Reclassification: Disparities, contributors, and outcomes

Federal reporting and studies alike have found disparities in the reclassification rates between ELs with and without disabilities (Kieffer & Parker, Reference Kieffer and Parker2016; Umansky et al., Reference Umansky, Thompson and Díaz2017; U.S. Office of English Language Acquisition, 2021). At a national level, ELs with disabilities were only 0.8% of the 14.2% of ELs reclassified in 2018 (U.S. Office of English Language Acquisition, 2021). These federal data comport with prior findings: ELs with disabilities are disproportionately less likely to be reclassified as FEP (Burke et al., Reference Burke, Morita-Mullaney and Singh2016; Slama et al., Reference Slama, Molefe, Gerdeman, Herrera, Brodziak de los Reyes, August and Cavazos2017; Umansky et al., Reference Umansky, Thompson and Díaz2017). As demonstrated in Kieffer and Parker’s (Reference Kieffer and Parker2016) reclassification study of New York City Public Schools, ELs with disabilities generally required an additional 4 years to meet reclassification criteria, whereas ELs with SLIs were reclassified after 6 years and ELs with SLDs were reclassified after 8 years. With so few reaching reclassification, ELs with disabilities effectively become “stuck” in language services—a phenomenon Umansky and colleagues (Reference Umansky, Thompson and Díaz2017) call a reclassification bottleneck.

As a result of these reclassification disparities, a disproportionate percentage of ELs with disabilities become long-term English learners (LTELs), a term for ELs who receive language services for approximately 5 or more years without being reclassified as FEP (Brooks, Reference Brooks2018; Burke et al., Reference Burke, Morita-Mullaney and Singh2016; Kieffer & Parker, Reference Kieffer and Parker2016; Sahakyan & Ryan, Reference Sahakyan and Ryan2018). In Sahakyan and Ryan’s (Reference Sahakyan and Ryan2018) study of 15 states, almost half of all ELs with disabilities became LTELs. As studies reported, those labeled as LTELs have limited access to high-quality content-area instruction, advanced courses, and ironically, to language services they need for ELP growth (Callahan & Shifrer, Reference Callahan and Shifrer2016; Dabach, Reference Dabach2014; Menken et al., Reference Menken, Kleyn and Chae2012; Thompson, Reference Thompson2015). In addition, LTELs often experience stigma and are subjected to deficit perspectives of their academic and linguistic abilities (Brooks, Reference Brooks2018; Flores, Kleyn & Menken, Reference Flores, Kleyn and Menken2015; Lee & Soland, Reference Lee and Soland2022). Collectively, these studies signal that limited opportunities to learn are concomitant with delayed reclassification for ELs. For this reason, research is beginning to examine the factors contributing to reclassification disparities among ELs with disabilities and their peers without disabilities.

One potential explanation for the low reclassification rates of ELs with disabilities is the reclassification policies themselves, specifically the criteria for FEP status. Meeting a cut score on a standardized ELP assessment, which is a foundational criterion across all states, is challenging for many ELs (Linquanti, Reference Linquanti2001; Robinson-Cimpian & Thompson, Reference Robinson-Cimpian and Thompson2016) but particularly for ELs with disabilities—many of whom have disabilities that have a language basis, such as SLDs in reading, SLIs, and intellectual disabilities (de Valenzuela, Pacheco & Shenoy, Reference de Valenzuela, Pacheco and Shenoy2022; Shenoy, de Valenzuela & Pacheco, Reference Shenoy, de Valenzuela and Pacheco2022; Umansky et al., Reference Umansky, Thompson and Díaz2017).

