Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jan 1.
Published in final edited form as: Int J Biling Educ Biling. 2023 Jan 18;27(2):240–252. doi: 10.1080/13670050.2023.2166397

Bilingual Children’s Vocabulary Skills at 5 Years Predict Reading Comprehension Development Within, Not Across, Languages

David Giguere a, Erika Hoff b
PMCID: PMC10901524  NIHMSID: NIHMS1872040  PMID: 38425731

Abstract

Extensive evidence indicates that early vocabulary skills predict later reading development among monolingually developing children. Some evidence suggests that a relationship between vocabulary and later reading also holds across languages among children whose home language differs from the school language. However, these findings have been mixed and it remains unclear if, and under what circumstances, vocabulary in one language supports reading comprehension development in another. The present study followed 84 Spanish-English bilingual children, assessing their vocabulary skills at 5 years and their reading comprehension at 6, 7, 8, and 9 years. Longitudinal multilevel models revealed significant within-language relations between early vocabulary knowledge and subsequent reading comprehension in both English and Spanish and no across-language relations. There were significant concurrent across-language relations between English and Spanish reading comprehension skills. These findings suggest that the contribution of vocabulary knowledge to reading comprehension is language specific but that there are also language general components to reading comprehension, which result in significant concurrent relations between reading comprehension skill across languages.

Keywords: bilingual development, reading comprehension, vocabulary, language transfer

Learning to read depends on multiple, earlier-acquired oral language skills (e.g., August & Shananhan, 2006). These dependencies are reflected in significant predictive relations between children’s oral language skills prior to reading instruction and their subsequent reading ability (Grimm et al., 2018; Hjetland et al., 2019; Lindsey et al., 2003). Oral language skills are the abilities to produce and comprehend spoken language. The majority of research that looks at early predictors of reading development has studied monolingual children who begin to read in the same language they have learned to speak. However, many children enter school as bilinguals who speak both the language of schooling and also a home language that differs from the language of reading instruction in school. Their oral language skills, specifically in the domain of vocabulary knowledge, are distributed across two languages. Thus, measures of single language vocabulary skills do not capture the full linguistic abilities of these children. An important question for understanding the process of learning to read in bilingual children concerns the degree to which their oral language skills in both languages will support reading in the language of schooling. While there is good evidence that phonological skills transfer and support the decoding component of reading across languages, the evidence for cross-linguistic transfer of vocabulary skills to reading comprehension is mixed.

Theoretical Background

Although the nuances of reading theories continue to be debated, several leading theories make vocabulary knowledge central to reading comprehension. The lexical quality hypothesis posits that the basis for the well-established statistical relations between children’s vocabulary knowledge and their reading comprehension is that a deep understanding of words allows for more automatic and efficient word recognition during reading which frees up cognitive resources for other comprehension processes (Perfetti, 2007). The reading systems framework also argues that vocabulary knowledge is central to reading comprehension and that word knowledge plays a causal role in reading comprehension (Perfetti & Stafura, 2014). Intervention studies have highlighted the potential causal role vocabulary knowledge has on reading comprehension—children with explicit vocabulary instruction outperform controls on measures of reading comprehension (Beck et al., 2002; Biemiller, & Boote, 2006). While the evidence from monolingual children make it clear that vocabulary plays a key role in reading comprehension, it is less understood if, and under what circumstances, bilingual children’s vocabulary skill in one language can support reading comprehension in their other language.

The common underlying proficiency model, which focuses on reading comprehension in bilingual children, remains a leading theory of cross-language transfer and posits that reading comprehension is shaped by common underlying linguistic and cognitive skills that are shared across all languages (Cummins, 1981; Geva & Ryan, 1993). It is argued that while each language has unique surface features, there is a single underlying language proficiency that reading draws upon. In this model, early oral language skills do not merely transfer from one language to another— they underlie the reading comprehension processes in both languages. This model predicts that bilingual children who struggle to read in one language will struggle to read in the other because the same underlying skills are tapped. Therefore, it can be predicted from this framework that strong oral skills in a home language will contribute to reading development in a school language, and vice versa.

Findings that reading skills themselves are related across languages provide support for a common underlying proficiency model (Hammer et al., 2020; Palermo et al., 2017; Wagley et al., 2022). Additionally, empirical studies have identified several oral language skills that cross-linguistically predict reading comprehension, including phonological awareness (Goodrich & Lonigan, 2017; Melby-Lervåg & Lervåg, 2011), morphological awareness (Pasquarella et al., 2015), and narrative ability (Miller et al., 2006). However, support for the common underlying proficiency model has been less consistent in the domain of vocabulary, and research has yielded ambiguous findings (Chung et al., 2019). In the present study we address the currently unclear relation between bilingual children’s oral vocabulary knowledge prior to reading instruction and their subsequent trajectories of reading comprehension in two languages.

