Long COVID for Public Schools: Chronic Absenteeism Before and After the Pandemic — Nat Malkus (January 2024)
Summary
A January 2024 report by Nat Malkus, from the American Enterprise Institute, titled Long COVID for Public Schools: Chronic Absenteeism Before and After the Pandemic. Its executive summary states that the national average chronic absenteeism rate increased from 15 percent in 2019 to 28 percent in 2022. The report draws on the American Enterprise Institute's Return to Learn Tracker, and it describes a student as chronically absent who misses 10 percent or more of a school year. It states that as of January 2024, data on the 2022–23 school year were limited to 38 states and Washington, DC.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
- Issuer
- Source documents
- Document type
- Long Covid For Public Schools Chronic Absenteeism Before And After The Pandemic
- Case
- Long Covid For Public Schools Chronic Absenteeism Before And After The Pandemic
Cite this document
“Long COVID for Public Schools: Chronic Absenteeism Before and After the Pandemic — Nat Malkus (January 2024).”
Original at www.aei.org. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/long-covid-for-public-schools-chronic-absenteeism-before-and-after-the-pandemic/
, SHA-256 6251b46c7c7194a1c217aee9f05727ee0aa09ffec90b0fcdb7c412938c63fc83.Full text
Long COVID for
Public Schools
C HRONI C AB S E N T E E I S M B EFO R E
A ND AFTE R THE PA N D E M I C
Nat Malkus
JA N UA RY 2024
A M E R I C A N E N T E R P R I S E I N S T I T U T E
Executive Summary
T his report documents chronic absenteeism over
the COVID-19 pandemic. Drawing on the Amer-
ican Enterprise Institute’s Return to Learn Tracker
elevated in 2023. I also use district-level data to report
variable rates of chronic absenteeism by district char-
acteristics including pre-pandemic achievement,
Chronic Absenteeism Data Collection, the most com- poverty, size, and the duration of remote instruc-
prehensive and current data collection on pandemic tion in the 2021 school year. Of particular concern,
and post-pandemic chronic absenteeism, I show that the percentage-point increases in chronic absentee-
pandemic increases in chronic absenteeism were ism were larger in districts and among groups that
widespread during the pandemic. More worrisome, already had higher chronic absenteeism rates before
using the most recent data for the 2022–23 school year, the pandemic.
I show that even after the pandemic subsided drasti- Given the potential for these rates of chronic
cally, the elevated rates of chronic absenteeism fell absenteeism to hamper urgently needed recovery
very little. from pandemic learning loss and its negative associa-
I find that the national average chronic absen- tion with school culture, chronic absenteeism is likely
teeism rate increased from 15 percent in 2019 to public schools’ greatest post-pandemic challenge.
28 percent in 2022 and remained substantially
1
Long COVID for Public Schools
CHRONIC ABSENTEEISM BEFORE AND AFTER
THE PANDEMIC
Nat Malkus
T he tumult of the COVID-19 pandemic made one
thing about education excruciatingly clear: Con-
sistently showing up to school is good for students.
rates.8 The available data from the 2022–23 school
year, arguably the first post-pandemic school year,
show those rates are falling, but not nearly fast
During the pandemic, extended closures and remote enough.
learning kept millions of students out of school for This report documents chronic absenteeism across
unprecedented periods of time, causing significant the pandemic. It shows that pandemic increases in
problems for students and schools. The academic con- chronic absenteeism were widespread and that they
sequences of these disruptions are glaringly evident in were larger in more-disadvantaged districts and dis-
students’ test scores,1 the social and emotional fallout tricts that had higher rates of chronic absenteeism
is reflected in numerous indicators,2 and the behav- before the pandemic—meaning that in many districts,
ioral challenges seen following the return to in-person current rates of chronic absenteeism are truly alarm-
schooling have made life difficult on students, teach- ing. All told, the data suggest that chronic absentee-
ers, and entire school communities.3 Indeed, as pro- ism is shaping up to be education’s long COVID and
nouncements from policymakers,4 headlines on op-ed that any hope of academic recovery will require get-
pages,5 and school district press releases have all sug- ting students to attend school consistently.
gested,6 if we are to catch students up from pandemic This report draws on data from the American
learning loss, we have to do more than just return to Enterprise Institute’s Return to Learn Tracker (R2L),
the status quo. which I created at the beginning of the pandemic to
Unfortunately, consistent school attendance was collect weekly data on the remote-learning status
a problem even before the pandemic struck. In 2018 of 8,600 school districts that collectively educated
and 2019, chronic absenteeism, measured as the per- 89 percent of public school students. Unfortunately,
centage of students who miss 10 percent or more of a nearly four years after the start of the pandemic,
school year, affected about 15 percent of K–12 public chronic absenteeism rates indicate that the “Return
school students. Well before the pandemic, attuned to Learn” name is still apt: Even though nearly every
observers and the US Department of Education alike school returned to in-person instruction over two
characterized chronic absenteeism as a “crisis.”7 years ago, many students have not fully returned to
The pandemic may have changed the terms of school in earnest.
what qualifies as a crisis in education, but current R2L has the most up-to-date national data on
chronic absenteeism levels would constitute a cri- chronic absenteeism available anywhere. Where avail-
sis under any terms. During the 2021–22 school year, able, I collected district-level data on chronic absen-
28 percent of students were chronically absent, an teeism from state education agencies for the 2016–17
increase of roughly 89 percent over pre-pandemic through 2022–23 school years.
2
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
As of January 2024, data on the 2022–23 school chronic absenteeism as one “nontraditional measure
year was limited to 38 states and Washington, DC. of student progress.”13
While some states make district chronic absentee- Even pre-COVID, the data on chronic absenteeism
ism data public a few months after the school year were troubling: One 2012 study estimated that, each
ends (e.g., Indiana and Rhode Island), others release year, between 5 and 7.5 million of the roughly 50 mil-
data 10 or more months after the school year ends.9 lion US students in pre-K–12 schools were chronically
In other words, waiting to publish this report until absent.14 Several analyses noted that low-income stu-
all states release data would mean publishing this dents, minority students, and students with disabil-
report many months into 2024. But that would be ities were more likely to be chronically absent than
unwise: Available chronic absenteeism rates for the were wealthier, white, or nondisabled students.15
2022–23 school year make clear that we do not have Moreover, chronic absenteeism is associated
time to wait until we have complete data. If we want with lower standardized test scores.16 Emma García
to address it with eyes wide open, chronic absentee- and Elaine Weiss found that eighth graders who
ism needs to be studied in real time. missed three or more days of school in the month
The remainder of this report consists of six sec- before being tested scored between 0.3 and 0.6 stan-
tions. In the first section, I review the basic findings dard deviations lower on the 2015 National Assess-
on chronic absenteeism. In the second section, I give ment of Educational Progress math test, depending
an overview of the data used in this report. Next, in on the number of days missed, than did those who
the third section, I review how chronic absenteeism did not miss any school days.17 Studies have shown
rates changed throughout the pandemic at the state larger associations between chronic absenteeism and
level. In the fourth section, I examine district charac- achievement in later grades than in earlier grades and
teristics that are associated with changes in chronic in math than in reading.18 Studies have also shown
absenteeism during and after the pandemic. The fifth that chronic absenteeism is negatively associated
section overviews chronic absenteeism gaps by stu- with social emotional measures, again with stron-
dent race. In the final section, I discuss these results ger effects in later grades and particularly in mid-
and suggest policy responses. dle school.19 In some contexts, chronic absenteeism
is also strongly predictive of course failure and high
school non-completion.20
Chronic Absenteeism Before the There is also evidence that when students are
Pandemic chronically absent, their non–chronically absent peers
suffer as well. Michael A. Gottfried found that chron-
According to the standard definition, a student is con- ically absent students may cause lower achievement
sidered chronically absent if he or she misses 10 per- in reading and math among their peers.21 Gottfried
cent or more of a school year, typically 18 days out of suggested that chronically absent students are more
a 180-day school year.10 Before the 2015 Every Stu- likely to have academic and behavioral problems
dent Succeeds Act, most states did not collect data when they return to school, forcing teachers to spend
on chronic absenteeism and instead reported only more time addressing their needs at the expense of
average daily attendance.11 Average daily attendance other students’ needs.