Standardized ELP assessments come in one of two forms: general and alternate ELP assessments. Most ELs with disabilities take a general ELP assessment, such as the WIDA ACCESS, ELPAC, ELPA, or TELPAS, among others. A small percentage of ELs with disabilities, such as those with significant cognitive disabilities, take the alternate ELP assessment used in their state (e.g., Alternate ACCESS, Alt ELPA, Alternate ELPAC). Scholars have raised concerns about the construct validity of general and alternate ELP tests (i.e., whether the tests measure what they purport to measure) and, relatedly, the validity of test score inferences (i.e., whether conclusions drawn from the test scores are accurate), questioning whether the tests conflate ELP and disabilities (de Valenzuela et al., Reference de Valenzuela, Pacheco and Shenoy2022; Schissel & Kangas, Reference Schissel and Kangas2018; Shenoy et al., Reference Shenoy, de Valenzuela and Pacheco2022). Shenoy et al.’s (Reference Shenoy, de Valenzuela and Pacheco2022) study highlights the complications arising from reliance on alternate ELP assessments in measuring ELP growth for ELs with significant cognitive disabilities: “By explicitly linking students’ level of language proficiency to their communication development, which is highly influenced by their disability, English language proficiency is thus conflated with disability” (p. 178). Moreover, commonly used accommodations for standardized ELP assessments often have a limited empirical basis and struggle to address the heterogeneity among ELs with disabilities (see Abedi, Reference Abedi2014; Randez & Cornell, Reference Randez and Cornell2023; Schissel & Kangas, Reference Schissel and Kangas2018). Taken together, this emergent body of scholarship suggests that an overreliance on standardized ELP assessments as the sole criterion for exiting may be a barrier to the reclassification of ELs with disabilities.

To mitigate such barriers within reclassification policies, some states have implemented what Kangas (Reference Kangas2024) calls exemption policies, which permit individual or population-wide exemptions for ELs with disabilities. Pennsylvania, the context of this study, is one state that has an active population-wide exemption policy, instituting separate reclassification criteria for all ELs with Individualized Education Programs (IEPs), legally required documents that outline a tailored educational plan for students with disabilities. Kangas and Schissel (Reference Kangas and Schissel2021) conducted an ethnographic study to examine the implementation of the state’s reclassification exemption policy at two middle schools in Pennsylvania. The results demonstrated that while the policy increased reclassification rates for ELs with disabilities, it created new inequities in the eyes of some educators: It discriminated against ELs with disabilities, setting the bar for reclassification too low and thereby essentially pushing them out of the language services they still needed.

Reclassification: Teacher decision-making

For reclassification of ELs in general, previous literature has primarily examined teachers’ decisions, problematizing the “data” they use. In state contexts in which teacher input is required for reclassification, studies report significant variation in the type and viability of the evidence used for reclassification decisions (Estrada & Wang, Reference Estrada and Wang2018; Hill, Weston & Hayes, Reference Hill, Weston and Hayes2014; Marvrogordato & White, Reference Marvrogordato and White2020). For example, in Hill et al.’s (Reference Hill, Weston and Hayes2014) analysis of reclassification in California, teachers considered several data sources, including students’ grades or grade point averages, behavior and motivation, discipline, attendance, and participation. These same studies indicated that teachers’ anecdotes about ELs were often used as “data” during reclassification (Estrada & Wang, Reference Estrada and Wang2018; Hill et al., Reference Hill, Weston and Hayes2014; Marvrogordato & White, Reference Marvrogordato and White2020). As an example, Marvrogordato and White (Reference Marvrogordato and White2020) found that when teacher input was a part of reclassification and was prioritized over other data, ELs were less likely to be reclassified.

The practices of the teachers in the aforementioned studies corroborate Vanlommel and Schildkamp’s (Reference Vanlommel and Schildkamp2019) broader findings of teachers’ high-stakes decision-making. Their analysis found that teachers operate as one of two types of decision-makers: process-oriented or intuitive. The former used predefined criteria for making interpretations about the data’s meaning, triangulated data sources, and considered alternative explanations. However, most teachers fall into the latter group, making decisions based on an intuitive judgment of spontaneously collected data.