Vocabulary and Reading Comprehension

There is robust evidence that early oral vocabulary skills play an important role in reading comprehension development (e.g., Quinn et al., 2015). For instance, among monolinguals, oral vocabulary skills at school entry have been found to predict reading comprehension in the 4th (Spira et al., 2005) and 7th grade (Tabors et al., 2001). The within-language relation between oral vocabulary knowledge and reading comprehension observed in monolingual children holds for bilingual children as well—in the majority language spoken at school (Babayiğit & Shapiro, 2019; Bellochi et al., 2017; Prevoo et al., 2016). The lexical quality hypothesis, that deep word knowledge frees cognitive resources and thereby explains the relation between vocabulary and reading comprehension, predicts that vocabulary skill will be a language-specific benefit for bilingual children. That is to say, apart from cognates, a strong mental representation of a word in Spanish is unlikely to facilitate the recognition of the same word in English. In contrast, the common underlying proficiency framework postulates that reading comprehension draws upon a single underlying ability—thus predictors of reading comprehension should operate within- and across-languages. The empirical findings from studies that have tested these frameworks have been mixed, raising the question of why vocabulary in one language predicts reading comprehension in the other in some studies and not others.

Close inspection of the literature suggests that the majority of the findings of cross-linguistic relations come from studies of children who are dominant in the home language and who have limited proficiency in the school language. These studies have found that home language vocabulary skills are related to reading comprehension in the school language, above and beyond the variance explained by within-language vocabulary skills (Howard et al., 2014; Proctor et al., 2006). One study examined the English reading comprehension skills of Spanish-English bilinguals in the 4th grade and found concurrent evidence that Spanish vocabulary skill was positively related to English reading comprehension above and beyond the variance explained by English vocabulary skills (Proctor et al., 2006). Another study examined English reading skills in kindergartners, third graders, and fifth graders. After accounting for SES, home language and literacy exposure, and English vocabulary skill, Spanish vocabulary skills uniquely predicted English reading outcomes in the kindergarten and third grade samples, but not in the fifth grade sample (Howard et al., 2014). Cross-linguistic associations between home language vocabulary skill and reading outcomes in the school language have been found in English-Spanish bilinguals (Davison et al., 2011), English-Chinese bilinguals (Li et al., 2012), young 4- to 5-year old children learning to read (Zhao et al., 2017) and 10-year-old children in later elementary years (Proctor et al., 2006).

In contrast, several studies have found that vocabulary is a language-specific predictor of reading comprehension (Farver et al., 2013; Melby-Lervåg & Lervåg, 2011). One longitudinal study followed a representative sample of Spanish-speaking English language learners in the United States and examined the influence of several oral language skills on later reading comprehension (Kieffer, 2012). The results revealed that early vocabulary skill was a stronger predictor of later reading comprehension than other measures of oral skill (listening comprehension, story recall), and that among these children, Spanish vocabulary skills in kindergarten did not uniquely predict English reading comprehension in the 8th grade (Kieffer, 2012). There is also concurrent evidence that vocabulary skills predict reading outcomes within languages but not across (Goodrich & Lonigan, 2017).

We hypothesize that in previous studies which found that home language vocabulary knowledge predicted subsequent school language reading comprehension, the home language measure may have been predictive because the vocabulary size—in addition to being a direct measure of vocabulary knowledge—works as an indirect indicator of learning abilities. Children exposed only to Spanish who have bigger vocabularies may have them because they are better at learning than children with smaller Spanish vocabularies, and those same learning abilities may serve them in learning to read in English. Of course, to the extent that vocabulary size functions as proxy for general learning ability, this also contributes to the within-language correlations between vocabulary size and reading comprehension. The best evidence that vocabulary knowledge contributes to reading comprehension would be a finding that vocabulary knowledge predicts reading comprehension within a language even when vocabulary knowledge in another language is also a predictor.

The present study provides that test of the relation between vocabulary knowledge and reading, asking whether English vocabulary knowledge at age 5 predicts English reading comprehension at ages 6 through 9, in a sample of English-dominant children who also hear Spanish at home. The present study provides an additional test of the common underlying proficiency hypothesis, asking whether reading comprehension ability in English, where children receive formal instruction, is related to reading comprehension ability in Spanish, where children receive no formal instruction. In sum, with data on the English and Spanish vocabulary knowledge at 5 years and English and Spanish reading comprehension from 6 to 9 years in 84 bilingual children, the present study tests two hypotheses of cross-linguistic transfer in bilingual children’s development of reading comprehension: (1) that prior vocabulary knowledge in one language will predict subsequent reading comprehension in the other, and (2) that reading comprehension ability will be correlated across languages.

Method

Participants

Eighty-four Spanish-English bilingually developing children (46 girls, 38 boys) participated. Their expressive and receptive vocabularies in English and in Spanish were assessed at age 5 years (Mean age = 60.4 months, SD = .5), and their reading comprehension skills in both languages were assessed at 6, 7, 8, and 9 years. Mean ages at each assessment point are provided in Table 1. These participants were drawn from a larger longitudinal study (Hoff et al. 2014). The criteria for inclusion in the present analyses were that the children have complete data on the measures of vocabulary skill at age 5 and at least two assessments of English and Spanish reading comprehension skill between the ages of 6 and 9 years. In the larger ongoing study, there were 103 children who met this criterion at age 5. Of these children, 19 did not have sufficient data between 6 and 9 years to participate. Independent samples t-tests were conducted to examine whether there were vocabulary differences at 5 years between the 19 participants not included in the present study and the 84 participants included. The results revealed no group-related differences in English expressive vocabulary nor in English or Spanish receptive vocabulary (all ps > .05, all 2-tailed tests). There was one difference: the 84 children included in the present study had significantly higher levels of Spanish expressive vocabulary than the 18 participants not included, t(101) = 2.3, p = .013, Cohen’s d = .58.

Table 1.