measures, however, are not necessarily reflective of Several studies have also noted that students who
chronic absenteeism numbers; for example, “it is pos- were chronically absent in earlier grades were more
sible for a school to have 90 percent average daily likely to continue to be chronically absent in later
attendance and still have as many as 40 percent of its grades.22
students chronically absent because on different days Pre-pandemic, certain school interventions have
different students are in school.”12 After 2015, though, proven capable of reducing absenteeism. These in-
“the vast majority of U.S. states” began reporting on clude low-cost interventions such as communication
3
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
efforts23 and more expensive interventions such as 2023 district-level data for 38 states and Washington,
transportation services and mentoring supports.24 DC. Data for additional states will be added to the
In general, lower-cost interventions prove to have R2L chronic absenteeism collection as they become
cost-effective but relatively small impacts on chronic available.27
absenteeism, while more resource-intensive inter- I supplement missing state-sourced data with
ventions can generate larger improvements. This pat- recently released district-level chronic absentee-
tern meshes with other research that suggests that ism counts from the US Department of Education’s
broader efforts, in addition to narrow individual and EDFacts data collection published through its ED
school interventions, may be required to substantially Data Express platform.28 These data include counts
improve chronic absenteeism.25 Whether and how of students chronically absent at the local educa-
much these interventions might influence the much tion agency (LEA) level for 2018 and 2019 and at
higher pandemic-era rates of chronic absenteeism the school and LEA level by student race for 2020
remains an open question. through 2022. The ED Data Express counts are the
best available numerators for calculating chronic
absenteeism rates but unfortunately do not include
Methods matching denominators. Therefore, after consult-
ing with National Center for Education Statistics
State-sourced data, which are both comparable (NCES) staff, I merged the ED Data Express count
over time and the most up-to-date large-scale data data with NCES Common Core of Data (CCD)
on chronic absenteeism, serve as the primary data K–12 student membership counts, which are the
in R2L’s chronic absenteeism tracker. I collected best available denominators for calculating chronic
district-level chronic absenteeism data from each absenteeism rates.
state, as available, for the 2016–17 through 2022–23 Note that these denominators are imperfect. At the
school years. (Henceforth, school years are referred direction of NCES, the EDFacts chronic absenteeism
to by the spring year—for example, as the 2017 counts include the number of all students enrolled
school year instead of the 2016–17 school year.) To in a site for at least 10 days who were absent for any
obtain rates, I collected the numerators (i.e., the reason for 10 percent or more of their enrollment.
number of students who were chronically absent) In contrast, most states use denominators for their
and denominators (i.e., the total number of students chronic absenteeism rates that include only stu-
eligible for inclusion in chronic absentee counts) for dents who were enrolled for a longer period of time,
each district.26 typically either at least a semester or 100 days in the
Some states posted these data only in select years. school year. In other words, the EDFacts counts for
As shown in Table A1, R2L includes state-sourced chronic absenteeism are drawn from a different, and
data for 21 states in 2017, 39 in 2018, and 43 in 2019. typically larger, pool of students than CCD enroll-
Thirty-three states provided data for 2020, and ment counts are drawn from.
43 states and Washington, DC, did for 2021; how- If all students at a site were enrolled for the entire
ever, remote instruction interfered with both atten- year, the two denominators would produce similar
dance itself and traditional attendance-collection rates. However, since many schools and districts only
data systems, making the quality and comparability count students who are enrolled for a longer period
of chronic absenteeism rates in the 2020 and 2021 of time, they will frequently have a smaller student
school years somewhat questionable. For 2022, R2L count in their enrollment data.29 Consequently, the
has state-sourced district-level data for 45 states and resulting rates lead to a potential upward bias in the
Washington, DC. EDFacts rates, and, though relatively small, that bias
Data from 2023 are still being reported, but at is likely to be higher in schools with higher intra-year
the time of publication, R2L includes state-sourced student mobility.
4
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
As such, I use estimates based on the ED Data Findings
Express collection for 2018 through 2022 only for
districts where no state-sourced data are available. Chronic absenteeism was a serious issue before the
The proportion of R2L data taken from ED Data pandemic, with meaningful variation across states
Express rather than state-sourced data are shown (Figure 1). In 2019, Alaska, at 28 percent, was the
in Table A1 and represent only a small share of dis- state with the highest rate of chronic absenteeism,
tricts in 2022 and relatively larger shares in 2018 and and only three states had average rates of 10 per-
2019. In most years, the vast majority of data were cent or less.32 (See Table A2.) The average chronic
drawn from state-sourced data. The differences in absenteeism rate across the nation was 15 percent,
average rates based on state-sourced data and the and the range across districts was rather limited. In
ED Data Express data are small but non-negligible 2019, the middle 50 percent of students attended
at the state level. I also remove all rates that exceed districts with chronic absenteeism rates between
100 percent. 9.4 and 18.8 percent, and the middle 80 percent of
I use several other data sources to compare rates students attended districts with chronic absentee-
of chronic absenteeism across types of districts. LEA ism rates between 6.4 and 25.1 percent.
location data come from CCD’s Education Demo- Figure 2 shows parallel data for 2022 and reflects
graphic and Geographic Estimates (EDGE) data the dramatic pandemic increase in chronic absen-
files. Urbanicity, school district size, and minority teeism. In 2022, Arizona, at 48 percent, had the high-
percentage figures also come from CCD data. Pov- est rate of chronic absenteeism, and Alabama, at
erty estimates for districts use Small Area Income 18 percent, had the lowest rate of chronic absentee-
and Poverty Estimates data from 2022. Duration of ism (Table A2). In 2022, the national average rate
in-person instruction in the 2021 school year and was 28 percent, an increase of 89 percent over 2019
duration of masking requirements in the 2022 school rates. The range across districts also grew by 2022,
year for about 8,500 districts come from R2L. Data when the middle 50 percent of students attended
on historical achievement come from the Stanford districts with chronic absenteeism rates between
Education Data Archive data for 2019.30 Data on the 19.3 and 36.8 percent. This 2022 interquartile range
prevalence of single-parent households come from is remarkable not only because it is larger than the
American Community Survey 2020 county-level 2019 mean but because the 25th percentile in 2022
estimates, which were translated to district mea- was far above the average chronic absenteeism in
sures using EDGE School District Geographic Rela- 2019. The range for the chronic absenteeism rates
tionship Files.31 for the middle 80 percent of students also grew for
All estimates in this report are student weighted 2022, ranging from 13.3 to 43.7 percent.
and represent the percentage of students attending Figure 3 displays 2023 data for the 39 states with
districts with a given rate of chronic absenteeism. data for 2023, arguably the first truly post-pandemic
Unweighted estimates would treat districts with school year. In 33 of 39 states, chronic absenteeism
equal weight, giving a district with 200 students the was lower in 2023 than in 2022; however, the dif-
same weight as a district with 200,000 students. ferences were relatively small. The average chronic
As such, percentages reflect the percentage of stu- absenteeism rate across these states in 2023 was
dents attending districts with a given average rate of 26 percent, down from 28 percent for the same
chronic absenteeism rather than the average of dis- 39 states in 2022. (See Table A3.)
trict rates. It is difficult to communicate the magnitude of
the change in chronic absenteeism rates during the
pandemic, but using pre-pandemic chronic absen-
teeism percentiles for districts to measure states’
post-pandemic averages is suggestive. Table 1
5
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Figure 1. State Average Chronic Absenteeism: 2019
0 48
WA
ME
MT ND
OR MN VT*
NH*
ID* WI
SD NY MA
MI CT
WY* RI
IA PA
NE NJ
NV
OH
IL IN DE
UT MD
CA CO WV
VA DC
KS MO
KY
NC
TN
AZ* OK
NM
AR* SC
AL GA
MS
TX LA*
AK
FL
HI
Note: *Estimates are based on data from Ed.gov.
