IES Blog

In part 1 of this blog series, we highlight the benefits and advantages of using federal education data for policymaking at the federal, state, and district levels. In part 2, we will explore the challenges of and opportunities afforded by using these data.

States, districts, and schools are inundated with requests for data. To manage the volume of requests and avoid overwhelming educators, many districts have established processes to vet and limit the number of surveys allowed in their schools and administrative offices. District clearance processes are also understandably meant to make sure everyone is in compliance with data privacy laws. NCES data collections are sometimes not cleared by district offices, which then means NCES is not allowed to contact schools or educators to learn from their perspectives or experiences.

Since many state or district policymakers prioritize local survey collections, federal surveys are occasionally rejected by district offices that are striving to keep from overburdening their educators with too many survey requests. Without district permission, NCES surveys won’t include the educators in those schools, meaning that those districts’ voices will be missing from the table when decisions are made. This is problematic for the entire education system for a few reasons.

1. Local data rarely reach federal policymakers. We know state and district decisionmakers derive a lot of value from state and/or district survey collections—since they’re designed to provide detailed local data—and do not always see value in federal data collection and reporting efforts. More localized data are critical to their day-to-day decisions. However, the presence of numerous state-run surveys—in addition to the myriad individualized district surveys that can exist within a single state—has begun to create a data silo where information remains frozen within a state or district system. Since NCES (and therefore the Department of Education and Congress) rarely receives data from state or local collections, these data sources cannot readily be used to generate national policies, which greatly limits opportunities for state and district systems to learn from each other. These data silos can, for example, impact the focus or breadth of federal grants or funding available for schools.
   
   Federal, state, and district education agencies serve different roles in the education sector but have mutually beneficial responsibilities that should complement and support one another. The solution isn’t to supplant federal data collections with local ones, or vice versa, but instead to supplement local collections with federal collections like the National Teacher and Principal Survey (NTPS) so education decisionmakers at all levels have access to necessary information to make good decisions for our schools.
    
2. Benchmarking and comparability are limited. Without federal data collections, it can be difficult or impossible for states, districts, and local policymakers to compare their schools and educators with those in other areas because of the lack of common focus and definitions across data collections. Even if the topics being collected are similar, individualized district or state surveys can differ widely in both content and wording.
   
   National data collections—like the NTPS—are excellent tools local policymakers should use when setting priorities on behalf of the students and staff in their state or district. Since the data from the NTPS are collected from educators in the same way across the entire country, they can be used to establish benchmarks against which local collections can measure themselves.
    
3. Lack of participation decreases the representativeness of storytelling. If districts do not approve NCES’s survey research applications, we are unable to reach educators in certain schools, which can limit the kinds of perspectives that are included in the data. To paint a true picture of the education landscape, our survey teams select districts, schools, and/or educators that are as representative of the education field as possible.
   
   Teachers and principals who participate in NCES studies are grouped in different ways—such as by age, race/ethnicity, or the type of school at which they work—and their information is studied to identify patterns of experiences that people in these different groups may have had. This is what makes our datasets representative, or similar enough to the demographics of the population to able to accurately reflect the characteristics of everyone (even those who aren’t sampled to participate).
   
   For example, the NTPS is designed to support analysis of a variety of subgroups, such as those by

• school level (i.e., elementary, middle, high, and combined);
   
• school community type (i.e., urban, suburban, town, and rural);
   
• teachers’ and principals’ years of experience in the profession; and
   
• race/ethnicity of teachers and principals (figure 1).

These diverse subgroups are critically important for both federal and local policymakers who want to make decisions using information that truly represents everyone in the field.

Figure 1. Percentage of K–12 public and private school teachers who reported that they have any control over various areas of planning and teaching in their classrooms, by school type and selected school characteristic: 2020–21 

NOTE: Data are weighted estimates of the population. Response options included “no control,” “minor control,” “moderate control,” and “a great deal of control.” Teachers who reported “minor control,” “moderate control,” or “a great deal of control” were considered to have reported having “any control.”
SOURCE: U.S. Department of Education, National Center for Education Statistics, National Teacher and Principal Survey (NTPS), “Public and Private School Teacher Data File,” 2020–21.

