A Longitudinal Investigation of the Relations Between Emergent Bilingual Children's Language Experience and Bilingual Language Outcomes

Abstract

Purpose:

This study aims to characterize Spanish–English emergent bilingual (EB) children's language experiences and to examine how these experiences contribute to their bilingual development over a 2-year period. All EB children were enrolled in dual language immersion (DLI) programs. This setting provides a unique opportunity to explore how bilingual language experiences shape bilingual language development across time.

Method:

Fifty-seven Spanish–English EB children participated in the study. Bilingual language skills were assessed at two time points: Time 1 (2019) and Time 2 (2021). Participants were in Grade 1 or 3 at Time 1 (M_age = 91.4 months, SD = 12.37) and Grade 3 or 5 at Time 2 (M_age = 113.43 months; SD = 14.07), respectively. Children were assessed on phonological awareness, morphological awareness, vocabulary, and syntactic awareness in both English and Spanish. Language experience was measured through child- and teacher-reported surveys.

Results:

Over time, teacher-reported Spanish experience at school increased and English experience decreased, while child-reported English experience at home remained stable. Except for Spanish vocabulary, significant gains were found for all language measures, indicating continuous development trajectories. Teacher-reported English experience at Time 1 positively predicted English vocabulary gains. Time 2 teacher-reported Spanish experience and child-reported English experience negatively predicted Spanish vocabulary growth, suggesting possible trade-offs.

Conclusions:

This study highlights the importance of balanced, high-quality bilingual experience across home and school. DLI programs support growth in multiple language domains, though Spanish expressive vocabulary may need targeted support. Findings have educational and clinical implications for assessment, early intervention, and promoting heritage language use without compromising English-language development.

The number of bilingual children is growing worldwide due to globalization and immigration trends (Huang & Kuo, 2020). Language proficiency is a well-established predictor of academic success, yet school-aged emergent bilingual¹ (EB) children face the dual challenge of acquiring two languages while simultaneously mastering academic content (Howard et al., 2014; Huang et al., 2022). Understanding their language development is therefore critical. In the United States, 22% of children in 2016 spoke a language other than English at home (Annie E. Casey Foundation, 2018), and approximately 80% of EB children spoke Spanish as their home language (Office of English Language Acquisition, 2023). This study focuses on Spanish–English EB children to better understand their language experiences and bilingual development trajectories.

Among the array of predictors of language development, language experience, defined as encompassing both language input and output, has emerged as one of the most powerful predictors (Choi & Shen, 2025; Coffey & Snedeker, 2025; Hammer et al., 2020; Hoff, 2018; Paradis, 2023; Paradis & Jia, 2017; Place & Hoff, 2011; Shen & Choi, 2024). Despite considerable evidence linking children's language experience to their language development, relatively fewer studies have focused on EB children who are developing two or more languages concurrently. Given the strong connection between language proficiency and academic success (Hoover & Gough, 1990), it is critical to investigate how EB children's language experiences contribute to their dual language development. Unlike less malleable predictors such as socioeconomic status, language experience is modifiable and, therefore, presents a meaningful target for interventions that aim to support EB children's linguistic and academic growth.

This study is guided by two complementary theoretical frameworks: bioecological theory (Bronfenbrenner & Morris, 1998) and sociocultural theory (Vygotsky, 1978). Bioecological theory conceptualizes child development as a product of dynamic, reciprocal interactions, known as proximal processes, between the child and the multiple layers of their environment, including home, school, and cultural context. Within this framework, language experiences such as child-directed speech are seen as key drivers of cognitive and linguistic development. Sociocultural theory complements this view by emphasizing the role of social interaction and culturally mediated tools (e.g., language and symbols) in shaping learning. According to this theory, language is acquired through meaningful, scaffolded interactions with more knowledgeable others, rather than mere exposure. Together, these frameworks inform the study's central hypothesis that bilingual children's language development is directly shaped by their language experiences. By integrating both perspectives, the study highlights how bilingual language learning is an active, context-dependent, and malleable process that unfolds over time through interactions within the child's social and environmental ecology.

Bilingual Children's Language Experience

Bilingual children's language skills vary widely, and a major contributor to this variability is differences in language experience, which consists of language input and output (Bedore et al., 2016; Hoff & Core, 2013; Pearson et al., 1997; Rowe, 2012). Language input is generally operationalized as child-directed speech or language exposure, whereas language output refers to children's active language use. Researchers have measured language input in terms of both quantity (e.g., number of words or utterances) and quality (e.g., lexical diversity, mean length of utterance), with consistent findings showing that richer input contributes to stronger language and later literacy skills (Coffey & Snedeker, 2025; Hirsh-Pasek et al., 2015; Paradis, 2011; Rowe, 2012). Input has also been operationalized as the frequency or amount of exposure to a target language, often measured via parent, teacher, or child self-report.

Language output, in contrast, reflects the child's active engagement in speaking and writing and has typically been measured using adult reports (Bedore et al., 2016; Hoff & Ribot, 2017). Research has shown that greater language output is associated with stronger expressive language skills, particularly in the language used most frequently. However, many EB children tend to respond in the societal/majority language, resulting in a pattern of stronger receptive skills but weaker expressive skills in their heritage language (HL; Hoff, 2018). These patterns highlight the importance of capturing both dimensions of language experience to understand bilingual children's developmental trajectories.

Previous studies have shown that bilingual children's language experience shifts over time (Huang & Kuo, 2020). Paradis (2011) showed that exposure to each language in different social contexts (e.g., home vs. school) influences language dominance and proficiency over time. In a recent study, Welsh and Hoff (2021) highlighted how Spanish–English bilingual preschoolers' exposure to the societal language (English in this case) increased significantly in out-of-home settings, such as school and social interactions. During the preschool years (ages 3–5 years), bilingual children's English exposure increased and Spanish exposure decreased, largely due to increased exposure to English in school and social environments. Although there are some variabilities based on parents' language backgrounds and education level, the general pattern gradually shifts children's language dominance. This shift underscores the need for structured support from parents and educators to maintain bilingual children's home language proficiency.

The Effect of Language Experience on Language Development

Language experience plays a fundamental role in shaping child language development, influencing their vocabulary, grammar, and overall proficiency. Rowe (2012) demonstrated that both the quantity and quality of child-directed speech significantly impact vocabulary growth, with diverse and sophisticated input fostering richer language development. Hoff (2006, 2018) further emphasizes that differences in language exposure, including frequency, complexity, and interactional context, account for variations in children's language outcomes, particularly in bilingual children from low-socioeconomic backgrounds. In a recent meta-analysis study, Coffey and Snedeker (2025) examined the relation between caregiver speech (i.e., language input) and child language development. Results from this study indicate that language input has a moderate but consistent influence across different study designs, input measures, and home language groups, including non–English-speaking populations. The results also showed no significant differences in effect size across four common language input measures: tokens, types, utterances, and mean length of utterance.