Source: Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute, January 2024, https://www.
returntolearntracker.net.
Figure 2. State Average Chronic Absenteeism: 2022
0 48
WA
ME
MT ND
OR MN VT*
NH*
ID* WI
SD NY MA
MI CT
WY* RI
IA PA
NE NJ
NV
OH
IL IN DE
UT MD
CA CO WV
VA DC
KS MO
KY
NC
TN
AZ* OK
NM
AR* SC
AL GA
MS
TX LA*
AK
FL
HI
Note: *Estimates are based on data from Ed.gov.
Source: Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute, January 2024, https://www.
returntolearntracker.net.
6
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Table 1. Post-Pandemic Chronic Absenteeism in Terms of 2019 Percentiles, by State: 2022 and 2023
Percentile in 2019 Percentile in 2019 Percentile in 2019
State 2022 2023 State 2022 2023 State 2022 2023
US 96 94
AK 93 92 LA 72 92 NY 94 —
AL 94 94 MA 99 90 OH 87 83
AR 83 — MD 91 91 OK 89 91
AZ 95 — ME 99 98 OR 99 99
CA 98 97 MI 90 85 PA 87 —
CO 88 83 MN 97 — RI 78 78
CT 94 93 MO 95 92 SC 99 100
DE 92 90 MS 100 99 SD 84 84
FL 99 98 MT 91 — TN 97 —
GA 99 99 NC 99 98 TX 98 —
IA 100 98 ND 89 90 UT 98 95
ID 100 100 NE 81 80 VA 98 98
IL 91 91 NH 100 — VT 99 —
IN 92 90 NJ 88 — WA 99 99
KS 94 91 NM 99 98 WI 87 —
KY 94 95 NV 93* 93* WV 88 88
Note: *Nevada had no 2019 data, so its percentiles are based on 2018.
Source: Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute, January 2024, https://www.
returntolearntracker.net.
displays states’ average chronic absenteeism rates Similarly, if a school district’s chronic absenteeism rate
in 2022 and 2023 as the student-weighted percen- in 2019 were the same as the average 2023 rate for the
tile from 2019. In simpler terms, the columns show 39 states with 2023 data, that district would have
where the state average post-pandemic would rank ranked in the 94th percentile of most chronically
in the 2019 distribution of chronic absenteeism for absent districts in the nation.
the same state. For example, Table 1 shows that if Looking at these comparisons shows that all
an Alaska school district’s 2019 chronic absentee- states have experienced large increases in chronic
ism rate were the same as Alaska’s average chronic absenteeism. Indeed, in 2022 and 2023, the average
absenteeism rate in 2022, that district would have chronic absenteeism rate in Rhode Island, the state
ranked in the 93rd percentile of most chronically that performed best in 2022 and 2023 relative to its
absent districts in Alaska in 2019. pre-pandemic performance, was the same as Rhode
Table 1 also shows how the nation as a whole per- Island’s 78th percentile district rate for 2019.33 These
formed in 2022 and 2023. If a school district’s chronic figures demonstrate both the enormous change in
absenteeism rate in 2019 were the same as the chronic absenteeism and how modest the declines in
national average chronic absenteeism rate in 2022, chronic absenteeism between 2022 and 2023 were.
that district would have ranked in the 96th percen- While the dramatic increase in chronic absen-
tile of most chronically absent districts in the nation. teeism between 2019 and 2022 is perhaps more
7
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Figure 3. State Average Chronic Absenteeism: 2023
0 48
WA
ME
MT ND
OR MN VT
NH
ID WI
SD NY MA
MI CT
WY RI
IA PA
NE NJ
NV
OH
IL IN DE
UT MD
CA CO WV
VA DC
KS MO
KY
NC
TN
AZ OK
NM
AR SC
AL GA
MS
TX LA
AK
FL
HI
Source: Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute, January 2024, https://www.
returntolearntracker.net.
startling, the most important data displayed here What Kinds of Districts Struggle the Most?
are the partial data from 2023. The 2023 data aren’t
more important simply because they are more recent With district-level data from across the county, I
but because they show the severity of chronic absen- next describe what district characteristics are asso-
teeism after the pandemic. The 2022 high mark in ciated with higher rates of chronic absenteeism. The
chronic absenteeism came during a school year overall pattern is simple: Increases in chronic absen-
marked by the omicron strain of COVID-19, which teeism were widespread, with 97 percent of students
caused substantial peaks in COVID cases and, pre- in 2022 attending a school district whose chronic
sumably, absenteeism related to those COVID cases. absenteeism rate was higher than it was in 2019.
In 2022, when students largely attended school in However, the percentage-point increases were larger
person yet COVID cases were very high, high absen- in districts that already had higher chronic absentee-
teeism may actually have been good. However, 2023 ism rates in 2019.
was a very different year, with a much lower COVID In the remainder of this section and the following
threat but only modest improvements in chronic one, I consider how chronic absenteeism varied by a
absenteeism. This suggests that chronic absentee- number of characteristics: historical district achieve-
ism during 2023 was not primarily driven by COVID ment, district poverty, urbanicity, and more. For all
infections. For this reason, the still very high chronic the characteristics I consider, I divide districts by ter-
absenteeism rates from 2023 are an ominous omen ciles weighted by student membership. For instance,
for schools. when I examine chronic absenteeism rates by pov-
erty levels, I look at chronic absenteeism rates for the
third of districts with lower poverty levels, the third
8
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Figure 4. Change in Chronic Absenteeism Rate by District Type: 2019 to 2022
District Type
Low Achievement 17%
Middle Achievement 14%
High Achievement 10%
Low Poverty 11%
Middle Poverty 14%
High Poverty 17%
Low Minority Percentage 11%
Middle Minority Percentage 12%
High Minority Percentage 15%
Most Remote 14%
Middle Remote 13%
Least Remote 12%
Lowest Single-Parent Percentage 11%
Middle Single-Parent Percentage 14%
Highest Single-Parent Percentage 15%
Small Districts 11%
Midsize Districts 12%
Large Districts 14%
10% 12% 14% 16% 18% 20% 22% 24% 26% 28% 30% 32% 34% 36% 38%
2019 2022
Note: The figure only includes student-weighted rates for districts with data for both 2019 and 2022.
Source: Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute, January 2024, https://www.
returntolearntracker.net.
of districts with higher poverty levels, and the third with the lowest achievement, chronic absenteeism
of districts in the middle. Each of these groups is quite jumped 17 percentage points, from 20 to 36 percent,
large: Higher-poverty districts, for example, include between 2019 and 2022. Meanwhile, in the third of
roughly one-third of American K–12 students, not just districts with the highest achievement, rates doubled
students in districts where the average family lives during the same period, increasing 10 points, from
well below the poverty line. 11 to 21 percent.
