Pandemic Learning: Technical Materials for Teacher Survey and Discussion Groups with Public K-12 School Principals, Teachers, and Parents
- Document type
- Audit
- Date
- 2021-07-09
Summary
A GAO electronic supplement, GAO-22-105817, titled Pandemic Learning: Technical Materials for Teacher Survey and Discussion Groups with Public K-12 School Principals, Teachers, and Parents. It serves as a companion to GAO-22-104487 and presents the methodology of a nationwide survey of K-12 public school teachers that Gallup conducted for GAO between June 18 and July 9, 2021. The supplement reports that 2,862 eligible teachers completed the survey from an initial sample of 45,792, with a weighted cumulative response rate of 8.2 percent, and it describes the survey design, sample frames and weighting. It also describes 18 virtual discussion groups with teachers, parents and principals and closes with the discussion groups held with Department of Defense Education Activity teachers and parents.
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
Cite this document
“Pandemic Learning: Technical Materials for Teacher Survey and Discussion Groups with Public K-12 School Principals, Teachers, and Parents.”
Original at www.gao.gov, dated 2021-07-09. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/2021-07-09-gao-22-105817-pandemic-learning-technical-materials-for-teacher-survey-and-disc/
, SHA-256 f0386efa7c338456d2a3b71dbb76d7594850f125d9b5f4a1c58484a8697da29d.Full text
Pandemic Learning: Technical Materials for Teacher Survey and Discussion Groups with
Public K-12 School Principals, Teachers, and Parents
This product is a supplement to Pandemic Learning: As Students Struggled to Learn, Teachers
Reported Few Strategies as Particularly Helpful to Mitigate Learning Loss (GAO-22-104487),
the first in a series of three reports on pandemic learning issuing in spring 2022.
Background
This electronic supplement serves as a companion to GAO-22-104487, Pandemic Learning: As
Students Struggled to Learn, Teachers Reported Few Strategies as Particularly Helpful to
Mitigate Learning Loss, as well as two forthcoming reports on pandemic learning. This
supplement presents technical information about our two methodologies—a nationwide survey
of K-12 public school teachers and discussion groups with educators and parents—as well as a
reproduction of the survey instrument and survey results in aggregate form.
Our survey asked teachers about instructional models, adult support provided to students
learning in a virtual environment, difficulties students faced when learning, their students’
academic progress, strategies used to mitigate learning loss, students who never showed up for
class, and students who became disengaged, among other topics. GAO contracted with Gallup
to conduct a nationally representative survey of elementary and secondary public school
teachers between June 18 and July 9, 2021.
To gain further insights about the topics covered in the survey and to get the perspective of
other K-12 stakeholders, we held 18 virtual discussion groups with public school teachers (six
groups), parents of students (six groups), and principals (six groups) between June 29 and July
14, 2021. We also held one discussion group with teachers at Department of Defense
Education Activity (DODEA) schools on July 19, 2021 and one with DODEA parents on July 21,
2021. DODEA teachers participated in discussion groups only; no DODEA teachers completed
the teacher survey.
Technical details of our survey and discussion group methodologies are linked below. Also
linked below is a reproduction of the web-based survey with questions and aggregate results.
We did not include respondent narrative answers that were provided to open-ended questions.
We conducted the work upon which this supplement is based in accordance with generally
accepted government auditing standards. Those standards require that we plan and perform the
audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings
and conclusions based on our audit objectives. We believe that the evidence obtained provides
a reasonable basis for our findings and conclusions based on our audit objectives.
Contents
The Technical Description of GAO’s Survey of K-12 Public School Teachers
A Reproduction of the Survey Instrument and Aggregate Survey Results
The Technical Description of GAO’s Discussion Groups with Educators and Parents
Page 1 GAO-22-105817 Pandemic Learning
Contacts
For more information, contact Jacqueline M. Nowicki at (617) 788-0580 or NowickiJ@gao.gov.
Office of Public Affairs; Chuck Young, Managing Director (202) 512-4800 or
YoungC1@gao.gov.
Copyright
This is a work of the U.S. government and is not subject to copyright protection in the United
States. The published product may be reproduced and distributed in its entirety without further
permission from GAO. However, because this work may contain copyrighted images or other
material, permission from the copyright holder may be necessary if you wish to reproduce this
material separately.
GAO Contact
Jacqueline M. Nowicki
Director
NowickiJ@gao.gov
(617) 788-0580
(105817)
Page 2 GAO-22-105817 Pandemic Learning
Technical Description of K-12 Public School Teacher Survey
We conducted a nationally representative survey of K-12 public school teachers to obtain their
perspectives about learning loss and strategies to mitigate it during the pandemic. We designed
the survey questions and contracted with Gallup to conduct a nationally representative survey of
elementary and secondary public school teachers between June 18 and July 9, 2021. We
obtained generalizable survey responses from 2,862 general education teachers. The initial
sample was selected from the Gallup Panel, a probability-based panel of U.S. adults, and a
national list of teachers.
Survey Design
We designed a survey of K-12 public school teachers as part of our body of work to understand
how COVID-19 has affected public K-12 education. The survey asked teachers about their
instructional models, adult support provided to their students, difficulties their students faced,
their students’ academic progress, strategies they used to mitigate learning loss, and the extent
to which their students were engaged in learning, among other topics. We wrote 41 questions
on these topics and asked respondents to reflect on the 2020-21 school year. To avoid having
respondents guess at answers, we pre-tested the survey to identify which questions needed
“not applicable” or “don’t know” response options.
The survey defined several terms, including:
• This school year: answer questions that refer to 'this school year' by considering your
experiences during the 2020-2021 school year.
• Typical school year: answer questions that refer to a ‘typical school year’ by considering
your experiences during a recent school year prior to the pandemic.
• Your students: answer each question based on your experiences with, and observations
of, the students whom you teach at least one subject.
• English Learners: English Learners (EL) are students who your school has identified as
learning English. Your school may refer to these students as English language learners.
• Learning loss: Learning loss is the extent to which a student is behind where they would
be at the same point in a typical school year.
• Unaccounted for: Students who were registered to be in your class(es) but who did not
attend all year. These students are sometimes referred to as “no shows.”
• Disengaged: Students whose limited participation in your class(es) greatly affected their
learning or grades. Examples of limited participation include:
- logging in but not participating,
- attending class but not completing or turning in work, or
- missing a substantial number of classes.
Page 1 GAO-22-105817 Pandemic Learning
We asked teachers to identify which of the following four models they taught in for the majority
of the 2020-21 school year: 1
• Full-time in-person learning – all or almost all students attended in person full-time.
• Full-time virtual learning – all or almost all students attended virtually full-time.
• Hybrid learning – students alternated between attending some days in person and other
days virtually.
• A mix of learning – some students generally attended in person full-time and other
students generally attended virtually full-time.
We then asked those teachers whose students were virtual, in person, or hybrid for the majority
of the year to consider their experiences teaching students in their respective models for many
of the questions, rather than to aggregate their experiences teaching across any of the models
they used during the 2020-21 school year. We randomly assigned teachers whose students
were in a mixed model to respond based on their in-person or virtual students. We asked
teachers to consider experiences teaching in specific models to reduce the length of the survey
session and the burden on respondents.
We pretested the survey with 10 teachers to evaluate their understanding of the terms used and
the intent of questions, the ease and time required to complete the survey, and their feedback
on additional topics to include. Gallup also conducted a pilot test prior to data collection and
obtained 10 completed web interviews to evaluate respondent comprehension, item relevance
to the sampled population, the extent to which it was feasible for respondents to answer a
survey question and the steps required to select a response, and any possible issues with the
visual design or survey navigation. We incorporated feedback from the pretests and pilot tests
prior to fielding the survey.
Sample Design
The survey was designed to make generalizable estimates for the population of general
education teachers at public elementary and secondary schools and who taught a core subject.
For the purpose of this survey, core subjects included: elementary school, math, science,
computer science/information technology, English/language arts/reading/writing, social studies
and world/foreign languages or English language learning. Gallup drew the sample from two
population frames: (1) its Gallup Panel, a probability-based panel of people who agree to
participate in surveys on various topics and (2) The Dun and Bradstreet listed frame of U.S.
educators, which supplemented the Gallup Panel sample with people known to work as
teachers. The probability sample was designed to achieve margin of error targets for key
subgroups by location, participation in free or reduced-price lunch programs, grade level, and
percentage of English learners (see table 1).
1We instructed respondents not to include any time that individual or small groups of students attended virtually in
order to quarantine or isolate.
Page 2 GAO-22-105817 Pandemic Learning
Table 1: Subpopulation by Margin of Error (MOE) requirement and Minimum Sample Size
MOE Minimum required Number of surveys
requirement sample size completed
Location
Rural/Town +/- 7% 400 768
Suburban +/- 7% 400 1,061
Urban +/- 7% 400 990
% Eligible for free or reduced-price lunch (FRPL)
Low (<=20%) +/- 7% 400 558
Mid-Range (21-80%) +/- 7% 400 1,305
High (81-100%) +/- 7% 400 757
Grade Level
K-2 +/- 10% 200 254
3-8 +/- 10% 200 1,536
9-12 +/- 10% 200 1,323
English Learners (EL)
25% or more are EL +/- 10% 200 525
Less than 25% are EL +/- 10% 200 2,336
Source: Gallup | GAO-22-105817
Note: The number of completed surveys provided margins of errors that met or exceeded all of the
requirements.
