Dual language supports for dual language learners? Exploring preschool classroom instructional supports for DLLs' early learning outcomes

Classified as probably not a substantial claim.

Preschool education is beneficial for DLLs.

“Although evidence suggests that preschool education is beneficial for DLLs”

• Also stated in: § Background: DLLs in preschool · boosted

  “Encouragingly, participation in publicly-funded preschool, including Head Start and public pre-k, has been shown to be particularly advantageous for DLLs’ school readiness (e.g., Bloom & Weiland, 2015; Cooper & Lanza, 2014; Gormley, 2008; McCoy, Morris, Connors, Gomez & Yoshikawa, 2016; Phillips et al., 2017).”

• cites Bloom & Weiland, 2015Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Cooper & Lanza, 2014 sourceSupported
• cites Gormley, 2008 sourceNot checked — no full text of the source

May combine two claims.

Both Head Start and public pre-K led to improvements in DLLs' early learning outcomes, including English literacy and quantitative reasoning skills.

“studies have found that both Head Start (Cooper & Lanza, 2014) and public pre-K (Gormley, 2008) led to improvements in DLLs’ early learning outcomes, including English literacy and quantitative reasoning skills.”

• cites Cooper & Lanza, 2014 sourceSupported
• cites Gormley, 2008 sourceNot checked — no full text of the source

May combine two claims.

Home language supports may foster DLLs' bilingual development, which in turn may strengthen executive functions important for quantitative reasoning and conceptual development.

“Specifically, it follows from the documented connection between bilingualism and executive function (e.g., Bialystok, 2015; 2017) that supports for home language would support DLLs’ developing bilingualism and, in turn, support the executive functioning skills that are critical for quantitative reasoning and conceptual development (e.g., Clark, Pritchard, & Woodward, 2010).”

• cites Bialystok, 2015 sourceSupported
• cites Bialystok, 2017 sourceSupported
• cites Clark, Pritchard, & Woodward, 2010 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Just one teacher who answered the survey said they were fluent in Spanish.

“Only one survey respondent reported speaking fluent Spanish.”

• AssumptionRests on an assumption: The 92% of teachers who completed the survey are representative of all teachers in the analytic sample.

May combine two claims.

In this sample, CASEBA and CLASS scores were generally low, but CLASS scores matched national estimates for Head Start programs that year.

“Scores on the CASEBA and CLASS were generally low but, at least for CLASS which has comparable national data, on par with other estimates (e.g., for Head Start programs in the same year; Office of Head Start OHS, 2017).”

• cites Office of Head Start OHS, 2017Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Every item in the home language support subscale averaged between 1 and 2, showing little evidence that teachers used these practices.

“Descriptive statistics for individual CASEBA items (see Table 2) reveal that, for all items that compose the supports for home language subscale, the average score was between 1 and 2, indicating minimal evidence that teachers are providing these supports.”

• AssumptionRests on an assumption: A single day of CASEBA observation adequately represents the classroom's typical DLL-specific instructional supports.

May combine two claims.

English acquisition support items scored higher overall, from 2 to 5, suggesting these practices were provided with minimal to good evidence.

“Individual items in the supports for English acquisition subscale were generally higher, ranging from 2 to 5, indicating minimal to good evidence that these supports are provided.”

• AssumptionRests on an assumption: A single day of CASEBA observation adequately represents the classroom's typical DLL-specific instructional supports.

May combine two claims.

T-tests indicated significant gains (P < .01) from fall to spring for English language, quantitative reasoning, and Spanish basic concepts.

“T-tests comparing the mean fall scores to mean spring scores showed that these improvements were statistically significant (P < .01) for English language and quantitative reasoning outcomes and Spanish basic concepts”

May combine two claims.

Home language supports did not predict English literacy scores or English expressive vocabulary scores.

“Supports for home language were not predictive of English literacy or expressive vocabulary scores.”

May combine two claims.

On average, Spanish expressive vocabulary grew from fall to spring, but the growth was not statistically significant.

“This finding emerged in the context of minimal growth from fall to spring in Spanish expressive vocabulary – although the average growth from fall to spring was positive (albeit non-significant), Spanish expressive vocabulary scores actually decreased from fall to spring for a third of the sample.”