Despite its complexities and significance, reclassification decision-making for ELs with disabilities has garnered little attention. Given the opaque intersection of language and disability, input from teams of teachers is recommended as a criterion for reclassification of dually identified students (see Kangas, Reference Kangas2024; Kangas & Schissel, Reference Kangas and Schissel2021; Park & Chou, Reference Park and Chou2019). However, the few prior studies on the reclassification of ELs with disabilities suggest that teachers may require additional support navigating this complex process (e.g., Burho & Thompson, Reference Burho and Thompson2021; Burho, Thompson, Bromley & Bovee, Reference Burho, Thompson, Bromley and Bovee2024; Kangas & Schissel, Reference Kangas and Schissel2021). Kangas and Schissel (Reference Kangas and Schissel2021) found that teachers attempted to “sort out” whether ELs’ disabilities affected their ability to progress in their ELP by relying on anecdotes. In their study on reclassification decision-making for ELs with disabilities, Burho and Thompson (Reference Burho and Thompson2021) examined educators’ engagement with parents. The researchers found a one-way communication process in which educators conveyed information to parents in ways that failed to consider parents’ important insights, answer their questions, and engage them as joint decision-makers. In the end, reclassification decisions were often rendered by educators before meetings with parents. In a more recent phenomenological study, Burho et al. (Reference Burho, Thompson, Bromley and Bovee2024) found that amid multiple constraints, educators attempted to balance “efficiency with individualized attention to dually identified students” (p. 175) during reclassification. Educators reported implementing strategies that would promote the reclassification of ELs with disabilities, believing this would expand these students’ opportunity to learn. Collectively, these studies are the only investigations into teachers’ decision-making while reclassification for ELs with disabilities. Yet none explore teachers’ use and understanding of data during this critical process.

Like all high-stakes decisions, reclassification determinations for ELs should be made after considering multiple data sources, and not from a single assessment score (see AERA/APA/NCME, 2014; Linquanti, Cook, Bailey & MacDonald, Reference Linquanti, Cook, Bailey and MacDonald2016). For ELs with disabilities, scholars recommend reclassification policies and procedures that require multiple evidence that enables fuller, complementary understandings of their ELP and multiple stakeholders of educators, parents, and even the students themselves engaging in decision-making (Kangas, Reference Kangas2024; Park & Chou, Reference Park and Chou2019; Shenoy et al., Reference Shenoy, de Valenzuela and Pacheco2022). For the former, Park and Chou (Reference Park and Chou2019) recommend that teachers draw from a “body of evidence” (p. 11), including student curriculum-based assessments, portfolios, student observations, and parent input, among others. Shenoy et al. (Reference Shenoy, de Valenzuela and Pacheco2022) also called for a broader array of ELP data from holistic assessments, ongoing assessments of a child’s communication skills, bilingual assessments, to name a few.

For the latter, scholarship has highlighted the importance of coupling multiple data sources with team-based approaches in reclassification decision-making (Kangas, Reference Kangas2024; Kangas & Schissel, Reference Kangas and Schissel2021; Park & Chou, Reference Park and Chou2019). Park and Chou (Reference Park and Chou2019) recommend that several stakeholders should be part of the decision-making team for ELs with disabilities, including the ELs’ special and general education teachers, parents/guardians, and individuals familiar with English language acquisition. Multi-stakeholder teams are especially critical given Burho and Thompson’s (Reference Burho and Thompson2021) findings about the lack of parent engagement and Kangas and Schissel’s (Reference Kangas and Schissel2021) finding that reclassification decisions are often made by one individual, the ESL teacher, with other key IEP members offering little to no input.

In sum, reclassification remains a critical issue in the education of ELs with disabilities, as their chances of exiting are alarmingly slim. The standardized ELP assessments upon which reclassification often hinges present challenges for dually identified students and complexities for the teachers who are charged with interpreting these data. Research, however, has yet to examine the ways in which teachers make sense of ELP data for ELs with disabilities while making reclassification recommendations. In response to this empirical gap, this qualitative case study investigated the data interpretation and decision-making practices of teachers during reclassification.

Interviews

We conducted 17 semi-structured interviews with the ESL staff of Tamarack. We developed a protocol prior to each interview as a foundational guide, but we also allowed for spontaneous questions that arose in response to the participants’ shared beliefs and experiences.

The interviews targeted information regarding the participants’ (a) use and interpretation of ELP data (RQ1), (b) approach to reclassification decisions (RQ2), and (c) beliefs and capacities to implement the reclassification policy (RQ3). Following each interview, Kangas wrote an analytic memo distilling its emerging patterns. The average length of the interviews was approximately 45 min. All interviews were audio-recorded and transcribed verbatim.

Longitudinal ELP assessment data

Given the vital role standardized ELP scores play in reclassification, we collected these data from the sample of all current and recently reclassified ELs with disabilities in the district. As Pennsylvania participates in WIDA’s ELP assessments, we collected all available WIDA ACCESS or Alternate ACCESS scores for the past 5 years (i.e., 2018–2023). These longitudinal data allowed us, first, to understand the scores of the ELs with disabilities over time and, second, to investigate teachers’ use and beliefs around these data (RQ1, RQ3).

