Chronic Absenteeism: 2017–2023

Consistent attendance is key to student success, but post-pandemic attendance has been far from consistent. Nationwide, chronic absenteeism—the percentage of students missing at least 10% of a school year—surged from 15% in 2018 to 28% in 2022, and remained high in 2023. Surging chronic absenteeism clearly stemmed from the pandemic, but limited data has obscured where, and for whom, the change is greatest.

Return to Learn’s (R2L) chronic absenteeism data span districts in 50 states from 2016–17 to 2022–23, where available, for the most current and comprehensive chronic absenteeism data available anywhere. Click for more on R2L’s methods, and scroll down to see how chronic absenteeism changed from 2017 to 2023.

Mapping Chronic Absenteeism: 2017–2023

R2L last added WI &NY 2022-23 data. Last update: Mar 21, 2024

Click Here For Key Findings
  • Chronic absenteeism includes excused and unexcused absences, and naturally rose during the pandemic.
  • In 2018 and 2019, chronic absenteeism held at 15% nationally. By 2022 chronic absenteeism hit 28%, 88% higher than the average rate three years earlier.
  • Down in 34 of 40 states reporting data, chronic absenteeism fell in 2023, but still remained 78% higher than the national pre-pandemic baseline.
  • In 2022, every state except Alabama had average chronic absenteeism that exceeded that state’s 75th percentile in 2018.
  • In 35 of 47 states analyzed, average chronic absenteeism in 2022 surpassed that state’s 90th percentile in 2018.
  • In historically low-achieving districts, chronic absenteeism increased 17 points—from 19% in 2018 to 36% in 2022. In high-achieving districts, chronic absenteeism increased 10 points—from 10% to 20%.
  • Chronic absenteeism in high-poverty districts increased 17 points—from 19% in 2018 to 36% in 2022—while the increase in low-poverty districts was 11 points—from 10% to 21%.
  • In high-minority districts, chronic absenteeism grew 15 points—from 17% in 2018 to 32% in 2022. In low-minority districts the increase was 10 points—from 13% to 23%.

Chronic Absenteeism by District Type: 2018–2022

District-level data reveal patterns in the kinds of districts that saw greater and lesser changes in chronic absenteeism over time. Click on the drop down menu below to see these differences by achievement, size, 2020–21 remote instruction, 2021–22 mask mandates and other factors.

State Change In Chronic Absenteeism: 2018–2023

Gauging the Change in Chronic Absenteeism: 2018–2022

Search R2L’s Chronic Absenteeism Data

The Return to Learn Tracker’s goal is to provide up-to-date data on how US school districts responded to the pandemic and are being affected by it. This tracker captures data on chronic absenteeism for over 14,700 school districts and charter schools nationwide and will be updated as more state data becomes available. We hope these data will help school communities as they face ongoing decisions, provide the basic knowledge necessary for shaping policy across states, and allow other researchers to more accurately study COVID-19’s impacts on schools.

We adhere to high standards in research methodology and practices, pursuing rigorous transparency in our approach to this work.

AEI would like to thank The Achelis and Bodman Foundation for its generous support that helped make the Return to Learn Tracker possible.