May combine two claims.

Classroom instructional supports for DLLs' home language may be important not only for maintaining children's home language but also for acquiring foundational pre-math skills.

“This study adds to an emerging body of research suggesting classroom instructional supports for DLLs’ home language may be important not just for the maintenance of children’s home language, but also for the acquisition of foundational pre-math skills.”

Classified as probably not a substantial claim.

Little is known about which instructional supports in the preschool classroom best promote DLLs' early learning skills.

“little is known about which instructional supports in the preschool classroom best promote DLLs’ early learning skills.”

May combine two claims.

The gap in English skills at school entry can have serious implications for DLLs because the U.S. education system typically uses English as the language of instruction and may struggle to meet the needs of the growing and diverse DLL population.

“This gap in English skills at school entry can have serious implications for DLLs, considering that the U.S. education system typically adopts English as the language of instruction and thus may struggle to meet the needs of the growing and diverse DLL population.”

Probably not a substantial claim: may not pass S1 (asserted), S2 (checkable). May combine two claims.

Given the mixed-delivery system of public preschool available to many DLLs, further research would benefit from considering the full landscape of public preschool settings by including both public pre-k and Head Start classrooms in one study, in addition to considering the implications of different classroom contexts where DLLs may or may not be the minority.

“Additionally, each paper focused exclusively on a Head Start or state-funded public pre-K classrooms. Given the mixed-delivery system of public preschool available to many DLLs, further research would benefit from considering the full landscape of public preschool settings by including both public pre-k and Head Start classrooms in one study, in addition to considering the implications of”

May combine two claims.

Research has linked classroom instructional support as measured by the CLASS to early learning outcomes among preschool children in general, although associations are typically small and occasionally null.

“Among preschool children in general, research has linked classroom instructional support as measured by the CLASS to early learning outcomes, although associations are typically small (e.g., Burchinal, 2018; Burchinal et al., 2011; Weiland & Yoshikawa, 2013) and occasionally even null (e.g., Pianta et al., 2017).”

• cites Burchinal, 2018 sourceNot checked — no full text of the source
• cites Burchinal et al., 2011Not checked — no full text of the source
• cites Weiland & Yoshikawa, 2013 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable).

The search for a broader set of DLL-specific instructional supports continues.

“Yet given the LISn’s exclusive focus on language interaction, the search for a broader set of DLL-specific instructional supports continues.”

Probably not a substantial claim: may not pass S2 (checkable).

The unique context of prior studies - centers in regions where the majority of students are DLLs with bilingual teachers - highlights a need for additional research to examine how DLL-specific instructional supports manifest in the more common preschool context where DLLs are the minority and teachers mostly speak only English.

“Although these results provide some preliminary evidence that preschool classroom supports for DLLs’ home language associate with their early learning skills, the unique context of these studies – centers in regions where the majority of students are DLLs with bilingual teachers – highlights a need for additional research to examine how these DLL-specific instructional supports manifest in the”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Only recently have studies begun to look inside the classroom to identify promising, malleable features of public preschool that may catalyze early skill gains.

“Yet despite this evidence, only recently have studies begun to look inside the classroom to identify promising, malleable features of public preschool that may catalyze these early skill gains.”

May combine two claims.

In 2011-2013, just 17% of preschool teachers spoke a language other than English, with 10% speaking Spanish.

“Indeed, in 2011–2013, just 17% of preschool teachers spoke a language other than English, with 10% speaking Spanish (Park, McHugh, Zong, & Batalova, 2015).”

• cites Park, McHugh, Zong, & Batalova, 2015Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

Castro and colleagues' inventory of instructional supports with promise for DLLs falls into three categories: conventional elements of instructional support important for all children, DLL-specific supports for home language development, and DLL-specific supports for English acquisition.

“We draw on Castro and colleagues’ inventory of instructional supports with promise for DLLs, which fall into three categories: (a) conventional elements of instructional support that are important for DLLs and all children; (b) DLL-specific instructional supports for home language development; and (c) DLL-specific instructional supports for English acquisition.”

