The State of Public Education in 2024 in Five Charts

Every year RAND sends surveys to members of the American Educator Panels to keep tabs on what is happening in U.S. classrooms.

By Sy Doan, Elizabeth D. Steiner, Melissa Kay Diliberti, Julia H. Kaufman, Anna Shapiro, Ashley Woo

Photo by pixdeluxe/Getty Images

Every year RAND sends more than a dozen surveys to teachers, principals, and superintendents who are members of the American Educator Panels to keep tabs on what is happening in classrooms around the country. What pressures are teachers feeling? Are they keeping up with technology? How is instruction changing?

When we looked back on the last 12 months of reports based on our surveys, these are the findings that told us the most about the state of American public schools.

Teachers Who Feel Underpaid Want a $17,000 Raise

The average teacher reported a base salary of $67,000 during the 2022–2023 school year. Only 34 percent of K–12 teachers considered their base salary to be “somewhat” or “completely” adequate, compared with 61 percent of working adults. Teachers considering their salaries inadequate wanted a $17,000 raise, on average. This is similar to the size of the teacher pay gap, that is, the difference in earnings between teachers and non-teachers with the same level of education.

Low pay and long hours are leading reasons that teachers consider leaving their jobs. But pay increases alone — without improvements in teachers’ hours worked or working conditions — are unlikely to induce large shifts in teachers’ intentions to leave.

— Sy Doan & Elizabeth Steiner

Figure 1: What K–12 Teachers Earn — and What They Think About It

Source: https://www.rand.org/pubs/infographics/IGA1108-1.html

AI Have and Have-Nots

Teachers were much more likely to have been trained on how to use AI if they worked in historically advantaged school districts. For example, 27 percent of districts serving mostly White students had provided AI training by fall 2023 compared with only 11 percent of districts serving mostly students of color. If districts’ plans for ‘23–’24 training panned out, 65 percent of majority-White districts provided training on AI use this past school year compared with 39 percent of districts serving mostly students of color — a 26 percentage — point gap.

— Melissa Diliberti

Figure 2: Districts’ AI Training for Teachers

NOTE: This figure depicts response data from the following survey question posed to districts: “Has your district provided training to your teachers about use of generative artificial intelligence (like ChatGPT)?” (n = 224). Bars may not sum to 100 percent because of rounding. Source: https://www.rand.org/pubs/research_reports/RRA956-21.html

Early Tracking Has Lasting Implications

Grouping students by achievement level, called tracking, is one way to target instruction for students who may be in the same grade but have different levels of mastery over grade-level content. According to principals surveyed in spring 2023, tracking within mathematics classes can be happening as early as elementary school. Tracking into math classes is increasingly common by middle school, especially in some states, such as Florida. To make matters more complex, methods for tracking students can lead to biases in who gets access to more challenging coursework, such being able to enroll in algebra in 8th grade. Among the questions this data raises are: How early is too early to track students? Are decisions made in an impartial way? Are students getting stuck on lower-level tracks even when they could handle more challenging content?

— Julia Kaufman

Figure 3: How Students Get Tracked According to Elementary and Middle School Principals

NOTE: This figure uses principal survey response data for the following question (K–5, n = 1,715; 6–8, n = 1,143): “Some schools organize mathematics instruction differently for students with different achievement levels. What is your school’s policy about how students are grouped in the following grade bands?” Principals could choose only one response; response options are mutually exclusive. Asterisks (*) indicate that the percentage of principals in a given state who reported grouping by achievement level either within or into mathematics classes is significantly different from percentages of principals in the rest of the country (p < 0.05). This figure includes only principals of elementary and middle school grades. Some bars do not sum to 100 percent because of rounding. Source: https://www.rand.org/pubs/research_reports/RRA2836-2.html

Reading Instruction, Even in Middle and High School

States across the country have implemented education reforms targeting foundational skill development, most often aimed at supporting K–3 teachers. But as revealed in the National Assessment of Educational Progress, many American students are not proficient or advanced readers even at higher grade levels. RAND’s research suggests that teachers — even of grades 6–12 — are spending time on foundational reading skills. The foundational reading activities we asked about in our survey included practicing basic features of print concepts, phonological awareness (i.e., being able to identify units of oral language such as words or syllables), phonics skills to recognize or decode words, and fluency in reading text. Our data suggest that additional resources, staff, and professional development are needed to help teachers deliver grade level–appropriate reading instruction.

— Anna Shapiro

Figure 4: Reading Foundations Still Being Taught in Upper Grades, Too

Note: This figure depicts response data from the following survey question: “In the last five lessons you taught this class (i.e., the past week, if you teach every day), how often did students engage in each of the following tasks, with or without your prompting?” Each bar represents the proportion of teachers who responded “3–4 lessons” or “every lesson.” Teachers who did not respond to the item or responded “Not relevant or not appropriate for the grade/class I teach” are excluded from the analysis. Sample sizes are presented as a range because of variance in the number of respondents for each item. Black bars represent 95-percent confidence intervals. Source: https://www.rand.org/pubs/research_reports/RRA134-23.html

Political Climate Cools Conversations in Classrooms

Two-thirds of teachers reported that they had decided on their own to limit classroom discussions of political and social issues. Among those subject to local restrictions coming from their school or district leaders, this percentage was even higher — about 80 percent or more, regardless of the political climate of the local community in which they taught. But teachers who were not subject to any formal state or local restriction were more likely to limit what they taught if their local political climate was more conservative. Ultimately, teaching and learning may be starting to diverge in liberal and conservative communities.

— Ashley Woo

Figure 5: Discussion of Political and Social Issues Curbed by Local Restrictions and Political Climate

NOTE: Results are from a logistic regression model predicting the likelihood of teachers deciding, on their own, to limit discussions about political and social issues in class as a function of whether teachers taught in a school where they were subject to state and/or local restrictions and the local political climate, proxied by the share of the presidential vote that went for President Trump in 2020 in the county where the respondent’s school is located. Controls include school demographics (i.e., school grade level, school locale, school poverty status, and student racial/ethnic composition) and teacher demographics (i.e., gender, years of teaching experience, and race/ethnicity). The shaded areas present 95-percent confidence intervals. We exclude respondents who were not sure whether they were subject to local restrictions and/or were not sure whether they had ever decided to limit discussions about political and social issues in class. n = 1,106. Source: https://www.rand.org/pubs/research_reports/RRA1108-10.html

The American Educator Panels (AEP)

This originally appeared on rand.org on August 30, 2024.