Larger datasets allow for more nuanced comparisons by school, principal, or teacher characteristics that aren’t possible in smaller datasets, allowing for more equity in national and local estimates and more distinct answers to key policy questions. But we need district-level support to provide these nuanced data.

Although the NTPS has a fairly large sample to support state representation, it still only includes a small percentage of all schools and educators in the country. Sampling is used to avoid collecting data from all systems, staff, and students, thus helping to limit overall respondent burden on our education system. For this reason, it’s important that all sampled schools and educators participate if selected. Since some districts also have formal review processes—through which a survey must be approved before any schools or educators can be contacted—it is also important that districts with sampled schools grant NCES surveys permission to collect data from their schools.

Both levels of participation will help us collect data that accurately describe a state or population. Otherwise, the story we are telling in the data is only augmenting some voices—and these are the experiences that will be reflected in federal policy and funding.

As the education sector strives to understand the needs of students and staff on the tails of the coronavirus pandemic, trustworthy data are only becoming more critical to the decisionmaking process. NCES datasets like the NTPS are critical resources that federal, state, and district policymakers can and have used for benchmarking strategic goals or conducting analyses on how a topic has (or hasn’t) changed over time.

The catch being, of course, that all data on the NTPS—and many other NCES surveys—come directly from schools, principals, and teachers themselves. These analyses and reports are not possible without district and educator participation. While it may seem counterintuitive that any one person could make such a large difference in federal education policy, the concept doesn’t differ from civic duties such as voting in federal or local elections.

By Maura Spiegelman and Julia Merlin, NCES

While federal, state, and district policymakers have used education data as the backbone of their policy and funding decisions for years, there’s nothing quite like a global pandemic to highlight the criticality of reliable data and illuminate the gaps in our collective knowledge. 

But where can policymakers find education datasets that are large enough for comparative or trend analyses while still specific enough to measure timely issues in local contexts? How can policymakers extract and interpret information from these education datasets? The National Center for Education Statistics (NCES) is a perfect resource for these data and information needs.  

As the primary statistical agency of the U.S. Department of Education, the National Center for Education Statistics (NCES) is mandated to report complete statistics on the condition of American education. While the condition of an education system is often assessed through indicators of achievement and attainment, NCES is also mandated to report on the conditions of the education workplace.

As such, NCES has reported timely information from schools. For example, this past December, NCES released data that indicated that public schools have experienced difficulty filling positions throughout the COVID-19 pandemic.¹ In order to understand the broader labor situation, NCES is utilizing the Job Openings and Labor Turnover Survey to describe the tightness of the job market.

JOLTS Design

The Job Openings and Labor Turnover Survey (JOLTS), conducted by the U.S. Bureau of Labor Statistics (BLS), provides monthly estimates of job openings, hires, and total separations. The purpose of JOLTS data is to serve as demand-side indicators of labor shortages at the national level.²

The JOLTS program reports labor demand and turnover estimates by industry, including education.³ As such, this analysis focuses on the public state and local education industry (“state and local government education” as referred to by JOLTS),⁴ which includes all persons employed by public elementary and secondary school systems and postsecondary institutions.

The JOLTS program does not produce estimates by Standard Occupational Classification.⁵ When reviewing these findings, please note occupations⁶ within the public state and local education industry vary⁷ (e.g., teachers and instructional aides, administrators, cafeteria workers, transportation workers). Furthermore, as the JOLTS data are tabulated at the industry level, the estimates are inclusive of the elementary, secondary, and postsecondary education levels.

Analysis

In this blog post, we present selected estimates on the number and rate of job openings, hires, and total separations (quits, layoffs and discharges, and other separations). The job openings rate is computed by dividing the number of job openings by the sum of employment and job openings. All other metric rates (hires, total separations, quits, layoffs and discharges, and other separations) are defined by taking the number of each metric and dividing it by employment. Fill rate is defined as the ratio of the number of hires to the number of job openings, and the churn rate is defined as the sum of the rate of hires and the rate of total separations.⁸

Table 1. Number of job openings, hires, and separations and net change in employment in public state and local education, in thousands: January 2020 through January 2023

*Significantly different from January 2023 (p < .05).
¹ Net employment changes are calculated by taking the difference between the number of hires and the number of separations. When the number of hires exceeds the number of separations, employment rises—even if the number of hires is steady or declining. Conversely, when the number of hires is less than the number of separations, employment declines—even if the number of hires is steady or rising.
NOTE: Data are not seasonally adjusted. Detail may not sum to totals because of rounding.
SOURCE: U.S. Department of Labor, Bureau of Labor Statistics, Job Openings and Labor Turnover Survey (JOLTS), 2020–2023, based on data downloaded April 5, 2023, from https://data.bls.gov/cgi-bin/dsrv?jt.

