EL Identification, Enrollment, Attendance, and Assessment

“We didn't have an assessment tool that was designed to work in a remote environment…Students were provisionally qualified based on their home language survey information.” —Kristin Percy Calaff

Remote Identification and Screening Procedures were Imperfect

Early on in the pandemic, the U.S. Department of Education (ED) issued clarifying guidance to state and local education agencies on their responsibility to continue identifying students who might need additional language instructional services.¹ State and local education agencies responded by developing provisional identification procedures to allow new English learners (ELs) to start receiving services without requiring them to be screened in person. Alternative procedures depended heavily on a few questions asked on home language surveys used during intake, which were sometimes supplemented with interviews with parents.

While schools and districts appreciated the flexibility offered by virtual screening processes, the people we spoke to said that there was a lot of room for error with these alternative procedures due to limited and imperfect information, which can ultimately affect service provision. And as Jorge Macias, chief of language and cultural education at Chicago Public Schools, stated, “although interviews are helpful, you really didn't get to actually assess the child's proficiency like you would traditionally.” In the end, many of our interviewees said that they chose to err on the side of caution to ensure no EL fell through the cracks.

Provisional identification procedures led to an inflation of the number of currently identified ELs going into the 2021–22 school year. This means that going into fall 2021, schools were cleaning up the data pool of students who had been provisionally screened and subsequently identified as English learners. Kristin Percy Calaff, director of multilingual education in Washington state, said, “numbers are probably overinflated right now and it's going to take a year until the dust settles and see where we are." Similarly, Macias said that going into the fall of 2021, Chicago Public Schools had to screen roughly 19,000 students between August and October in order to clean up these provisional identifications. Although overall student enrollment did decrease, he said, EL enrollment in CPS did not decrease.

This inflation may also be concentrated in earlier grades. Percy Calaff said that essentially all kindergarteners were provisionally qualified in 2020–21 if their home language survey showed that they had a language other than English at home. Additionally, it is expected that a small proportion of ELs in Washington state may continue to be provisionally identified because some families may still not feel comfortable returning to in-person schooling and schools cannot force them to bring the child in to be officially screened and tested. According to Percy Calaff, this is expected to affect about 5 to 10 percent of ELs in the state. Despite the fact that there was a lot of room for error, “the alternative to provisional screening is to not serve these kids and that is not okay, so they were erring on the side of caution and provisionally screening until they know otherwise,” she said.

Although many of our interviewees spoke of the important role parents played in the execution of remote education, this level of engagement also came with some unintended consequences. For example, Kelly Alvarez, Title III program consultant for the Michigan Department of Education, noted that giving parents a bird’s eye view of identification and screening procedures resulted in higher than usual screener denials as well as parents opting out of the English language proficiency assessment and language services altogether.

English Learner Enrollment and Attendance Slowed

This overidentification trend took place at the same time that national enrollment in K–12 schools dropped by roughly 3 percent—a decline of roughly 1.5 million students between 2020 and 2021.² This decline has not been spread out evenly across grades, as the number of preschool and kindergarten students decreasing by a combined 13 percent.³

While provisional data is not disaggregated by student subgroups such as English learners, interviewees confirmed declines in their EL populations. Michigan, for example, observed a 6 percent reduction in 2021, according to Jennifer Paul, EL and accessibility assessment specialist for the Michigan Department of Education. She said that the state would normally see about 4 to 5,000 new ELs every year. Paul attributed the decline, anecdotally, to restrictive national immigration policies compounded by the pandemic.⁴ California saw a major decline in enrollment of dual language learners (DLLs) in early childhood settings on top of already low enrollment, since some families simply disappeared, according to Anya Hurwitz, executive director of the Sobrato Early Academic Language (SEAL) model.

Beyond declines in enrollment, school systems also observed decreased attendance among ELs. Indeed, The 74 Million investigated attendance rates in multiple school districts across the country and found large increases in rates of chronic absence among EL students. Chronic absence is commonly defined as missing 10 percent or more of school days. Districts such as Houston observed a 158 percent increase in chronic absenteeism among ELs.⁵ Although reasons behind decreased attendance are difficult to define, Marguerite Lukes, director of research and innovation of the Internationals Network, said, “we're seeing lower attendance, and it's a range across schools, but lower than we normally would. And I think some of this is an on-going demand on students’ schedules due to work and the stress of the transition back.” School systems also struggled with EL attendance due to communication challenges and increases in transience among families, according to Robert Measel, ESL/bilingual education advisor for the Pennsylvania Department of Education.