As such, it is reasonable to assume that chronic Increases by poverty were similar, with rates in
absenteeism can vary not just by tercile but also within high-poverty districts increasing from 20 to 37 per-
terciles. For example, since chronic absenteeism rates cent, while rates in low-poverty districts increased
were substantially higher in the lower-achieving from 12 to 23 percent. The pattern was less stark
third of districts, it is not unreasonable to assume by the percentage of non-white students: Chronic
that chronic absenteeism rates might have been even absenteeism rose by 15 points, from 17 to 32 percent,
higher in the lowest-achieving tenth of districts. In in high-minority districts but by less in middle- and
other words, although the chronic absenteeism rates low-minority districts—from 12 to 25 percent and
shown below are conspicuous, the trends discussed from 13 to 24 percent, respectively.
below suggest that chronic absenteeism rates are even There are smaller but still substantial differ-
more conspicuous for large numbers of students. ences by other measures, such as the duration of
Increases in chronic absenteeism were most glar- remote instruction during 2021, the first full pan-
ing in districts with low pre-pandemic academic demic school year. The third of districts in the R2L
achievement. (See Figure 4.) In the third of districts instructional status dataset that were least remote
9
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
and those in the middle third saw similar increases absenteeism. COVID infections certainly explain
between 2018 and 2022, while the third of districts some of the increase in chronic absenteeism rates,
that were most remote saw slightly larger increases: especially in 2022, but they do not explain the varia-
Chronic absenteeism rates in the least remote dis- tion seen across district types nor the elevated rates
tricts increased 12 points, from 13 to 25 percent; rates seen in 2023. And third, chronic absenteeism has
in the middle third of districts increased 13 points, increased the most for disadvantaged students, those
from 13 to 26 percent; and chronic absenteeism rates who also experienced the greatest learning losses
in the most remote districts increased 14 points, during the pandemic and can least afford the harms
from 17 to 31 percent. Note that these raw differences that come with chronic absenteeism.35
may or may not be due to the duration of remote
instruction, as the duration of remote instruction
was longer in districts that had lower achievement, Chronic Absenteeism by Race
higher poverty, and other characteristics that might
explain part or all of these gaps. The previous section described how chronic absen-
Increases in chronic absenteeism were relatively teeism differed by district characteristics, but now I
weakly related to district size and county-level per- turn to examine how it varied across student char-
centage of single-parent households. Chronic absen- acteristics. The EDFacts data provide chronic absen-
teeism rates in the third of districts with the highest teeism counts for each district and school by student
percentage of single-parent households increased race for 2020, 2021, and 2022. I aggregate these num-
15 points, from 17 to 31 percent, while rates in the mid- bers to estimate chronic absenteeism rates for white,
dle third of districts increased by 14 points, from 14 to black, and Hispanic students using counts from CCD
28 percent, and rates in the lowest third of districts files. EDFacts data by student race are available begin-
increased by 11 points, from 12 to 23 percent. Addi- ning in 2020, and Table 2 compares chronic absentee-
tionally, there were no real differences between small ism rates overall and for each race group.
and midsized districts, which both had chronic absen- These data should be interpreted with caution
teeism rates that increased roughly 12 points, from for two reasons. First, the 2020 data come from the
14 to 26 percent; however, chronic absenteeism rates 2019–20 school year, which was interrupted by the
in large districts increased somewhat more, from 15 to outbreak of the pandemic and thus reflect
29 percent, or by 14 points. pre-pandemic rates, albeit with some uncertainty.
As shown on the R2L website, these patterns tend Second, as previously noted, the numerators from
to be similar within states, but these patterns do not EDFacts and denominators from CCD are not well
hold for all states. The R2L website also shows the matched. Nonetheless, these data provide the best
same patterns of change not only from 2018 to 2022 available look on race differences nationwide.
but also from 2018 to 2023.34 By and large, declines in The changes in chronic absenteeism rates by race
chronic absenteeism rates in 2023 were slightly larger resemble the patterns in chronic absenteeism rates by
for the types of districts that had the largest increases district type: Race groups with higher pre-pandemic
between 2018 and 2022, but these differences are rel- rates also had the highest percentage-point increases.
atively small. In the 78,000 schools that have chronic absentee-
There are three broad and sobering conclusions ism data sorted by race for both 2020 and 2022, rates
that can be drawn from these comparisons. First, increased from 7 to 15 percent among Asian students
the spike in chronic absenteeism is both severe and and from 11 to 24 percent among white students. Mean-
widespread, with no district type spared an increase while, increases in chronic absenteeism for both black
of fewer than 10 points, or a proportional increase of and Hispanic students were more than 20 percentage
75 percent. Second, it does not appear that COVID points: Among Hispanic students, chronic absentee-
infections can fully explain these increases in chronic ism increased from 16 to 36 percent, and among black
10
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Table 2. Chronic Absenteeism by Student Race: 2020 and 2022
Percentage- Percentage N 2020 N 2022
2020 2022 Point Change Increase (Millions) (Millions)
All Students 14% 29% 15.5 114% 41.8 41.7
Asian 7% 16% 8.2 112% 1.8 1.8
Black 19% 39% 20.7 112% 6.7 6.5
Hispanic 15% 36% 21.1 139% 10.2 10.1
White 11% 24% 12.8 114% 20.9 19.7
Note: Data come from 78,000 schools with data in both 2020 and 2022.
Source: Author’s calculations using Ed Data Express, Chronically Absent Students School Data, 2019–20 and 2021–22; and National
Center for Education Statistics, Common Core of Data, Public Elementary/Secondary School Universe Survey Data, version 1a, https://
nces.ed.gov/ccd/pubschuniv.asp.
students, chronic absenteeism increased from 18 to more advantaged terciles from the middle and less
39 percent. advantaged ones.
These averages by race also differ across district Because black and Hispanic students are not evenly
types. Table 3 displays changes between 2020 and distributed across terciles, these patterns are even
2022 for Asian, black, Hispanic, and white students worse for black and Hispanic students than Table 3
by district terciles of poverty, pre-pandemic achieve- suggests. Table A5 shows that the low-poverty and
ment,36 urbanicity, and remote instructional sta- high-achievement terciles—which are defined based
tus during the 2021 school year. Changes in chronic on the entire population of students—include rela-
absenteeism by district characteristic affected each tively smaller shares of black and Hispanic students.
race group similarly, with larger percentage-point For example, 54 percent of black students go to school
increases for districts that had higher rates of chronic in low-achievement districts, while just 22 percent of
absenteeism pre-pandemic—that is, for high-poverty, white students do.
lower-historic-achievement, and urban districts and Suburban and rural rates of chronic absenteeism
for districts that offered extended remote instruction within a race group were similar for Asian, black, and
in 2021. Hispanic students in 2022, but urban rates were much
With rates near or above 40 percent, chronic higher for all three groups of students. For example,
absenteeism in 2022 was greatest for black and His- in 2022 the chronic absenteeism rate for Asian stu-
panic students in the highest-poverty and lowest- dents was 13 percent in both suburban and rural dis-
achievement terciles. However, middle-poverty and tricts but 20 percent in urban districts.
middle-achievement terciles also saw substantial Differences in chronic absenteeism rates by 2021
gains in chronic absenteeism, leading to 2022 rates remote instructional status also show markedly dif-
that were similar to rates in the highest-poverty and ferent rates for Asian, black, and Hispanic students.
lowest-achievement terciles and much higher than In the most remote districts, chronic absenteeism
the rates in the lowest-poverty and highest- rates were markedly higher for Asian, black, and His-
achievement terciles. The pattern for Asian and white panic students than they were in the least remote and
students is somewhat similar, though less consistent middle third of districts, though a similar pattern of
and with smaller absolute differences. In general, smaller differences existed before the pandemic. For
chronic absenteeism rates in 2022 show a demarca- white students, there were relatively small differences
tion by poverty and achievement that separates the by urbanicity and remote instructional status.
11
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Table 3. Chronic Absenteeism by Student Race and District Type: 2020 and 2022
2020 2022
Poverty Low Middle High Low Middle High
Asian 7% 6% 9% 13% 16% 19%
Black 16% 18% 19% 32% 40% 41%
Hispanic 16% 15% 15% 32% 38% 38%
White 10% 11% 13% 20% 25% 27%
Achievement Low Middle High Low Middle High
Asian 9% 9% 6% 21% 20% 11%
Black 20% 18% 12% 43% 38% 27%
Hispanic 16% 16% 13% 40% 37% 29%
White 16% 12% 8% 31% 26% 19%
Town or Town or
District Urbanicity Urban Suburban Rural Urban Suburban Rural
Asian 10% 6% 6% 20% 13% 13%
Black 23% 16% 12% 46% 34% 32%
Hispanic 17% 14% 12% 41% 34% 32%
White 13% 10% 12% 26% 21% 25%
Most Least Most Least
2021 Instruction Remote Middle Remote Remote Middle Remote
Asian 9% 7% 4% 18% 13% 13%
Black 21% 17% 15% 44% 36% 34%
Hispanic 19% 14% 12% 41% 34% 32%
White 13% 11% 10% 25% 23% 24%
Note: The table excludes California entirely due to missing 2020 data, and it excludes districts with missing achievement data from Stan-
ford Education Data Archive data and 2021 instructional data from AEI’s Return to Learn instructional tracker. Sample sizes are shown in
Table A5.
