Teacher Attrition and Mobility in the Pandemic

Abstract

We contextualize the magnitude of teacher attrition during the pandemic, including from the 2020–2021 school year to the 2021–2022 school year, using longitudinal data on teachers in Washington since the 1984–1985 school year. The teacher attrition rate after the 2020–2021 school year (7.3%) increased by almost one percentage point from the attrition rate after the 2019–2020 school year (6.4%), but these rates are well within the range of turnover rates observed during pre-pandemic years. The increase in turnover during the pandemic was also smaller than pre-pandemic differences in turnover between high- and low-poverty classrooms in the state, and these inequities in turnover between high- and low-poverty classrooms decreased during the pandemic relative to pre-pandemic years.

Teacher attrition is squarely in the policy spotlight. Major news outlets such as CNN, the New York Times, Newsweek, NPR, the Wall Street Journal, and the Washington Post all have featured recent stories highlighting the impact of the COVID-19 pandemic on teacher shortages. All these news outlets point to teacher attrition as a primary culprit, using phrases like “mass exodus,” “in crisis,” and “Great Resignation” to describe the current state of affairs in the teacher labor market. Below is a small sampling of their characterization of the situation:

Burned out teachers are leaving the classroom for jobs in the private sector, where talent-hungry companies are hiring them—and often boosting their pay—to work in sales, software, healthcare and training, among other fields. (Dill, 2022)

55% of [teachers] say they will leave teaching sooner than they had originally planned, according to a poll of its members by the nation’s largest teachers union. (Kamenetz, 2022)

The educator profession—a critical cornerstone of American life—is in crisis. (Maxouris & Zdanowicz, 2022)

A mass exodus could worsen existing staff shortages in schools and cripple the education system in the U.S. (Rahman, 2022)

The coronavirus is vastly exacerbating that shortfall, experts say, by prompting many teachers to leave the profession or take early retirement. (Singer, 2021)

But the evidence of increased teacher attrition in these media reports is thin, as they cite anecdotes from specific former teachers or surveys about what current teachers say they might do in the future. Moreover, recent evidence from data collected before the pandemic (Nguyen et al., 2022) finds that only about a third of teachers who report an intent to leave actually do leave their position at the end of the year. So, what do we know at this point about rates of teacher attrition during the pandemic? Contrary to the definitive takes from press reports, the existing large-scale, quantitative evidence suggests that teacher attrition after the first year of the pandemic (i.e., after the 2019–2020 school year) dropped relative to the preceding school years (e.g., Bacher-Hicks et al., 2021; Diliberti & Schwartz, 2021; Goldhaber & Theobald, 2022). This likely reflects teachers hunkering down after the 2019–2020 school year in the midst of the uncertainty of a pandemic (Ayaita & Stürmer, 2020; van Huizen & Alessie, 2019).

However, there are reasons to think that things could be different after the 2020–2021 school year and coming into the 2021–2022 school year. For example, surveys of teachers highlight the considerable challenges of teaching during the pandemic (e.g., Diliberti et al., 2021; Kraft et al., 2021), and teacher attrition rates also tend to be inversely related to unemployment rates (Eagan et al., 2022; Goldhaber & Theobald, 2022). In 2021, the economy continued to improve, labor markets were tight, and, arguably, there was less COVID-related economic uncertainty (Baker et al., 2020). All of this suggests that we might expect to see teacher attrition increase coming the 2021–2022 school year. Moreover, given mounting evidence that the COVID-19 pandemic disproportionately impacted learning gains for historically marginalized student groups (Goldhaber et al., 2022), we should be particularly attuned to teacher turnover across different schools serving different populations of students.

