]: What shall we do with the Boo-Hoo Baby?
Jai [speaking]: The English rhymes with the Punjabi!
[Jai commented on the similar sounds of the English ‘Boo-Hoo’ and the Punjabi ‘Hoonhoon’.]
Reader [reading]: Aino dud pilao kutta bolaya. (‘Feed him,’ said the dog)
Teacher [reading]: ‘Feed him,’ said the dog.
Ashprit [repeating]: Kutta (dog)
Teacher [repeating]: Kutta (dog)
Aanisah: In Urdu too.
[Aanisah noted the similarity between the Punjabi and Urdu words for dog.]
Reader [reading, then asking]: Bacha booliya kiday bacha boliya? (The baby said . . . What did the baby
say?) Hoonhoon (Boo Hoo)
Ashprit: Ronay lag gaya (He started crying)
Reader: Haan (Yes)
Teacher [reading]: ‘Boo Hoo Hoo,’ said the baby.
Reader [reading]: Anoo sulao batakh ney kayha (‘Put him to bed,’ said the duck.)
Aanisah [pointing at ‘ZZZZZZZZ’ in the reader’s book ]: In Urdu it goes like this [Aanisah writes the Urdu
letter ‘zāl’ with her finger in the air]. . . and a dot that means zāl. (ذ). By doing so, she shares that Urdu has
an equivalent sound that is represented by the letter ‘zāl’].
Reader: Haan (Yes).
by guest on January 29, 2013ecr.sagepub.comDownloaded from
Naqvi et al. 11
In the immediately preceding segment, Aanisah noted the form of the word in her native Urdu
language and the reader nods his head in agreement. This Punjabi reader is also fluent in Hindi and,
consequently, knows that Urdu and Hindi are jointly comprehensible on an oral level but not on a
written level (i.e. a speaker of Hindi understands spoken Urdu and vice versa but neither can read
the other’s script as Urdu is derived from Persian and Arabic and Hindi comes from Sanskrit). In
other words, he acknowledges the unique symbols Aanisah describes and traces in the air to repre-
sent ZZZZZZZ.
We interpret these examples as demonstrating the children’s metalinguistic awareness as they
are clearly working with the forms the languages are expressed through rather than the content
expressed through the language. Laurent and Martinot (2009) decomposed metalinguistic aware-
ness into metaphonological abilities (letter or strings of letters that represent phonemes) and
metasyntactic abilities (expectations of conventions). It is the former that we believe to be at
play here.
A second question that emerged for us from the results of this study was as follows. Why did we
not see significant results on the conventions subtest, which taps into Laurent and Martinot’s
(2009) second type of metalinguistic ability, metasyntactic abilities? We reasoned that because the
teachers read to the children and did not provide a time for them to interact physically with the
books, they were not able to learn such things as left to right versus right to left text reading (i.e.
print and book conventions). Further, despite the differences in text conventions from English to
Urdu, these elements were not explicitly mentioned during the dual language book reading, so
children in the treatment group may not have been aware of the difference. Dual language books
written in English and Urdu/Punjabi offer excellent opportunities to focus on such conventions,
however. Elements such as this should be included in future research.
We also considered why there was no significant difference between the comparison and
treatment groups’ meaning change scores, as reading aloud to children has been shown to help
them relate words and meanings (Gold and Gibson, 2001). The current lack of significant find-
ings might be explained by the nature of the TERA-3 meaning subtest, which requires children
to identify items that might not be familiar to them, such as a photo of Jello, which contains gela-
tin made from animal fat rendering and is thus inappropriate for consumption for most Muslim
children and many Punjabi/Sikh children. Additionally, the meaning subtest required children to
identify cultural icons, such MacDonald’s yellow ‘M’. Because many of the children in the study
do not eat this type of food, they may not be as familiar with the icon as mainstream North
American children, evidenced in there consistently incorrect responses. Moreover, the dual lan-
guage books used in the program did not focus on these cultural icons and so did not impact this
knowledge base. Given that many of these students come from homes where English is not the
primary language and that a number of students had recently immigrated to Canada from other
countries, these cultural references may have unfairly disadvantaged our participants.