• cites Castro et al. (2011)Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Both studies used a DLL-specific instructional support measure, the Classroom Assessment of Supports for Emergent Bilingual Acquisition (CASEBA), and focused on samples of Spanish-speaking DLLs in centers with high proportions of DLLs and majority Spanish-speaking teachers.

“Both studies used a DLL-specific instructional support measure, the Classroom Assessment of Supports for Emergent Bilingual Acquisition (CASEBA), and focused on samples of Spanish-speaking DLLs in centers with high proportions of DLLs and majority Spanish-speaking teachers.”

Classified as probably not a substantial claim.

This body of literature has produced mixed results.

“This body of literature has produced mixed results.”

May combine two claims.

Recent results linked complex teacher-child English language interactions captured by the LISn with children's English and Spanish bilingual receptive vocabulary but found no association between Spanish language use and children's outcomes.

“Recent results linked complex teacher-child English language interactions captured by the LISn with children’s English and Spanish bilingual receptive vocabulary but found no association between Spanish language use and children’s outcomes (Franco et al., 2019).”

• cites Franco et al., 2019 sourceNot checked — no full text of the source

May combine two claims.

White, Fernandez, & Greenfield (2019) found positive associations between classroom supports for home language and DLLs' home language outcomes, as well as for other early learning skills.

“However, they did find positive associations between classroom supports for home language and DLLs’ home language outcomes, as well as for other early learning skills (i.e., executive function and science readiness assessed in the child’s dominant language; White, Fernandez, & Greenfield, 2019).”

• cites White, Fernandez, & Greenfield, 2019Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

Researchers have begun to call for new measures that may better capture DLL-specific classroom instructional supports.

“In part due to the small and relatively weak research base linking CLASS instructional support to DLLs’ early learning outcomes, researchers have begun to call for new measures that may better capture DLL-specific classroom instructional supports (Burchinal, 2018; Burchinal et al., 2011; Castro, Espinosa & Páez, 2011; Espinosa, 2020; National Academies of Science, Engineering & Medicine, 2017).”

• cites Burchinal, 2018 sourceNot checked — no full text of the source
• cites Burchinal et al., 2011Not checked — no full text of the source
• cites Castro, Espinosa & Páez, 2011Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim. May combine two claims.

Instructional features that support DLLs' English acquisition include having teachers who are trained in instructional practices to support English language development for DLLs, who have knowledge of children's developing English language backgrounds, are able to adapt English instruction according to children's skills, provide explicit instruction about the English language to facilitate

“Instructional features that support DLLs’ English acquisition include having teachers who are trained in instructional practices to support English language development for DLLs, who have knowledge of children’s developing English language backgrounds, are able to adapt English instruction according to children’s skills, provide explicit instruction about the English language to facilitate”

• cites Castro et al., 2011Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S1 (asserted), S2 (checkable).

Emerging research on preschool classroom instructional supports for DLLs should consider whether and to what extent the benefits of these supports depend on the proportion of DLLs in the classroom.

“Acknowledging that DLLs’ classroom contexts are quite diverse, emerging research on preschool classroom instructional supports for DLLs should consider whether and to what extent the benefits of these supports depend on the proportion of DLLs in the classroom.”

Classified as probably not a substantial claim.

Associations between classroom supports and DLLs' skill development may be moderated by the proportion of their peers who are also DLLs.

“Thus, associations between classroom supports and DLLs’ skill development may be moderated by the proportion of their peers who are also DLLs.”

Classified as probably not a substantial claim. May combine two claims.

According to Castro and colleagues, instructional features that support DLLs' home language development include having teachers who speak the child's home language, are trained in instructional practices to support children's development in their home language, and have knowledge of the child's language background.

“According to Castro and colleagues, instructional features that support DLLs’ home language development include having teachers who speak the child’s home language, are trained in instructional practices to support children’s development in their home language, and have knowledge of the child’s language background (Castro et al., 2011).”

• cites Castro et al., 2011Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable).

To better understand how to support DLLs' early learning skills when teachers who speak Spanish or other non-English languages are unavailable, researchers must look beyond the single measure of using home language for instruction.

“To better understand how to support DLLs’ early learning skills when teachers who speak Spanish (or other non-English languages) are unavailable, researchers must look beyond the single measure of using home language for instruction (Castro et al., 2011).”