On the other hand, language output, the productive dimension of language experience, has also been shown to impact language development. In a recent review, Paradis (2023) found that language use with different interlocutors (e.g., parents, siblings) shapes EB children's outcomes in both their HL and societal/majority language. Specifically, increased use of the HL at home supports HL development, while use of the societal/majority language with siblings or peers can enhance societal/majority language abilities but potentially undermine HL proficiency.

Hoff and Ribot (2017) examined the relation between language use and vocabulary development among Spanish–English EB preschoolers between the ages of 2.5 and 5 years. All children were living in South Florida in the United States at the time of the study. The authors found that greater English use at home was associated with stronger English expressive vocabulary growth among Spanish–English EBs. However, this relation was nonlinear: Benefits of increased English exposure were most pronounced in homes where English was already dominant. Conversely, early exposure to Spanish did not guarantee Spanish development, as bilingual children generally showed stronger growth in English than in Spanish. Ribot et al. (2018) also worked with Spanish–English EB preschoolers from Spanish-speaking families in South Florida and found that language output, operationalized as active language use, is crucial for expressive language development. Extending this line of inquiry, Ryan (2021) worked with early elementary students in a French–English dual language immersion (DLI) program in the United States. She found that out-of-school active use of French, but not listening exposure, plays a significant role in vocabulary growth. Their sample consisted of both French HL and non-French HL speakers.

Together, these findings reinforce the idea that bilingual children must actively use their home language to maintain their HL proficiency. Also, active use of the societal/majority language also promotes development in that language. Collectively, these studies underscore that language development is not solely biologically determined but shaped by the richness, consistency, and diversity of language experiences across different learning environments.

Language Experience and HL Maintenance

Language experience also plays a critical role in the maintenance of bilingual children's HL, especially in contexts where the societal language dominates. Without deliberate efforts, bilingual children risk losing proficiency in their HL over time. Paradis (2011) emphasizes that consistent exposure and use of the HL within the family are essential for sustaining bilingual children's HL proficiency, with reduced input and exposure often leading to attrition over time. Similarly, Hoff et al. (2012) demonstrate that bilingual children from Spanish-speaking homes often experience diminished Spanish proficiency when English exposure dominates in school and peer interactions. In their study, bilingual children who receive consistent and high-quality input in Spanish at home maintain stronger proficiency compared to those who primarily interact in English.

Existing research on bilingual children's language output and HL development has demonstrated its independent role in HL development, beyond the influence of language input (Bedore et al., 2012; Hammer et al., 2012; Hoff, 2018). Bedore et al. (2012) underscored the importance of language output in their study of Spanish–English EB children. They found that current language output was a stronger predictor of language dominance and proficiency than age of first exposure. In a recent study with Mandarin–English bilingual children by Song et al. (2024), the researchers found that Mandarin–English bilinguals in the United States exhibited stagnation in their Chinese vocabulary growth, despite continued home language exposure. Their study also revealed that structured language experiences, such as storytelling and book reading, significantly predicted stronger home language development. The results suggest that mere exposure is insufficient and active engagement is key. These findings also align with broader evidence that HL maintenance requires not only frequent exposure but also meaningful interactions and literacy-rich activities, reinforcing the need for parents and educators to actively support bilingual children's linguistic development across both their languages.

Language Development Among Bilingual Children in Dual Language Programs

While much of the existing research on school-aged EB children focuses on those in English-only instructional programs, where bilingual children's HL is minimal or absent, the current study offers a unique perspective by examining EBs enrolled in DLI programs. In contrast to English-only programs, bilingual children in DLI program receive content-area instruction in their HL.

Burkhauser et al. (2016) analyzed data from a longitudinal study on DLI programs in Oregon to examine students' proficiency growth in Spanish, Japanese, and Mandarin Chinese between third and eighth grades. Most students in the Spanish DLI programs came from English-speaking homes, with approximately 30% (N = 230) reporting Spanish as their home language, while nearly all students in the Japanese and Mandarin Chinese DLI programs were from English-speaking households. Using the Standards-Based Measurement of Proficiency, the authors found that the majority of students across all three language tracks reached at least intermediate low proficiency by eighth grade. These outcomes underscore the effectiveness of DLI programs, particularly in comparison to traditional foreign language instruction. Notably, EB students from Spanish-speaking homes outperformed their English-speaking peers in Spanish listening and writing, but not in Spanish reading or speaking, suggesting nuanced patterns of language development influenced by home language background.

In another study that utilized statewide assessment data, Watzinger-Tharp et al. (2018) examined language outcomes in Utah's DLI programs. The vast majority of the students were from English-speaking households. The authors analyzed the language performance of students in Grades 3 through 6 who were enrolled in Spanish, French, or Mandarin Chinese DLI program. They assessed students' speaking, reading, and reading using standardized language measures developed by the American Council on the Teaching of Foreign Languages. Results showed that most students reached intermediate low proficiency by sixth grade, with higher performance in Spanish and French than in Mandarin Chinese. Across languages, students performed better in receptive skills (reading and listening) than productive skills (speaking). Taken together, both studies highlight the long-term effectiveness of DLI programs in fostering partner language proficiency compared to traditional foreign language instruction. They also underscore the importance of language typology, instructional consistency, and home language use in shaping bilingual development within DLI contexts.

Turning now to studies that focused on Spanish–English EB children in DLI programs, Lucero (2018) examined how Spanish–English EB children in DLI programs develop their narrative retelling skills in both languages. Over a 2-year period from kindergarten to second grade, the study found significant improvements in vocabulary in both languages but no significant gains in grammar in either language. Children's macrostructure skills (i.e., story organization) improved in English over time but stagnated in Spanish, despite continued DLI instruction. The results suggest that narrative development may diverge across languages, with growth increasingly favoring English, a trend that may reflect shifts in language experiences or language dominance over time.

Miller et al. (2006) followed a group of Spanish–English EB children in DLI programs from kindergarten through third grade to study their English and Spanish oral language and reading development. Their results demonstrated that oral language abilities predicted reading achievement within and across languages, highlighting the idea that strong language skills in one language can positively impact literacy in the other language. In another study, Huang et al. (2021) also found that Spanish–English EBs' narrative skills in their HL Spanish significantly contributed to their English reading outcomes, reinforcing the role of linguistic interdependence in bilingual literacy.

Howard et al. (2014) worked with third-grade Spanish–English EBs enrolled in DLI programs and found that language experiences at home and in school played an important role in shaping children's oral vocabulary development, which in turn was a strong predictor of English reading outcomes. Their study revealed that EB children's language experience in Spanish and English varied across contexts. At school, most children received instruction predominantly in Spanish, while at home, Spanish was more commonly spoken with parents, and English was more often used with siblings. This variation in language experience correlated with vocabulary development: Children demonstrated stronger receptive vocabulary in Spanish than in English, reflecting the cumulative effects of early and sustained exposure to Spanish at both home and school. These findings point to the importance of assessing oral vocabulary in both of EBs' languages and taking their language experiences into account when interpreting their vocabulary knowledge.