Mean W Scores and (Standard Deviations) of Passage Comprehension in English and in Spanish, and Mean Age in Months at 6, 7, 8 and 9 Years

6 years 7 years 8 years 9 years
Reading comprehension, English 447.61 (28.80) n = 62 475.94 (22.99) n = 62 491.72 (17.83) n = 74 500.24 (14.85) n = 71
Reading comprehension, Spanish 401.16 (38.75) n = 62 428.23 (30.36) n = 62 445.15 (27.90) n = 71 450.66 (38.46) n = 70
Mean age in months 75.0 (3.1) 85.8 (2.1) 97.8 (1.3) 110.1 (1.5)
Open in a new tab

Note. Passage comprehension was assessed by the Passage Comprehension Subtest in the Woodcock Muñoz Language Survey-Revised.

All participants were born in the United States, resided in South Florida, and had at least one parent who was born in a Spanish speaking country. All participants were spoken to in Spanish at home; most also were spoken to in English. There was one exception, one child only heard English at home. That child was born to a Spanish-speaking father and English-speaking mother. During the course of the study, the child’s father passed away and the paternal grandparents, who did not live with the child, become the child’s only source of Spanish input.

At each assessment point, the primary caregiver was administered an extensive questionnaire in interview and was asked to provide estimates of the child’s relative home language exposure in English and Spanish. The interview included questions about who lived at home with the child and what languages they spoke to the child. For the entire sample, the mean percentage of speech addressed to the child at home in Spanish, as reported by the primary caregiver, was 61% at 5 years, 56% at 6 years, 61% at 7 years, 56% at 8 years, and 47% at 9 years. The average age of arrival to the United States for the foreign-born parents was 22.4 years (SD = 8.1) for mothers and 23.0 (SD = 9.0) for fathers. Forty-three mothers (53%) had a 4-year college degree or higher; 30 had degrees earned in a Spanish-speaking country, 19 had degrees earned in the U.S., and 6 had earned college or advanced degrees in both a Spanish-speaking country and in the U.S.

All children were full term and healthy at birth, had normal hearing based on parental report of otoacoustic emissions testing at birth, and showed no sign of communicative delay at 30 months, based on the Ages and Stages screening instrument (Squires, Potter, & Bricker, 1999). Parental consent and child assent were obtained prior to study participation.

Procedure

Children were assessed annually, typically within 2 months of their birthdays between 5 and 9 years. Children were assessed in their homes, or, in a few cases a university lab or public library space, when parents were not comfortable with home visits. All tests were individually administered by fully bilingual, trained, and experienced examiners who were familiar to the participants. Tests of English and Spanish were administered on separate days, with the order of administration counterbalanced across participants. Oral vocabulary skills were assessed in English and in Spanish at age 5. English and Spanish reading comprehension were measured at 6, 7 8, and 9 years.

Measures

Expressive vocabulary

English and Spanish expressive vocabulary skills were measured in separate administrations of the Expressive One-Word Picture Vocabulary Test– Spanish Bilingual Edition (EOWPVT; Brownell, 2001). This is an examiner administered standardized test in which participants view an image that depicts an action, object, category or concept and are asked to provide a label. The bilingual version is an adaptation of the English test, created by excluding items that the developers judged to be culturally biased (e.g. fireplace). The standard administration of the bilingual version allows the child to label the picture in either language and yields a conceptual score. We modified the standard conceptual administration procedure to allow only English labels during the English assessment and only Spanish labels during Spanish assessment in order to obtain separate assessments of the children’s skills in each language. There are no standard scores for this use; thus we used raw scores as the measure.

Receptive vocabulary

Receptive vocabulary in both languages was measured using standard scores on the Peabody Picture Vocabulary Test (PPVT-4; Dunn & Dunn, 2012) and its Spanish counterpart, Test de Vocabulario en Imágenes Peabody (TVIP; Dunn, Padilla, Lugo, & Dunn, 1986). In both versions of the test, the interviewer says a word and shows a flip book with four distinct pictures presented on the page. The participant must point to the matching illustration. For instance, the examiner would say the word pencil and the child must choose the appropriate picture from the set of four. The difficulty of the word increases with each item, and the test discontinues when the participant reaches the ceiling. The test has been normed on English (PPVT-4) and Spanish (TVIP) monolingual populations and is widely used is language development research.

Reading comprehension

English and Spanish reading comprehension ability were assessed with the passage comprehension subtest of the Woodcock-Muñoz Language Survey-Revised (WMLS-R; Woodcock et al., 2005), separately administered in each language. The tests require participants to read a sentence or a small paragraph with a blank space representing a missing word. The participants must fill in the missing word. The test stops when the ceiling is reached. The test was designed to assess the reading comprehension skills of individuals until late adulthood. The outcome measures yield a W score, which is a developmental standard score that is centered on a value of 500, which is set to approximate the average performance of a 10-year-old (Jaffe, 2009). The English and Spanish subtests from the Woodcock-Muñoz Language Survey-Revised were co-normed and equated to produce a Rasch-based W score with an equal interval scale that is comparable across languages. Extensive research has documented strong reliability and validity (Woodcock et al., 2005).