Additionally, the TERA-3 was standardized on 875 children from 22 states in the United States
and demographic characteristics were representative of the United States. Specifically, the stand-
ardization sample was 68 percent European American, 15 percent African American, 13 percent
Hispanic American, 3 percent Asian/Pacific Islander, and 1 percent Native American/Eskimo/
Aleut, which clearly differs starkly from the ethnic composition of our sample. Taken together,
these points suggest that the TERA-3 may not have been the most appropriate measure to use
with this population. However, the authors thoroughly researched the available instruments and
were unable to locate a more appropriate reliable, valid, and standardized measure of early read-
ing skills, which they felt was critical to conducting one of the first empirically valid studies of
dual language books.
by guest on January 29, 2013ecr.sagepub.comDownloaded from
12 Journal of Early Childhood Research 11(1)
Implications
The results of this study hold significant implications for the use of dual language books with chil-
dren whose first language is other than English. Because alphabetic knowledge is an essential ele-
ment of early literacy development, the current finding that young Urdu and Punjabi speaking
participants achieved significantly greater gains than children who heard stories in English only is
noteworthy. Moreover, the empirical evidence in support of the use of dual language books in
elementary classrooms offers teachers and policy-makers quantitative evidence that supports qual-
itative evidence that has already informed practice.
Although findings are promising, they should be interpreted with caution. Our sample size was
small and not demographically representative of the typical Canadian classroom. The logistics of
working in schools and classrooms presented some challenges to this research. For example, dis-
ruption and noise was sometimes evident during sessions. Some readers reported difficulties with
some of the translations. For instance, many of the translated words were too formal and not used
in everyday language, which resulted in the readers thinking that the dual language speaking stu-
dents might be unfamiliar with the translated readings. Readers also mentioned some difficulties
making the readings engaging for all students, especially those children who did not speak the
second language. Because readers also hoped to extend the home language and English literacy of
the dual language learners, they sometimes found it difficult to balance extension and discussion
of the text with ensuring the attentiveness and interest of all children. Clearly more research is
needed to improve the program and to replicate these findings. As well, longitudinal work is neces-
sary to determine if gains in early literacy skills are maintained. In conclusion, we concur with
Gutiérrez et al. (2010: 338) that there is a need for ‘a robust research agenda that focuses on young
simultaneous bilinguals . . . [and] encourage[s] the development of language and literacy interven-
tions that serve as cultural amplifiers’.
Acknowledgements
This research was generously supported by the Alberta Center for Child Family and Community Research.
We wish to gratefully acknowledge the participation of the research team and partner educators in the partici-
pating schools in the Calgary Board of Education. We also thank all the parents, volunteer readers, teachers,
and support staff, as well as the student participants who took part in the dual language book reading
program.
Appendix Table. Storybooks used in the dual language book reading program in order of presentation
Title Author and/or translators Publisher, place, date
1 Brown Bear, Brown Bear,
What Do You See?
E: B. Martin Henry Holt, New York, 1992
F: L. Bourguignon Mantra Lingua, London, 2008
P: none indicated Mantra Lingua, London, 2004
U: Q. Zamani Mantra Lingua, London, 2004
2 The Very Hungry Caterpillar E: E. Carle Philomel Books, New York, 1987
F: L. Bourguignon Mantra Lingua, London,2004
(Continued)
by guest on January 29, 2013ecr.sagepub.comDownloaded from
Naqvi et al. 13
Title Author and/or translators Publisher, place, date
P: K. Manku Mantra Lingua, London, 2004
U: Q. Zamani Mantra Lingua, London, 2004
3 Floppy’s Friends E: G. Van Genechten Mantra Lingua, London, 2004
F: G. Orio-Glaunec Mantra Lingua, London, 2004
P: none indicated Mantra Lingua, London, 2004
U: Q. Zamani Mantra Lingua, London, 2004
4 Dear Zoo E: R. Campbell Simon & Schuster, New York, 1999
F: none indicated Mantra Lingua, London, 2004
P: none indicated Mantra Lingua, London, 2004
U: none indicated Mantra Lingua, London, 2004
5 What Shall We Do With
the Boo Hoo Baby?