The Gallup Panel is a panel of people selected with probability-based methods to represent the
population of adults in the United States. Established in 2004, the Gallup Panel consists of more
than 110,000 members who have agreed to complete surveys regularly. Its sampling frames
cover more than 95 percent of the U.S. adult population, including approximately 2,700 K-12
teachers. Gallup recruits members using address-based sampling and random-digit-dial
telephone sampling procedures for both landline and cell phones. Random recruitment allows
calculation of selection probabilities, and Gallup Panel members are required to provide
demographic data (date of birth, gender, education level, race/ethnicity, and ZIP code) that can
be used to carry out post-stratification adjustments that ensure the sample represents selected
population characteristics. The design allows Gallup to calculate selection probability and
response rates so that results can be projected to the target population with known levels of
precision.
Gallup specifically collects teacher status from all panel members and regularly updates this
demographic in the panel member database to maintain a subpanel of teachers. Gallup uses a
tightly controlled process for inviting panel members to participate in research, ensuring that
respondents are not professional panelists. Members are invited to no more than two to three
surveys per month. While Gallup regularly conducts surveys of teachers, Gallup ensures that no
individual teacher in its panel takes more than four teacher-specific surveys annually. This
keeps teachers from being overburdened and taking multiple teacher surveys concurrently.
Gallup supplemented the Gallup Panel sample with teachers sampled from a quasi-probability
listed frame of 42,938 teachers because the number of teachers in Gallup panel was not
sufficient to achieve targeted precision levels. This comprehensive database of U.S. educators
Page 3 GAO-22-105817 Pandemic Learning
is maintained by an established list provider and includes email addresses. The list is regularly
updated and audited, has more than 5 million K-12 personnel in the U.S., and includes
educators from more than 13,000 districts. Gallup thoroughly vetted many sample sources for
comprehensiveness and quality and has worked with this list source for nearly 10 years for
projects that require larger sample sizes than what can be achieved with the Gallup Panel
alone.
Survey Administration
We contracted with Gallup to administer our survey. Gallup fielded the web survey between
June 18 and July 9, 2021. Gallup began inviting potential respondents on June 18, 2021. Non-
responders received email reminders on June 21, June 24, June 26, June 28, July 1, July 3,
July 5, and July 8.
A total of 2,862 eligible teachers completed the survey from an initial sample of 45,792 teachers
(see table 2). The average response time was 26 minutes and 25 seconds, excluding outliers.
The survey achieved an overall weighted cumulative response rate of 8.2 percent when
adjusted for estimated eligibility of 76.6 percent. 2 The Gallup Panel sample had a response rate
of 49.6 percent and the listed sample had a 5.4 percent response rate. For further details on the
reliability of estimates produced using this survey, see these two sections below: 1) sample
weighting and 2) survey exclusions, quality, and error.
Table 2: Response Outcomes of Nationally Representative Survey of K-12 Public School Teachers Administered by Gallup
Gallup panel Listed sample Total sample
1 Total sampled cases 2,886 42,938 45,792
2 Completed survey (weighted cases) 803 2,059 2,862
3 Ineligible 705 311 1016
4 Partial complete 96 376 472
5 Attempted survey (2+3+4) 1,604 2,746 4,350
6 Unknown (1-5) 1,282 40,192 41,442
7 Eligibility rate ((2+4)/5) 56% 89% 77%
8 Estimated eligible unknowns (6*7) 719 35,640 31,763
9 Total eligible sample (2+4+8) 1,618 38,075 35,097
Source: Gallup | GAO-22-105817
Note: The American Association for Public Opinion Research’s response rate 3 (unknown eligibility)
was used to calculate the estimated number of eligible respondents.
Sample Weighting
Gallup adjusted the survey weights to account for potential nonresponse biases by accounting
for relevant school characteristics for non-respondents and re-weighting (post-stratifying) the
sample to match the number and regional distribution of teachers and teacher demographics.
More specifically, the base weights for both samples were set to one because samples were
2The American Association for Public Opinion Research’s response rate 3 (unknown eligibility) was used to calculate
the estimated number of eligible respondents.
Page 4 GAO-22-105817 Pandemic Learning
selected with equal probability from both sources. The responses were weighted to minimize
bias independently for each source and for the sources combined. Responses from each source
were first weighted for non-response by region. Gallup then used post-stratification weighting
adjustments to adjust the sample to match national demographics of teacher age, sex, race,
and school characteristics such as school location (urban, suburban, rural). Data from the
National Center for Education Statistics National Teacher and Principal Survey for 2017-2018
were used as targets for post-stratification weighting. Last, the final weights were calculated by
combining the weights of the completed surveys from both samples using composite weighting
procedures and trimming to minimize the effect of extreme weights on variance of estimates.
Variances were adjusted for a calculated design effect of 2.03.
Survey Exclusions, Quality, and Error
Based on the survey and weighting adjustment methods described above, we determined that
estimates from this survey are generalizable to the population of K-12 public school teachers,
and are sufficiently reliable for the purposes of our report.
All estimates in this report have a margin of error less than or equal to +/- 10 percent at the 95
percent confidence level. Margins of error for subpopulations of interest ranged from 2.9 percent
(teachers in schools with less than 25 percent ELL, the subpopulation with the largest sample
size) to 7.2 percent for teachers of K-2 students (the subpopulation with the smallest sample
size; see table 3). Margins of error for individual questions varied depending on the number of
responses. All reported margins of error include the computed design effects in the weighting.
The quality of the two sample frames and the information used to address non-sampling errors
are important contributors to the quality of survey results. The Gallup Panel is a probability-
based panel that is representative of adults in the U.S. The Dun and Bradstreet list of teachers
does not cover all teachers and may under-represent teachers in small rural schools. However,
this under-coverage is small relative to the population of teachers and is unlikely to bias national
population estimates. Methodologies used to minimize and address non-response are also
important contributors to the quality of results. This survey’s focus on a narrow topic–teacher
experiences related to COVID-19 during the 2020-21 school year–is beneficial for non-response
bias adjustments with covariates that are correlated to the key outcome variables. The
covariates used for non-response bias adjustments (teacher age, gender, race, and location)
are collected from Panel members The data used to adjust the responses to the target
population are provided from a survey of the target population that was high quality in terms of
sample frame, coverage, design, and analysis. The most recently available data from the
National Center for Education Statistics National Survey of Teachers and Principals were used
to perform post-stratification. However, as with all non-probability adjustment methods, the
inferences made from these data require reliance on modeling assumptions.
Table 3: Margin of Error Calculation for Subpopulation Estimates from a Nationally Representative Survey of K-12 Public
School Teachers Administered by Gallup
Sample size Margin of error
Location
Rural/town 768 4.9%
Suburban 1,061 4.3%
Urban 990 4.5%
Page 5 GAO-22-105817 Pandemic Learning
Sample size Margin of error
Percent eligible for free or reduced-price lunch
Low (0-20% eligible) 558 5.8%
Mid-range (21-80% eligible) 1,305 3.9%
High (81-100% eligible) 757 4.9%
Grade level (teacher)
K–2 254 7.2%
3–8 1,536 3.5%
9 – 12 1,323 3.3%
Percent English learners (EL)
Less than 25% EL 2,336 2.9%
25% or more EL 525 5.9%
Source: Gallup. | GAO-22-105817
Analysis of Survey Data
We began by analyzing the aggregate responses for all 2,862 teachers who completed the
survey. Their responses are generalizable to the population of all general education public
school teachers in the U.S.
We then disaggregated responses by
• Locale, as defined by teachers’ reports of whether they lived in
• a rural area or small town
• suburb, or
• urban area
• Percentage of their students who participated in the free or reduced-price lunch (FRPL)
program, as reported by the teacher
• High-poverty: teachers reported 81-100 percent of their students participated in
FRPL
• Mid-poverty: teachers reported 21-80 percent of their students participated in
FRPL
• Low-poverty: teachers reported 0-20 percent of their students participated in
FRPL
• Percentage of a teacher’s students were English learners, as reported by the teacher
Page 6 GAO-22-105817 Pandemic Learning
• Low: less than 20 percent of a teacher’s students were English learners 3
• High: 20 percent or more of a teacher’s students were English learners
• Grade level, as reported by the teacher 4
• Grades K-2 5
• Grades 3-8
• Grades 9-12
• Instructional model
• Virtual: teachers who reported their students spent the majority of the year in a
virtual environment
• Hybrid: teachers who reported their students spent the majority of the year in a
hybrid environment
• In person: teachers who reported their students spent the majority of the year in
person.
Odds Ratio Analysis
Odds ratios are a measure of association between two variables. They approximate how much
more likely or unlikely (in terms of odds) it is for the outcome to be present among subjects in
populations that differ by a characteristic. For example, K-5 teachers in a virtual environment
with high-poverty students had an odds ratio of 4.3 in response to a question about students
lacking a school meal compared to a typical year. In other words, these teachers were 4.3 times
more likely than all other K-5 teachers to report that their students lacked a school meal in 2020-
21 compared to a typical pre-pandemic year. Odds ratios assume the responses are not
correlated, and the interpretation is dependent on model specification. We did not control for
multiple effects, and results could differ if we had included multiple effects in our models. 6 We
took statistical uncertainty into account when drawing conclusions from the results.