The Current Study

This study aims to characterize Spanish–English EB children's language experience and the contributions of experience to their bilingual development by leveraging a longitudinal data set that includes 57 Spanish–English EB children in DLI program. DLI programs provide more balanced and structured experience in both languages, allowing us to explore how such experiences shape bilingual language outcomes over a 2-year period. Language experience was measured using both child and teacher report, while language proficiencies in the domains of phonological awareness (PA), morphological awareness (MA), vocabulary, and syntactic awareness (SA) in both languages were evaluated using standardized or research-developed assessments. The three research questions are as follows:

1. How does Spanish–English EBs' language experience change over 2 years (Time 1 to Time 2)?

2. To what extent do Spanish–English EBs develop proficiency in both Spanish and English over 2 years (Time 1 to Time 2)?

3. How does EBs' language experience at Time 1 and Time 2 predict their bilingual language outcomes at Time 2, controlling for child age, sex, and socioeconomic status?

Method

The current study is a secondary analysis of data from 57 Spanish–English EB children drawn from a multiyear investigation of Spanish–English EBs' language and literacy development (Huang et al., 2022). All participants were recruited from Spanish–English one-way DLI programs² at five elementary schools in a Southwestern U.S. city. The educational goals of the district's DLI program are to promote bilingualism, biliteracy, academic achievement, and cross-cultural awareness. The district implements a 50/50 immersion model, in which students receive instruction in both English and Spanish. Literacy instruction is provided in both languages through dedicated Spanish Language Arts and English Language Arts periods. Content areas are divided by language: Mathematics is taught in Spanish, while science and social studies are taught in English. All the schools serve a high concentration of students of Mexican heritage and are located in neighborhoods with a significant percentage of economically disadvantaged households. School records consisting of parents' responses to a home language survey indicated that all participants spoke Spanish at home and were classified as EBs at the time of data collection.

The 57 children were tracked over 2 years and assessed on bilingual language skills at two time points: Time 1 in fall 2019 and Time 2 in fall 2021. They were a mixed grade group: At Time 1, they were either in Grade 1 (n = 25) or Grade 3 (n = 32), with the mean age being 91.4 months (SD = 12.37). At Time 2, they all progressed to Grade 3 or Grade 5, respectively (M_age = 113.43 months, SD = 14.07).³ Children's sex is roughly balanced in the sample (45% girls). None of the children had significant speech, language, or developmental disorders, according to school reports.

Instruments

Bilingual language skills (i.e., PA, MA, vocabulary, and SA) in English and Spanish were measured at both time points. Percentages of correct scores were derived and used as variables.

English and Spanish PA

The Elision subtest from the Comprehensive Test of Phonological Processing–Second Edition (CTOPP-2; Wagner et al., 2013) was administered to assess PA in English. Similarly, the Elision subtest from the Test of Phonological Processing in Spanish (TOPPS; Francis et al., 2001) was used to measure Spanish PA. Both tests exhibit high reliability: The CTOPP-2 reports reliability indices ranging from .88 to .94 for children aged 4–17 years, while the TOPPS has a reliability index of .83 for the full test.

English and Spanish MA

The Early Lexical Morphology Measure (ELMM) in English and its Spanish counterpart (ELMM-S; Marks et al., 2023) were used to assess participants' implicit MA in both languages. In these tasks, participants were provided with a word (e.g., quickly) and asked to complete a sentence using part of that word (e.g., That lion was really ___ [quick]). Both measures focused on compound and derivational morphology but did not include items targeting inflectional morphology. The reliability indices (α) for ELMM and ELMM-S are .93 and .95, respectively.

English and Spanish Vocabulary

The English Picture Vocabulary subtest (Test 1) and its Spanish counterpart, Vocabulario sobre dibujos (Test 10), from the Woodcock–Johnson IV Tests of Oral Language (Schrank et al., 2014), were administered to assess expressive vocabulary knowledge. In both subtests, participants were presented with a picture of an object and required to name the object in the picture. The items in the two language versions are not direct translations, though the constructs are equivalent. The reliability coefficient ranges from .77 to .94 for both the English and Spanish subtests.

English and Spanish SA

The Morphosyntax subtest of the Bilingual English–Spanish Assessment: Middle Extension (BESA-ME; Peña et al., 2010) was used to evaluate participants' SA in both English and Spanish. The BESA-ME is normed for children aged 7;0–12;6 (years;months), making it suitable for the participants in this study. The test includes cloze and sentence repetition tasks, where participants completed and repeated sentences featuring fundamental morphosyntactic constructions, such as regular and irregular past tense and relative clauses. Participants were required to demonstrate understanding by selecting the correct picture from a set of options. The coefficient α for this subtest is .87 and .88 in Spanish and English, respectively.

Language Experience

To gather information about the child participants' demographic backgrounds and their language experience, we administered a “child survey” and a “teacher survey” using items adapted from previous research (e.g., Huang et al., 2021, 2022). The child survey included questions about participants' demographic information (e.g., date of birth, home language), language experience (e.g., “How frequently do you use English at breakfast in the morning?”), and the frequency of reading various genres in English and Spanish (e.g., “How often do you read magazines in English/Spanish outside of school?”), assessed on a Likert-type scale (see Appendix A for the full survey). To create the “child reported relative English experience” variable, we derived a composite score by averaging participants' self-reported ratings across seven survey items. Out of the seven items, only one is related to school experience. The other six primarily capture language experience at home. Responses were rated on a 5-point Likert scale (1 = no English, 2 = a little English, 3 = some English, 4 = mostly English, 5 = only English).

The teacher survey provided data on the child participants' frequency of Spanish and English experience at school, including interactions with teachers, peers, and other adults. Questions included items such as “How often does the child speak with you, the teacher, in Spanish/English?” Responses were recorded on a 6-point scale (0 = never, 1 = rarely, 2 = sometimes, 3 = very often, 4 = usually, 5 = always, 6 = don't know). Each language had three items, for a total of six questions (see Appendix B for the full survey). Composite scores for “teacher-reported English experience at school” and “teacher-reported Spanish experience at school” were calculated by averaging teacher ratings across the three items corresponding to each language.

Procedure

At each time point, participants completed language assessments during two separate sessions, spaced at least 4 days apart to minimize practice effects. The order of the two language sessions was counterbalanced at both time points.⁴ In each session, a native or heritage speaker research assistant fluent in the target language (English or Spanish) provided instructions in the corresponding language. The same research assistant also administered the child survey at the end of the Spanish session, with the survey language adjusted based on each child's preference. Participants were offered multiple short breaks during each session. Teacher surveys were distributed to teachers of the child participants via e-mail and were collected after the completion of all child assessments.