States, schools, and districts adopted more flexible attendance policies during the pandemic in an attempt to deal with attendance issues. In Michigan, “the legislature approved a flexible attendance policy that gave districts some real flexibility for how they could serve kids,” Alvarez said, which allowed them to provide focused attention to certain groups that needed it. And Center City Public Charter Schools in Washington DC significantly altered the threshold for what constituted being “present,” according to data and assessment director Laura Berger. She said, “before the pandemic, in-person learning students had had to be present for 80 percent of the day in order for them to be present that day,” but for virtual learning, students were only required to log into at least one class and complete work for each class in order to get credit for the whole day. These accommodations, however, were eliminated once schools resumed in-person instruction.

“Students have experienced a slowdown in their normal rate of growth as a result of lack of access to education.” —Robert Measel

Assessment was Disrupted Unevenly Across School Years

Interviewees said that they had either finished administering their ELP exam, or were close to finishing the test when schools shut down in March 2020. The ELP assessment is often administered between January and March, before students take academic assessments such as math, reading/language arts, and/or science closer to the end of the school year.⁸ Many districts were already towards the end of their normal ELP testing window when schools shut down, which means that few domains (see Box: Four Domains of Language) were left incomplete in 2020, if at all. For example, in Washington, DC writing was the only domain that was left unfinished.

ED expected statewide standardized testing in spring 2021 to move forward and the Department granted significant flexibility for how states could carry that out. States could extend the testing window through the summer or fall; administer assessments remotely, where feasible; and/or shorten the state assessment.⁹ Despite this flexibility, or possibly because of it, test administration was uneven within and across states, which made for a more interrupted ELP assessment in 2021 compared to the first year of the pandemic.

In October 2021, WIDA released the most comprehensive report to date quantifying the impact remote learning has had on ELs' testing, proficiency, and growth during the pandemic. The report, which compared WIDA ACCESS data for academic years 2018–19, 2019–20, and 2020–21, found that the number of ELs who were tested across WIDA states in 2021 was 30 percent lower than for previous years.¹⁰ Those we interviewed who administered the test in 2021 said that statewide participation on the ELP assessment ranged from below 50 percent (in Harrisonburg, VA and Chicago, IL) to 70–75 percent (in Michigan), to slightly less than usual, about 10 percent fewer (in Pennsylvania).

Because the WIDA Consortium decided not to allow for remote administration of the WIDA ACCESS,¹¹ the 35 states and the District of Columbia which use this assessment did not have the option to remotely test for language progress. Interviewees said that their ability to test ELs was often hindered by the fact that many schools were still operating remotely and parents were not comfortable with bringing their kids in just to test them.¹² On the other hand, California did allow for remote administration of its English Language Proficiency Assessments for California (ELPAC), but as Renae Bryant, director of plurilingual services at Anaheim Union High School District told us, even with this flexibility her district was only able to assess about 60 percent of its ELs.

Several interviewees said that they took advantage of the flexibility offered by ED by extending the testing window through September and October of 2021. Some attributed higher participation rates as well as higher than expected scores to this decision, because students simply had more time to receive instruction and language support services before they were tested. This flexibility does not come without drawbacks, however, as interviewees expressed concerns over the implications that testing well into the fall of 2021 would have on spring 2022 assessments.

Linguistic Growth was Affected in Positive and Negative Ways

The October 2021 WIDA report found that though student proficiency and growth did go down overall in 2021, decreased performance varied by grade and domain. Elementary and middle school grades were hit the hardest, specifically first and sixth grades. Among the four domains tested, speaking was affected the most.¹³ However, because the report represents aggregated and averaged test scores across years for a fragment of the EL population (roughly 1.1 million of the over five million ELs students nationwide), the authors note that “to ensure a consistent and fair year-to-year comparison of average EL proficiency and growth, states and districts should examine the demographic makeup of students tested in their locale and consider the ELs who did not get the opportunity to take the assessment.”¹⁴

From local and state data, education leaders we consulted were able to observe and confirm that the pandemic did not impact all grades or linguistic domains evenly. Various interviewees said that ELs’ productive skills, such as speaking and writing, were most negatively impacted by remote instruction, with scores in the speaking domain actually going negative in some grades in Pennsylvania. At the same time, ELs’ receptive skills, such as listening and reading, remained relatively steady, with reading growth actually increasing for some students in Pennsylvania. Several of our interviewees attributed this uneven impact to the nature of remote education, which relied heavily on certain skills while stifling the use of others.