In this brief, we explore teacher attrition during the pandemic, including from the 2020–2021 school year to the 2021–2022 school year, using publicly available data on the public teaching workforce in Washington. This dataset, the S-275, includes a snapshot of the state’s public school workforce (including teachers and other certificated personnel) from October 1 of each school year since 1984–1985, and does not include data on late teacher hires or short-term substitute teachers. Teacher attrition and mobility data from the last year of available national data of teacher turnover (2011–2012; Goldring et al., 2014) suggest that teacher turnover rates in Washington in this year (15.0%) are comparable to the national estimate (15.8%). The state introduced new employment categories in 2021 to 2022 for teachers funded by federal special purpose aid related to the coronavirus pandemic (i.e., ESSER funds); we consider these as “teachers” for the purposes of this analysis (i.e., teachers who move into one of these ESSER-funded positions not counted as moving into a non-teaching category), though we note that the funding of these positions may be temporary.

The stacked bar plots in Figure 1 show the proportion of teachers in different years who at the end of the year: (a) left their schools and the state’s public school workforce entirely (black), (b) left their current teaching position for a nonteaching position (e.g., administration or instructional coach; dark gray) within the state’s public school system, or (c) left their school for another public school teaching position in the state (light gray). We present these three statistics separately because they address different policy questions. From the perspective of the state, the overall attrition rate (i.e., the black and dark gray bars) represents the number of teachers who need to be replaced for the next school year, though attrition from the workforce altogether (black bars) is likely a larger problem than movement into noninstructional positions (dark gray bars) that also play important roles in public schools. But from the perspective of an individual school, all teachers who leave the school (i.e., also including teachers in the light gray bars) need to be replaced next year, so school mobility is also an important factor in terms of school teaching stability and hiring demands. The total teacher turnover rate (i.e., the sum of these three proportions) is shown at the top of each bar.

Figure 1.

Selected teacher turnover rates in Washington, 1984–1985 through 2020–2021 school years.

Consistent with findings from other states from the first 2 years of the pandemic—for example, from Arkansas (Camp et al., 2022) and Massachusetts (Bacher-Hicks et al., 2022)—the overall takeaway from Figure 1 is that the teacher attrition rate after the 2020–2021 school year increased from the attrition rate after the 2019–2020 school year, but these rates are well within the range of turnover rates observed during pre-pandemic years. Specifically, Figure 1 shows four important points:

1. The attrition rate of teachers from the public school workforce after the 2020–2021 school year (7.3%, last set of bars) increased by almost one percentage point from the attrition rate after the 2019–2020 school year (6.4%, next-to-last set of bars), and is also higher than the last pre-pandemic year (6.7%, fourth set of bars). The proportional increases in attrition were relatively consistent for early- and late-career teachers, as well as for teachers of color and White teachers (not shown in Figure 1 but available upon request).

2. Combined with the increased movement of teachers into non-teaching positions in public school districts (which increased from 2.0% after 2019–2020 to 2.7% after 2020–2021), the percentage of teacher “leavers” as defined by federal reports increased by about 1.6 percentage points, which represents a nearly 20% increase in the proportion of teacher leavers compared to the first pandemic year.

3. Rates of school-to-school mobility also increased by nearly a percentage point (to 7.8% from 6.9%). Thus, the total teacher turnover rate (17.8%) increased by 2.5 percentage points in the second pandemic year relative to the first—this is the second largest change in teacher turnover rates seen since 1984 to 1985—and was over one percentage point higher than in the average pre-pandemic year (16.7%, first set of bars).

4. The overall teacher turnover rate after the 2020–2021 school year is well within the range of turnover rates observed during pre-pandemic years (second and third sets of bars), and the rate of attrition from the workforce is the highest since the 2006–2007 school year.

Do these statistics support recent assertions of a “Great Resignation” or “Teacher Exodus” from public schools (Rahman, 2022)? These terms are, of course, in the eye of the beholder, but it is simultaneously true that (a) teacher attrition in Washington increased substantially after the second year of the pandemic, resulting in hundreds of additional open teaching positions relative to previous school years and (b) even these increased attrition rates are not inconsistent with what we have seen in past years.