• cites Castro et al., 2011Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

The CLASS was not developed to measure supports that are uniquely relevant to DLLs' diverse language backgrounds.

“Perhaps this is not surprising, given that the CLASS was not developed to measure supports that are uniquely relevant to DLLs’ diverse language backgrounds.”

May combine two claims.

Existing research on peer effects on language development suggests that for DLLs, peer scaffolding and more frequent language interactions play important roles in supporting DLLs' language growth.

“More specifically, existing research on peer effects on language development suggests that for DLLs, peer scaffolding that occurs when students collaborate and support one another’s skills (Cole, 2014) and more frequent language interactions wherein children may be exposed to higher quality language and diverse vocabulary via their peers (Gámez, Griskell, Sobrevilla & Vazquez, 2019) play important”

• cites Cole, 2014 sourceNot checked — no full text of the source
• cites Gámez, Griskell, Sobrevilla & Vazquez, 2019 sourceNot checked — no full text of the source

May combine two claims.

Figueras-Daniel and Li (2021) found no association between supports for English acquisition and DLLs' English or Spanish receptive vocabulary, nor between supports for home language and Spanish receptive vocabulary.

“In their sample of state-funded pre-K classrooms, Figueras-Daniel and Li (2021) found no association between supports for English acquisition and DLLs’ English or Spanish receptive vocabulary, nor between supports for home language and Spanish receptive vocabulary.”

• cites Figueras-Daniel and Li (2021) sourceNot checked — no full text of the source

May combine two claims.

Downer et al. (2012) found that CLASS instructional support was equally as predictive of early literacy and quantitative reasoning skills for DLLs as for non-DLLs, although associations for both groups were modest.

“Downer et al. (2012) found that CLASS instructional support was equally as predictive of early literacy and quantitative reasoning skills (administered in the child’s preferred language) for DLLs as for non-DLLs, although associations for both groups were modest.”

• cites Downer et al. (2012) sourceNot checked — no full text of the source

Classified as probably not a substantial claim.

Castro et al. (2011) identified elements that may be especially important in promoting DLLs' early learning, according to evidence from language development research and literature on classroom quality.

“In an effort to provide a roadmap toward promising DLL-specific measures of preschool classroom instructional supports, Castro et al. (2011) identified elements that may be especially important in promoting DLLs’ early learning, according to evidence from language development research and literature on classroom quality.”

• cites Castro et al. (2011)Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

The first category, conventional elements of instructional support, includes quantity and quality of language interactions and small-group and individualized instructional opportunities, both of which are captured by the CLASS instructional support subscale.

“The first category, conventional elements of instructional support, includes quantity and quality of language interactions and small-group and individualized instructional opportunities, both of which are captured by the CLASS instructional support subscale (Pianta et al., 2008).”

• cites Pianta et al., 2008Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Raikes et al. (2019) found that the use of Spanish in early childhood classrooms was beneficial for the Spanish language skills of DLLs but found no association with English language skills.

“For instance, Raikes et al. (2019) found that the use of Spanish in early childhood classrooms was beneficial for the Spanish language skills of DLLs but found no association – positive nor negative – with English language skills.”

• cites Raikes et al. (2019) sourceNot checked — no full text of the source

May combine two claims.

Together, these two studies suggest that some level of home language for instruction is beneficial for DLLs' English development, but to prepare DLLs for a K-12 education system with English as the main language of instruction, teachers may need to provide targeted supports for English acquisition as well.

“Together, these two studies suggest that some level of home language for instruction is beneficial for DLLs’ English development, but to prepare DLLs for a K-12 education system with English as the main (and usually only) language of instruction, teachers may need to provide targeted supports for English acquisition as well.”

Classified as probably not a substantial claim. May combine two claims.

These mesosystem contextual factors enable teachers to incorporate the child's home language into the classroom environment, support the development of the child's home language, administer appropriate and holistic assessments in the home language, and provide opportunities for DLLs to communicate in their home language.