Many of our interviewees noticed that what ELs might not have gained in English they gained in their home languages by virtue of spending more time at home. Both Alvarez and Alicia Passante, ESL program manager at Center City Public Charter Schools in Washington, DC, said that increased home time helped ELs develop closer ties to family, which is beneficial to their overall development and in shaping their identity. Crystal Gonzales, executive director at the English Learners Success Forum, said that her organization observed a shift in the narrative about ELs and the importance of their home languages. This led to a spike in the use of certain resources such as videos, newspapers, and activities in students’ home languages that schools were able to share with families. Unfortunately, it is difficult to quantify this progress because, as several interviewees noted, schools do not typically assess ELs in their home language, which makes it difficult to capitalize on this asset.

“In theory, data delays shouldn't have an impact on instruction [because] on a day-to-day basis, their academic program should be whatever is required for that student.” —Robert Measel

Uneven administration of the ELP assessment means that some schools and districts will have data gaps for many of their English learners, particularly those who disengaged from school altogether in 2021 and were harder to reach when the testing window came around. In Harrisonburg, for example, although district-wide about 50 percent of ELs were tested, this number went down to 15 percent among those in high school. This means that, though not widespread, there will be students who have incomplete or nonexistent ELP scores from both 2020 and 2021, particularly in the higher grades. These students should be prioritized in efforts to gather as much information as possible about where they currently stand in their language development.

The ELP assessment is particularly meaningful for ELs because “they are essentially gate-keeper assessments,” said Paul, for students required to take it. This is because, depending on the educational program that guides instruction at the local level, ELs may not have access to the full academic curriculum needed to prepare them for postsecondary prospects until they are deemed proficient in English and reclassified.¹⁵ Data gaps due to delayed or missing ELP scores during the pandemic may have implications on various programmatic decisions that affect ELs, including when they are reclassified, their expected growth targets, and the maximum number of years these students will have to reach proficiency.

Various interviewees said that reclassification decisions were affected by the pandemic in 2021. Alvarez said the Michigan statewide reclassification rate went down to 4 percent from a usual 11 percent. Passante said that the EL “bubble” at Center City schools in DC expanded slightly between 2019–20 and 2020–21 because the district has not reclassified students for the last two academic years.

These data gaps may also require schools to make changes to EL timelines to proficiency and growth targets so as to not derail the accountability system. As Measel, at the Pennsylvania Department of Education, articulated, there is no differentiation in an accountability system where everyone fails.

Although these data gaps may affect policy decisions at a system level, all of the people we spoke to indicated that they may not necessarily affect what ELs are experiencing in the classroom. This is because “state assessments are not the things that drive instruction on an ongoing basis, they are only one piece of the puzzle,” as Joe Luft, former executive director of the Internationals Network, said. Others added that scores on statewide ELP assessments may not necessarily reflect the actual skills of kids in the classroom, as they often score their proficiency lower than what teachers are seeing. As a result, many of the people we consulted placed a heavy emphasis on the ability to measure outputs in real time through more authentic assessments like performance tasks, portfolios, capstone projects, and informal checks. In fact, Bryant said that even before the pandemic standardized tests were not necessarily the driving force in her school district, and that leadership had invested heavily in formative assessment tools and strategies, which educators were able to carry through the pandemic.

However, the impact of ELs’ declining English skills as measured by the ELP assessment should not be understated, as it can have downstream effects and compound challenges in every other academic area. As Measel told us, “there is no such thing as a gap in a development process” like language development, which means schools and districts must ensure they are accurately assessing ELs as they return to in-person learning, to determine where they are along the language development spectrum. Other interviewees said that they have observed increased interest in benchmarking and formative assessments that can be used to fill data holes for students that were not able to test in 2021 and/or 2020. However, these should be undertaken with caution, because according to Paul, there is no standard that establishes what these tools must have to ensure accessibility for ELs, which means they come with a range of quality in terms of the supports and accommodations that are provided.