Data analysis plan

A series of longitudinal multilevel models was conducted to examine the within- and across-language relations of vocabulary at 5 years to subsequent reading comprehension development from 6 to 9 years using the HLM 7 software (Raudenbush et al., 2011). Multilevel modeling analysis allows researchers to model growth over time as well as examine factors related to individual differences in growth trajectories. Previous research has used multilevel models to examine individual difference in language development and reading comprehension (Relyea & Amendum, 2020; Rowe et al., 2012). With multilevel modeling, only one outcome variable can be predicted; therefore, separate models of English and Spanish reading comprehension were calculated. The hierarchical structure of the data is repeated measures nested within individuals. Age was centered at 6 years and predictors were mean centered. Vocabulary scores were time-invariant predictors, and across-language reading comprehension was as a time-varying predictor. Missing data were handled using full information maximum likelihood (FIML). FIML is recommended when comparing models in multilevel modeling (Anguinis et al., 2013) and FIML has been shown to produce unbiased parameter estimates (Enders & Bandalos, 2001). All models included random effects of intercept and linear slope.

Model building for reading comprehension in English and Spanish proceeded in the following steps: In the first, base model, Age (with 4 time points) was entered, and in Model 1 the quadratic effect of Age (i.e., Age2) was added. Age2 is added to the base model to examine whether the rate of reading comprehension growth changes over time. Previous research has found that reading comprehension and language tend to grow at a decelerated pace over time (e.g., Kieffer, 2012; Rowe et al., 2012). Thus, the Age2 predictor was included to account for potential nonlinear growth rates. In Model 2, within-language vocabulary at 5 years was added as a fixed effect, along with the interaction of vocabulary with Age and Age2. In Model 3, nonsignificant predictors were trimmed and the across-language vocabulary and reading comprehension scores were added to the model. In Model 4, nonsignificant predictors were again trimmed. Model fit was assessed after each model and comparisons of fit between nested models were accomplished using chi-square difference test from the −2 log likelihood. Akaike information criteria (AIC) and Bayesian information criteria (BIC) were used to compare non-nested models. More parsimonious models were preferred.

Results

Descriptive Statistics

Mean raw scores for expressive vocabulary at 5 years were 48.3 (SD = 13.5) in English and 22.6 (SD = 18.8) in Spanish. Mean standard scores for receptive vocabulary were 104.2 (SD = 17.7) in English and 100.9 (SD = 15.3) in Spanish. Expressive vocabulary scores were higher in English than in Spanish, t(83) = 10.2, p <.001, d = 1.1; receptive vocabulary scores were not significantly different, t(82) = 1.6, p =.11, d = .175.

Table 1 presents the mean W scores for English and Spanish reading comprehension at ages 6 through 9 years. Paired samples t-tests revealed that the W scores for reading comprehension were significantly better in English than in Spanish at each time point (all ps <.001; Cohen’s d ranged from 1.4 to 2.1).

Expressive and receptive vocabulary scores were highly correlated within each language, r(82) = .67, p < .001 for the English measures and r(82) = .76, p < .001 for the Spanish measures. Expressive vocabulary scores were not correlated across languages, r(82) = .001, p = .991, and receptive vocabulary scores were correlated, r(82) = .33, p =.002. For data reduction purposes, composite scores of English and Spanish vocabulary were created by averaging the z-scores for the expressive and receptive measures in each language. The English and Spanish composite vocabulary scores were not significantly related, r(82) = .201, p (2-tailed) = .067.

English Reading Comprehension

Table 2 presents the models of English reading comprehension growth. English reading comprehension skill grew over time (base model) at a decelerating rate (Model 1). The −2 log likelihood difference tests revealed that the inclusion of Age2 in Model 1 significantly improved model fit when compared to the base model, χ2 (1) = 37, p<.001. Model 2 added the English vocabulary composite score at age 5 and its interactions with Age and Age2, which improved model fit when compared to model 1, χ2 (3) = 28, p<.001. In this model, English vocabulary was a significant positive predictor of the height of the English reading comprehension trajectory and a significant negative predictor of the slope. The interaction between English vocabulary and Age2 was not significant. Model 3 trimmed the nonsignificant interaction and added the Spanish vocabulary composite score and its interactions with Age and Age2 and concurrent Spanish reading comprehension. Model 2 and Model 3 are not nested models because Model 3 both trimmed a predictor and added new predictors. Therefore, the −2 log likelihood difference test could not compare model fits. For these models, model fit was evaluated using the AIC and BIC indices—both indices indicated that model fit was improved in Model 3. In this model, Spanish vocabulary did not have a significant main effect on English reading comprehension, nor did it interact with Age or Age2. However, Spanish reading comprehension was a significant time-varying predictor. Additionally, in Model 3, the interaction between English vocabulary and Age was no longer significant. The final model of English reading comprehension (Model 4) trimmed the nonsignificant predictors and included English vocabulary at 5 years as positive predictor of the height of the reading trajectory and positive concurrent relations with Spanish reading comprehension. Removing the nonsignificant predictors did not significantly reduce model fit when compared to Model 3, χ2 (4) = 8, p =.09. Model 4 was the best fitting and most parsimonious model. Figure 1 plots the estimated English reading comprehension trajectories of children with high and low English vocabulary composite scores at 5 years.

Table 2.