Analytical Approach

First, to examine the differences of language experience between Time 1 and Time 2, we specified paired-samples t tests comparing English- and Spanish-language experience in school reported by teachers and relative English experience reported by children across two time points. Composite variables averaged from teacher-reported English-language experience, Spanish-language experience, and child-reported relative English-language experience were used in the paired-samples t tests. Second, to investigate whether bilingual children grew significantly in their bilingual language outcomes from Time 1 to Time 2, we conducted paired-samples t tests comparing their language scores in Time 1 and those in Time 2. For both these analyses, Bonferroni correction was applied to correct for Type 1 error due to multiple t-test comparisons.

Linear regression was used to examine the relations between language experience and bilingual language outcomes over time. We specified eight regression models predicting each of the language outcomes in English and Spanish at Time 2 (i.e., PA, vocabulary, MA, and SA). In each of the regression model, we used teacher-reported English-language experience and Spanish-language experience as well as child-reported relative English-language experience at both Time 1 and Time 2 as independent variables. To estimate the predictive relations between language experience and residualized gains in language outcomes, we included the corresponding bilingual language outcomes at Time 1 in the regression models. Moreover, child age, sex, and socioeconomic status were further controlled for in the models.

Assumptions of normality, homoscedasticity, and linearity were met for all regression models. We used Cook's distance (Cook, 1977) to detect outliers. Cook's distance combines standardized residuals and leverage values to identify influential observations, with a distance of over 1 signaling an influential observation. Cook's distance ranged from 0 to 0.70, indicating that there were no outliers in the sample. Variance inflation factors (VIFs) were used to identify multicollinearity issues, with a value of 10 indicating a collinearity problem (O'Brien, 2007) and a value of 5 indicating considerable collinearity (Menard, 2001). The VIFs of the independent variables ranged from 1.25 to 3.91, suggesting no multicollinearity issues.

Results

Changes of Language Experience and Bilingual Language Outcomes Over Time

Table 1 displays descriptive statistics and paired-samples t-test comparisons between children's language experience at Time 1 and Time 2. Figure 1 presents the changes from Time 1 to Time 2. Teachers reported that bilingual children had a significant decrease in their English experience at school, t(56) = −4.12, p < .001, Cohen's d = 0.55, from Time 1 (M = 2.92, SD = 1.28) to Time 2 (M = 1.82, SD = 1.02). However, teacher-reported children's Spanish-language experience at school had a significance increase, t(56) = 2.89, p = .005, Cohen's d = 0.38, from Time 1 (M = 3.29, SD = 1.25) to Time 2 (M = 4.02, SD = 0.99). There was no significant difference of child-reported relative English experience, t(50) = 1.24, p = .222, Cohen's d = 0.17, between Time 1 (M = 2.60, SD = 0.92) and Time 2 (M = 2.77, SD = 0.83). Results stayed consistent after correcting for multiple t-test comparisons using Bonferroni correction.

Table 1.

Descriptive statistics and comparisons of language experience and language skills over time.

Variable	Descriptive statistics				Paired-samples t test		Grade 1 subsample (n = 25)	Grade 3 subsample (n = 32)
	M	SD	Min	Max	t	p	t	t
Language experience
Teacher-reported English experience in school (N = 57)
Time 1	2.92	1.28	0	5
Time 2	1.82	1.02	0.67	5
Change	−1.11	2.03	−4	5	t(56) = −4.12***	< .001	t(24) = −0.39	t(31) = −6.85***
Teacher-reported Spanish experience in school (N = 57)
Time 1	3.29	1.25	0.67	5
Time 2	4.02	0.99	1	5
Change	0.74	1.93	−4	4.33	t(56) = 2.89**	.005	t(24) = −0.78	t(31) = 5.28***
Child-reported relative English experience (N = 51)
Time 1	2.6	0.92	1	5
Time 2	2.77	0.83	1	4.33
Change	0.17	0.98	−1.67	3.33	t(50) = 1.24	.222	t(23) = 1.61	t(26) = 0.07
Bilingual language outcomes
English phonological awareness (N = 57)
Time 1	0.56	0.25	0	0.97
Time 2	0.72	0.21	0.21	0.97
Change	0.16	0.16	−0.15	0.56	t(56) = 7.54***	< .001	t(24) = 7.04***	t(31) = 4.82***
English vocabulary (N = 57)
Time 1	0.37	0.08	0.11	0.54
Time 2	0.44	0.07	0.28	0.59
Change	0.07	0.05	−0.07	0.17	t(56) = 11.21***	< .001	t(24) = 7.67***	t(31) = 8.07***
English morphological awareness (N = 57)
Time 1	0.28	0.2	0	0.78
Time 2	0.54	0.22	0.12	0.88
Change	0.26	0.15	−0.1	0.57	t(56) = 12.72***	< .001	t(24) = 8.75***	t(31) = 9.58***
English syntactical awareness (N = 57)
Time 1	0.68	0.22	0.13	0.98
Time 2	0.78	0.19	0.13	1
Change	0.1	0.14	−0.15	0.57	t(56) = 5.58***	< .001	t(24) = 4.49***	t(31) = 3.73***
Spanish phonological awareness (N = 57)
Time 1	0.54	0.34	0	1
Time 2	0.74	0.31	0	1
Change	0.2	0.27	−0.2	0.85	t(56) = 5.72***	< .001	t(24) = 6.83***	t(31) = 2.26*
Spanish vocabulary (N = 57)
Time 1	0.43	0.1	0.23	0.6
Time 2	0.45	0.12	0.19	0.64
Change	0.02	0.07	−0.13	0.21	t(56) = 1.69	.096	t(24) = 1.74	t(31) = 0.69
Spanish morphological awareness (N = 57)
Time 1	0.51	0.28	0.03	0.93
Time 2	0.68	0.24	0.13	0.98
Change	0.17	0.19	−0.38	0.68	t(56) = 6.59***	< .001	t(24) = 6.91***	t(31) = 3.29**
Spanish syntactical awareness (N = 57)
Time 1	0.53	0.24	0.05	0.83
Time 2	0.61	0.25	0.04	0.94
Change	0.07	0.13	−0.31	0.38	t(56) = 4.43***	< .001	t(24) = 3.74**	t(31) = 2.58*

^*p < .05.

^**p < .01.

^***p < .001.

Figure 1.

Changes in language experience from Time 1 to Time 2.

Table 1 also presents descriptive statistics and the paired-samples t-test comparisons of children's bilingual language outcomes between Time 1 and Time 2 (also see Figure 2). From Time 1 to Time 2, bilingual children showed significant growth in all bilingual language outcomes except for Spanish vocabulary, t(56) = 1.69, p = .096, Cohen's d = 0.22. For English-language measures, there was a 16% increase of English PA, t(56) = 7.54, p < .001, Cohen's d = 1.00, from Time 1 (M = 0.56, SD = 0.29) to Time 2 (M = 0.72, SD = 0.21); a 7% increase of English vocabulary, t(56) = 11.21, p < .001, Cohen's d = 1.49, from Time 1 (M = 0.37, SD = 0.08) to Time 2 (M = 0.44, SD = 0.07); a 26% increase of English MA, t(56) = 12.72, p < .001, Cohen's d = 1.68, from Time 1 (M = 0.28, SD = 0.20) to Time 2 (M = 0.54, SD = 0.22); and a 10% of gain in English SA, t(56) = 5.58, p < .001, Cohen's d = 0.74, from Time 1 (M = 0.68, SD = 0.22) to Time 2 (M = 0.78, SD = 0.19).