3Gallup’s margin of error target for whether a teacher had a high percentage of English learners was 25 percent or
more.
4If the teacher reported working with students across our grade level categories (e.g., worked with both second- and
third-grade students, we assigned the teacher to the grade level category of the earliest grade taught).
5Our odds ratio analysis and regression modeling, described later in this section, grouped grade levels as K-5, 6-8,
and 9-12 due to our sampling size limitations.
6D.W. Hosmer, S. Lemeshow, and R.X. Sturdivant. 2013. Applied Logistic Regression (John Wiley & Sons. Hoboken,
New Jersey).
Page 7 GAO-22-105817 Pandemic Learning
Regression Modeling
We developed a series of statistical models to describe the associations between teachers’ use
of various strategies to address learning loss and their perceived effectiveness.
Our survey asked two closed-response questions about learning loss strategies. We reproduce
the question wording and response options in the survey results section of this supplement. The
first question was:
Q29: Approximately, how often during this school year have you used the following
strategies to support learning or address learning loss?
(Learning loss is the extent to which a student is behind where they would be at the
same point in a typical school year.)
For each strategy that a teacher reported using, we then asked:
Q30: How many of your students improved their academic progress as a result of each
of the following strategies?
Our models estimated the probability that a teacher would report that “about half” or more of
their students “improved their academic progress” varied across instructional models. 7 We
grouped responses to each question, in order to increase the sample sizes. We estimated these
probabilities separately by grade level, in-person or hybrid and virtual instructional models, and
a three-way categorization of the school’s free or reduced-price lunch (FRPL) participation. 8 We
limited the survey respondents to those who had observed values on all variables—a sub-
sample that could vary across learning strategies.
Specifically, we estimated the following logistic regression model:
Y i is a binary random variable that indicated whether teacher i said that a strategy helped “about
half” or more of their students; x i was a vector of fixed, discrete covariates measuring
instructional model, grade level, and FRPL participation, as described above; and β was a
vector of coefficients. We estimated each model among respondents who said they used the
strategy.
We applied the final survey weights to estimate the coefficients, in order to compensate for
possible bias from sample selection, non-response, and other sources of error, as described
7The survey asked respondents to select one of the following five options: none/almost none, some, about half, most,
and almost all/all. We grouped these responses into two categories for modeling: less than half and about half or
more.
8We asked Gallup to construct a three-way categorization of FRPL participation for which each category produced
margin of error estimates no larger than +/- 7 percent. Gallup constructed the low category to include teachers who
responded 20 percent or less of their students participated in FRPL, the mid category to include teachers who
responded 21 through 80 percent of their students participated in FRPL, and the high category to include teachers
who responded more than 80 percent of their students participated in FRPL.
Page 8 GAO-22-105817 Pandemic Learning
above. However, we did not use complex sample variance estimators to estimate the
uncertainty of the parameters. Instead, we applied “robust” variance estimators for the logistic
model under simple random sampling that were consistent with arbitrary patterns of
heteroskedasticity. This is appropriate for a “structural” modeling application, where we assume
the outcome to vary randomly conditional on fixed covariates. This contrasts with typical survey
analysis variance estimators, which assume the data are fixed and estimator variance occurs
through random sample selection. 9
We summarized the fitted model by using the parameters to estimate the mean predicted
probability of perceived improvement across teachers, conditional on grade level and
instructional model and averaging over sample values of FRPL participation. We estimated 95
percent confidence intervals for this derived quantity using the Delta Method and Normal
approximations, as implemented in Stata 17. Overall samples sizes of about 2,800 teachers
satisfied the large sample assumptions required for robust variance estimation methods of the
parameters and the derived quantities.
9Jeffrey M. Wooldridge, Econometric Analysis of Cross Section and Panel Data (Cambridge: MIT Press, 2002), 592-
596.
Page 9 GAO-22-105817 Pandemic Learning
Letter
K-12 Public School
Teachers Survey:
Survey Questions
and Results
Introduction
Teacher Survey on Pandemic The survey takes approximately 20 minutes to complete and will provide
Learning the U.S. Congress with important information about students’ learning
during the pandemic. You have been selected from among thousands of
teachers nationally to complete this survey.
This survey is sponsored by the U.S. Government Accountability Office
(GAO). GAO is an independent, nonpartisan research agency of the U.S.
Congress. This survey is intended to provide information for federal
policymakers. It will be used to provide Congress information on national
trends related to:
• students’ school day settings,
• student learning during the pandemic,
• strategies for supporting learning and addressing learning loss during
the pandemic, and
• disconnected students.
This survey is not a review of schools or school districts or whether they
are meeting legal or educational requirements.
Page 1 GAO-22-105817 Pandemic Learning
S1 According to Gallup’s records, you work as a K-12 teacher. Did you teach in a public, K-12 school during the 2020-2021
school year?
95 percent confidence 95 percent confidence
Estimated number of interval – lower bound interval – upper bound
Response teachers (number) (number)
Yes 2,350,753 2,263,273 2,438,233
Q1 How many total years of K-12 teaching experience do you have?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Less than 1 year 0 0 1
1 year 3 2 4
2-3 years 7 5 8
4-6 years 14 12 16
7-9 years 12 10 14
10 or more years 65 62 67
Q2 Which grade level(s) did you teach during the 2020-2021 school year? Select all that apply.
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Kindergarten 8 7 10
1st grade 12 10 14
2nd grade 12 10 14
3rd grade 15 13 17
4th grade 16 14 19
5th grade 21 19 24
6th grade 12 11 13
7th grade 16 14 18
8th grade 15 13 16
9th grade 18 17 20
10th grade 20 19 22
11th grade 23 21 25
12th grade 22 20 24
Page 2 GAO-22-105817 Pandemic Learning
Q3 What subject(s) did you teach during the 2020-2021 school year? Select all that apply.
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
General education, such as 24 21 26
elementary education
Math 28 25 30
Science (e.g., biology, 23 21 25
engineering)
Computer science/information 4 3 5
technology
English/language 34 32 37
arts/reading/writing
Social studies (e.g., history, 23 21 25
psychology, economics)
World/Foreign languages 2 1 2
English language learning 7 6 9
Special education 4 3 5
P.E./Health 1 1 2
Music, Visual or Performing Arts 1 1 2
Technical or vocational (e.g., 1 1 2
industrial arts, family and
consumer science)
Substitute teacher 2 1 2
Other 3 2 4
Q4 Over the current school year, to how many students, in total, have you taught these subjects?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
60 or fewer 41 38 44
61 to 100 23 21 25
101 to 150 23 21 24
More than 150 13 12 15
Note: In our survey, teachers answered the question with a number of students. We grouped the
responses into these four categories.
Page 3 GAO-22-105817 Pandemic Learning
Q5 Approximately how many of these students are English Learners (EL)?
(English Learners (EL) are students who your school has identified as learning English. Your school may refer to these
students as English language learners.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
5 or fewer 54 51 56
6 to 10 15 13 17
11 to 20 12 11 14
More than 20 20 18 22
Note: In our survey, teachers answered the question with a number of students. We grouped the
responses into these four categories.
Q5A Which of the following best describes the district in which you taught during the 2020-2021 school year?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
I taught in the same school 97 96 98
district as I did during the 2019-
2020 school year
I taught in a different school 3 2 4
district than the 2019-2020
school year
Q5B Which of the following best describes the grade level you taught during the 2020-2021 school year?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
I taught the same grade as I did 85 83 87
during the 2019-2020 school
year
I taught a different grade during 15 13 17
the 2019-2020 school year
Page 4 GAO-22-105817 Pandemic Learning
Q6_T
Students’ School Day Settings
First, we have some questions about your experiences teaching
during the 2020-2021 school year. If you teach at more than one
school, for the rest of the survey, please think about the school with
which you spent the most time this school year.
Please consider the students for whom you taught the subjects you
selected earlier in the survey.
Q7 Which of the following best describes the learning situation of your students at the beginning of the 2020-2021 school
year?
(Do not include any time that individual or small groups of students attended virtually in order to quarantine or isolate.)
Response 95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
teachers (percentage) (percentage)
Full-time in-person learning – all 15 13 17
or almost all students attended in
person full-time.
Full-time virtual learning – all or 48 46 51
almost all students attended
virtually full-time.
Hybrid learning – students 15 14 17
alternated between attending
some days in person and other
days virtually.
A mix of learning – some 22 19 24
students generally attended in
person full-time and other
students generally attended
virtually full-time.
Page 5 GAO-22-105817 Pandemic Learning
Q8 Which of the following best describes the learning situation of your students at the end of the 2020-2021 school year?
(Do not include any time that individual or small groups of students attended virtually in order to quarantine or isolate.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Full-time in-person learning – all 35 33 38
or almost all students attending
in person full-time.
Full-time virtual learning – all or 10 8 11
almost all students attending
virtually full-time.
Hybrid learning – students 12 11 14
alternating between attending
some days in person and other
days virtually.