E: C. Cowell Scholastic Press, New York, 2000
F: M. Michaelides Mantra Lingua, London, 2002
P: none indicated Mantra Lingua, London, 2002
U: Q. Zamani Mantra Lingua, London, 2002
6 My Daddy is a Giant E: C. Norac Macmillan Children’s Books, London,
2004
F: G. Orio-Claunec Mantra Lingua, London, 2004
P: none indicated Mantra Lingua, London, 2004
U: T. Aajmi Mantra Lingua, London, 2004
7 Splash! E: F. McDonnell Candlewick Press, Somerville, MA, 2003
F: A. Arnold Mantra Lingua, London, 2004
P: S. Attariwala Mantra Lingua, London, 2004
U: Q. Zamani Mantra Lingua, London, 2007
8 The Wheels on the Bus E: none indicated Child’s Play International, Swindon, UK,
2003
F: none indicated Mantra Lingua, London, 2005
P: none indicated Mantra Lingua, London, 2005
U: none indicated Mantra Lingua, London, 2005
9 The Swirling Hijaab E: N. B. Robert Mantra Lingua, London, 2002
F: M. Michaelides Mantra Lingua, London, 2002
P: P. Dave Mantra Lingua, London, 2002
U: Q. Zamani Mantra Lingua, London, 2002
10 The Little Red Hen and the
Grains of Wheat
E: L. Hen Mantra Lingua, London, 2005
F: Annie Arnold Mantra Lingua, London, 2005
P: none indicated Mantra Lingua, London, 2005
U: Q. Zamini Mantra Lingua, London, 2005
11 We’re Going on a Bear
Hunt
E: M. Rosen Simon & Schuster, New York, 1989
F: none indicated Mantra Lingua, London, 2007
P: none indicated Mantra Lingua, London, 2007
U: none indicated Mantra Lingua, London, 2007
Note: All books were originally written in English and subsequently translated into other languages. E = English,
F = French, P = Punjabi, U = Urdu.
Appendix Table. (Continued)
by guest on January 29, 2013ecr.sagepub.comDownloaded from
14 Journal of Early Childhood Research 11(1)
Notes
1. Readers interested in exploring the cultural benefits of exemplary instructional practices using dual lan-
guage books are encouraged to read Naqvi et al. (2010).
2. In the jurisdiction in which the study was conducted ‘Kindergarten’ refers to an education program for
children prior to the first year of formal schooling (i.e. grade 1, which is mandatory for children who
have reached 6 years of age, as of 1 September. Attendance is optional although the vast majority of
parents choose to register their children in kindergarten.
3. Readers interested in a microgenetic analysis of interactions among readers, teachers, and students using
dual language books are referred to Naqvi et al. (in preparation).
References
Bialystok E, McBride-Chang C and Luk G (2005) Bilingualism, language proficiency, and learning to read in
two writing systems. Journal of Educational Psychology 97: 580–590 doi: 10.1037/0022-0663.97.4.580.
Brisk ME and Harrington MM (2007) Literacy and Bilingualism: A Handbook for All Teachers, 2nd edn.
Mahwah, NJ: Lawrence Erlbaum.
Campbell R and Sais E (1995) Accelerated metalinguistic (phonological) awareness in bilingual children.
British Journal of Developmental Psychology 13: 61–68.
Canadian Council on Learning (2006) First language not necessarily linked to reading proficiency, May.
Available at: http://www.ccl-cca.ca/pdfs/LessonsInLearning/May-12-06-first-lang-not-li.pdf.
Capps R, Fix M, Murray J, Passel JS and Herwantoro S (2005) The New Demography of America’s Schools:
Immigration and the No Child Left Behind Act. Washington, DC: The Urban Institute.
Cazden CR (1974) Play with language and metalinguistic awareness: One dimension of language experience.
The Urban Review 7: 28–29.
Conteh J (2007) Opening doors to success in multilingual classrooms: Bilingualism, codeswitching and the
professional identities of ethnic minority primary teachers. Language and Education 21: 457–472 doi:
10.2167/le711.0.
Cummins J (2007) Rethinking monolingual instructional strategies in multilingual classrooms. Canadian
Journal of Applied Linguistics 10: 221–240. Available at: http://www.aclacaal.org/RevueAn.htm.
Dickinson DK and Tabors PO (eds) (2001) Beginning Literacy with Language: Young Children Learning at
Home and School. Baltimore, MD: Paul H. Brookes Publishing.
Fort P and Stechuk R (2008) The cultural responsiveness and dual language education project. Zero to Three
29: 24–28. Available at: http://www.zerotothree.org/site/PageServer?pagename=est_journal.
Gold J and Gibson A (2001) Reading aloud to build comprehension. Reading Rockets. Available at: http://
www.readingrockets.org/article/343.
Gutiérrez KD, Zepeda M and Castro DC (2010) Advancing early literacy learning for all children:
Implications of the NELP report for dual-language learners. Educational Researcher 39: 334–339 doi:
10.3102/0013189X10369831.