Figure 2.

Changes in bilingual language outcomes from Time 1 to Time 2. PA = phonological awareness; MA = morphological awareness; SA = syntactic awareness.

For Spanish-language measures, EBs demonstrated a 20% of growth of Spanish PA, t(56) = 5.72, p < .001, Cohen's d = 0.76, from Time 1 (M = 0.54, SD = 0.34) to Time 2 (M = 0.74, SD = 0.31); a 17% increase of Spanish MA, t(56) = 6.59, p < .001, Cohen's d = 0.87, from Time 1 (M = 0.51, SD = 0.28) to Time 2 (M = 0.68, SD = 0.24); and a 7% of gain in their Spanish SA, t(56) = 4.43, p < .001, Cohen's d = 0.59, from Time 1 (M = 0.53, SD = 0.24) to Time 2 (M = 0.61, SD = 0.25). The significant growth from Time 1 to Time 2 stayed robust after correcting for multiple t-test comparisons using Bonferroni correction.

To account for the age differences, we ran some additional paired-samples t-test comparisons separately for the younger (i.e., first graders at Time 1) and older (i.e., third graders at Time 1) cohorts. Results are presented in Table 1, revealing differences in teacher-reported language use across cohorts. Specifically, children in the younger cohort did not show changes in teacher-reported language use at school, whereas children in the older cohort showed significant decrease in use of English at school and significant increase in use Spanish at school from Time 1 to Time 2. Child-reported language use and all bilingual language outcomes did not show different changing/growing patterns between the two cohorts.

Relations Between Language Experience and Bilingual Language Outcomes Over Time

Table 2 presents the results of the regression models for each bilingual language outcome. We report significant findings of the associations between language experience and residualized gains in bilingual language outcomes. First, teacher-reported English experience at Time 1 positively predicted EB children's gain in English vocabulary from Time 1 to Time 2 (β = .27, p = .032). Second, teacher-reported English experience (β = .46, p < .001) and Spanish experience (β = .33, p = .014) at Time 1 both positively predicted Spanish vocabulary gain. However, teacher-reported Spanish experience (β = −.32, p = .015) at Time 2 and more English experience reported by children (β = −.30, p = .003) at Time 2 both negatively predicted children's residualized gain in Spanish vocabulary. Furthermore, more English experience reported by children in Time 2 negatively predicted children's Spanish SA gain (β = −.22, p = .011).

Table 2.

Linear regressions predicting bilingual language outcomes over time.

Variable	B	SE	β	p
Predicting Time 2 English phonological awareness
Time 1 teacher-reported English experience	0.02	0.03	.23	.424
Time 2 teacher-reported English experience	0.01	0.04	.03	.855
Time 1 teacher-reported Spanish experience	0.03	0.03	.20	.221
Time 2 teacher-reported Spanish experience	0.03	0.03	.14	.397
Time 1 child-reported relative English experience	−0.01	0.03	−.06	.627
Time 2 child-reported relative English experience	−0.002	0.03	−.01	.953
Age in months	−0.003	0.002	−.16	.226
Male	0.12	0.05	.29**	.009
Time 1 skill	0.78	0.10	.88***	< .001
Predicting Time 2 English vocabulary
Time 1 teacher-reported English experience	0.02	0.01	.27*	.032
Time 2 teacher-reported English experience	0.01	0.01	.18	.243
Time 1 teacher-reported Spanish experience	−0.01	0.01	−.14	.310
Time 2 teacher-reported Spanish experience	0.002	0.01	.03	.843
Time 1 child-reported relative English experience	0.01	0.01	.12	.235
Time 2 child-reported relative English experience	0.01	0.01	.07	.492
Age in months	0.000	0.001	−.02	.825
Male	−0.003	0.01	−.02	.809
Time 1 skill	0.55	0.11	.62***	< .001
Predicting Time 2 English morphological awareness
Time 1 teacher-reported English experience	0.001	0.03	.004	.979
Time 2 teacher-reported English experience	−0.01	0.04	−.03	.864
Time 1 teacher-reported Spanish experience	−0.03	0.03	−.16	.361
Time 2 teacher-reported Spanish experience	−0.02	0.04	−.08	.649
Time 1 child-reported relative English experience	0.02	0.03	.07	.572
Time 2 child-reported relative English experience	−0.01	0.03	−.04	.777
Age in months	0.003	0.002	.15	.304
Male	−0.04	0.05	−.09	.433
Time 1 skill	0.65	0.14	.59***	< .001
Predicting Time 2 English syntactical awareness
Time 1 teacher-reported English experience	0.04	0.02	.24	.116
Time 2 teacher-reported English experience	0.04	0.04	.19	.319
Time 1 teacher-reported Spanish experience	0.03	0.03	.19	.248
Time 2 teacher-reported Spanish experience	0.03	0.03	.16	.319
Time 1 child-reported relative English experience	0.02	0.02	.08	.494
Time 2 child-reported relative English experience	0.02	0.03	.08	.487
Age in months	0.001	0.002	.05	.661
Male	−0.03	0.04	−.07	.529
Time 1 skill	0.64	0.11	.71***	< .001
Predicting Time 2 Spanish phonological awareness
Time 1 teacher-reported English experience	0.04	0.04	.15	.378
Time 2 teacher-reported English experience	0.04	0.07	.12	.589
Time 1 teacher-reported Spanish experience	0.02	0.05	.06	.756
Time 2 teacher-reported Spanish experience	0.08	0.06	.25	.208
Time 1 child-reported relative English experience	−0.10	0.05	−.28*	.043
Time 2 child-reported relative English experience	0.04	0.05	.10	.444
Age in months	−0.01	0.004	−.19	.203
Male	0.03	0.08	.05	.708
Time 1 skill	0.69	0.13	.73***	< .001
Predicting Time 2 Spanish vocabulary
Time 1 teacher-reported English experience	0.04	0.01	.46***	< .001
Time 2 teacher-reported English experience	−0.02	0.02	−.13	.376
Time 1 teacher-reported Spanish experience	0.03	0.01	.33*	.014
Time 2 teacher-reported Spanish experience	−0.04	0.02	−.32*	.015
Time 1 child-reported relative English experience	0.002	0.01	.02	.848
Time 2 child-reported relative English experience	−0.04	0.01	−.30**	.003
Age in months	0.00	0.001	−.01	.956
Male	−0.02	0.02	−.10	.228
Time 1 skill	0.73	0.11	.65***	< .001
Predicting Time 2 Spanish morphological awareness
Time 1 teacher-reported English experience	0.03	0.03	.14	.359
Time 2 teacher-reported English experience	0.03	0.04	.13	.485
Time 1 teacher-reported Spanish experience	−0.03	0.03	−.18	.265
Time 2 teacher-reported Spanish experience	−0.02	0.04	−.07	.676
Time 1 child-reported relative English experience	−0.04	0.03	−.15	.201
Time 2 child-reported relative English experience	−0.06	0.03	−.20	.091
Age in months	−0.003	0.003	−.17	.216
Male	0.001	0.05	.002	.984
Time 1 skill	0.56	0.11	.67***	< .001
Predicting Time 2 Spanish syntactical awareness
Time 1 teacher-reported English experience	0.02	0.02	.08	.429
Time 2 teacher-reported English experience	−0.01	0.03	−.05	.717
Time 1 teacher-reported Spanish experience	−0.02	0.02	−.10	.414
Time 2 teacher-reported Spanish experience	−0.03	0.03	−.13	.255
Time 1 child-reported relative English experience	−0.04	0.02	−.14	.106
Time 2 child-reported relative English experience	−0.07	0.03	−.22*	.011
Age in months	−0.001	0.002	−.05	.621
Male	−0.03	0.04	−.06	.470
Time 1 skill	0.78	0.09	.74***	< .001