Estimates of Fixed and Random Effects from a Series of Multilevel Models Predicting English Reading Comprehension Development from 6 to 9 years with English and Spanish Oral Skills at 5 Years and Concurrent Spanish Reading Comprehension (N=84)

Base Model Model 1 Model 2 Model 3 Model 4
γ SE γ SE γ SE γ SE γ SE
Fixed effects
 Intercept 453.4 3.1 447.1 3.2 447.3 2.9 456.5 2.8 455.1 2.8
 Age 17.2 1.0 34.1 2.8 34.1 2.8 24.3 2.9 25.8 2.9
 Age2 −5.5 0.9 −5.5 0.9 −3.8 0.8 −4.1 0.8
 English vocabulary 15.1 3.3 9.6 2.7 6.8 1.2
 English vocabulary ×Age −6.6 3.1 −1.1 1.2
 English vocabulary ×Age2 1.4 0.9
 Spanish vocabulary −1.2 2.9
 Spanish vocabulary ×Age −4.1 3.0
 Spanish vocabulary ×Age2 1.3 0.9
 Spanish reading comprehension 0.3 0.03 0.3 0.03
Random effects
 Intercept 280 349.2 302.3 258.2 267.2
 Slope (linear) 127 140.5 134.4 160.9 157.4
Goodness of fit
 −2LL 2346 2309 2281 2188 2196
 AIC 2358 2323 2301 2214 2214
 BIC 2380 2348 2337 2261 2246
Open in a new tab

Note. Bold numbers represent significant effects; SE = Standard error.

†

p < .1

Reading comprehension was assessed by the passage comprehension subtest in the Woodcock Muñoz Language Survey-Revised; Vocabulary measures are a composite of same language receptive and expressive vocabulary scores; Expressive vocabulary was assessed by the Expressive One-Word Picture Vocabulary Test– Spanish Bilingual Edition (EOWPVT); Receptive vocabulary was assessed by Peabody Picture Vocabulary Test (PPVT-4) and its Spanish counterpart, Test de Vocabulario en Imágenes Peabody (TVIP); −2LL = −2 log likelihood; AIC = Akaike information criteria; BIC = Bayesian information criteria.

Figure 1.

Figure 1

Open in a new tab

Caption: Estimated trajectories of English reading comprehension growth from 6 to 9 years for children with high and low levels of English vocabulary at age 5 (N = 84).

Note. Low vocabulary = 25th percentile and high vocabulary = 75th percentile; Vocabulary is a composite score of the EOWPVT and PPVT; Reading comprehension was assessed by the passage comprehension subtest in the WMLS-R.

Alt Text: There are two lines that are plotting the estimated English reading comprehension trajectories from 6 to 9 years. Both lines are improving at the same rate but those with high vocabulary skills at age 5 are consistently above those with low vocabulary skills at age 5.

Spanish Reading Comprehension

Table 3 presents the models of Spanish reading comprehension growth. Spanish reading comprehension grew over time (base model) at a decelerating rate (Model 1). The −2 log likelihood difference tests revealed that the inclusion of Age2 in Model 1 significantly improved model fit when compared to the base model, χ2 (1) = 16, p<.001. Model 2 added the Spanish vocabulary composite score at age 5 and its interactions with Age and Age2, which improved model fit when compared to model 1, χ2 (3) = 35, p<.001. In this model, Spanish vocabulary was a positive predictor of the height of the Spanish reading trajectory, its interactions with Age and Age2 indicated that children with lower levels of Spanish vocabulary at 5 years grew in Spanish reading comprehension at a slower rate than did children with higher Spanish vocabulary scores, but their rate of growth was accelerating rather than decelerating. Model 3 added the English vocabulary composite score and its interactions with Age and Age2 and concurrent English reading comprehension, which improved model fit when compared to model 2, χ2 (4) = 68, p<.001. In this model, English vocabulary did not have a significant main effect on Spanish reading comprehension, nor did it interact with Age or Age2. However, English reading comprehension was a significant time-varying predictor. The final model of Spanish reading comprehension (Model 4) trimmed the nonsignificant effects of English vocabulary and included Spanish vocabulary at age 5 as a positive predictor of the height of the Spanish reading trajectory, a positive predictor of the trajectory’s slope, and a negative predictor of its deceleration. Concurrent English reading comprehension was a significant positive predictor. Removing the nonsignificant predictors did not significantly reduce model fit when compared to Model 3, χ2 (3) = 3, p =.392. Model 4 was the best fitting and most parsimonious model. Figure 2 plots the estimated Spanish reading comprehension trajectories of children with high and low levels of Spanish vocabulary composite scores at 5 years.

Table 3.

Estimates of Fixed and Random Effects from a Series of Multilevel Models Predicting Spanish Reading Comprehension Development from 6 to 9 years with Spanish and English Vocabulary at 5 Years and Concurrent English Reading Comprehension (N=84)

Base Model
 Model 1
 Model 2
 Model 3
 Model 4
γ SE γ SE γ SE γ SE γ SE
Fixed effects
 Intercept 407.1 4.0 401.0 4.3 401.2 4.0 424.8 4.3 423.9 4.1
 Age 16.4 1.7 32.8 4.2 33.0 4.1 9.0† 4.8 9.9 4.7
 Age2 −5.3 1.3 −5.5 1.2 −1.8 1.2 −2.0 1.3
 Spanish vocabulary 12.8 4.5 9.9 3.8 10.0 3.3
 Spanish vocabulary ×Age 8.8† 4.6 10.1 4.5 9.8 4.3
 Spanish vocabulary ×Age2 −2.7 1.4 −3.0 1.3 −3.0 1.4
 English vocabulary 0.4 4.0
 English vocabulary ×Age −2.1 4.5
 English vocabulary ×Age2 0.1 1.3
 English reading comprehension 0.7 0.08 0.7 0.08
Random effects
 Intercept 262.9 297.4 265.4 184.9 185.6
 Slope (linear) 158.2 174.7 182.1 184.2 184.5
Goodness of fit
 −2LL 2547 2531 2496 2428 2431
 AIC 2559 2545 2516 2456 2453
 BIC 2580 2570 2552 2506 2492
Open in a new tab

Note. Bold numbers represent significant effects; See Table 2 caption for further explanations.