I collected district level data on chronic absenteeism from each state, as available, for the 2016-17 through 2022-23 school years (henceforth years are referred to by the spring year, such as the 2017 through 2023 school years). I also collected the numerators and denominators for each district where available, and derive numerators or denominators when rates and either numerators or denominators were available. These data serve as the primary data in R2L’s chronic absenteeism data collection. All estimates on this site are weighted by numbers of students (specifically, by the number included in the denominator of  the chronic absenteeism rate). Some states posted data only for select years.  Many state did not post data for the 2019-20 and 2020-21 school years, as remote instruction interfered with attendance and traditional attendance collection data systems. Other states did not report chronic absenteeism data at the district level in 2017, though most did by 2018, and in 2019 and 2022.  Data from 2023 are still being reported, and will be added as they come available. Fortunately, shortly before our data was set to be released, the US Department of Education released district chronic absenteeism counts from EDFacts that can supplement these data. ED Data Express released counts of students chronically absent at the LEA level for 2018 and 2019, and at the school and LEA level for 2020 through 2022.  I merged these data with membership counts from CCD data files for K-12 students.  These membership counts serve as the best available denominators to produce rates from the chronically absent student counts.  However, these denominators are imperfect and are subject to error because the counts for membership and for chronic absenteeism may have been taken at different times and under different rules.  Additionally, most state only include students in their chronic absenteeism data (or regular attendance data, which is the inverse percentage)  if the students were enrolled for a specified time, often defined as a quarter or 100 days in the school year, while the EDFacts data includes students enrolled for at leas 10 days. Likely driven by these differential inclusion criteria, in most instances of difference estimates from the ED Data Express data yield rates that are slightly higher than those reported by states with more more conservative inclusion criteria and accurate denominators. As such, I use estimates based on the ED Data Express collection for 2018 through 2022, only for districts with no data are available from the state. The proportion of R2L data taken from ED Data Express, rather than state supplied data, are depicted in the table below. These affect a small share of districts in 2022, and relatively larger shares in 2018 and 2019.  In most years the vast majority were drawn from state data as detailed below, and averages with and without the ED Data Express data are minimal (less than 0.6 percentage points). More work to assess these differences is ongoing. I also remove all rates that exceed 100%. I use several other data sources in this tracker. LEA location data come from the EDGE data files provided by CCD, and CCD data also provide urbanicity, school district size, and minority percentage.  Poverty estimates for districts use SAIPE data from 2022. Return to Learn data provide information on the duration of In-person instruction in 2020-21 and the duration of masking requirements in 2021-22 for about 8,500 districts. Data on historic achievement come from the Stanford Education Data Archive (SEDA) data for 2019. Data on county prevalence of single parent households comes from Census and is cross walked to districts weighted by districts average student membership.  
R2L Chronic Absenteeism Data based on ED Data Express Estimates
State 2017 2018 2019 2020 2021 2022 2023
AL 0% 0% 0% 100% 0% 0% n.a.
AR n.a. 100% 100% 100% 100% 100% n.a.
AZ n.a. 100% 100% 100% 100% 100% n.a.
CA 0% 0% 3% n.a 3% 3% 0%
CO 0% 5% 11% 2% 0% 0% 0%
DC n.a. 0% 100% 100% 0% 0% n.a.
DE 0% 0% 0% 100% 3% 3% 0%
HI n.a. 100% 0% 0% 0% 0% n.a.
IA n.a. 100% 100% 0% 0% 0% n.a.
ID n.a. 100% 100% 100% 2% 6% n.a.
KY n.a. 0% 0% 100% 0% 14% n.a.
LA n.a. 100% 100% 100% 100% 3% 0%
MD n.a. 0% 0% 0% 1% 0% n.a.
MI 0% 1% 1% 1% 17% 1% 0%
MN 0% 0% 0% 100% 100% 0% n.a.
MO 0% 0% 1% 1% 0% 0% n.a.
MS n.a. 1% 1% 100% 1% 1% n.a.
MT n.a. 0% 0% 1% 0% 0% n.a.
NC n.a. 3% 4% 4% 5% 5% n.a.
ND n.a. 5% 5% 5% 5% 0% 0%
NE n.a. 100% 5% 3% 6% 5% n.a.
NH n.a. 100% 100% 100% 100% 100% n.a.
NJ 0% 0% 0% 100% 0% 0% n.a.
NM n.a. 100% 0% 0% 0% 0% 0%
NV n.a. 100% n.a. 100% n.a. n.a. 0%
NY n.a. 2% 3% 100% 4% 4% n.a.
OH 0% 1% 1% 1% 1% 2% 0%
OK n.a. 1% 1% 100% 100% 1% n.a.
OR 0% 0% 0% 100% 0% 0% n.a.
SC n.a. 1% 2% 3% 0% 0% 0%
SD n.a. 0% 0% 100% 0% 0% 0%
TN n.a. 1% 0% 0% 0% 0% n.a.
TX n.a. 100% 0% 0% 0% 100% n.a.
VT 0% 0% 33% 32% 31% 100% n.a.
WA 0% 0% 0% 1% 0% 0% n.a.
WY n.a. 100% 100% 100% 100% 100% n.a.
Note: R2L use no ED Data Express data for 15 states (omitted)

I collected district level data on chronic absenteeism from each state, as available, for the 2016-17 through 2022-23 school years (henceforth years are referred to by the spring year, such as the 2017 through 2023 school years). I also collected the numerators and denominators for each district where available, and derive numerators or denominators when rates and either numerators or denominators were available. These data serve as the primary data in R2L’s chronic absenteeism data collection. All estimates on this site are weighted by numbers of students (specifically, by the number included in the denominator of  the chronic absenteeism rate).

Some states posted data only for select years.  Many state did not post data for the 2019-20 and 2020-21 school years, as remote instruction interfered with attendance and traditional attendance collection data systems. Other states did not report chronic absenteeism data at the district level in 2017, though most did by 2018, and in 2019 and 2022.  Data from 2023 are still being reported, and will be added as they come available.