Note. English-language experience and Spanish-language experience in school were reported by teachers; relative English experience was reported by children.

^*p < .05.

^**p < .01.

^***p < .001.

We also report significant associations between other control variables and bilingual language outcomes. Time 1 corresponding skills consistently predicted Time 2 bilingual language outcomes for all measures (all ps < .001), suggesting the importance of prior skills. In addition, sex predicted children's English PA gain (β = .29, p = .009). That is, boys tended to grow more in English PA from Time 1 to Time 2, compared to girls.

Discussion

There is a growing number of Spanish–English EB children in the United States, yet research efforts remain limited in understanding their language experience and bilingual language outcomes, particularly those who are enrolled in DLI programs. As these EB children navigate both languages in educational and home environments, it becomes crucial to examine how their exposure to and use of both languages shapes their bilingual development trajectories.

This study focuses on Spanish–English EB children in a DLI program, offering unique insights into changes in their language experiences over 2 years and how these experiences relate to bilingual language outcomes. Understanding the evolution of these experiences, both at home and in school, can inform educational practices that support optimal bilingual development.

Language Experience Changes Over Time and Vary by Context

We found an interesting shift in language experience patterns at school, such that teachers reported a significant decrease in English experience alongside an increase in Spanish experience over time. This suggests a shift in the school environment that emphasizes Spanish-language environment over time, reflecting the effort with the DLI program to enhance Spanish exposure and use as part of its bilingual education goals. This shift may be a result of responding to observed language needs among students.

In contrast, child-reported English experience, primarily reflecting home language use, remained relatively stable. This discrepancy between teacher and child reports highlights the importance of context in shaping language use and perception. These findings align with prior studies (e.g., Choi & Shen, 2025; Paradis, 2011; Welsh & Hoff, 2021) showing that EB children's language experiences vary significantly across settings. They underscore the need to consider both home- and school-based experience when evaluating bilingual development.

The different developmental patterns between the younger and older cohorts are worth mentioning. While younger children did not exhibit significant changes in their language use at school over time, older children showed a notable shift, specifically a decrease in English use and an increase in Spanish use at school from Time 1 to Time 2. These findings suggest that older bilingual children may become more comfortable or intentional in using their HL in academic settings as they progress through school. It is also possible that changes in classroom dynamics, peer interactions, or teacher practices over time may support greater use of Spanish among older students. Interestingly, no age-related differences emerged in child-reported language use, indicating that while perceived or observed use at school may vary with age, self-reported language practices follow similar patterns across cohorts. These findings underscore the importance of considering both the reporter and the child's developmental stage when examining bilingual language use in school contexts.

Growth in Bilingual Language Proficiency Over Time

We observed a significant increase in English PA, vocabulary, MA, and SA over time, which demonstrates the effectiveness of their learning environments in fostering English-language development. The most pronounced growth was observed in English MA, which increased by 26%, suggesting that children are making substantial gains in understanding and using morphological structures in English. Additionally, the 16% growth in English PA and 10% growth in English SA indicate meaningful progress in skills that support literacy and comprehension in DLI programs. Similarly, the significant improvements in Spanish PA, MA, and SA further emphasize the children's bilingual language development. The 20% growth in Spanish PA and 17% growth in Spanish MA suggest that children were acquiring foundational language skills in Spanish, potentially driven by the DLI program's emphasis on bilingualism and biliteracy. The more modest 7% growth in Spanish SA indicates steady, albeit slower, development in syntactic complexity, which may require longer periods of exposure and practice to achieve substantial gains (Brown, 1973; Ortega, 2009). Brown (1973) documented the gradual emergence of syntactic structures in early language development, and Ortega (2009) argued that extended time and practice are required for syntactic development. These findings suggest that the DLI program is supporting children's bilingual development across multiple domains, although rates of growth differ between languages and skills.

Notably, Spanish vocabulary did not show significant growth, which may indicate challenges in expressive vocabulary acquisition despite the DLI context. This lack of significant growth could stem from limited exposure to diverse vocabulary in Spanish compared to English or a greater focus on other linguistic skills in the program, such as MA development. Alternatively, it may reflect children's reliance on English vocabulary development, given its utility in broader social and academic contexts. These results align with a previous study by Mancilla-Martinez and Lesaux (2011), who followed Spanish–English EBs longitudinally from preschool through fifth grade and found decelerating growth in Spanish vocabulary. The rate of Spanish vocabulary growth began to slow down around 8 years of age. By fifth grade, EBs showed substantially lower vocabulary levels in Spanish than in English.

However, the current results contrast with those of Lucero (2018), who found significant growth in expressive vocabulary in both Spanish and English among Spanish–English EBs enrolled in a DLI program from kindergarten to second grade. One possible reason for the divergent findings is the age of participants at the second time point—second grade in Lucero's study versus Grades 3 and 5 in the current study. Additionally, the vocabulary measures were also different: Lucero used a narrative retelling task, while the current study employed a standardized expressive vocabulary test. The current results also differ from Howard et al. (2014), who reported stronger receptive Spanish vocabulary among third-grade EB children in DLI settings. However, the current study assessed expressive rather than receptive vocabulary—skills that are more demanding and reflective of word retrieval and depth of knowledge (Dixon et al., 2023). Previous studies have shown that among Spanish–English EBs, receptive vocabulary scores were consistently higher than expressive ones (Mancilla-Martinez et al., 2020). The use of an expressive rather than receptive vocabulary measure in the study could have underestimated children's Spanish vocabulary knowledge.