As we will explicate in the findings, Tamarack School District lacked a student data management system that properly tracked students’ ELP data. Thus, we carefully combed through 5 years of score reports provided to Tamarack by WIDA, tracking each student from the dataset across the annual reports, culling their data from the reports, and inputting them into a database in preparation for data analysis.

Reclassification reports

In accordance with the state’s reclassification policy, Tamarack’s ESL administrative office typically completed brief reports when an EL with a disability was exited from language services, with the ESL teacher providing their reclassification decision (RQ1) as well as the evidence they used to inform their decision (RQ2). We collected all available reclassification reports for recently reclassified ELs with disabilities, for a total of 66.

Analysis of ELP assessment score trends

To understand the ELP assessment score trends of each EL with a disability (RQ1)—the anchoring data in reclassification—we created and analyzed growth gradients, which chart changes in longitudinal data across time (Miles, Huberman & Saldaña Reference Miles, Huberman and Saldaña2019). In this study, we specifically tracked the WIDA ACCESS or Alt ACCESS ELP scores for each student over time in the sample. This analytic approach was multi-step. First, we compiled all ELP scores for the ELs with disabilities in the sample from 2018 to 2023. Second, we developed growth gradients for each student’s ELP scores, including their composite scale score and domain-specific scale scores (i.e., listening, speaking, reading, and writing). Third, we compared trends in the composite and domain-specific scores to students’ demographics, particularly their identified disability, L1, and grade level. This analysis allowed us to identify the ELP score trends for an individual EL with a disability and for the sample overall.

Analysis of reclassification eligibility and decisions

To understand how ELs with disabilities were reclassified (RQ2), we analyzed the data through a partially ordered meta matrix (Miles et al, Reference Miles, Huberman and Saldaña2019). The matrix allowed us to extract, condense, and display together multiple variables from across the study’s data sources in one place, including the reclassification eligibility of the students, reclassification decisions rendered by the ESL teacher, and the data they cited using to make their recommendations. Through this matrix, we were able to detect the relationships among the timing of reclassification, decisions made by the teachers, and data they used, or in many cases, did not use to inform their decision-making.

Coding

To understand ESL teachers’ data interpretation (RQ1), reclassification decisions (RQ2), and their will and capacity to reclassify ELs with disabilities (RQ3), we used a two-cycle hybrid coding process (Fereday & Muir-Cochrane, Reference Fereday and Muir-Cochrane2006) to analyze the interview data. In the first cycle, we applied open codes (Saldaña, Reference Saldaña2021) that both were inductive, emerging from the data, and deductive, derived from the theoretical framework. These codes were refined and compiled into a codebook during the first cycle. In the second cycle of coding, we employed pattern coding (Saldaña, Reference Saldaña2021), grouping the initial codes based on their connections to one another and the theoretical framework of local capacity and will.

Trustworthiness

This study implemented numerous techniques to increase the trustworthiness, or credibility, of the data and our interpretations. First, we kept detailed audit trails of our data collection and analytic approaches (Creswell & Miller, Reference Creswell and Miller2000; Lincoln & Guba, Reference Lincoln and Guba1985). Second, we strove to triangulate our data sources during both data collection and analysis (Lincoln & Guba, Reference Lincoln and Guba1985; Tracy, Reference Tracy2010). As an example, we conducted preliminary analyses of the longitudinal ELP assessment scores and the reclassification reports and then during interviews, we inquired with teachers about the patterns that arose. Third, we embedded member checking into the interviews through sharing interpretations of the data with participants and asking for them to corroborate, refute, or expand these interpretations (Lincoln & Guba, Reference Lincoln and Guba1985; Tracy, Reference Tracy2010). Finally, we sought out disconfirming evidence throughout all data analyses, noting particular students, teachers, and evidence that rivaled the predominant patterns we found in the data (Creswell & Miller, Reference Creswell and Miller2000; Lincoln & Guba, Reference Lincoln and Guba1985; Tracy, Reference Tracy2010). In doing so, our analyses showcased the complexities of reclassifying ELs with disabilities.