A mix of learning – some 43 41 46
students generally attending in
person full-time and other
students generally attending
virtually full-time.
Q9 Which of the following best describes the learning situation of your students for the majority of the 2020-2021 school
year?
(Do not include any time that individual or small groups of students attended virtually in order to quarantine or isolate.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Full-time in-person learning – all 25 22 27
or almost all students attended in
person full-time.
Full-time virtual learning – all or 28 26 31
almost all students attended
virtually full-time.
Hybrid learning – students 20 18 22
alternated between attending
some days in person and other
days virtually.
A mix of learning – some 27 25 30
students generally attended in
person full-time and other
students generally attended
virtually full-time.
Page 6 GAO-22-105817 Pandemic Learning
Q10A For how many months during the 2020-2021 school year did you teach students in full-time in-person learning – all or
almost all students attending in person full-time.
(Do not include any time that individual or small groups of students attended virtually in order to quarantine or isolate.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
0 1 0 3
1 0 0 1
2 0 0 1
3 1 0 2
4 2 1 4
5 4 2 7
6 5 3 8
7 8 6 12
8 13 10 18
9 32 27 38
10 32 27 38
11 0 0 1
12 1 0 3
Q10B For how many months during the 2020-2021 school year did you teach students in full-time virtual learning – all or
almost all students attending virtually full-time.
(Do not include any time that individual or small groups of students attended virtually in order to quarantine or isolate.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
0 0 0 1
1 1 0 3
2 0 0 1
3 1 0 3
4 1 1 3
5 5 3 7
6 13 10 17
7 24 20 28
8 16 13 19
9 12 9 16
10 24 20 29
Page 7 GAO-22-105817 Pandemic Learning
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
11 1 0 1
12 2 1 5
Q10C For how many months during the 2020-2021 school year did you teach students in hybrid learning – students
alternating between attending some days in person and other days virtually.
(Do not include any time that individual or small groups of students attended virtually in order to quarantine or isolate.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
0 1 0 3
1 1 0 1
2 2 1 5
3 8 6 13
4 16 12 21
5 16 12 20
6 18 14 23
7 14 11 18
8 9 7 12
9 8 6 11
10 6 5 8
11 0 0 1
12 0 0 1
Q10D For how many months during the 2020-2021 school year did you teach students in a mix of learning – some students
generally attending in person full-time and other students generally attending virtually full-time.
(Do not include any time that individual or small groups of students attended virtually in order to quarantine or isolate.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
0 2 1 4
1 1 0 2
2 2 1 4
3 6 4 9
Page 8 GAO-22-105817 Pandemic Learning
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
4 5 3 8
5 9 6 13
6 8 6 11
7 8 6 11
8 9 7 13
9 23 19 27
10 24 20 29
11 1 0 3
12 1 1 3
Q11 Do you generally know whether your students have another person present when learning virtually, or is this something
you don’t know?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes, for most or many of my 20 17 23
students, I know if they have
another person present.
Yes, for some of my students, I 34 31 37
know if they have another person
present.
No, I don’t know this information 46 43 49
about most or all of my students.
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
in-person learning environment did not answer this question.
Page 9 GAO-22-105817 Pandemic Learning
Q12 From your observations during the current school year, how many of your students learning virtually usually had the
following people present?
A. An adult but I’m unable to determine whom
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 16 12 21
Some 45 38 51
About half 8 5 12
Most 15 11 21
Almost all/All 14 10 19
Don’t know 2 1 5
Note: This question was asked of teachers who responded yes to Q11 only.
B. Student’s parent or guardian
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 7 4 11
Some 45 39 52
About half 11 7 16
Most 26 21 32
Almost all/All 9 5 13
Don’t know 3 1 6
Note: This question was asked of teachers who responded yes to Q11 only.
C. An adult who is not the student’s parent or guardian
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 21 16 27
Some 58 51 64
About half 6 3 9
Most 2 0 4
Almost all/All 1 0 3
Don’t know 13 9 18
Note: This question was asked of teachers who responded yes to Q11 only.
Page 10 GAO-22-105817 Pandemic Learning
D. A trained learning support person/coach/other professional
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 75 69 81
Some 14 10 19
About half 1 0 3
Most 1 0 3
Almost all/All 1 0 4
Don’t know 8 5 13
Note: This question was asked of teachers who responded yes to Q11 only.
E. A person under the age of 18
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 23 18 28
Some 50 44 57
About half 9 5 13
Most 5 2 8
Almost all/All 2 1 5
Don’t know 12 8 17
Note: This question was asked of teachers who responded yes to Q11 only.
F. No one
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 35 29 41
Some 39 33 46
About half 4 2 7
Most 8 5 13
Almost all/All 3 1 7
Don’t know 10 7 15
Note: This question was asked of teachers who responded yes to Q11 only.
Page 11 GAO-22-105817 Pandemic Learning
Q13 Do you have any students who need substantial support or assistance from people at home to participate fully in virtual
learning?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 69 66 72
No 31 28 34
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
in-person learning environment did not answer this question.
Q14 How many of your virtual students who need substantial support or assistance from people at home are…
A. Receiving adequate support during the school day?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 14 12 17
Some 51 48 55
About half 10 8 12
Most 11 9 14
Almost all/All 5 4 7
Don’t know 8 6 10
Note: Teachers who answered yes to question 13 were then given question 14 to answer.
B. Receiving adequate support outside the school day, after school (e.g., with homework)?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 21 18 24
Some 46 43 50
About half 8 6 10
Most 7 6 9
Almost all/All 2 1 4
Don’t know 15 13 18
Note: Teachers who answered yes to question 13 were then given question 14 to answer.
Page 12 GAO-22-105817 Pandemic Learning
Student Learning During the COVID-19 Pandemic
Q15_T
GAO note: For this set of questions, teachers who spent the majority of the school year in an in-
person environment, virtual environment, or hybrid environment were asked to consider their
experiences in that particular instructional model. Teachers who spent the majority of the school year
in a mix of learning—some students generally attended in person full-time and other students
generally attended virtually full-time—were randomly assigned to answer these questions from either
the in-person learning perspective or the virtual learning perspective. Teachers proceeded through
question 15 after seeing one of the three forms below:
1. FORM_A
For each of the questions in this section, please consider your
experiences teaching students who were primarily learning virtually
(learning remotely).
Thinking about the time you spent teaching virtually, how many of
your students regularly experienced the following difficulties?
2. FORM_B
For each of the questions in this section, please consider your
experiences teaching students who were primarily learning in person.
Thinking about the time you spent teaching in person, how many of
your students regularly experienced the following difficulties?
3. FORM_C
For each of the questions in this section, please consider your
experiences teaching students who were primarily learning in a hybrid
setting.
Thinking about the time you spent teaching hybrid, how many of your
students regularly experienced the following difficulties?
A. Absence from school or class
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 14 12 16
Some 62 60 65
About half 15 14 17
Most 6 5 8
Almost all/All 2 1 3
Page 13 GAO-22-105817 Pandemic Learning
B. Late arrival to or early departure from class
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 17 15 19
Some 62 60 65
About half 13 12 15
Most 6 5 7
Almost all/All 2 1 3
C. Difficulty completing class assignments
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 9 8 11
Some 49 47 52
About half 24 22 27
Most 13 11 14
Almost all/All 4 3 6
D. Difficulty completing homework
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 7 6 9
Some 40 37 42
About half 19 17 21
Most 13 12 15
Almost all/All 7 6 9
Does not apply 13 12 16
Page 14 GAO-22-105817 Pandemic Learning
E. Difficulty understanding lessons
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 10 9 12
Some 64 61 66
About half 18 16 21
Most 5 4 7
Almost all/All 2 1 3
F. Limited or no class participation
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 17 15 19
Some 45 42 47
About half 18 17 20
Most 15 14 17
Almost all/All 5 4 6
G. Failing a class or being significantly behind meeting academic standards for a subject
Estimated percentage of 95 percent confidence 95 percent confidence
teachers interval – lower bound interval – upper bound
Response (percentage) (percentage)
None/Almost none 15 13 17
Some 56 54 59
About half 18 16 20
Most 9 7 10
Almost all/All 2 2 3
H. Lack of a dedicated device to use for learning (e.g., personal computer, tablet)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 77 74 80
Some 20 17 22
About half 2 1 3
Most 1 1 2
Almost all/All 0 0 1
Page 15 GAO-22-105817 Pandemic Learning
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
in-person learning environment did not answer this question.
I. Lack of reliable internet service
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 30 27 33
Some 59 55 62
About half 8 6 9
Most 3 2 4
Almost all/All 1 0 2
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
in-person learning environment did not answer this question.