Overall, these findings highlight the robust development of Spanish–English EB children's language skills over time, underscoring the dynamic nature of their bilingual language growth in English and Spanish. While the DLI program appears successful in fostering growth in PA, MA, and SA in both languages, targeted efforts may be needed to enhance Spanish vocabulary acquisition. Future research should explore the instructional and environmental factors contributing to these patterns and identify strategies to promote more equitable growth across all bilingual language outcomes.

Relations Between Language Experience and Bilingual Language Outcomes Over Time

Our findings also provide valuable insights into the complex roles of language experiences in shaping bilingual children's language development. For English-language development, teacher-reported English experience at Time 1 positively predicted gains in children's English vocabulary, underscoring the importance of use and exposure to English in supporting vocabulary growth. This result aligns with research emphasizing the role of language experience in language acquisition (Hoff & Ribot, 2017; Paradis, 2023; Ribot et al., 2018), particularly for bilingual children in structured educational settings (Huang et al., 2022; Ryan, 2021).

While the findings suggested a straightforward relation between language experience and EB children's second-language development (i.e., English), the results revealed more complexity in the relations between language experience and HL (i.e., Spanish). Interestingly, teacher-reported English and Spanish experiences at Time 1 positively predicted gains in Spanish vocabulary, suggesting that initial exposure to both languages supports HL vocabulary development. However, the negative associations between Time 2 teacher-reported Spanish experience and child-reported English experience with residualized gains in Spanish vocabulary highlight potential trade-offs between language experience during this period. Increased emphasis on English at home or reduced focus on Spanish in school might constrain children's vocabulary growth in Spanish. Similarly, child-reported English experience at Time 2 negatively predicted Spanish SA gains, indicating that greater reliance on English may inhibit more complex language development in Spanish. These patterns align with previous findings (e.g., Choi & Shen, 2025; Hoff et al., 2012; Song et al., 2024), showing that while HL use does not negatively impact English development, excessive reliance on English experience may hinder HL development.

Finally, the consistent predictive power of Time 1 baseline skills for all Time 2 language outcomes reinforces the foundational role of prior skills in bilingual language development. This finding underscores the importance of early interventions and support to build strong linguistic foundations, which can have cascading effects on subsequent growth. Overall, these findings emphasize the multifaceted nature of bilingual language development and the need for context-based support in educational and home environments.

Conclusions and Implications

To conclude, this study highlights the dynamic and context-specific nature of bilingual language experience among Spanish–English EB children in DLI programs. Over time, school-based language experience shifted toward increased Spanish use and reduced English use, aligning with the DLI program's bilingual education goals. In contrast, child-reported home language experience remained relatively stable, revealing distinct roles that school and home environments play in shaping bilingual trajectories. Children showed significant growth in both English and Spanish in areas such as PA, MA, and SA, suggesting that the DLI context effectively supports the development of bilingual skills. However, despite the program's emphasis on bilingualism, the limited growth in Spanish expressive vocabulary points to an area where more intentional and explicit support is needed.

The study also found that early language experiences, particularly teacher-reported English exposure at Time 1, predicted later English vocabulary development. In contrast, the relations between language experience and Spanish outcomes were more complex. While increased Spanish experience did not hinder English development, a disproportionate reliance on English was associated with reduced growth in Spanish language. These patterns point to potential trade-offs and emphasize the importance of maintaining balanced, high-quality exposure to both languages across contexts. Supporting strong bilingual outcomes requires coordinated efforts at home, in school, and within the broader sociocultural environment. Children need ample opportunities and meaningful incentives to use both languages, especially their HL (López & Paez, 2020). There should also be instructional support, such as via DLI programs, or societal support for HL to counteract prestige imbalances. Importantly, valuing and maintaining the HL does not detract from English-language acquisition. A dual-focus bilingual approach that strengthens both the HL and the societal/majority language can promote synergies that benefit overall language development.

These findings have significant implications for both educational and clinical practices. In educational settings, the uneven growth across language domains, particularly the limited gains in Spanish expressive vocabulary, points to a need for targeted instructional support. Educators should provide rich, diverse vocabulary input in Spanish and create authentic opportunities for students to use the language in both academic and social contexts. Additionally, the strong predictive role of Time 1 baseline skills suggests that early screening and targeted language support can have lasting effects on bilingual development (Crowe et al., 2021; Durán et al., 2016).

In clinical settings, these findings reinforce the importance of considering children's language experiences across both home and school contexts when assessing bilingual development. Clinicians should be cautious in interpreting slower vocabulary growth in one language, particularly expressive vocabulary, as indicative of a delay or disorder if the child demonstrates strengths in other domains or in the other language (Bedore & Peña, 2008). Language evaluations should also include multiple domains and account for the child's unique language experience at home and in school. Intervention planning should also consider strategies that promote the use and development of the child's HL.

Limitations and Future Directions

Several limitations of the current study must be acknowledged. First, there were constraints in how language experience was measured. While we collected child and teacher reports, parent report data were not included in the study. Although we initially sought to collect parent report data through phone interviews, extensive missing data ultimately prevented its meaningful inclusion in our analysis. Relying exclusively on children's self-reports introduces reliability challenges given their potentially limited ability to recall language experiences from birth. Future research would benefit from incorporating parent report data to provide a more comprehensive picture of children's language experiences.

Second, language experiences evolve over time, but our study lacked cumulative longitudinal data regarding children's language experiences from birth to the time of testing. Such cumulative data would yield deeper insights into bilingual language development trajectories.

Third, our measures emphasized the frequency rather than the quality of language use. Prior research (e.g., Paradis, 2023; Song et al., 2024) has emphasized that bilingual children's language development is shaped not only by how much each language is used but also by who is using it and in what contexts. Future studies should adopt more nuanced approaches to capturing language experience, including the quality of interactions, the types of communicative activities, and the language proficiency of interlocutors.

Fourth, we only included child-reported measures of relative English use. This decision was based on efforts to keep the child interview brief and accessible for younger participants. However, we acknowledge that omitting child-reported Spanish use limits our understanding of the full picture of bilingual experience. Future work should explore ways to capture both English and Spanish experiences from children in a developmentally appropriate and engaging manner. Additionally, our reliance on self-report introduces potential recall biases. Future work should incorporate more objective measures of language experience, such as direct observation of home and school language environments. For example, researchers can conduct direct observations to evaluate classroom quality, such as using the Classroom Assessment Scoring System (Pianta et al., 2008). Technologies such as the Language Environment Analysis (LENA) system (LENA Foundation, 2017) can also offer valuable, ecologically valid data to supplement and triangulate self-reports.

Our language measures were also limited to standardized or researcher-developed assessments administered in structured contexts. These types of measures may not fully capture EB children's language use or pragmatic abilities in naturalistic settings. This is particularly relevant for Spanish expressive vocabulary, which showed limited growth over a 2-year period and may have been underestimated due to test format or contextual constraints.