Will: Data skepticism

As ESL teachers examined student data and made reclassification recommendations, there was one belief that was influential—that standardized ELP assessment data were inaccurate and thus could not be fully trusted. We came to call this belief data skepticism. Throughout interviews ESL teachers relayed the layered nature of their skepticism, questioning the accuracy of the general ELP assessment scores for ELs overall but with additional concerns levied against both general and alternate ELP assessments for ELs with disabilities.

A foundational skepticism held by the teachers was that the emotions and motivations of ELs and ELs with disabilities alike influenced their performance on the ELP assessments, often in ways that underreported their capabilities. In interviews, teachers reported observing a range of student emotions during the annual assessment—from apathy to anxiety to frustration—that affected students’ efforts. Maribel, who taught in elementary grades, shared how some of her students just clicked random responses on the screen while taking the computerized assessment: “Once we know what the student’s abilities are, it doesn’t always match with the WIDA scores. Some kids, it’s just willy-nilly clicking on whatever.” Similarly, the ESL coordinator intimated:

For other teachers, however, they observed ELs’ hesitance and anxiety, especially during the speaking domain of the assessment, which requires ELs to speak aloud, recording into the computer’s microphone. Rubi, a middle school ESL teacher observed: “I’ve seen kids just push the record button and sit there silently. And then you wonder why their scores don’t move.” Consequently, most teachers in the department felt that the ELs’ composite scores (i.e., overall proficiency score) were dragged down by their speaking performance, as Nora relayed: “Kids can always speak much, much more than they let on with the computer.” The growth gradients of the longitudinal ELP scores corroborated this perspective, with listening as the lowest trending domain (n = 35) among dually identified students in the sample regardless of their identified disability. Reading was the second lowest trending domain (n = 34), which was unsurprising given that one-third of the students in the sample were ELs with SLDs in reading.

Teachers especially in the secondary grades observed that beyond feelings of reluctance, ELs often intentionally underperformed out of a desire for the comfort and security that was afforded in ESL classes. More specifically, Christine described her ELs’ trepidation knowing they may exit ESL after passing the test, and as such, one of her students “totally bombed” the test in an effort to stay. Another secondary ESL teacher observed ELs deliberately performing poorly on the assessment to be with their family members: “A kid will find out that his cousin’s in the program and he’s like, well, I’m totally just going to screw up this WIDA ACCESS test. And then I’ll meet you there [in ESL], you know, next year.” Observing ELs’ engagement with and emotions during the ELP assessment, ESL staff questioned the accuracy of their students’ scores, citing its mismatch with the realities of the students’ capabilities in the classroom. In total, 14 of 17 participating staff expressed concerns over the construct validity of the test and the conclusions that could be made based on its scores. Tati, for example, said: “I don’t think it’s a good test. It doesn’t tell the true score.” Corroborating this perspective, Vicky believed: “It’s not really accurate.” In a departure from her colleagues, however, Fabiana thought the following of the construct validity of the WIDA ACCESS: “It’s close to, but it’s not quite.”

While teachers possessed these beliefs about the WIDA ACCESS for all ELs, they held added skepticism toward the scores of ELs with disabilities. Rachel reasoned:

Other teachers acknowledged the ELP scores of ELs with disabilities as “mostly flat” with dually identified students demonstrating minimal progress on the assessment. Recalling when the assessment was revised in 2017, some teachers felt it was now too challenging for all ELs, particularly those with disabilities. As Christine opined: “They did change the test at some point—where it gets harder. That’s very, I think, discouraging for the kids with disabilities. It’s like, ‘I’m never going to get out of here, you know?” Rubi, one of the most vocal critics of the WIDA ACCESS, also voiced strong objections to WIDA’s Alternate ACCESS. She described the Alternate ACCESS assessment as a “joke,” detailing her experience:

In sum, the interviews revealed a pervasive belief among teachers—that data from ELP assessments offered inaccurate depictions of students’ ELP and thus could not be trusted and used. Their skepticism of ELP assessment data was prominent for all ELs but was particularly acute for ELs with disabilities.