J. Difficulty using devices or other technology to participate in class or complete assignments
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 42 39 45
Some 47 45 50
About half 7 6 8
Most 3 2 4
Almost all/All 1 0 2
K. Difficulty obtaining supplies or educational materials (e.g., pencils, notebooks, paper packets, manipulatives, lab kits,
library books)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 56 53 59
Some 34 32 37
About half 6 5 8
Most 3 2 4
Almost all/All 1 0 2
Page 16 GAO-22-105817 Pandemic Learning
L. Lack of an appropriate workspace, free from distractions or other difficulties
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 25 23 28
Some 42 40 45
About half 19 17 21
Most 11 9 12
Almost all/All 3 2 5
M. Difficulty getting assistance, support, or supervision at their workspace
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 29 27 32
Some 45 43 48
About half 13 12 15
Most 10 8 11
Almost all/All 2 2 3
N. Being distracted/busy providing assistance to siblings or younger students
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 8 6 10
Some 56 53 59
About half 21 19 24
Most 11 9 13
Almost all/All 4 3 6
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
in-person learning environment did not answer this question.
O. Lack of adequate breakfast/lunch
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 49 46 52
Some 36 33 38
About half 5 4 6
Most 2 2 3
Almost all/All 1 1 2
Page 17 GAO-22-105817 Pandemic Learning
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Does not apply 7 6 8
What other factors in students’ workspaces, if any, created
difficulties in learning?
This question was open-ended. Responses are omitted to preserve
respondents’ confidentiality.
P. Signs of emotional distress
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 20 18 22
Some 61 59 64
About half 13 11 15
Most 5 4 6
Almost all/All 1 1 2
Q. Focus or attention issues
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 4 3 5
Some 46 43 48
About half 28 26 30
Most 18 16 20
Almost all/All 4 3 6
R. Disruptive behaviors (verbal or physical)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost none 45 43 48
Some 49 46 51
About half 4 3 6
Most 1 1 2
Almost all/All 0 0 1
Page 18 GAO-22-105817 Pandemic Learning
Q16 Compared to a typical school year, how frequently have each of
the following occurred among your students this school year?
GAO note: For this set of questions, teachers who spent the majority of the school year in an in-
person environment, virtual environment, or hybrid environment were asked to consider their
experiences in that particular instructional model. Teachers who spent the majority of the school year
in a mix of learning—some students generally attended in person full-time and other students
generally attended virtually full-time—were randomly assigned to answer these questions from either
the in-person learning perspective or the virtual learning perspective. Teachers proceeded through
question 16 after seeing one of the three forms below:
1. FORM_A For each of the questions in this section, please consider
your experiences teaching students who were primarily learning
virtually (learning remotely).
Compared to a typical school year, how frequently have each of the
following occurred among your students this school year?
2. FORM_B For each of the questions in this section, please consider
your experiences teaching students who were primarily learning in
person.
Compared to a typical school year, how frequently have each of the
following occurred among your students this school year?
3. FORM_C For each of the questions in this section, please consider
your experiences teaching students who were primarily learning in a
hybrid setting.
Compared to a typical school year, how frequently have each of the
following occurred among your students this school year?
A. Absence from school or class
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 8 7 10
Somewhat less 9 7 10
About the same amount 19 17 22
Somewhat more 38 36 41
Much more 25 23 28
Page 19 GAO-22-105817 Pandemic Learning
B. Late arrival to or early departure from class
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 7 6 9
Somewhat less 7 6 9
About the same amount 30 28 33
Somewhat more 33 31 36
Much more 22 20 25
C. Difficulty completing class assignments
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 3 2 4
Somewhat less 5 3 6
About the same amount 29 26 31
Somewhat more 36 33 38
Much more 28 26 31
D. Difficulty completing homework
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 3 2 4
Somewhat less 4 3 6
About the same amount 24 22 26
Somewhat more 22 20 25
Much more 29 26 31
Does not apply 18 16 21
Page 20 GAO-22-105817 Pandemic Learning
E. Difficulty understanding lessons
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 3 2 4
Somewhat less 5 4 6
About the same amount 41 39 44
Somewhat more 38 35 40
Much more 13 12 15
F. Limited or no class participation
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 4 3 5
Somewhat less 6 5 8
About the same amount 30 27 32
Somewhat more 28 26 31
Much more 32 30 35
G. Failing a class or being significantly behind meeting academic standards for a subject
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 4 3 5
Somewhat less 4 3 5
About the same amount 29 27 31
Somewhat more 39 37 42
Much more 24 22 26
H. Difficulty using devices or other technology to participate in class or complete assignments
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 12 10 14
Somewhat less 16 14 18
About the same amount 42 39 44
Somewhat more 22 20 25
Much more 8 6 10
Page 21 GAO-22-105817 Pandemic Learning
I. Difficulty obtaining supplies or educational materials (e.g., pencils, notebooks, paper packets, manipulatives, lab kits,
library books)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 11 10 13
Somewhat less 12 10 14
About the same amount 53 51 56
Somewhat more 18 16 20
Much more 6 5 8
J. Lack of an appropriate workspace, free from distractions or other difficulties
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 6 5 7
Somewhat less 5 4 6
About the same amount 31 29 34
Somewhat more 33 31 36
Much more 25 23 28
K. Difficulty getting assistance, support, or supervision at their workspace
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 5 4 7
Somewhat less 6 5 8
About the same amount 36 33 38
Somewhat more 32 30 35
Much more 20 18 23
L. Lack of school/district-provided breakfast/lunch
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 29 26 31
Somewhat less 9 8 11
About the same amount 40 38 43
Somewhat more 15 13 17
Much more 7 6 8
Page 22 GAO-22-105817 Pandemic Learning
M. Responsibility for caring for others in the household
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 3 2 5
Somewhat less 3 2 4
About the same amount 21 19 24
Somewhat more 42 38 45
Much more 31 28 34
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
in-person learning environment did not answer this question.
N. Signs of emotional distress
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 3 2 4
Somewhat less 6 5 7
About the same amount 30 28 33
Somewhat more 47 45 50
Much more 14 13 16
O. Focus or attention issues
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 2 1 3
Somewhat less 3 2 4
About the same amount 26 23 28
Somewhat more 41 38 43
Much more 29 27 31
P. Disruptive behaviors (verbal or physical)
Estimated percentage of 95 percent confidence 95 percent confidence
teachers interval – lower bound interval – upper bound
Response (percentage) (percentage)
Much less 22 20 24
Somewhat less 17 15 19
About the same amount 41 39 44
Somewhat more 15 13 17
Much more 5 4 6
Page 23 GAO-22-105817 Pandemic Learning
Q17_T For this next set of questions, please think about all of your
students.
Q17 How many of your students were at or above grade level at the beginning of a typical school year?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 5 4 6
Some 30 28 33
About half 28 26 30
Most 35 33 38
All 2 1 3
Note: Teachers who responded that they have been teaching for less than 2 years did not answer this
question.
Q18 How many of your students were at or above grade level during the beginning of the 2020-2021 school year?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 7 5 8
Some 46 43 48
About half 25 23 27
Most 21 20 23
All 1 1 2
Q19 How many of your students were below grade level at the beginning of this school year compared to a typical school
year?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Much less 1 1 2
Somewhat less 3 2 4
About the same amount 43 41 46
Somewhat more 39 36 41
Much more 13 11 15
All 2 1 3
Note: Teachers who responded that they have been teaching for less than 2 years did not answer this
question nor did teachers who responded that all of their students were at or above grade level at the
beginning of the 2020-21 school year and at the beginning of a typical school year.
Page 24 GAO-22-105817 Pandemic Learning
Q20 Currently, or at the end of the 2020-2021 school year…How many of your students were behind grade level?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 4 3 4
Some 52 49 54
About half 23 21 26
Most 19 17 21
All 3 2 4
GAO note: There was no Q21 in the survey.
Q22 Currently, or at the end of the 2020-2021 school year…How many of your students were at grade level?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 3 2 4
Some 38 36 41
About half 27 25 30
Most 30 28 33
All 1 1 2
GAO note: There was no Q23 in the survey.
Q24 Currently, or at the end of the 2020-2021 school year…How many of your students were above grade level?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 13 11 15
Some 78 76 80
About half 5 4 6
Most 4 3 4
All 0 0 1
Page 25 GAO-22-105817 Pandemic Learning
Q25 Still thinking about all of your students, which of the following best describes the amount of academic progress your
students have made this year, compared to a typical school year?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Significantly less progress 19 17 21
Somewhat less progress 45 42 47
About the same amount of 25 23 27
progress
Somewhat more progress 9 8 11
Significantly more progress 2 1 3
Note: Teachers who responded that they have been teaching for less than 2 years did not answer this
question.
Q26 During the 2020-2021 school year, did you have students who have made less academic progress than you would have
expected in a typical school year?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 84 82 86
No 16 14 18
Note: Teachers who responded that they have been teaching for less than 2 years did not answer this
question.
Q27 Please think about your students who made less academic
progress this year compared to a typical school year. Did each of
the following factors contribute to these students making less
progress?
GAO note: Only teachers who answered yes to Q26 were asked Q27.
A. Difficulties related to learning virtually (e.g., weak or no internet connection, not having a dedicated workspace for
learning)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 79 77 82
No 16 14 18
Does not apply 5 4 7
Page 26 GAO-22-105817 Pandemic Learning
B. Students starting off the school year below grade level
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 82 80 84
No 16 14 18
Does not apply 2 2 3
C. Changes to curriculum
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 24 22 27
No 56 53 59
Does not apply 20 18 22
D. Absence of consequences for a lack of academic progress
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 77 74 79
No 19 17 21
Does not apply 5 3 6
E. Inability to cover as much content or go into as much depth
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 74 71 77
No 23 20 25
Does not apply 3 2 4
F. Changes at home caused by the pandemic (e.g., job loss in family, pandemic-related move)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 80 78 82
No 16 13 18
Does not apply 4 3 6
Page 27 GAO-22-105817 Pandemic Learning
G. Social or emotional issues
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 83 80 85
No 14 12 17
Does not apply 3 2 4
Q28 Thinking about your students who have made more academic
progress this year compared to a typical school year, explain what
factors contributed to that progress.