Source: Author’s calculations using Ed Data Express, Chronically Absent Students School Data, 2019–20 and 2021–22; and National
Center for Education Statistics, Common Core of Data, Public Elementary/Secondary School Universe Survey Data, version 1a, https://
nces.ed.gov/ccd/pubschuniv.asp.
These data make clear that chronic absenteeism students and parents, teachers and principals, and
has dramatically increased for all students. They district, state, and federal leaders will do to help stu-
also make clear that the chronic absenteeism crisis dents attend school more consistently again.
is dramatically more severe for black and Hispanic
students, for students in historically disadvantaged
districts, and even more so for black and Hispanic Discussion
students in historically disadvantaged districts. While
these data come from 2022, the apex of pandemic While it may seem premature to publish a report on
chronic absenteeism, the limited improvements we chronic absenteeism in January 2024, when data for
have seen in the available data from 2023 suggest many states are still outstanding, the urgent need
these patterns are durable. The question now is what to address this problem during the 2024 school year
12
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
outweighs the upside of waiting for more complete academic recovery and further widen achievement
data. The reason the 2023 data are so important is gaps in the coming years.
simple. While complete 2022 data showing chronic A viable response to the problem of chronic
absenteeism rates are important in their own right— absenteeism will have to match not only its scale
especially given the alarming differentials described and severity but also its nature. While the nature
above—they show rates during the pandemic. Indeed, of pandemic learning loss requires instructional
elevated chronic absenteeism rates in 2022, when the solutions—likely in terms of both instructional
pandemic was at its omicron height, were concern- approaches and instructional time—post-pandemic
ing but perhaps warranted, and one could have rea- chronic absenteeism looks more like a cultural prob-
sonably hoped that they would not be predictive of lem. During the pandemic, altered school practices
post-pandemic trends. loosened established norms for school attendance,
Unfortunately, the 2023 data clearly show that and over the past few years, students and families
such hopes would have been unfounded. While have grown accustomed to these new norms. The
chronic absenteeism did decline in 33 of the 39 states pandemic surge in chronic absenteeism may not sup-
with available 2023 data, those declines are far too port this cultural diagnosis, but the post-pandemic
small. Even if rates were to fall at the same pace for durability of that surge does.
the foreseeable future, public schools would not To the extent that post-pandemic chronic absen-
return to pre-pandemic rates—rates that were already teeism reflects a cultural problem for schools,
problematic—until roughly 2030. efforts to fix the problem should target cultural
But even then, there is no obvious rationale for change. Before the pandemic, many interventions
supposing that chronic absenteeism would fall at that proved effective at reducing chronic absen-
such a rate instead of remaining at or near current teeism had small marginal effects, which, though
levels. Indeed, there were strong reasons to expect worthwhile, are not likely sufficient for address-
chronic absenteeism rates to fall between 2022, ing the current problem schools and districts are
when COVID case rates were high, and 2023, the first facing. Many of the antecedents of chronic absen-
post-pandemic school year, but what strong reasons teeism frequently mentioned before the pandemic—
are there to suppose that chronic absenteeism will fall including student employment, bullying, transporta-
at a similar rate between 2023 and 2024 or between tion, and various markers of disadvantage—remain
2028 and 2029? Why not suppose instead that 2023 worthwhile concerns after the pandemic, but there is
chronic absenteeism rates are the new post-pandemic no evidence these have increased to the extent that
baseline? In short, surging chronic absenteeism is not chronic absenteeism has. Addressing the current
a pandemic problem but a post-pandemic problem, chronic absenteeism crisis with the pre-pandemic
one that educators will need to grapple with in the playbook does not look like a viable solution to the
coming years. root problem facing schools today.
Chronic absenteeism harms students even in nor- What might more plausible antecedents of post-
mal times, but it is even worse in the context of pan- pandemic chronic absenteeism include? Though more
demic learning loss because it will hamper academic research is required, it seems plausible that the pan-
recovery and likely already has.37 Moreover, the demic weakened the culture of attendance in several
groups of students with the highest rates of chronic ways: Remote schooling and quarantines removed
absenteeism—black and Hispanic students and students from the routines of going to school (i.e., in
those in low-achievement and high-poverty school a school building) and parents from the routines of
districts—are the same students whose pandemic making sure their children get to school, disrupted
learning losses tend to be larger.38 Chronic absentee- learning and lower-quality academic offerings weak-
ism is arguably the number-one problem facing pub- ened parents’ and students’ perceptions of the impor-
lic schools today, and it will likely hamper pandemic tance of school and sense of obligation regarding
13
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
school attendance, pandemic-era online instructional stop. Certainly, such proclamations won’t be suffi-
supports made the consequences of missing school cient for cultural change in schools, but they will
seem less consequential, and being away from teach- provide the political cover and moral standing for
ers and peers worsened students’ sense of belong- schools to mount muscular responses to the chronic
ing at school. If this assessment is correct, turning absenteeism crisis. Governors can provide cover for
the tide will require conspicuous efforts to return district superintendents and subsequently super
to pre-pandemic norms that parents, students, and intendents for principals to push students and fam-
schools adhered to just a few years ago. ilies to change.
This enormous challenge demands a response Absent from the list above are teachers, but only
of equal magnitude. Though this report focuses on because they warrant their own mention. Teachers
describing the problem, some observations regard- are an indispensable component of the change that
ing what may be part of a necessary, if not sufficient, must happen in schools. No combination of coun-
treatment are warranted. The first is that school selors, central office staff, emailed announcements,
districts will need to use both carrots and sticks truancy officers, and superintendents will be able to
to address the problem. Surely that means adding provide the effective communication and demon-
supports targeting student belonging, transporta- strate the concern that students’ teachers can. We can
tion challenges, communication efforts, and posi- empathize with beleaguered teachers who bear sub-
tive reinforcements, but those inducements should stantial daily burdens and have limited capacity for
be supported with expectations and appropriate another responsibility and still see that their engage-
consequences for both truant students and parents ment is necessary for substantial cultural change in
who prove unwilling to fulfill their moral and legal schools and the accompanying improvements in stu-
obligations to make sure their child attends school dent attendance.
regularly. This generation of students already faces
setbacks too large to be satisfied with marginal inter-
ventions, however well-intentioned. About the Author
Ameliorating this cultural challenge will be diffi-
cult without bold leadership. Leaders—from prin- Nat Malkus is a senior fellow and the deputy direc-
cipals to the president—should use the bully pulpit tor of education policy at the American Enterprise
to make plain that, just as the pandemic is over, Institute, where he specializes in empirical research
pandemic-era exceptionalism in schools must also on K–12 schooling and, since the COVID-19 pan-
be over, meaning that school attendance must return demic struck, has tracked its various impacts on
to normal and that other unwanted behaviors must schools.