Figure 2.

Figure 2

Open in a new tab

Caption: Estimated trajectories of Spanish reading comprehension growth from 6 to 9 years for children with high and low levels of Spanish vocabulary at age 5 (N = 84).

Note. Low vocabulary = 25th percentile and high vocabulary = 75th percentile; Vocabulary is a composite score of the EOWPVT and TVIP; Reading comprehension was assessed by the passage comprehension subtest in the WMLS-R.

Alt Text: There are two lines plotting the estimated Spanish reading trajectories from 6 to 9 years for those with high and low levels of Spanish vocabulary skill at 5 years. The line that represents those with high skills is consistently above the line that represents those with low skills, but the bottom line appears to be catching up to the top line.

Discussion

The present study examined Spanish-English bilingual children who were English-dominant in expressive vocabulary and reading comprehension and tested (1) the predictive relations within and across languages between vocabulary knowledge at 5 years and subsequent reading comprehension, assessed from 6 to 9 years, and (2) the concurrent relations during the period from 6 to 9 years between reading comprehension skill in English and Spanish. The central findings were (1) that vocabulary knowledge predicts reading comprehension only in the same language, and (2) that reading comprehension skills are concurrently related across languages. These findings obtained in the models of both English and Spanish reading comprehension. The only difference between the final models of English and Spanish reading comprehension was that English vocabulary predicted only the height of the English reading trajectories whereas Spanish vocabulary also predicted the shape of Spanish reading trajectories.

The current finding adds one more finding to the existing evidence that the English oral skills bilingual children possess at school entry matter (e.g., Kieffer, 2012). Although English-only measures do not completely capture bilingual children’s linguistic abilities, they do predict future English reading comprehension trajectories. The present findings further suggest that early vocabulary knowledge in a home language supports future reading comprehension in the home language. These within-language relations are consistent with the findings of other studies (Melby-Lervåg & Lervåg, 2011).

The present study did not find cross-linguistic transfer of vocabulary skills to reading comprehension in either language. This null finding is consistent with studies that have examined reading comprehension among samples of bilingually developing children who receive formal reading instruction only in English (Mancilla-Martinez & Lesaux, 2010). The language-specific nature of the relations observed argues that if there is a common underlying proficiency that supports reading comprehension in both of a bilingual’s languages, vocabulary knowledge is not part of that proficiency. Rather, these findings suggest that vocabulary knowledge supports reading comprehension only in reading material that makes use of that vocabulary—and cognates. This finding is consistent with other studies that find that vocabulary knowledge provides a language-specific benefit to reading comprehension (Goodrich & Lonigan, 2017). The contrary findings of correlations between Spanish vocabulary skills and English reading comprehension come largely from studies of children who were Spanish dominant (Davison et al., 2011; Proctor et al., 2006). In those samples, the measures of Spanish vocabulary may have been predictive of reading skill in English only because vocabulary knowledge also serves as an index of general learning ability. In the present English-dominant sample, their English vocabulary knowledge captured both that general learning ability and the size of the vocabulary that they needed to understand the texts they read. There was nothing else that Spanish vocabulary indexed that English reading relied on. Future research might directly compare the within- and across-language relations between vocabulary knowledge and reading comprehension in both Spanish-dominant and English-dominant samples.

In contrast to the lack of cross-linguistic relation between vocabulary and reading, there were cross-linguistic relations between measures of reading. This is consistent with findings of other studies (Hammer et al., 2020; Wagley et al., 2022) and provides evidence of a common language-general skill underlying the ability to draw meaning from text. In that the children in the present study received formal instruction in English reading and not in Spanish, it seems most plausible that the tutored skills in English supported reading in Spanish. Anecdotally, many of the children (and their parents) were surprised that they were able to read in Spanish, having not done it before we asked them to give it a try.

The present study has limitations. The findings rely entirely on two measures of oral vocabulary knowledge and a single measure of reading comprehension in each language. With only two observed predictor variables in each language, the present study did not create a latent measure of vocabulary skill. There are several advantages to using latent variables are predictors (e.g., absence of error). Nonetheless, the present findings underscore the language-specific importance of vocabulary knowledge to reading comprehension while also suggesting that reading comprehension skills may transfer across languages. The implication of these findings for reading instruction in bilingual children is clear: the benefit of vocabulary knowledge in Spanish is to speaking Spanish and to reading Spanish, not to reading English. However, the reading comprehension skills taught in one language will transfer to the other language. These findings provide empirical support for the educational practice of initially teaching young bilingual children to read in their home language—if that is their stronger language (Relyea & Amendum, 2020). Learning to read will be more successfully accomplished in the stronger language, and those skills will later transfer.

Acknowledgements:

This article represents a portion of David Giguere’s doctoral dissertation.