“These mesosystem contextual factors, linking children’s home experiences with their preschool classroom environments (Bronfenbrenner & Morris, 2006), enable teachers to incorporate the child’s home language into the classroom environment, support the development of the child’s home language, administer appropriate and holistic assessments in the home language, and provide opportunities for DLLs to”

• cites Bronfenbrenner & Morris, 2006Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Castro et al., 2011Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

The Spanish expressive vocabulary subtest had an internal consistency of α = 0.85.

“Spanish expressive vocabulary was assessed using the CELF-Spanish Vocabulario Expresivo (Expressive Vocabulary) subtest (α = 0.85)”

May combine two claims.

Parental English and Spanish skills were not controlled, though they may shape home language exposure and relate to preschool selection.

“Further, we were unable to control for the Spanish or English language skills of parents, which could contribute to the quality of language exposure at home and may also correlate with selection into certain preschool classrooms.”

Classified as probably not a substantial claim.

Including the fall outcome in models yields estimates of preschool-year effects net of initial skill levels.

“The inclusion of the lagged outcome in models predicting spring skills produces estimates of preschool year experiences, net of where children started in the fall of their preschool year.”

Classified as probably not a substantial claim.

The CLASS instructional support subscale measures teacher feedback quality, concept development techniques, and modeling of language and higher-order thinking.

“The CLASS uses direct observation to assess three domains of classroom supports: emotional support, classroom organization, and instructional support. The present study used the instructional support subscale to assess the quality of teachers’ feedback to students, techniques used to develop learning concepts, and modeling of language and higher-order thinking skills (α = 0.91).”

• cites Pianta et al., 2008Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

Teacher Spanish proficiency was measured only by whether the lead teacher was fluent, not in detail.

“Additionally, we did not have a detailed measure of teacher’s Spanish language skills – just an indicator of whether or not the lead teacher was fluent in Spanish.”

Classified as probably not a substantial claim.

To prevent non-random missingness, children failing the CELF-P2 screener received a score two SD below the sample mean, or zero if that was below the possible range.

“To capture this and avoid non-random missingness in child assessments, children who failed the screener were assigned a score of two standard deviations below the mean of the entire study sample (DLLs and non-DLLs). If two standard deviations below the mean was a number less than zero (i.e., outside the possible range of scores), a score of zero was assigned instead.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

About 77% of the analytic sample attended TPS pre-k, close to the 78% in the full sample.

“Approximately 77% of children in the analytic sample were in TPS pre-k, which is similar to the distribution in the overall sample (78%).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Imputation was limited to children with complete classroom observation and spring outcome data.

“Values were only imputed for participants with full classroom observation and spring child outcome data.”

May combine two claims.

With higher classroom DLL proportions, the positive link between home language support and quantitative reasoning weakens, and at 70% DLLs or more the effect is no longer distinguishable from zero.

“As shown in Fig. 1, the negative coefficient on the interaction term suggests that as the proportion of DLLs in the classroom increases, home language support becomes less positively predictive of quantitative reasoning, to the point where the effect becomes indistinguishable from zero when the classroom has 70% or more DLLs.”

Probably not a substantial claim: may not pass S3 (load-bearing).

After observations, anchor observers checked every score sheet to confirm the evidence supported the assigned scores.

“For data collection, these anchor observers reviewed each score-sheet post-observation to verify that observers had provided sufficient evidence for the scores assigned.”

Classified as probably not a substantial claim.

CASEBA was created to assess preschool classroom supports for DLLs' cognitive, social-emotional, and language development.

“The CASEBA was designed to evaluate preschool classroom supports for the development of DLLs’ cognitive, socio-emotional, and language skills (Freedson, Figueras-Daniel, Frede, Jung & Sideris, 2009).”

• cites Freedson, Figueras-Daniel, Frede, Jung & Sideris, 2009Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Home language support from teachers may help DLLs build more complex language understanding, improving their quantitative reasoning performance.

“Thus, teacher’s support for home language may allow DLLs to develop more complex linguistic understanding that leads them to perform better on the quantitative reasoning assessment, since they can better comprehend the word problems.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The quantitative reasoning subtest had an internal consistency of α = 0.93.

“Quantitative reasoning was assessed in English using the Woodcock Johnson Applied Problems subtest (α = 0.93)”

Probably not a substantial claim: may not pass S3 (load-bearing).