Capacity: Resources and training

While teachers were skeptical toward standardized ELP data, they also had a lack of capacity to make reclassification decisions, as they were equipped with limited resources and training to progress monitor ELP data. One of the most significant resource constraints was that the ESL teachers did not have a tool to manage and track ELP data, whether formative or summative. Much to the frustration of the ESL staff, the district’s designated data management system was not “built” for ELs; thus, certain data were hard to find or missing altogether. Brett characterized the district database as “not great” for ELs. The ESL coordinator wanted the district to purchase an EL-specific data management system, but the current price of such a system was too costly.

To compensate for the inadequacies of the current system, the ESL administrative office developed its own ad hoc tool—a Google sheet—containing a snapshot of relevant data for the district’s ELs, including their primary language, year of EL classification, initial ELP level, and immigration status, among others. The ESL coordinator shared about the incompatibility of their current system and the needs of the teachers: “It’s easier for an ESL teacher to find that data because we have it in our like, office drive. But if like a classroom teacher or a special ed teacher wanted to find that data, they could not find that data right now.” While the Google sheet offered ESL staff quick, current information about their ELs, the ESL department still did not have a system for collecting and monitoring ELP data, either formative or summative. For summative data (i.e., standardized ELP assessment data), the ESL staff relied on annual reports provided by WIDA in the form of PDFs. Understanding how an EL performed on the assessment overtime was labor-intensive; teachers had to search the PDFs, which were organized by year and therein by school. Many times, ESL teachers did not know the previous schools the student attended in the district, and thus their efforts to find these data fell flat, as Donna relayed:

Indeed, to examine the standardized ELP data for this study, we constructed a database of ELP scores for the WIDA ACCESS and Alternate ACCESS, tracing each individual EL with a disability across multiple years of PDF reports—a task that required a hefty investment of time. Examining ELs’ standardized ELP data became untenable for an ESL staff that was already stretched too thin.

For formative ELP data, ESL staff reported not having the resources nor the training to progress monitor their ELs’ ELP. When asked how the district could better support them, for instance, secondary ESL teacher, Rachel commented needing both a tool and training:

Beyond these resources, Rachel was eager for any training to support her professional development, admitting she needed a “refresh” after receiving her ESL certification 15 years prior. Corroborating the need for a “refresh” among the staff, Rubi insisted that her colleagues were unable to understand ELP score changes. She described a specific instance when she discovered that some of her ELs with disabilities were in error not considered for reclassification:

Echoing this perspective, Maribel, who teaches in one of the elementary schools, reported being confronted with more questions—and no concrete answers—when examining ELP scores of ELs with disabilities and then needing to make instructional decisions: “I don’t know. I don’t know,” she repeated. “They are who they are. And I think, like I said, what we’re doing here, is it really helping them? You know what I mean?” Sharing similar uncertainty when examining the ELP scores for dually identified students, Vicky described: “And the problem is, there’s never going to be a cut-and-dry for every kid. We’re [ESL teachers] all coming from different places.” Compounding their challenges in interpreting ELP data during reclassification, 7 of 17 teachers reported unawareness of and confusion about the reclassification criteria needed for ELs with disabilities to be eligible to exit.

Despite being charged with making reclassification decisions, the ESL staff lacked the capacity to understand and interpret ELP data—key skills in progress monitoring—because of a scarcity of training and an inadequate data management system. At the same time, they also possessed skepticism toward standardized ELP data. Their skeptical beliefs and constrained capacity had a reinforcing effect on one another (Figure 1). With a profound distrust in standardized ELP data for ELs with disabilities, teachers had little incentive to develop their skills in interpreting and using these data, and the need to secure a more useful data management system for the department waned. Moreover, with limited resources and skills to collect, understand, and use ELP data rendered these data an afterthought among ESL staff.

On-the-ground policy implementation: Reclassification decisions

With skepticism and constrained capacity to interpret and monitor ELP data, ESL teachers primarily made reclassification decisions not based in data but rather on their values and instincts (Figure 1). Specifically, they were guided by their values—to advocate for their students’ learning opportunities—and instincts regarding whether students’ disabilities were the root cause of their “difficulties” in making ELP growth.

When asked which data they used to make reclassification decisions, teachers reported consulting any combination of the following non-linguistic evidence, including grades, standardized reading and academic assessment scores, student behaviors, participation, along with general academic skills. Strikingly, only one reported examining standardized ELP scores over time and few considered students’ academic skills in reading, writing, speaking, and listening. In short, ESL teachers took an idiosyncratic and non-systematic approach to what data they reviewed and how they reviewed it.