Koda K (2008) Impact of prior literacy experience on second-language learning to read. In:Koda K and
Zehler A (eds) Learning to Read Across Languages: Cross-linguistic Relationships in First- and Second-
language Literacy Development. New York: Cambridge University Press.
Laurent A and Martinot C (2009) Bilingualism and phonological segmentation of speech: The case of English-
French pre-schoolers. Journal of Early Childhood Literacy 9: 29–49 doi: 10.1177/1468798408101102.
Lesaux N and Gava E (2006) Synthesis: Development of literacy in language-minority students. In: August
D and Shanahan T (eds) Developing Literacy in Second-Language Learners: Report of the National
Literacy Panel on Language-minority Children and Youth. Mahwah, NJ: Lawrence Erlbaum, 53–74.
Liang P (2010) Portrait of the Immigrant Community. Edmonton, AB: Statistics Canada.
Ma J (2008) ‘Reading the word and the world’: How mind and culture are mediated through the use of dual-
language storybooks. Education 36: 237–251 doi: 10.1080/03004270802217686.
Martin P, Arvind B, Bhojani N and Creese A (2006) Managing bilingual interaction in a Gujarati complemen-
tary school in Leicester. Language and Education 20: 5–22 doi: 10.1080/09500780608668707.
by guest on January 29, 2013ecr.sagepub.comDownloaded from
Naqvi et al. 15
Mcquillan K (2009) Teachers reading aloud. Principal Leadership 9(9): 30–31.
Ministry of Education Alberta (2006) A Review of K-12 ESL Education in Alberta: Final Report. Edmonton,
AB: Ministry of Education Alberta.
Ministry of Education Alberta (2009) Working with Young Children Who are Learning English as a New
Language. Edmonton, AB: Ministry of Education Alberta.
Mullis L (2007) Research project promoting children’s home languages in an East London nursery school.
University of East London, London, UK.
Naqvi R (2009) Building bridges: Acknowledging children’s first languages. Paper presented at the American
Educational Research Association Annual Meeting, San Diego, CA, April.
Naqvi R, Thorne KJ, McKeough AM and Pfitscher CM (2010) Building Bridges: Acknowledging
Children’s First Languages. A Final Report for the Alberta Centre for Child, Family and Community
Research, May. Available at: http://www.research4children.com/admin/contentx/dpDocuments/launch.
cfm?ItemId=5817.
Naqvi R, Thorne KJ, Pfitscher CM and McKeough AM (in preparation) Explemary practice in dual language
book reading.
Neuman SB, Copple C and Bredekamp S (2000) Learning to Read and Write. Developmentally Appropriate
Practice for Children. Washington, DC: National Association for the Education of Young Children.
Paez MM, Tabors PO and Lopez LM (2007) Dual language and literacy development of Spanish-speaking
preschool children. Journal of Applied Developmental Psychology 28: 85–102 doi: 10.1016/j.app-
dev.2006.12.007.
Paris S (2005) Re-interpreting the development of reading skills. Reading Research Quarterly 40: 184–202
doi: 10.1598/RRQ.40.2.3.
Reid DK, Hresko WP and Hammill DD (2001) Test of Early Reading Ability, 3rd edn. Austin, TX: Pro-Ed.
Robertson LH (2006) Learning to read ‘properly’ by moving between parallel literacy classes. Language and
Education 20: 44–61 doi: 10.1080/09500780608668709.
Scarborough HS and Dobrich W (1994) On the efficacy of reading to preschoolers. Developmental Review
14: 245–302.
Sneddon R (2009) Bilingual Books – Biliterate Children: Learning to Read through Dual Language Books.
London: Trentham Books.
Statistics Canada (2010) 2006 Census of Canada: Immigrant Status for the Population of Canada, Provinces
and Territories, 1911 to 2006 Censuses. Catalogue no. 97-557-XCB2006006.
Taylor LK, Bernhard JK, Garg S and Cummins J (2008) Affirming plural belongings: Building on students’
family-based cultural and linguistic capital through multiliteracies pedagogy. Journal of Early Childhood
Literacy 8: 269–294 doi: 10.1177/1468798408096481.
Tsow M (1986) Dual language text or not. Language Matters 3: 13–17.
Whitehurst GJ, Epstein JN, Angell AL, Payne AC, Crone DA and Fischel JE (1994) Outcomes of an emergent
literacy intervention in Head Start. Journal of Educational Psychology 86: 542–555.
by guest on January 29, 2013ecr.sagepub.comDownloaded from
