Texas provides an early, ominous look at pandemic learning loss

Nat Malkus June 30, 2021

This week, Texas became the first state to publish a full set of student test scores for this past school year. This provides an early look at what we might expect to see across the country after over a year of pandemic schooling. The news was not good.

For some subjects, test scores did not just drop, but plummeted. The Texas Education Agency (TEA) released math and reading tests results for grades 3 through 8, as well as high school students’ results on multiple end-of-course assessments. As one example, Algebra I assessments show that in 2021, just 41 percent of students met grade-level expectations, down from 62 percent in 2019. Scores also dropped for most reading assessments, but by much smaller margins. Scores for science and social studies fell somewhere in-between.

The size of these drops is even more startling when viewed over time. The figure above shows three years of steady growth from 2016 to 2019, where grade-level performance grew in math and reading in grades 3 to 8 and in high school Algebra I. The pandemic wiped out those gains and then some. Yes, reading performance declines were smaller than those in math, but the earlier gains were also smaller and were also lost to the pandemic school year. For both subjects in elementary grades, other numbers published by TEA show that the pandemic erased the last seven years of academic progress.

It’s important to look closely at these results, and those that come from other states, to make sure declines are not overstated. Some differences between years could stem from a different population of test-takers, which is particularly likely during the pandemic. If, for instance, this year’s test-takers were more disadvantaged, the declines would likely be overstated. Unfortunately, that does not appear to be the case.  

The number of 2021 test-takers was 15 percent lower than in 2019, enough to bias the results if high achievers were more likely to be absent this year. However, advantaged students appear to be, if anything, over-represented. Compared to 2019, higher proportions of gifted and talented students took the STARR tests this year, and lower proportions of at-risk students, migrant students, and Title I students took this year’s test. If absent students are having an effect on these results, they are under-stating the pandemic damage on student learning.

The pandemic introduced a new group of students — remote learners — that may warrant more attention this year than other, more conventional subgroups. Indeed, TEA data show the biggest drops occurred for students who spent more of the year learning remotely, and by a huge margin. In districts that were mostly in person, at least three-quarters of the year, students not meeting math grade-level expectations increased by nine percentage points. In districts that were mostly remote, that increase was a whopping 32 percentage points. While all students faced some degree of learning loss, students learning from home fared much, much worse.  

5th graders at Louise Elementary School study in a classroom with plastic partitions, during the coronavirus disease (COVID-19) pandemic in Louise, Texas, U.S., November 20, 2020. Via REUTERS/Go Nakamura

The differentials in Texas suggest ominous results will be coming for the rest of the country. TEA’s report shows the impact of remote schooling in Texas, but does not show much more in-person schooling in Texas compared to other states. Initial analyses of AEI’s Return to Learn Tracker data show Texas was well within the top 10 states for offering in-person instruction across this school year, meaning the fallout will be greater elsewhere.

Now there could be other good reasons to discount the importance of these declines. Perhaps students, particularly remote students, simply cared less about the test this year. That would be cold comfort however, as those less engaged with tests were likely disengaged with school more broadly. Another possibility is that students had less direct test prep this school year, and thus lower scores are not meaningful. I am sure that is part of the picture, but it doesn’t explain differences for remote learners. Blithely dismissing these dramatic declines is wishful thinking. 

Disclaimers aside, the picture in Texas is bleak. There is no sugarcoating it. If past progress is any indicator, the recovery will take years, and even longer in districts that spent more time in remote learning. These flailing scores don’t just reflect lost gains, but lost student potential. What is bad for Texas is likely even worse for states that spent much of the year with schools remote. These early data from Texas are likely the first in a string of devastating test score results from other states, and I am afraid the worst reports lie ahead.

Learn more: Nearly all school districts finally offer some in-person instruction: We should not be satisfied | The steady march to in-person learning is happening more quickly for white students | 8,000 districts point to 5 trends in instructional offerings