To make reclassification decisions, teachers primarily relied heavily on teacher input. Approximately half of the ESL teachers sought out input informally from their general education, special education, and ESL colleagues during reclassification. Although ESL teachers consulted their colleagues, the ESL coordinator intimated that these colleagues often deferred to the decisions the ESL teacher recommended, commonly responding: “I don’t know anything about language.” While ESL teachers may have informally sought input from their colleagues, none of the ESL teachers reported consulting the IEP teams of ELs with disabilities, despite the mandate from the state for such school-based teams to render a reclassification decision. The district’s reclassification reports offered corroboration of this practice, which commonly featured recommendations from ESL teachers, with none including the perspectives or even signatures of IEP team members. Our analysis of reclassification records indicated ESL teachers’ input was often the determinative criterion that resulted in dually identified students being reclassified or remaining in language services.

As ESL teachers made reclassification decisions, they often grounded their decisions not in data but in their own values regarding the learning opportunities of their ELs. For instance, the middle school ESL teachers often decided to keep ELs with disabilities in language services because they would otherwise sit in what the district called an “advisory period”—effectively a study hall. In light of ELs’ options, Rachel reported: “I’d rather you stay with me and let me help you out the best that I can, because the alternative is you’re going to be sitting in advisory.” She explained further:

Her colleague Nora, too, believed language services were more advantageous, and thus, in many instances, she did not recommend reclassification.

A fifth-grade ESL teacher, Eugenia, thought remaining in language services was preferred over the general education classroom: “If I can hold them in the ELD [English language development] classroom even one more year rather than just leaving them in the regular classroom, I bring them back in.” In short, these teachers believed that the support ELs with disabilities would continue to receive through language services could only have a buoying effect on their academic and ELP growth.

At the high school, however, ESL teachers were more likely to recommend reclassification because remaining in language services often meant that ELs had little room to take other courses. Morgan shared how she approached reclassification with a “be free” approach. Espousing the values of ESL teachers, reclassification reports for some ELs with disabilities included statements regarding the benefit of language services: “He has a speech IEP and his language acquisition does not interfere with his speech. He does not benefit from ELD [English language development] services.” Other reports echoed: “No longer benefits from ELD service” and “Does not benefit from ELD services.”

While making decisions based on their values, ESL veteran teachers—those with 15 years or more of experience in education—also relied on their instincts regarding whether disability was the source of ELs with disabilities’ academic and ELP “difficulties.” Across the interviews, these seasoned teachers referred to this sense in a number of ways: professional judgment, feelings, or gut. Maribel shared how she relied upon her gut because, as an ESL teacher, she knew her students on a deeper level:

Feeling skeptical about certain data, Rubi reported similarly relying on her instincts: “There’s a lot that I don’t buy into here. So, I’m hesitant to use some of the stuff that. So I think here it would be more like, it would be more of a feeling.” She explained further how these feelings guided her decision-making: “But there have definitely been times that I’ve looked at a kid and been like, ‘I don’t really think this student should reclassify.’” For Vicky, a veteran teacher, disentangling language learning from a disability was intuitive to her:

Reclassification reports indicated that teachers’ instincts were commonly used as evidence to reclassify. In the section of the reclassification report that prompted IEP teams to list the evidence used to make a reclassification decision, teachers would identify either language or disability as a root cause of academic and linguistic “problems.” For instance, one reclassification report listed the following so-called evidence for an EL with an intellectual disability: “Limited language in both English and Spanish—no identifiable language needs related to English. Does not have acquisition needs. Speech and disability are his concerns—not language.” Instances in which teachers believed disability was the primary cause of ELs with disabilities’ “difficulties,” recommendations to reclassify followed suit. In another reclassification report, the teacher wrote: “Mental needs along with his education diagnosis supersede all his English language needs.” Similarly, another report indicated for an EL with an SLD in reading: “His issues are more related to his IEP rather than his second language abilities. She [ESL teacher] reports him exiting.” In other reports, no evidence was provided whatsoever, further underscoring how data were divorced from the decision-making of ESL teachers. The lack of data in the reports and the decision-making of ESL teachers is unsurprising given their skepticism toward standardized ELP on one hand and their constrained capacity to progress monitor formative and summative ELP data on the other.