The English expressive vocabulary subtest had an internal consistency of α = 0.84.

“Children’s English expressive vocabulary was assessed used the CELF-P2 English Expressive Vocabulary subtest (α = 0.84)”

May combine two claims.

The CASEBA is new, and its home language and English acquisition support scales have not been extensively validated.

“Another limitation of the present study is that the CASEBA is a relatively new measure. While it represents a promising attempt to address DLL-specific instructional supports (see Figueras-Daniel & Li, 2021; Freedson et al., 2009; White, Fernandez, & Greenfield, 2019), its measures of supports for home language and supports for English acquisition have not been extensively studied and validated.”

• cites Figueras-Daniel & Li, 2021 sourceNot checked — no full text of the source
• cites Freedson et al., 2009Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites White, Fernandez, & Greenfield, 2019Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

English acquisition support may matter less for DLLs' early learning than home language support.

“One possible explanation is that support for English acquisition may simply be less important for DLLs’ early learning than supports for home language.”

Probably not a substantial claim: may not pass S3 (load-bearing).

One possible explanation for the home language support-quantitative reasoning link is that students feel more supported and enjoy school more when teachers support their home language.

“One reason for this finding may be the supportiveness of the classroom environment; that is, a student may feel more supported and like school more in classrooms where the teacher supports their home language, such as through providing materials in the child’s home language, which was the most common home language support provided in our sample (see Table 2).”

Classified as probably not a substantial claim. May combine two claims.

Given low home language support and few home-language-speaking teachers, basic Spanish concepts may have been promoted less than expressive vocabulary, which could be taught through Spanish books and props.

“Thus, it is possible that in a context with so little support for home language development and so few teachers who speak the home language, basic Spanish concepts were not promoted as much as expressive vocabulary, which could be explicitly taught though the provision of books and props in Spanish.”

Classified as probably not a substantial claim.

Because of restricted range, CELF-P2 raw scores were used along with child age as a control.

“Due to restricted range owing to CELF-P2 scaled scores in-sufficiently accounting for language diversity, raw scores were used in conjunction with a control for child age.”

May combine two claims.

In classrooms with 80-90% DLLs, home language supports predicted higher Spanish expressive vocabulary, but in most classrooms the effect could not be distinguished from zero.

“This suggests that for classrooms where the vast majority of students – 80%–90% – were DLLs, home language supports predicted greater Spanish expressive vocabulary; however, the effect was indistinguishable from zero in most classrooms (Fig. 2).”

Probably not a substantial claim: may not pass S3 (load-bearing).

The CLASS instructional support subscale had an internal consistency of α = 0.91.

“The present study used the instructional support subscale to assess the quality of teachers’ feedback to students, techniques used to develop learning concepts, and modeling of language and higher-order thinking skills (α = 0.91).”

May combine two claims.

When dual immersion classrooms were controlled, the home language support-quantitative reasoning link weakened, suggesting that DLL-focused classroom environments may support early learning beyond language.

“Further, our sensitivity tests where we controlled for dual immersion classrooms weakened the relationship between home language supports and quantitative reasoning, suggesting that classroom environments that are focused on dual language learners may provide important support for their early learning skills in areas beyond language.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Initial certification showed two anchor observers reached 90% agreement with a trainer across three classroom visits.

“During initial certification, two gold standard anchor observers attained 90% overall agreement across three classroom visits with a trainer from the tool development team.”

Classified as probably not a substantial claim.

The extent of Spanish use in non-immersion classrooms is uncertain.

“While we can reasonably assume the classrooms in immersion schools provided some instruction in Spanish, we cannot be certain whether and to what extent Spanish was used in other classrooms in non-immersion schools.”

• AssumptionThe authors state this as an assumption: Classrooms in dual language immersion schools provided some instruction in Spanish.

The quote only partly states this claim.

The supports for home language scale, averaging 10 items, had an internal consistency of α = 0.83.

“The supports for home language scale includes the average of 10 items that capture the extent to which teachers provide supports for children’s home language, such as using the home language for instruction, using high quality talk in the home language, providing “rich” read-aloud experiences in the home language, providing books, print, and literacy props in the home language, and supporting home”

Classified as probably not a substantial claim.