Footnotes

Declarations of interests: none

References

  1. Aguinis H, Gottfredson RK, & Culpepper SA (2013). Best-practice recommendations for estimating cross-level interaction effects using multilevel modeling. Journal of Management, 39, 1490–1528. DOI: 10.1177/0149206313478188 [DOI] [Google Scholar]
  2. August D, & Shanahan T (2006). Developing literacy in second-language learners: Report of the National Literacy Panel on Language-Minority Children and Youth. Mahwah, NJ: Lawrence Erlbaum. [Google Scholar]
  3. Babayiğit S, and Shapiro L (2020) Component skills that underpin listening comprehension and reading comprehension in learners with English as first and additional language. Journal of Research in Reading, 43, 78–97. 10.1111/1467-9817.12291. [DOI] [Google Scholar]
  4. Bellocchi S, Tobia V, & Bonifacci P (2017). Predictors of reading and comprehension abilities in bilingual and monolingual children: A longitudinal study on a transparent language. Reading and Writing, 30, 1311–1334. 10.1007/s11145-017-9725-5 [DOI] [Google Scholar]
  5. Brownell R (2001). Expressive one-word picture vocabulary test—Spanish-English bilingual edition. Novato, CA: Academic Therapy Publications. [Google Scholar]
  6. Chung SC, Chen X, & Geva E (2019). Deconstructing and reconstructing cross-language transfer in bilingual reading development: An interactive framework. Journal of Neurolinguistics, 50, 149–161. 10.1016/j.jneuroling.2018.01.003 [DOI] [Google Scholar]
  7. Cummins J (1981). The role of primary language development in promoting educational success for language minority students. In California State Department of Education (Ed.), Schooling and language minority students: A theoretical framework. Evaluation, Dissemination and Assessment Center, California State University, Los Angeles. [Google Scholar]
  8. Davison MD, Hammer C, & Lawrence FR (2011). Associations between preschool language and first grade reading outcomes in bilingual children. Journal of Communication Disorders, 44, 444–458. doi: 10.1016/j.jcomdis.2011.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dunn LM, & Dunn DM (2007). Peabody Picture Vocabulary Test: PPVT-4. Pearson Assessments. [Google Scholar]
  10. Dunn L, Padilla R, Lugo S, & Dunn L (1986). Test de vocabulario en imágenes Peabody. Circle Pines, MN: American Guidance Service. [Google Scholar]
  11. Enders CK, & Bandalos DL (2001). The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural equation modeling, 8, 430–457 [PubMed] [Google Scholar]
  12. Farver JAM, Xu Y, Lonigan CJ, & Eppe S (2013). The home literacy environment and Latino head start children's emergent literacy skills. Developmental Psychology, 49, 775. doi: 10.1037/a0028766 [DOI] [PubMed] [Google Scholar]
  13. Geva E, & Ryan EB (1993). Linguistic and cognitive correlates of academic skills in first and second languages. Language Learning, 43, 5–42. [Google Scholar]
  14. Goodrich JM, & Lonigan CJ (2017). Language-independent and language-specific aspects of early literacy: An evaluation of the common underlying proficiency model. Journal of Educational Psychology, 109, 782. 10.1037/edu00001794 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Grimm RP, Solari EJ, & Gerber MM (2018). A longitudinal investigation of reading development from kindergarten to grade eight in a Spanish-speaking bilingual population. Reading and Writing, 31, 559–581. 10.1007/s11145-017-9798-1 [DOI] [Google Scholar]
  16. Hammer CS, Burchinal M, Hong SS, LaForett DR, Páez M, Buysse V, ... & López LM (2020). Change in language and literacy knowledge for Spanish–English dual language learners at school-entry: Analyses from three studies. Early Childhood Research Quarterly, 51, 81–92. 10.1016/j.ecresq.2019.07.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hjetland HN, Lervåg A, Lyster SAH, Hagtvet BE, Hulme C, & Melby-Lervåg M (2019). Pathways to reading comprehension: A longitudinal study from 4 to 9 years of age. Journal of Educational Psychology, 111, 751. 10.1037/edu0000321 [DOI] [Google Scholar]
  18. Hoover WA, & Gough PB (1990). The simple view of reading. Reading and Writing, 2, 127–160. [Google Scholar]
  19. Howard ER, Páez MM, August DL, Barr CD, Kenyon D, & Malabonga V (2014). The importance of SES, home and school language and literacy practices, and oral vocabulary in bilingual children’s English reading development. Bilingual Research Journal, 37, 120–141. [Google Scholar]
  20. Jaffe LE (2009). Development, interpretation, and application of the W score and the relative proficiency index (Woodcock-Johnson-III Assessment Service Bulletin No. 11). Rolling Meadows, IL: Riverside Publishing. [Google Scholar]
  21. Kieffer MJ (2012). Early oral language and later reading development in Spanish-speaking English language learners: Evidence from a nine-year longitudinal study. Journal of Applied Developmental Psychology, 33, 146–157. doi: 10.1016/j.appdev.2012.02.003 [DOI] [Google Scholar]
  22. Li T, McBride-Chang C, Wong A, & Shu H (2012). Longitudinal predictors of spelling and reading comprehension in Chinese as an L1 and English as an L2 in Hong Kong Chinese children. Journal of Educational Psychology, 104, 286. [Google Scholar]