14
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Appendix
Table A1. Percentage of R2L Chronic Absenteeism Data Drawn from EDFacts, by State: 2018–22
2018 2019 2020 2021 2022 2018 2019 2020 2021 2022
AL 0% 0% 100% 0% 0% ND 5% 5% 5% 5% 0%
AR 100% 100% 100% 100% 100% NE 100% 1% 0% 0% 0%
AZ 100% 100% 100% 100% 100% NH 100% 100% 100% 100% 100%
CA 0% 3% 100% 4% 4% NJ 0% 0% 100% 0% 0%
CO 0% 0% 2% 0% 0% NM 100% 0% 0% 0% 0%
DC 0% 100% 100% 0% 0% NV 100% 0% 100% 0% 0%
DE 0% 0% 100% 3% 3% NY 2% 3% 100% 4% 4%
HI 100% 0% 0% 0% 0% OH 1% 2% 2% 1% 3%
IA 100% 100% 0% 0% 0% OK 3% 4% 100% 100% 7%
ID 100% 100% 0% 2% 6% OR 0% 0% 100% 0% 0%
KY 0% 0% 100% 0% 0% SC 1% 1% 3% 0% 1%
LA 100% 100% 100% 100% 4% SD 0% 0% 100% 0% 0%
MD 0% 0% 0% 1% 0% TN 1% 0% 0% 0% 0%
MI 1% 1% 1% 16% 1% TX 100% 0% 0% 0% 0%
MN 0% 0% 100% 100% 0% VT 0% 33% 39% 38% 100%
MO 0% 1% 1% 0% 1% WA 0% 0% 1% 0% 0%
MS 1% 1% 100% 1% 2% WY 100% 100% 100% 100% 100%
NC 3% 4% 4% 5% 5%
Note: No EDFacts data were available for 2017 or 2023 or used for 2018–22 in Arkansas, Connecticut, Florida, Georgia, Illinois, Indiana,
Kansas, Massachusetts, Maine, Montana, Pennsylvania, Rhode Island, Utah, Virginia, Wisconsin, or West Virginia.
Source: Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute, January 2024, https://www.
returntolearntracker.net.
15
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Table A2. Chronic Absenteeism Rates by State: 2017–23
School Year (Spring) School Year (Spring)
State 2017 2018 2019 2020 2021 2022 2023 State 2017 2018 2019 2020 2021 2022 2023
US 13% 15% 15% 13% 19% 28% 26% MS — 17% 14% 9% 21% 28% 24%
AK 24% 26% 28% 17% 22% 46% 43% MT — 21% 23% 14% 27% 38% —
AL 18% 18% 11% 9% 8% 18% 18% NC — 15% 16% 9% 26% 32% 27%
AR* — 19% 19% 23% 16% 27% — ND — 11% 12% 11% 16% 21% 19%
AZ* — 25% 25% 16% 25% 48% — NE — 16% 15% 15% 21% 24% 22%
CA 11% 12% 13% — 14% 30% 25% NH* — 16% 14% 13% 18% 35% —
CO 19% 24% 23% 23% 26% 35% 31% NJ 10% 11% 11% 8% 13% 18% —
CT — 11% 10% 12% 19% 24% 20% NM — 17% 17% 15% 30% 41% 40%
DC — 29% 33% 29% 31% 48% 44% NV — 20% 19% 20% 31% 36% 35%
DE 11% 11% 11% 17% 18% 20% 20% NY — 18% 19% 21% 24% 32% —
FL — 20% 20% 17% 25% 32% 31% OH 13% 15% 16% 10% 23% 29% 26%
GA — 13% 13% 9% 21% 25% 23% OK — 13% 13% 18% 14% 19% 21%
HI — 19% 19% 12% 18% 37% 30% OR 20% 20% 20% 22% 28% 36% 38%
IA — 11% 12% 17% 21% 26% 21% PA 15% 15% 15% 15% 18% 26% —
ID* — 3% 3% 8% 16% 21% 18% RI 19% 19% 19% 18% 28% 34% 29%
IL — 17% 18% 11% 21% 30% 28% SC — 12% 13% 14% 8% 20% 25%
IN 10% 12% 12% 11% 20% 23% 21% SD — 13% 14% 11% 18% 22% 21%
KS — 14% 14% 14% 18% 26% 23% TN — 13% 13% 13% 15% 20% —
KY — 18% 18% 12% 22% 28% 30% TX — 13% 11% 7% 15% 26% —
LA* — 16% 16% 11% 25% 18% 22% UT 12% 13% 14% 9% 19% 29% 25%
MA 13% 13% 13% 13% 18% 28% 24% VA 12% 11% 11% 12% 11% 20% 19%
MD — 20% 20% 18% 24% 31% 30% VT* 11% 12% 14% 16% 22% 35% —
ME 16% 16% 17% 18% 21% 30% 26% WA 15% 15% 15% 9% 19% 33% 30%
MI 16% 21% 20% 21% 20% 39% 32% WI 12% 13% 13% 13% 16% 22% —
MN 13% 14% 14% 15% 17% 30% — WV — 19% 20% 19% 13% 29% 28%
MO 11% 12% 13% 15% 17% 24% 23% WY* — 2% 2% 2% 2% 37% —
Note: * Estimates for these states are based in whole or in part on data from Ed.gov.
Source: Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute, January 2024, https://www.
returntolearntracker.net.
16
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Table A3. State Data–Sourced Chronic Absenteeism Rates by State: 2017–23
School Year (Spring) School Year (Spring)
State 2017 2018 2019 2020 2021 2022 2023 State 2017 2018 2019 2020 2021 2022 2023
US 13% 15% 15% 12% 19% 28% 26% MS — 17% 14% — 21% 28% 24%
AK 24% 26% 28% 17% 22% 46% 43% MT — 21% 23% 14% 27% 38% —
AL 18% 18% 11% — 8% 18% 18% NC — 15% 16% 10% 27% 32% 27%
CA 11% 12% 13% — 15% 30% 25% ND — 11% 12% 10% 15% 21% 19%
CO 19% 24% 23% 23% 26% 35% 31% NE — — 15% 15% 21% 24% 22%
CT — 11% 10% 12% 19% 24% 20% NJ 10% 11% 11% — 13% 18% —
DC — 29% — — 31% 48% 44% NM — — 17% 15% 30% 41% 40%
DE 11% 11% 11% — 18% 21% 20% NV — — 19% — 31% 36% 35%
FL — 20% 20% 17% 25% 32% 31% NY — 18% 19% — 24% 32% —
GA — 13% 13% 9% 21% 25% 23% OH 13% 15% 16% 10% 23% 29% 26%
HI — — 19% 12% 18% 37% 30% OK — 13% 13% — — 19% 21%
IA — — — 17% 21% 26% 21% OR 20% 20% 20% — 28% 36% 38%
ID — — — 8% 16% 22% 18% PA 15% 15% 15% 15% 18% 26% —
IL — 17% 18% 11% 21% 30% 28% RI 19% 19% 19% 18% 28% 34% 29%
IN 10% 12% 12% 11% 20% 23% 21% SC — 12% 13% 14% 8% 20% 25%
KS — 14% 14% 14% 18% 26% 23% SD — 13% 14% — 18% 22% 21%
KY — 18% 18% — 22% 28% 30% TN — 13% 13% 13% 15% 20% —
LA — — — — — 18% 22% TX — — 11% 7% 15% 26% —
MA 13% 13% 13% 13% 18% 28% 24% UT 12% 13% 14% 9% 19% 29% 25%
MD — 20% 20% 18% 24% 31% 30% VA 12% 11% 11% 12% 11% 20% 19%
ME 16% 16% 17% 18% 21% 30% 26% VT 11% 12% 13% 16% 23% — —
MI 16% 21% 20% 21% 24% 39% 32% WA 15% 15% 15% 9% 18% 33% 30%
MN 13% 14% 14% — — 30% — WI 12% 13% 13% 13% 16% 22% —
MO 11% 12% 13% 15% 17% 24% 23% WV — 19% 20% 19% 13% 29% 28%
Note: Only data taken from state departments of education are included in this table.
Source: Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute, January 2024, https://www.
returntolearntracker.net.