The null findings for English acquisition supports may reflect an internal validity problem with CASEBA rather than a genuine absence of effect.

“This raises the possibility that our lack of association between supports for English acquisition and English language and quantitative reasoning outcomes represents an internal validity issue with the measure rather than a true null finding.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Prior research has reported no links between CLASS instructional support and early learning outcomes for DLLs.

“Finally, although not a key predictor of interested in the current study, it is worth noting that we also did not find any association between the traditional, general measure of instructional support as assessed using the CLASS; this finding is also in accord with recent research that found no associations between CLASS instructional support and early learning outcomes among DLLs (White, Fern”

• cites White, Fernandez, & Greenfield, 2019Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

Classroom observation data from CASEBA and CLASS were not imputed.

“No classroom observation data (CASEBA or CLASS) were imputed.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The English literacy subtest had an internal consistency of α = 0.94.

“Children’s English literacy skills were assessed using the Woodcock-Johnson Letter-Word Identification subtest (α = 0.94)”

Probably not a substantial claim: may not pass S3 (load-bearing).

The Spanish basic concepts subtest had an internal consistency of α = 0.79.

“Spanish basic concepts were assessed using the CELF-P2 Spanish Conceptos Básicos (Basic Concepts) subtest (α = 0.79)”

Classified as probably not a substantial claim. May combine two claims.

Quantitative reasoning draws on multiple early skills, including language and executive function, which together may be best supported by home language supports.

“This could be particularly relevant for our measure of quantitative reasoning, which relies on a host of early learning skills, including language (e.g., Foster et al., 2019; Goodrich & Namkung, 2019; Méndez et al., 2019) and executive functioning (e.g., Clark, Pritchard, & Woodward, 2010), which, in aggregate, may be best supported by supports for the home language.”

• cites Foster et al., 2019 sourceNot checked — no full text of the source
• cites Goodrich & Namkung, 2019 sourceNot checked — no full text of the source
• cites Méndez et al., 2019 sourceNot checked — no full text of the source

May combine two claims.

With only four dual language immersion classrooms, deeper analysis was impossible and the sensitivity test was underpowered.

“However, the small number of dual language immersion classrooms in our sample (n = 4) prohibits more in-depth analysis and underpowers our sensitivity test.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The observed home language support-outcome association could be an artifact of parental selection, language of instruction, or both if engaged parents choose dual language schools with more home language supports.

“If particularly involved or engaged parents select schools with dual language instruction, and those schools are more likely to have classrooms with more supports for home language for instance, then perhaps the association we observed between home language supports and outcomes was an artifact of parental selection, language of instruction, or both.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Missing covariate data were handled by multiple imputation, creating 25 datasets.

“Covariate data were multiply imputed to address item-level non-response.”

• Also stated in: § Analytic strategy · plain

  “Multiple imputation in Stata 16 relied on the ice command to create 25 imputed data sets”

• AssumptionRests on an assumption: The missing covariate and fall assessment data are missing at random, so multiple imputation yields unbiased estimates.

Classified as probably not a substantial claim.

The English Woodcock-Johnson Applied Problems measure of quantitative reasoning may be a more sensitive gauge of DLLs' spoken English understanding because it requires processing abstract numerical information while language is still developing.

“Considering the sensitivity of measurement, it is possible that because quantitative reasoning was assessed through the Woodcock-Johnson Applied Problems in English – which is an area of development for DLLs – the demand in processing abstract numerical information in light of a developing language may offer a more sensitive estimate of DLLs’ general understanding of spoken English.”

Probably not a substantial claim: may not pass S3 (load-bearing).

A second trained observer reviewed all scores to confirm support and prevent coding drift.

“Additionally, all scores were checked by another trained observer to ensure sufficient support for scores and protect against coding drift.”

Classified as probably not a substantial claim.

Every model adjusted for CLASS instructional support to address general teaching quality as a confounder.

“We also control for the CLASS instructional support scale to account for general instructional support that could confound our associations between DLL-specific instructional support and outcomes.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Disagreements were resolved by consensus between observers and the anchor.

“When disagreements occurred, observers were required to meet with the anchor to come to consensus on the score.”