  23. Lindsey KA, Manis FR, & Bailey CE (2003). Prediction of first-grade reading in Spanish-speaking English-language learners. Journal of Educational Psychology, 95, 482–494. doi: 10.1037/0022-0663.95.3.482 [DOI] [Google Scholar]
  24. Mancilla-Martinez J, & Lesaux NK (2010). Predictors of reading comprehension for struggling readers: The case of Spanish-speaking language minority learners. Journal of Educational Psychology, 102, 701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Melby-Lervåg M, & Lervåg A (2011). Cross-linguistic transfer of oral language, decoding, phonological awareness and reading comprehension: A meta-analysis of the correlational evidence. Journal of Research in Reading, 34, 114–135. 10.1111/j.1467-9817.2010.01477.x [DOI] [Google Scholar]
  26. Miller JF, Heilmann J, Nockerts A, Iglesias A, Fabiano L, & Francis DJ (2006). Oral language and reading in bilingual children . Learning Disabilities Research & Practice, 21, 30–43 [Google Scholar]
  27. Palermo F, Mikulski AM, Fabes RA, Martin CL, & Hanish LD (2017). Cross-language associations and changes in Spanish-speaking preschoolers’ English and Spanish academic abilities. Applied Psycholinguistics, 38, 347–370. 10.1017/S0142716416000217 [DOI] [Google Scholar]
  28. Pasquarella A, Chen X, Lam K, Luo YC, & Ramirez G (2011). Cross-language transfer of morphological awareness in Chinese–English bilinguals. Journal of Research in Reading, 34, 23–42. 10.1111/j.1467-9817.2010.01484.x [DOI] [Google Scholar]
  29. Perfetti C (2007). Reading ability: Lexical quality to comprehension. Scientific Studies of Reading, 11, 357–383. 10.1080/10888430701530730 [DOI] [Google Scholar]
  30. Perfetti C & Stafura J (2014). Word knowledge in a theory of reading comprehension. Scientific Studies of Reading, 18, 22–37. 10.1080/10888438.2013.827687 [DOI] [Google Scholar]
  31. Prevoo MJ, Malda M, Mesman J, & van IJzendoorn MH (2016). Within-and cross-language relations between oral language proficiency and school outcomes in bilingual children with an immigrant background: A meta-analytical study. Review of Educational Research, 86, 237–276. doi: 10.1080/15235882.2017.1383949 [DOI] [Google Scholar]
  32. Proctor CP, August D, Carlo MS, & Snow C (2006). The intriguing role of Spanish language vocabulary knowledge in predicting English reading comprehension. Journal of Educational Psychology, 98, 159. doi: 10.1037/0022-0663.98.1.159 [DOI] [Google Scholar]
  33. Proctor CP, Silverman RD, Harring JR, & Montecillo C (2012). The role of vocabulary depth in predicting reading comprehension among English monolingual and Spanish–English bilingual children in elementary school. Reading and Writing, 25, 1635–1664. [Google Scholar]
  34. Quinn JM, Wagner RK, Petscher Y, & Lopez D (2015). Developmental relations between vocabulary knowledge and reading comprehension: A latent change score modeling study. Child Development, 86, 159–175. 10.1111/cdev.12292 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Raudenbush SW, Bryk AS, Cheong AS, Fai YF, Congdon RT, & du Toit M (2011). HLM 7: Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International. [Google Scholar]
  36. Relyea JE, & Amendum SJ (2020). English reading growth in Spanish-speaking bilingual students: Moderating effect of English proficiency on cross-linguistic influence. Child Development, 91, 1150–1165. 10.1111/cdev.13288 [DOI] [PubMed] [Google Scholar]
  37. Rowe ML, Raudenbush SW, & Goldin-Meadow S (2012). The pace of vocabulary growth helps predict later vocabulary skill. Child Development, 83, 508–525. 10.1111/j.1467-8624.2011.01710.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Spira EG, Bracken SS, & Fischel JE (2005). Predicting improvement after first-grade reading difficulties: the effects of oral language, emergent literacy, and behavior skills. Developmental Psychology, 41, 225. [DOI] [PubMed] [Google Scholar]
  39. Squires J, Potter L & Bricker D (1999). Ages and stages questionnaire: Parent-completed,.child monitoring system, 2nd ed. Baltimore: Brookes Publishing. [Google Scholar]
  40. Tabors PO, Snow CE, & Dickinson DK (2001). Homes and schools together: Supporting language and literacy development. In Dickinson DK & Tabors PO (Eds.), Beginning literacy with language: Young children learning at home and school (pp. 313–334). Baltimore: Brookes. [Google Scholar]
  41. Wagley N, Marks RA, Bedore LM, & Kovelman I (2022). Contributions of bilingual home environment and language proficiency on children’s Spanish–English reading outcomes. Child Development. Advanced online publication. 10.1111/cdev.13748 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Woodcock RW, Muñoz-Sandoval AF, Ruef M, & Alvarado CG (2005). Woodcock-Muñoz Language Survey–Revised. Itasca, IL: Riverside Publishing. [Google Scholar]
  43. Zhao J, Dixon LQ, Quiroz B, & Chen S (2017). The Relationship Between Vocabulary and Word Reading Among Head Start Spanish–English Bilingual Children. Early Childhood Education Journal, 45, 27–34. Doi: 10.1007/s10643-015-0764-8-015-0764-8 [DOI] [Google Scholar]

RESOURCES