Fortunately, shortly before our data was set to be released, the US Department of Education released district chronic absenteeism counts from EDFacts that can supplement these data. ED Data Express released counts of students chronically absent at the LEA level for 2018 and 2019, and at the school and LEA level for 2020 through 2022.  I merged these data with membership counts from CCD data files for K-12 students.  These membership counts serve as the best available denominators to produce rates from the chronically absent student counts.  However, these denominators are imperfect and are subject to error because the counts for membership and for chronic absenteeism may have been taken at different times and under different rules.  Additionally, most state only include students in their chronic absenteeism data (or regular attendance data, which is the inverse percentage)  if the students were enrolled for a specified time, often defined as a quarter or 100 days in the school year, while the EDFacts data includes students enrolled for at leas 10 days. Likely driven by these differential inclusion criteria, in most instances of difference estimates from the ED Data Express data yield rates that are slightly higher than those reported by states with more more conservative inclusion criteria and accurate denominators.

As such, I use estimates based on the ED Data Express collection for 2018 through 2022, only for districts with no data are available from the state. The proportion of R2L data taken from ED Data Express, rather than state supplied data, are depicted in the table below. These affect a small share of districts in 2022, and relatively larger shares in 2018 and 2019.  In most years the vast majority were drawn from state data as detailed below, and averages with and without the ED Data Express data are minimal (less than 0.6 percentage points). More work to assess these differences is ongoing. I also remove all rates that exceed 100%.

I use several other data sources in this tracker. LEA location data come from the EDGE data files provided by CCD, and CCD data also provide urbanicity, school district size, and minority percentage.  Poverty estimates for districts use SAIPE data from 2022. Return to Learn data provide information on the duration of In-person instruction in 2020-21 and the duration of masking requirements in 2021-22 for about 8,500 districts. Data on historic achievement come from the Stanford Education Data Archive (SEDA) data for 2019. Data on county prevalence of single parent households comes from Census and is cross walked to districts weighted by districts average student membership.

 

R2L Chronic Absenteeism Data based on ED Data Express Estimates
State 2017 2018 2019 2020 2021 2022 2023
AL 0% 0% 0% 100% 0% 0% n.a.
AR n.a. 100% 100% 100% 100% 100% n.a.
AZ n.a. 100% 100% 100% 100% 100% n.a.
CA 0% 0% 3% n.a 3% 3% 0%
CO 0% 5% 11% 2% 0% 0% 0%
DC n.a. 0% 100% 100% 0% 0% n.a.
DE 0% 0% 0% 100% 3% 3% 0%
HI n.a. 100% 0% 0% 0% 0% n.a.
IA n.a. 100% 100% 0% 0% 0% n.a.
ID n.a. 100% 100% 100% 2% 6% n.a.
KY n.a. 0% 0% 100% 0% 14% n.a.
LA n.a. 100% 100% 100% 100% 3% 0%
MD n.a. 0% 0% 0% 1% 0% n.a.
MI 0% 1% 1% 1% 17% 1% 0%
MN 0% 0% 0% 100% 100% 0% n.a.
MO 0% 0% 1% 1% 0% 0% n.a.
MS n.a. 1% 1% 100% 1% 1% n.a.
MT n.a. 0% 0% 1% 0% 0% n.a.
NC n.a. 3% 4% 4% 5% 5% n.a.
ND n.a. 5% 5% 5% 5% 0% 0%
NE n.a. 100% 5% 3% 6% 5% n.a.
NH n.a. 100% 100% 100% 100% 100% n.a.
NJ 0% 0% 0% 100% 0% 0% n.a.
NM n.a. 100% 0% 0% 0% 0% 0%
NV n.a. 100% n.a. 100% n.a. n.a. 0%
NY n.a. 2% 3% 100% 4% 4% n.a.
OH 0% 1% 1% 1% 1% 2% 0%
OK n.a. 1% 1% 100% 100% 1% n.a.
OR 0% 0% 0% 100% 0% 0% n.a.
SC n.a. 1% 2% 3% 0% 0% 0%
SD n.a. 0% 0% 100% 0% 0% 0%
TN n.a. 1% 0% 0% 0% 0% n.a.
TX n.a. 100% 0% 0% 0% 100% n.a.
VT 0% 0% 33% 32% 31% 100% n.a.
WA 0% 0% 0% 1% 0% 0% n.a.
WY n.a. 100% 100% 100% 100% 100% n.a.
Note: R2L use no ED Data Express data for 15 states (omitted)