Finally, while the study controlled for Time 1 baseline language skills and child sex, it did not account for other potentially influential factors that shape bilingual development. Variables such as children's motivation and aptitude for language learning (Sun et al., 2016, 2022), as well as teachers' and parents' bilingual proficiency (Choi & Shen, 2025), may moderate the relation between language experiences and language developmental outcomes. The exclusion of these variables limits our ability to fully explain individual differences in growth trajectories. Future research should incorporate such individual and contextual characteristics to better understand the mechanisms through which language experiences shape bilingual development.

Author Contributions

Becky H. Huang: Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing. Ye Shen: Conceptualization, Methodology, Formal analysis, Writing – original draft, Writing – review & editing.

Ethics Statement

This study was approved by The University of Texas at San Antonio Institutional Review Board (IRB No. 19-075). All participants provided written informed consent before participating in the study.

Data Availability Statement

The data for the study are available from the corresponding author, Becky H. Huang, upon reasonable request.

Artificial Intelligence Statement

ChatGPT was used during the writing process to enhance clarity and readability. The authors carefully reviewed and edited the content and take full responsibility for the final version of the article.

Appendix A

Child Survey

• 1) Your Name: ________________________________

• 2) Your Date of Birth (mm/dd/yyyy): _____________________________

• 3) Which of the following best describes you? Check one or more.

  Hispanic or Latino _____ White _____ Asian or Pacific Islanders _____

  Black or African American _____American Indian or Alaska Native _____Middle Eastern _____

• 4) What language(s) do you speak at home? ____________________________________________

• 5) A. Other than English, what language(s) do you speak (i.e., can carry on a conversation in that language)? ________________________________________

       B. How old were you when you began speaking these languages? ________________________________________

• 6) Have you ever been in an English as a Second Language (ESL) class? ___Yes, when I was in __ grade. ___ No

• 7) A. Were you born in the United States? ___ Yes ___ No

       B. If you were not born in the U.S., at what age did you move to the U.S.? When I was ________ years old.

8) For each question below, please tell us how frequently you use English by circling the number:
a. What language do you use at breakfast in the morning?
1	2	3	4	5
No English	A little English	Some English	Mostly English	Only English
b. What language do you use after you get home from school?
1	2	3	4	5
No English	A little English	Some English	Mostly English	Only English
c. What language do you use at dinner in the evening?
1	2	3	4	5
No English	A little English	Some English	Mostly English	Only English
d. What language do you use when you ask for help with homework at home?
1	2	3	4	5
No English	A little English	Some English	Mostly English	Only English
e. What language do you use with your teachers and classmates in school?
1	2	3	4	5
No English	A little English	Some English	Mostly English	Only English
f. What language do you use on the weekend?
1	2	3	4	5
No English	A little English	Some English	Mostly English	Only English
g. I watch movies, TV, and listen to Music in:
1	2	3	4	5
No English	A little English	Some English	Mostly English	Only English
For questions 9–10, we are interested in learning about your reading activities in ENGLISH outside of school.

• 9) Each day, about how much time do you usually spend reading in ENGLISH outside of school? Mark one box with a check mark (√).

  • ❒  I do not read ENGLISH for enjoyment

  • ❒  30 minutes or less each day

  • ❒  More than 30 minutes to less than 60 minutes each day

  • ❒  1 to 2 hours each day

  • ❒  More than 2 hours each day

• 10) How often do you read these materials in ENGLISH outside of school? Mark one box with a check mark (√).

	Never or almost never	Once or twice a month	Once or twice a week	Every day or almost every day
A.  Magazines
B.  Comic books or graphic novels
C.  Fiction books (such as stories, novels, or chapter books)
D.  Nonfiction books (books that explain things or give information, such as history, biographies or books about animals)
E.  Emails, web pages, social media sites
For questions 11–12, we are interested in learning about your reading activities in SPANISH outside of school.

• 11) Each day, about how much time do you usually spend reading in SPANISH outside of school? Mark one box with a check mark (√).

  •   ❒ I do not read SPANISH for enjoyment

  •   ❒ 30 minutes or less each day

  •   ❒ More than 30 minutes to less than 60 minutes each day

  •   ❒ 1 to 2 hours each day

  •   ❒ More than 2 hours each day

• 12) How often do you read these materials in SPANISH outside of school? Mark one box with a check mark (√).

	Never or almost never	Once or twice a month	Once or twice a week	Everyday or almost every day
A.  Magazines
B.  Comic books or graphic novels
C.  Fiction books (such as stories, novels, or chapter books)
D.  Nonfiction books (books that explain things or give information, such as history, biographies or books about animals)
E.  Emails, web pages, social media sites

Appendix B

Teacher Survey

Today's date: ____________________________ School: _______________________________________________________________

Your name: ___________________________ The grade level you teach: _________ 1^st grade __________ 3^rd grade___________

Instruction: We are interested in what language(s) you use for reading instruction (questions 1–2), and how much the child uses and hears Spanish and English in the classroom and the contexts in which each is used (questions 3–4). Please complete questions below.

#1. What reading program do you use? _________________________

#2. I provide reading instruction in: ______ Spanish ______ English ______ both Spanish and English

#3: For each hour, please document the activity and indicate using a “x” whether the students in your class hear Spanish, English or both during that hour, and what language the students speak/use during that hour.

Time	Activity (e.g., Reading, math, lunch). If not applicable, please leave it blank.	What do your students hear?			What do (most of) your students speak/use?
		Spanish	Both	English	Spanish	Both	English

#4: For each of the students listed below, please fill in the number for both Spanish and English in each context.

	Speaks with you, the teacher, in Spanish	Speaks with other children, in Spanish	Speaks with other adults (e.g., classroom assistant), in Spanish	Speaks with you, the teacher, in English	Speaks with other children, in English	Speaks with other adults (e.g., classroom assistant), in English
	0 = Never 1 = Rarely 2 = Sometimes 3 = Very often 4 = Usually 5 = Always 6 = Don't know	0 = Never 1 = Rarely 2 = Sometimes 3 = Very often 4 = Usually 5 = Always 6 = Don't know	0 = Never 1 = Rarely 2 = Sometimes 3 = Very often 4 = Usually 5 = Always 6 = Don't know	0 = Never 1 = Rarely 2 = Sometimes 3 = Very often 4 = Usually 5 = Always 6 = Don't know	0 = Never 1 = Rarely 2 = Sometimes 3 = Very often 4 = Usually 5 = Always 6 = Don't know	0 = Never 1 = Rarely 2 = Sometimes 3 = Very often 4 = Usually 5 = Always 6 = Don't know
Example: John Doe	2	2	1	5	5	5

Funding Statement

The study was supported by the National Institutes of Health (NIH) under Award Number SC2HD100362 (awarded to Becky H. Huang). The content is solely the responsibility of the authors and does not necessarily reflect the official views of the NIH.