The first point is important, as teacher attrition really does predict district staffing challenges. To show this in more concrete terms, we calculate teacher turnover rates by district and compare them to recently collected data by Goldhaber and Gratz (2022) on school district teacher vacancy rates in October 2021. We find that the relationship between district attrition rates after the 2020 –2021 school year and the percentage of open teaching positions in the district in October 2021 is positive and statistically significant (r = 0.23, t = 3.11). In other words, districts that had more teachers leave the workforce after the 2020–2021 school year also had more difficulty hiring teachers by the start of the 2021–2022 school year. And there is no doubt that these staffing challenges are of concern given evidence that teachers hired late tend to be less effective (Papay & Kraft, 2016) and that staffing challenges have led schools to have to close (e.g., Velez, 2021).

But it is also important to keep this increase in teacher turnover in perspective. Not only are teacher turnover rates after the 2020–2021 school year lower than turnover rates from the mid-2000s, but the increase in these overall turnover rates is actually smaller than the average difference in turnover rates between high- and low-poverty classrooms in Washington state in a typical school year. To illustrate this, we use data from the three most recent school years to calculate the turnover rates in high-poverty classrooms (defined as the top quartile of the percentage of students in the classroom receiving free or reduced-priced lunch [FRL]) compared to turnover rates in low-poverty (bottom quartile FRL) classrooms. We present these rates in Figure 2. In the most recent pre-pandemic year, the difference in turnover rates between high- and low-poverty classrooms was 3.4 percentage points, a larger difference than we observed in turnover rates between the two pandemic school years.

Figure 2.

Teacher turnover rates in Washington by classroom poverty level, 2018–2019 through 2020–2021 school years.

Figure 2 shows that teacher turnover rates from both types of classrooms increased after the second pandemic year. But it is also notable from Figure 2 that the increase in teacher turnover we document after the second pandemic year was driven disproportionately by increased attrition rates from relatively advantaged classrooms in the state. In other words, consistent with evidence from Massachusetts (Bacher-Hicks et al., 2022), inequities in turnover between high- and low-poverty classrooms decreased during the pandemic relative to pre-pandemic years; in fact, after the most recent year of the pandemic, the difference in overall turnover rates between high- and low-poverty classrooms was less than one percentage point. These trends are consistent for other measures of classroom and school disadvantage (e.g., based on the percent of students of color; results available upon request).

In sum, the trends in teacher attrition during the pandemic in Washington suggest that many of the recent media stories about rising teacher attrition rates are accurate in direction but, arguably, not in magnitude. In particular, while there was clearly an uptick in attrition in Washington and other states, we would not characterize the attrition rates in Washington after the second year of the pandemic as a “mass exodus” of teachers. But these results should only be generalized to these contexts, and importantly, they do not imply that we should be unconcerned; we agree with Will (2022) that we should take reports of teacher burnout and dissatisfaction seriously, even if they do not lead to attrition. Instead, we would argue that some of the recent reporting on teacher attrition has mischaracterized the extent to which teacher attrition is driving staffing challenges and has not done enough to highlight some of the long-term issues of differential attrition in different school and classroom settings. The lack of nuance in some media reports is problematic to the degree that policymakers react to these reports, rather than crafting solutions to more specific staffing challenges that have existed in the teacher labor market long before the pandemic.

Authors

DAN GOLDHABER, PhD, is the director of the National Center for Analysis of Longitudinal Data in Education Research (CALDER) at American Institutes for Research, and also the director of the Center for Education Data and Research (CEDR) at the University of Washington. His research focuses on issues of educational productivity and reform at the K–12 level, the broad array of human capital policies that influence the composition, distribution, and quality of teachers in the workforce, and connections between students’ K–12 experiences and postsecondary outcomes.

RODDY THEOBALD, PhD, is a principal researcher in CALDER at American Institutes for Research. His research focuses on teacher education, teacher licensure, special education, and career and technical education.