17
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Table A4. Change in Chronic Absenteeism Rate by District Type: 2018–23
District Type 2018 2019 2022 2023
Low 19% 20% 36% 33%
Achievement Medium 16% 16% 30% 28%
High 11% 11% 21% 21%
Low 12% 12% 23% 22%
Poverty Medium 16% 15% 29% 28%
High 20% 20% 37% 33%
Low 14% 13% 24% 22%
Minority
Medium 13% 12% 25% 23%
Percentage
High 17% 17% 32% 30%
Most Remote 17% 17% 31% 29%
In-Person Instruction
Middle 13% 13% 26% 24%
in 2021
Least Remote 13% 13% 25% 23%
Low 12% 12% 23% 22%
Single-Parent
Medium 15% 14% 28% 26%
Percentage
High 17% 17% 31% 29%
Small 15% 15% 26% 24%
Size Medium 15% 14% 26% 24%
Large 15% 15% 29% 27%
Source: Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute, January 2024, https://www.
returntolearntracker.net.
18
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Table A5. Ns for Table 3. Chronic Absenteeism by Student Race: 2020 and 2022
2020 2022
Poverty Low Middle High Low Middle High
Asian 899,000 468,000 421,000 891,000 454,000 395,000
Black 1,417,000 1,877,000 2,798,000 1,392,000 1,815,000 2,636,000
Hispanic 2,379,000 3,098,000 3,868,000 2,441,000 3,101,000 3,758,000
White 7,970,000 6,930,000 5,312,000 7,469,000 6,548,000 4,998,000
Achievement Low Middle High Low Middle High
Asian 310,000 479,000 971,000 285,000 455,000 972,000
Black 3,137,000 1,823,000 1,127,000 2,951,000 1,753,000 1,128,000
Hispanic 3,877,000 3,381,000 1,950,000 3,780,000 3,357,000 2,015,000
White 3,841,000 6,925,000 9,177,000 3,560,000 6,517,000 8,682,000
District Urbanicity Urban Suburban Rural Urban Suburban Rural
Asian 644,000 1,093,000 121,000 605,000 1,090,000 119,000
Black 2,860,000 2,672,000 1,026,000 2,705,000 2,625,000 988,000
Hispanic 3,931,000 4,174,000 1,685,000 3,821,000 4,228,000 1,737,000
White 3,655,000 9,004,000 7,989,000 3,361,000 8,418,000 7,675,000
Most Most In- Most Most In-
2021 Instruction Remote Middle Person Remote Middle Person
Asian 929,000 446,000 386,000 887,000 440,000 387,000
Black 2,935,000 1,327,000 1,806,000 2,767,000 1,289,000 1,762,000
Hispanic 3,675,000 2,330,000 3,178,000 3,598,000 2,355,000 3,180,000
White 5,026,000 7,302,000 6,668,000 4,573,000 6,889,000 6,379,000
Note: The table excludes California entirely due to missing 2020 data, and it excludes districts with missing achievement data from
Stanford Education Data Archive data and 2021 instructional data from AEI’s Return to Learn instructional tracker.
Source: Author’s calculations using Ed Data Express, Chronically Absent Students School Data, 2019–20 and 2021–22; and National
Center for Education Statistics, Common Core of Data, Public Elementary/Secondary School Universe Survey Data, version 1a, https://
nces.ed.gov/ccd/pubschuniv.asp.
19
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
Notes
1. Donna St. George, “National Test Scores Plunge, with Still No Sign of Pandemic Recovery,” Washington Post, June 21, 2023,
https://www.washingtonpost.com/education/2023/06/21/national-student-test-scores-drop-naep.
2. US Department of Education, Institute of Education Sciences, National Center for Education Statistics, “More Than 80 Percent
of U.S. Public Schools Report Pandemic Has Negatively Impacted Student Behavior and Socio-Emotional Development,” press release,
July 6, 2022, https://nces.ed.gov/whatsnew/press_releases/07_06_2022.asp.
3. US Department of Education, Institute of Education Sciences, National Center for Education Statistics, “Source of the Data and
Accuracy of the Estimates for the School Pulse Panel: May 2022 Data Collection,” July 2023, https://ies.ed.gov/schoolsurvey/spp/SPP_
May2022.xlsx.
4. US Department of Education, “Fact Sheet: Biden-Harris Administration Announces New Actions to Build on Work to Accelerate
Learning and Improve Student Achievement,” press release, September 13, 2023, https://www.ed.gov/news/press-releases/fact-sheet-
biden-harris-administration-announces-new-actions-build-work-accelerate-learning-and-improve-student-achievement.
5. Editorial Board, “The Startling Evidence on Learning Loss Is In,” New York Times, November 18, 2023, https://www.nytimes.
com/2023/11/18/opinion/pandemic-school-learning-loss.html.
6. Los Angeles Unified School District, “Los Angeles Unified Continues Addressing Pandemic Learning Loss (10-17-23),” press
release, October 17, 2023, https://www.lausd.org/site/default.aspx?PageType=3&DomainID=4&ModuleInstanceID=4466&ViewID=
6446EE88-D30C-497E-9316-3F8874B3E108&RenderLoc=0&FlexDataID=138488&PageID=1.
7. Hedy Chang, “Absent an Education,” US News & World Report, June 8, 2016, https://www.usnews.com/opinion/articles/
2016-06-08/chronic-absenteeism-is-a-national-crisis; and US Department of Education, “Chronic Absenteeism in the Nation’s
Schools,” January 2019, https://www2.ed.gov/datastory/chronicabsenteeism.html.
8. Previously, Return to Learn reported this 2022 percentage as 29 percent, in part based on federal data. Texas data for 2022
reduced that percentage by 0.3 percentage points, enough to round the 2022 estimate down to 28 percent, which is based on the most
current state data.
9. As of publication in late January 2024, New Jersey, New York, Pennsylvania, Tennessee, and Texas have yet to release data for
2022–23. Several other states, including Arkansas, Montana, New Hampshire, and Wyoming, do not report chronic absenteeism rates,
and Return to Learn estimates these states’ rates using federal data, as discussed in the methods section.
10. Hedy N. Chang and Mariajosé Romero, Present, Engaged, and Accounted For: The Critical Importance of Addressing Chronic
Absence in the Early Grades, National Center for Children in Poverty, September 2008, https://www.nccp.org/wp-content/
uploads/2008/09/text_837.pdf.
11. Chang and Romero, Present, Engaged, and Accounted For.
12. Robert Balfanz and Vaughan Byrnes, The Importance of Being in School: A Report on Absenteeism in the Nation’s Public Schools,
Johns Hopkins University Center for Social Organization of Schools, May 2012, https://new.every1graduates.org/wp-content/
uploads/2012/05/FINALChronicAbsenteeismReport_May16.pdf.
13. Emma García and Elaine Weiss, Student Absenteeism: Who Misses School and How Missing School Matters for Performance, Eco-
nomic Policy Institute, September 25, 2018, https://files.eric.ed.gov/fulltext/ED593361.pdf.
14. Balfanz and Byrnes, The Importance of Being in School.
15. Chang and Romero, Present, Engaged, and Accounted For; García and Weiss, Student Absenteeism; and Mariajosé Romero and
Young-Sun Lee, A National Portrait of Chronic Absenteeism in the Early Grades, Columbia University National Center for Children in
Poverty, October 2007, https://academiccommons.columbia.edu/doi/10.7916/D89C7650.
16. Chuansheng Chen and Harold W. Stevenson, “Motivation and Mathematics Achievement: A Comparative Study of Asian-
American, Caucasian-American, and East Asian High School Students,” Child Development 66, no. 4 (August 1995): 1215–34, https://srcd.
onlinelibrary.wiley.com/doi/abs/10.1111/j.1467-8624.1995.tb00932.x; Michael A. Gottfried, “Excused Versus Unexcused: How Student
Absences in Elementary School Affect Academic Achievement,” Educational Evaluation and Policy Analysis 31, no. 4 (2009): 392–415,
20
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
https://journals.sagepub.com/doi/10.3102/0162373709342467; and Michael A. Gottfried and J. Jacob Kirksey, “‘When’ Students Miss
School: The Role of Timing of Absenteeism on Students’ Test Performance,” Educational Researcher 46, no. 3 (April 2017): 119–30,
https://journals.sagepub.com/doi/epub/10.3102/0013189X17703945.