This question was open-ended. Responses are omitted to preserve
respondents’ confidentiality. Also, teachers who responded that they have
been teaching for less than 2 years did not answer this question.
Strategies for Supporting Learning and Addressing Learning Loss During
the Pandemic
Q29 Approximately, how often during this school year have you
used the following strategies to support learning or address learning
loss?
GAO note: For this set of questions, teachers who spent the majority of the school year in an in-
person environment, virtual environment, or hybrid environment were asked to consider their
experiences in that particular instructional model. Teachers who spent the majority of the school year
in a mix of learning—some students generally attended in person full-time and other students
generally attended virtually full-time—were randomly assigned to answer these questions from either
the in-person learning perspective or the virtual learning perspective. Teachers proceeded through
question 29 after seeing one of the three forms below:
1. FORM_A
For each of the questions in this section, please consider your
experiences teaching students who were primarily learning virtually
(learning remotely).
2. FORM_B
For each of the questions in this section, please consider your
experiences teaching students who were primarily learning in
person.
3. FORM_C
For each of the questions in this section, please consider your
experiences teaching students who were primarily hybrid learning.
(Learning loss is the extent to which a student is behind where they would
be at the same point in a typical school year.)
Page 28 GAO-22-105817 Pandemic Learning
A. Use of live, virtual instruction
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 2 1 3
A few times during the school 4 3 6
year
About once or twice per month 3 2 4
About once or twice per week 11 9 13
Every day or almost every day 80 77 82
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
in-person learning environment did not answer this question.
B. Use of live, in-person instruction
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 3 2 4
A few times during the school 4 3 5
year
About once or twice per month 1 1 2
About once or twice per week 9 8 11
Every day or almost every day 83 80 85
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
virtual learning environment did not answer this question.
C. Use of asynchronous learning (recordings, independent work time)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 6 5 7
A few times during the school 17 15 19
year
About once or twice per month 9 8 11
About once or twice per week 27 25 30
Every day or almost every day 41 39 44
Page 29 GAO-22-105817 Pandemic Learning
D. Small group work connecting over devices
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 19 17 21
A few times during the school 23 20 25
year
About once or twice per month 12 11 14
About once or twice per week 25 23 27
Every day or almost every day 21 19 24
E. Small group work in person
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 29 27 32
A few times during the school 18 16 20
year
About once or twice per month 8 7 9
About once or twice per week 22 20 25
Every day or almost every day 23 20 25
F. Use of educational apps or platforms for subject learning activities (e.g., Lexia, Canvas)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 9 8 11
A few times during the school 7 6 8
year
About once or twice per month 7 6 8
About once or twice per week 18 16 20
Every day or almost every day 59 56 61
Page 30 GAO-22-105817 Pandemic Learning
G. Use of apps or platforms to submit assignments and provide feedback (e.g., Seesaw, Google Classroom)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 4 3 5
A few times during the school 4 3 5
year
About once or twice per month 4 3 5
About once or twice per week 11 9 13
Every day or almost every day 78 75 80
H. Use of paper packets, manipulatives, or other physical instructional materials
Estimated percentage of 95 percent confidence 95 percent confidence
teachers interval – lower bound interval – upper bound
Response (percentage) (percentage)
Never 26 24 28
A few times during the school 21 19 23
year
About once or twice per month 9 7 10
About once or twice per week 21 19 23
Every day or almost every day 24 22 27
I. Increased number of teachers or staff to support student learning
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 61 59 64
A few times during the school 14 12 15
year
About once or twice per month 5 4 6
About once or twice per week 9 7 11
Every day or almost every day 11 10 13
Page 31 GAO-22-105817 Pandemic Learning
J. Team building/get-to-know-you activities
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 11 9 12
A few times during the school 44 42 47
year
About once or twice per month 15 13 17
About once or twice per week 18 16 20
Every day or almost every day 12 11 14
K. One-on-one check-ins between teacher and student
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 5 4 6
A few times during the school 21 19 24
year
About once or twice per month 25 22 27
About once or twice per week 28 26 31
Every day or almost every day 21 19 23
L. One-on-one or small group meetings with counselor
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 35 33 38
A few times during the school 38 35 40
year
About once or twice per month 15 13 18
About once or twice per week 9 8 11
Every day or almost every day 2 1 3
Page 32 GAO-22-105817 Pandemic Learning
M. Reaching out to parents or guardians to discuss social/emotional needs
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 6 5 7
A few times during the school 34 32 37
year
About once or twice per month 32 30 35
About once or twice per week 21 19 24
Every day or almost every day 6 5 8
N. Use of individual or small group tutoring sessions during the school day
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 14 12 16
A few times during the school 18 16 20
year
About once or twice per month 14 13 16
About once or twice per week 31 29 34
Every day or almost every day 23 21 25
O. Providing after school or weekend tutoring
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 41 38 43
A few times during the school 15 13 17
year
About once or twice per month 8 7 10
About once or twice per week 19 17 21
Every day or almost every day 9 8 11
Does not apply 8 7 10
Page 33 GAO-22-105817 Pandemic Learning
P. Movement breaks
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 14 13 16
A few times during the school 11 10 13
year
About once or twice per month 7 6 8
About once or twice per week 16 14 18
Every day or almost every day 52 49 55
Q. Providing extended school day (e.g., allowing students to learn during times other than the typical school day hours)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 58 54 61
A few times during the school 10 8 12
year
About once or twice per month 6 4 7
About once or twice per week 11 9 13
Every day or almost every day 15 13 18
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
virtual learning environment did not answer this question.
R. Providing flexible school day (e.g., allowing students to learn during times other than the typical school day hours)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Never 33 30 36
A few times during the school 12 10 14
year
About once or twice per month 8 7 10
About once or twice per week 15 13 18
Every day or almost every day 32 29 35
Note: Teachers who responded that they spent the majority of the 2020-21 school year in a full-time
in-person learning environment did not answer this question.
Page 34 GAO-22-105817 Pandemic Learning
Q30 How many of your students improved their academic progress
as a result of each of the following strategies?
GAO note: For this set of questions, teachers who spent the majority of the school year in an in-
person environment, virtual environment, or hybrid environment were asked to consider their
experiences in that particular instructional model. Teachers who spent the majority of the school year
in a mix of learning—some students generally attended in person full-time and other students
generally attended virtually full-time—were randomly assigned to answer these questions from either
the in-person learning perspective or the virtual learning perspective. Teachers were not asked about
strategies they responded to having never used in Q29.
A. Use of live, virtual instruction
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 8 6 9
Some 37 34 40
About half 13 11 15
Most 26 23 29
Almost All/All 18 15 20
B. Use of live, in-person instruction
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 1 1 2
Some 14 12 16
About half 9 7 11
Most 33 30 37
Almost All/All 42 39 46
C. Use of asynchronous learning (recordings, independent work time)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 14 12 15
Some 49 46 52
About half 17 15 19
Most 14 12 16
Almost All/All 7 6 8
Page 35 GAO-22-105817 Pandemic Learning
D. Small group work connecting over devices
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 17 15 19
Some 46 44 49
About half 13 11 15
Most 17 14 19
Almost All/All 7 6 9
E. Small group work in person
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 9 7 11
Some 31 28 33
About half 14 12 16
Most 28 25 31
Almost All/All 19 16 21
F. Use of educational apps or platforms for subject learning activities (e.g., Lexia, Canvas)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 7 6 9
Some 37 34 40
About half 16 14 18
Most 22 20 24
Almost All/All 18 16 20
Page 36 GAO-22-105817 Pandemic Learning
G. Use of apps or platforms to submit assignments and get feedback (e.g., Seesaw, Google Classroom)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 6 5 7
Some 29 27 32
About half 13 11 15
Most 26 24 29
Almost All/All 26 24 28
H. Use of paper packets, manipulatives, or other physical instructional materials
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 12 10 14
Some 35 32 38
About half 14 12 16
Most 25 22 28
Almost All/All 15 13 17
I. Increased number of teachers or staff to support student learning
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 14 11 18
Some 42 38 47
About half 12 9 15
Most 19 16 23
Almost All/All 13 10 16
Page 37 GAO-22-105817 Pandemic Learning
J. Team building/get-to-know-you activities
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 21 19 24
Some 47 44 49
About half 10 9 12
Most 13 11 15
Almost All/All 8 7 10
K. One-on-one check-ins between teacher and student
Estimated percentage of 95 percent confidence 95 percent confidence
teachers interval – lower bound interval – upper bound
Response (percentage) (percentage)
None/Almost None 5 4 6
Some 43 40 46
About half 15 13 17
Most 23 21 25
Almost All/All 14 12 16
L. One-on-one or small group meetings with counselor
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 17 14 20
Some 55 52 59
About half 9 7 11
Most 14 11 16
Almost All/All 5 3 7
M. Reaching out to parents or guardians to discuss social/emotional needs
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 7 6 9
Some 57 55 60
About half 14 12 16
Most 15 13 17
Almost All/All 7 6 9
Page 38 GAO-22-105817 Pandemic Learning
N. Use of individual or small group tutoring sessions during the school day
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 10 8 11
Some 50 47 52
About half 12 10 14
Most 18 16 20
Almost All/All 11 9 13
O. Providing after school or weekend tutoring
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 17 14 20
Some 58 54 61
About half 9 7 11
Most 12 10 14
Almost All/All 5 4 7
P. Movement breaks
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 16 14 18
Some 41 38 44
About half 12 10 14
Most 18 16 21
Almost All/All 13 11 15
Page 39 GAO-22-105817 Pandemic Learning
Q. Providing extended school day or flexible school day (e.g., allowing students to learn during times other than the typical
school day hours)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 13 10 17
Some 55 49 60
About half 14 11 18
Most 10 8 14
Almost All/All 7 5 11
R. Providing flexible school day (e.g., allowing students to learn during times other than the typical school day hours)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None/Almost None 14 11 17
Some 49 45 53
About half 12 10 15
Most 15 12 18
Almost All/All 10 8 13
Q31A What other strategies were most useful for helping your
students make academic progress?