17. García and Weiss, Student Absenteeism.
18. Esteban M. Aucejo and Teresa Foy Romano, “Assessing the Effect of School Days and Absences on Test Score Performance,”
Economics of Education Review 55 (December 2016): 70–87, https://www.sciencedirect.com/science/article/abs/pii/S0272775715301655;
Gottfried, “Excused Versus Unexcused”; Michael A. Gottfried, “The Detrimental Effects of Missing School: Evidence from Urban Sib-
lings,” American Journal of Education 117, no. 2 (February 2011): 147–82, https://www.journals.uchicago.edu/doi/10.1086/657886; Michael
A. Gottfried, “Chronic Absenteeism and Its Effects on Students’ Academic and Socioemotional Outcomes,” Journal of Education for
Students Placed at Risk 19, no. 2 (2014): 53–75, https://www.tandfonline.com/doi/abs/10.1080/10824669.2014.962696; Lucrecia Santi-
bañez and Cassandra M. Guarino, “The Effects of Absenteeism on Academic and Social-Emotional Outcomes: Lessons for COVID-19,”
Educational Researcher 50, no. 6 (August/September 2021): 392–400, https://journals.sagepub.com/doi/epub/10.3102/0013189X21994488;
García and Weiss, Student Absenteeism; and Seth Gershenson, Alison Jacknowitz, and Andrew Brannegan, “Are Student Absences
Worth the Worry in U.S. Primary Schools?,” Education Finance and Policy 12, no. 2 (Spring 2017): 137–65, https://direct.mit.edu/edfp/
article-abstract/12/2/137/10267/Are-Student-Absences-Worth-the-Worry-in-U-S.
19. Santibañez and Guarino, “The Effects of Absenteeism on Academic and Social-Emotional Outcomes”; Aucejo and Romano,
“Assessing the Effect of School Days and Absences on Test Score Performance”; Kate Finning et al., “Review: The Association Between
Anxiety and Poor Attendance at School—a Systematic Review,” Child and Adolescent Mental Health 24, no. 3 (September 2019): 205–16,
https://pubmed.ncbi.nlm.nih.gov/32677217; and Arya Ansari and Michael A. Gottfried, “Early Childhood Educational Settings and
School Absenteeism for Children with Disabilities,” AERA Open 4, no. 2 (April 2018), https://journals.sagepub.com/doi/epub/
10.1177/2332858418785576.
20. Balfanz and Byrnes, The Importance of Being in School; David Silver, Marisa Saunders, and Estela Zarate, “What Factors Predict
High School Graduation in the Los Angeles Unified School District?,” University of California, Santa Barbara, June 2008, https://www.
issuelab.org/resources/11619/11619.pdf; Elaine Allensworth and John Q. Easton, “What Matters for Staying On-Track and Graduating
in Chicago Public Schools,” University of Chicago Consortium on School Research, July 2007, https://consortium.uchicago.edu/
publications/what-matters-staying-track-and-graduating-chicago-public-schools; and Baltimore Education Research Consortium, Des-
tination Graduation: Sixth Grade Early Warning Indicators for Baltimore City Schools: Their Prevalence and Impact, February 2011,
https://baltimore-berc.org/pdfs/SixthGradeEWIFullReport.pdf.
21. Gottfried, “Chronic Absenteeism in the Classroom Context.”
22. For example, see Balfanz and Byrnes, The Importance of Being in School; and Romero and Lee, A National Portrait of Chronic
Absenteeism in the Early Grades.
23. Todd Rogers and Avi Feller, “Reducing Student Absences at Scale by Targeting Parents’ Misbeliefs,” Nature Human Behaviour 2
(2018): 335–42, https://www.nature.com/articles/s41562-018-0328-1; Ken Smythe-Leistico and Lindsay C. Page, “Ready . . . Set . . . Text!
Reducing School Absenteeism Through Parent-School Two-Way Text Messaging,” in Absent from School: Understanding and Address-
ing Student Absenteeism, ed. Michael A. Gottfried and Ethan L. Hutt (Cambridge, MA: Harvard Education Press, 2019), 165–80; and
Carly D. Robinson et al., “Reducing Student Absenteeism in the Early Grades by Targeting Parental Beliefs,” American Educational
Research Journal 55, no. 6 (2018): 1163–92, https://journals.sagepub.com/doi/10.3102/0002831218772274.
24. Michael A. Gottfried and Jay Stratte Plasman, “Linking the Timing of Career and Technical Education Coursetaking with High
School Dropout and College-Going Behavior,” American Educational Research Journal 55, no. 2 (2018): 325–61, https://journals.sagepub.
com/doi/abs/10.3102/0002831217734805; and Joshua Childs and Ain A. Grooms, “Improving School Attendance Through Collabora-
tion: A Catalyst for Community Involvement and Change,” Journal of Education for Students Placed at Risk 23, no. 1–2 (2018): 122–38,
https://eric.ed.gov/?id=EJ1174330.
25. Jeremy Singer et al., “Advancing an Ecological Approach to Chronic Absenteeism: Evidence from Detroit,” Teachers College
Record 123, no. 4 (April 2021), https://journals.sagepub.com/doi/10.1177/016146812112300406.
21
LONG COVID FOR PUBLIC SCHOOLS NAT MALKUS
26. Not all state provided both numerators and denominators. Where necessary, I algebraically derived numerators or denominators
when rates were available but either numerators or denominators were unavailable.
27. For the most up-to-date data, see Return to Learn Tracker, “Chronic Absenteeism: 2017–2023,” American Enterprise Institute,
www.ReturntoLearnTracker.net/ChronicAbsenteeism.
28. US Department of Education, Ed Data Express, website, https://eddataexpress.ed.gov.
29. Ideally, this report would use enrollment counts that include all students who are enrolled for at least 10 days, but such data are
not currently available from any source.
30. Educational Opportunity Project at Stanford University, website, https://edopportunity.org.
31. US Department of Education, Institute of Education Sciences, National Center for Education Statistics, “School District Geo-
graphic Relationship Files,” https://nces.ed.gov/programs/edge/geographic/relationshipfiles.
32. Washington, DC, had a higher rate in 2019, at 33 percent, but it is not a state and thus a poor comparison. Of those with state-
supplied data, Connecticut had the lowest rate at 10 percent in 2019. See Table A2.
33. Unfortunately, Rhode Island’s performance on this metric is largely due to its unusually high rates in 2019.
34. See these data at Return to Learn Tracker, website, https://www.returntolearntracker.net.
35. Education Opportunity Project at Stanford University, “2019–2022 Change in Average Math Achievement in the U.S.,” https://
edopportunity.org/recovery/#/chart/none/districts/mth1922/frl/all/3.15/37.39/-96.78.
36. District terciles are student weighted and thus include similar numbers of students in each.
37. Karyn Lewis and Megan Kuhfeld, “Education’s Long COVID: 2022–23 Achievement Data Reveal Stalled Progress Toward Pan-
demic Recovery,” Center for School and Student Progress, Northwest Evaluation Association, July 2023, https://www.nwea.org/
uploads/Educations-long-covid-2022-23-achievement-data-reveal-stalled-progress-toward-pandemic-recovery_NWEA_Research-
brief.pdf.
38. Education Opportunity Project at Stanford University, “2019–2022 Change in Average Math Achievement in the U.S.”
© 2024 by the American Enterprise Institute for Public Policy Research. All rights reserved.
The American Enterprise Institute (AEI) is a nonpartisan, nonprofit, 501(c)(3) educational organization and
does not take institutional positions on any issues. The views expressed here are those of the author(s).
22
File and source
- File
- long-covid-for-public-schools-chronic-absenteeism-before-and-after-the-pandemic.pdf
- Size
- 6,879,263 bytes
- SHA-256
- 6251b46c7c7194a1c217aee9f05727ee0aa09ffec90b0fcdb7c412938c63fc83
- Original
- www.aei.org