This question was open-ended. Responses are omitted to preserve respondents’ confidentiality.
Q31B What was the most useful strategy for addressing learning
loss?
This question was open-ended. Responses are omitted to preserve respondents’ confidentiality.
Q32 Briefly describe what strategies have been the most helpful for
teaching English Learners this school year. For example, this may
include strategies for teaching vocabulary or sentence framing,
such as using directions that include images or allowing students to
record responses rather than spell them out in writing.
This question was open-ended. Responses are omitted to preserve
respondents’ confidentiality. Teachers who responded they had no
English Learners earlier in the survey did not answer this question.
Page 40 GAO-22-105817 Pandemic Learning
Q33 Given your experiences teaching during the pandemic, what
additional resources would help students to learn successfully in
any future disruptions to the typical learning environment?
This question was open-ended. Responses are omitted to preserve respondents’ confidentiality.
This section of the survey asks about students with whom your school
had limited contact this school year. First, we’ll ask you about students
who were unaccounted for and then we will ask you about students who
were disengaged.
Unaccounted for: Students who were registered to be in your class(es)
but who did not attend all year. These students are sometimes referred to
as “no shows”.
Disengaged: Students whose limited participation in your class(es) greatly
affected their learning or grades. Examples of limited participation
include:
- logging in but not participating,
- attending class but not completing or turning in work, or
- missing a substantial number of classes.
Q34 Thinking about the current school year, how many unaccounted-for students did you have?
(Students who were registered to be in your class(es) but who did not attend all year.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
No unaccounted-for students 52 50 55
1 or more unaccounted-for 48 45 50
students
Note: We asked teachers to provide a number for how many unaccounted-for students they had, but
we report the results here in two groupings: no students and 1 or more students.
Page 41 GAO-22-105817 Pandemic Learning
Q35 How many disengaged students did you have?
(Students whose limited participation in your class(es) greatly affected their learning or grades.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
No disengaged students 15 13 17
1 or more disengaged students 85 83 87
Note: We asked teachers to provide a number for how many disengaged students they had, but we
report the results here in two groupings: no students and 1 or more students.
Q36 Still thinking about the current school year, how does the number of unaccounted-for students compare to a typical
school year?
(Students who were registered to be in your class(es) but who did not attend all year.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Substantially fewer 9 7 10
Somewhat fewer 2 2 3
About the same 46 43 49
Somewhat more 19 17 22
Substantially more 24 22 26
Note: Teachers who responded that they have been teaching for less than 2 years did not answer this
question.
Q37 Still thinking about the current school year, how does the number of disengaged students compare to a typical school
year?
(Students whose limited participation in your class(es) greatly affected their learning or grades.)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Substantially fewer 4 3 6
Somewhat fewer 2 1 3
About the same 20 18 23
Somewhat more 29 27 31
Substantially more 44 42 47
Note: Teachers who responded that they have been teaching for less than 2 years did not answer this
question.
Page 42 GAO-22-105817 Pandemic Learning
Q38 How much of a factor were each of the following reasons in
explaining why these students were unaccounted for?
Note: Teachers who responded they did not have any unaccounted for students did not answer this
question.
A. Student had difficulty learning in or adapting to the virtual environment
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 6 5 8
Not too much of a factor 8 6 10
Somewhat of a factor 22 19 25
Significant factor 38 35 41
Don’t know 26 23 29
B. Student had limited or no adult assistance or support at home
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 3 2 5
Not too much of a factor 3 2 4
Somewhat of a factor 17 15 20
Significant factor 56 53 60
Don’t know 20 18 23
C. Student was providing care to a family member
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 12 10 15
Not too much of a factor 11 9 13
Somewhat of a factor 25 22 29
Significant factor 20 18 23
Don’t know 32 28 35
Page 43 GAO-22-105817 Pandemic Learning
D. Student had work commitments that interfered with school
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 31 27 34
Not too much of a factor 10 8 12
Somewhat of a factor 15 13 17
Significant factor 15 13 17
Don’t know 30 26 33
E. Student did not have reliable internet access
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 18 16 21
Not too much of a factor 18 16 21
Somewhat of a factor 27 24 31
Significant factor 18 15 21
Don’t know 18 15 21
F. Student did not have a device for accessing the internet (e.g., personal computer, tablet) or had to share the device
Estimated percentage of 95 percent confidence 95 percent confidence
teachers interval – lower bound interval – upper bound
Response (percentage) (percentage)
Not at all a factor 56 52 59
Not too much of a factor 15 13 18
Somewhat of a factor 9 7 12
Significant factor 7 6 10
Don’t know 13 10 15
Page 44 GAO-22-105817 Pandemic Learning
Q39 How much of a factor were each of the following reasons in
explaining why these students were disengaged?
Note: Teachers who responded they did not have any disengaged students did not answer this
question.
A. Student had difficulty learning in or adapting to the virtual environment
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 6 5 8
Not too much of a factor 8 6 9
Somewhat of a factor 28 26 31
Significant factor 52 49 54
Don’t know 7 5 8
B. Student had limited or no adult assistance or support at home
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 3 2 4
Not too much of a factor 5 4 7
Somewhat of a factor 23 20 25
Significant factor 60 58 63
Don’t know 9 7 10
C. Student was providing care to a family member
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 17 15 19
Not too much of a factor 18 16 21
Somewhat of a factor 28 26 31
Significant factor 16 14 18
Don’t know 20 18 23
Page 45 GAO-22-105817 Pandemic Learning
D. Student had work commitments that interfered with school
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 48 45 50
Not too much of a factor 11 9 13
Somewhat of a factor 14 13 16
Significant factor 11 10 13
Don’t know 16 14 18
E. Student did not have reliable internet access
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 22 19 24
Not too much of a factor 25 23 28
Somewhat of a factor 30 27 33
Significant factor 16 14 19
Don’t know 7 5 8
F. Student did not have a device for accessing the internet (e.g., personal computer, tablet) or had to share the device
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Not at all a factor 65 62 68
Not too much of a factor 15 13 17
Somewhat of a factor 9 8 11
Significant factor 6 5 8
Don’t know 4 3 5
Q40 How many of your unaccounted-for and disengaged students had each of the following characteristics? Please select
“Don’t know” if you are not familiar with the students’ characteristics.
A. From a low-income household
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 5 4 6
Some 20 18 22
Page 46 GAO-22-105817 Pandemic Learning
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
About half 9 8 11
Most 29 27 32
All 19 17 21
Don’t know 18 16 20
B. From a food-insecure household
Response 95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
teachers (percentage) (percentage)
None 10 9 12
Some 23 21 25
About half 8 7 10
Most 16 14 18
All 6 4 7
Don’t know 37 34 39
C. A member of a racial or ethnic minority group
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 11 10 13
Some 26 24 29
About half 10 9 12
Most 24 21 26
All 20 17 22
Don’t know 8 7 10
D. English learner
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 37 34 39
Some 37 35 40
About half 8 6 9
Most 7 5 8
All 5 4 7
Don’t know 7 6 8
Page 47 GAO-22-105817 Pandemic Learning
E. Has an IEP
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 24 22 27
Some 55 52 58
About half 8 7 10
Most 4 3 6
All 3 2 4
Don’t know 5 4 7
F. Has a 504 plan
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 34 31 37
Some 52 49 54
About half 3 2 4
Most 2 1 3
All 0 0 1
Don’t know 10 8 11
G. Homeless or has been homeless this school year
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 40 37 43
Some 21 19 24
About half 0 0 1
Most 0 0 1
All 0 0 1
Don’t know 38 35 41
Page 48 GAO-22-105817 Pandemic Learning
H. Failing or potentially failing a class or grade; or significantly behind meeting academic standards for a subject
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 5 4 6
Some 34 31 36
About half 12 10 14
Most 27 24 29
All 17 15 20
Don’t know 6 4 7
I. Has a parent or guardian whose job requires them to work outside the home (e.g., an essential worker)
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 5 4 7
Some 23 21 26
About half 10 8 12
Most 23 21 25
All 6 5 8
Don’t know 32 30 35
J. Other
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
None 40 34 46
Some 37 31 43
About half 9 6 13
Most 9 6 12
All 6 4 9
Q41 Thinking about your unaccounted-for students, what additional
resources would you have needed to reengage them?
This question was open-ended. Responses are omitted to preserve respondents’ confidentiality.
Teachers who responded they did not have any unaccounted for students did not answer this
question.
Q41a Thinking about your disengaged students, what additional
resources would you have needed to reengage them?
Page 49 GAO-22-105817 Pandemic Learning
This question was open-ended. Responses are omitted to preserve respondents’ confidentiality.
Teachers who responded they did not have any disengaged students did not answer this question.
Finally, we have a few questions that will be used for demographic
purposes only.
D1 What is your age?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Aged 21-30 15 13 17
Aged 31-40 29 26 31
Aged 41-50 30 28 33
Aged 51-60 19 17 21
Aged 61-70 6 5 8
Aged 71-80 1 0 1
Aged 81-90 0 0 0
Aged 91-100 0 0 0
Note: While we asked this as an open-ended questions, we aggregated responses by 10 year age
bands.
D2 What is your gender?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Male 25 23 27
Female 74 72 76
Other 0 0 0
Prefer not to answer 1 1 2
D3 Are you of Hispanic, Latino, or Spanish origin? For example, Mexican, Mexican American, Chicano, Puerto Rican, Cuban,
or another Hispanic, Latino, or Spanish origin?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
Yes 9 7 11
No 91 89 93
Page 50 GAO-22-105817 Pandemic Learning
D4 Which of the following describes your race? You may select one or more.
95 percent confidence 95 percent confidence
Estimated number of interval – lower bound interval – upper bound
Response teachers (number) (number)
White 1,932,644 1,848,516 2,016,771
Black or African-American 190,740 153,795 227,686
Asian 56,204 36,698 75,710
American Indian or Alaska 27,882 16,592 39,172
Native
Native Hawaiian or Pacific 6,500 0 13,200
Islander
Other (please list) 19,950 12,517 27,382
D5 What is the highest level of school you have completed or the highest degree you have received?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
High school graduate 0 0 0
Technical, trade, vocational or 0 0 0
business school or program after
high school
Some college – college, 0 0 1
university or community college
– but no degree
Two-year associate degree from 1 0 1
a college, university, or
community college
Four-year bachelor’s degree 22 20 25
(e.g., BS, BA, AB)
Some postgraduate or 15 14 17
professional schooling but no
degree (e.g., some graduate
school)
Postgraduate or professional 61 58 64
degree, including master’s or
doctorate (e.g., MA, MS, PhD,
MD, JD)
Page 51 GAO-22-105817 Pandemic Learning
D6 In which state do you teach?
Responses are omitted to preserve respondents’ confidentiality.
D7 In which city do you teach?
Responses are omitted to preserve respondents’ confidentiality.
D8 What is the name of the school where you currently teach? We
will use information about your school for demographic purposes
only. We will not share results by school and your results will never
be shared with your school.
Responses are omitted to preserve respondents’ confidentiality.
D9 What is the ZIP Code where your school is located?
Responses are omitted to preserve respondents’ confidentiality.
Q2_2 Which of the following best describes the location of the school you taught at during the 2020-2021 school year?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
A rural area or small town 33 30 35
A large city 28 26 31
A suburb of a large city 39 36 41
D11 Approximately what percentage of students in your classes receive free or reduced lunch?
95 percent confidence 95 percent confidence
Estimated percentage of interval – lower bound interval – upper bound
Response teachers (percentage) (percentage)
0 – 20% 18 16 20
21 – 40% 16 14 18
41 – 60% 17 15 19
61 – 80% 13 12 15
81 – 100% 36 34 39
Page 52 GAO-22-105817 Pandemic Learning
Technical Description of Discussion Groups with Educators and Parents
To gain further insights about the topics covered in our survey of K-12 public school teachers
and to get the perspectives of other K-12 stakeholders, we held 18 virtual discussion groups
with public school teachers (six groups), parents of students (six groups), and principals (six
groups) between June 29 and July 14, 2021. 1 We contracted with Gallup to recruit and arrange
the K-12 public school groups. In addition, to respond to a provision in the conference report
accompanying the National Defense Authorization Act of Fiscal Year 2021 for GAO to conduct a
review of virtual learning in Department of Defense Education Activity (DODEA) schools, we
held two additional discussion groups. One group was with teachers and the other with parents
of students in DODEA schools on July 19, 2021 and July 21, 2021, respectively. We worked
with DODEA to recruit and arrange these two discussion groups. To select discussion group
comments for the reports, we first analyzed and coded the discussion group transcripts to
identify common themes among the groups. We then compared these themes to our survey
results to identify comments that were illustrative of the key themes across the survey and
discussion group analyses. Comments, information, and views obtained from these discussion
groups are not generalizable to other educators and parents.
Discussion Groups with K-12 Public School Educators and Parents
GAO moderators structured and guided the discussions using a standardized list of questions to
encourage participants to share their thoughts and experiences on students’ learning during
school year 2020-21 and strategies used to mitigate learning loss. We developed discussion
guides tailored to each stakeholder group (teachers, principals, and parents) without Gallup’s
input. Prior to conducting any of the discussion groups with participants recruited by Gallup, we
pretested our discussion guide with one teacher, two parents, and two principals.
Gallup recruited and screened participants and managed the logistics of convening the groups
with K-12 public school educators and parents. In total, Gallup arranged 18 virtual discussion
groups with three participant categories: teachers, parents, and principals. Each participant
category was further segmented based on their school’s geographic location (urban, suburban
or rural), with two groups for each participant type and location. In the end, we held six groups
with K-12 public school teachers, six groups with parents of children in K-12 public schools, and
six groups with K-12 public school principals.
Teachers participating in the groups had also responded to our generalizable teacher survey in
June 2021. After each teacher discussion group, Gallup emailed teacher participants to thank
them for their participation and requested that they share referrals with parents and principals.
These referrals accounted for about 5 percent of parents and 10 percent of principals. To
supplement the referrals, Gallup first recruited parents from surrounding communities (an
additional 45 percent of parents) and then from the same state as the teachers (the remaining
50 percent of parents). Similarly, Gallup recruited additional principals from the teachers’
schools or other schools in the same school district (about 20 percent of principals) and then
from the same state as the teachers (another 60 percent of principals). Gallup leveraged its
network for the final 10 percent of principals.
1Following completion of the survey, teachers received additional screening questions to determine their eligibility,
interest, and availability to participate in our discussion groups.
Page 1 GAO-22-105817 Pandemic Learning
We initially set a minimum group size of four and a maximum of eight to create a group dynamic
that would allow sufficient interaction among participants and enable them to share their
perspectives in the allotted time period. In the end, all of our discussion groups had between
four and nine participants. Discussion groups were held by video conference in the evenings
and lasted one hour each to accommodate the schedules of participants. The contractor also
created a written transcript of each group. Comments, information and views obtained from
these discussion groups are not generalizable to other educators and parents.
Discussion Groups with DODEA Teachers and Parents
DODEA operates the Department of Defense’s school system and is responsible for planning,
directing, coordinating, and managing educational programs for military dependent children from
pre-kindergarten through grade 12. As of March 2022, the majority of DODEA’s 160 accredited
schools were located on overseas installations, but DODEA also operated 45 schools on 13
installations in seven states. Similar to K-12 public school students, many students at DODEA
schools spent time learning in a virtual environment during school year 2020-21. For our
discussion groups with DODEA teachers and parents (one group each), we selected a non-
generalizable sample from DODEA’s Southeast district in the U.S., which had the highest
percentage of stateside students in full-time virtual status as of October 2020.
For consistency across groups, we recruited groups of the same size for our K-12 public school
discussion groups. To solicit participants for our discussion groups, we asked DODEA to send
an email to teachers and parents to inform them of our request. Teachers and parents
interested in participating responded to us directly. We received more responses than we could
accommodate, so we considered five factors in selecting participants: (1) military installation
within the Southeast district, (2) grade levels teachers taught or grade level students were in, (3)
subjects teachers taught, (4) gender of teachers and parents, and (5) race or ethnicity of
teachers and parents. In the end, five DODEA teachers participated in one group and five
DODEA parents participated in the other. As with our K-12 public school groups, we held the
group discussions by video conference in the evenings and each lasted one hour. We used the
same discussion guides as we did for our K-12 public school teacher and parent groups,
although we made limited revisions to ensure our questions were relevant in the DODEA
context. Comments, information, and views obtained from these discussion groups are not
generalizable to other DODEA teachers and parents.
Page 2 GAO-22-105817 Pandemic Learning
File and source
- File
- 22-105817-pandemic-learning-technical-materials-for-teacher-survey.pdf
- Size
- 1,247,546 bytes
- SHA-256
- f0386efa7c338456d2a3b71dbb76d7594850f125d9b5f4a1c58484a8697da29d
- Original
- www.gao.gov