[2] U.S. Bureau of Labor Statistics. Job Openings and Labor Turnover Survey. Retrieved March 28, 2023, from https://www.bls.gov/jlt/jltover.htm.

[3] For more information about these estimates, see https://www.bls.gov/news.release/jolts.tn.htm.

[4] JOLTS refers to this industry as state and local government education, which is designated as ID 92.

[5] For more information on the reliability of JOLTS estimates, see https://www.bls.gov/jlt/jltreliability.htm.

[6] North American Industry Classification System (NAICS) is a system for classifying establishments (individual business locations) by type of economic activity. The Standard Occupational Classification (SOC) classifies all occupations for which work is performed for pay or profit. To learn more on the differences between NAICS and SOC, see https://www.census.gov/topics/employment/industry-occupation/about/faq.html.

[7] JOLTS data are establishment based, and there is no distinction between occupations within an industry. If a teacher and a school nurse were hired by an establishment coded as state and local government education, both would fall under that industry. (From email communication with JOLTS staff, April 7, 2023.)

[8] Skopovi, S., Calhoun, P., and Akinyooye, L. Job Openings and Labor Turnover Trends for States in 2020. Beyond the Numbers: Employment & Unemployment, 10(14). Retrieved March 28, 2023, from https://www.bls.gov/opub/btn/volume-10/jolts-2020-state-estimates.htm.

The Common Core of Data (CCD) contains basic information on public elementary and secondary schools, local education agencies (LEAs), and state education agencies (SEAs) in the United States. The CCD collects fiscal and nonfiscal data about all public schools, public school districts, and state education agencies in the United States. Both IPEDS and CCD provide a sampling frame to many survey collections, including many conducted by NCES and the Department of Education. This blog post, one in a series of posts about CCD nonfiscal data, focuses on CCD’s two major releases and their corresponding components. For information on how to access and use CCD data, read the blog post Accessing the Common Core of Data (CCD).
 

Data Releases

CCD nonfiscal data are published in two releases every school year—as preliminary files and as provisional data files—within the CCD Data File tool. Understanding the differences between the two releases is important to understand how CCD nonfiscal data are released.

• The preliminary files contain basic information about schools and districts, such as name, address, phone number, status, and NCES ID number. Many schools and districts utilize information from the directory file, such as the NCES ID, to apply for grants or other opportunities for their schools. Therefore, it is important that these files are released first, even if the data are still preliminary. 
   
• The provisional data files are the full release of the CCD nonfiscal data. These data files provide school-, district-, and state-level data on topics like enrollment, staffing, and free or reduced-price lunch. These files are much more detailed and include data that are broken down by characteristics such as grade, race/ethnicity, and gender—as well as by combinations of these characteristics. These files are not updated unless there is a significant change to the data.

Each file release includes a version that indicates the type of release. The first preliminary files have “0a” in the file names, and revised preliminary files include “0b,” “0c,” and so on. The first provisional files have “1a,” in the file names, and revised provisional files include “1b,” “1c,” and so on. Note, however, that releasing revised files is rare.
 

Components of a Release

It is important to utilize the various components that accompany each release to find additional information that is specific to the file and can help you better understand the data. In addition, there are other resources available that provide more ways to access and understand the data.
 

Documentation Components

Every data file will have documentation files that provide information about the data. These include the following:

• release notes—basic information about the data release, including details about any changes to the files, such as a change in a variable’s description or a variable that was added to the file; summary tables that include national totals and tables with selected frequencies are also included.
   
• state data notes—information on data anomalies that are discovered during NCES’s collaboration with the states; broken down by state and by file type, these notes describe things like changes to how data were collected by the state.
   
• companion files—included in each data file component, these files include a list of all the variables in the data file—including a brief description—and frequency tables; you should start with the companion files to better understand what variables are in each data file.

Resources and Tools

Along with the release of the CCD nonfiscal data files, additional resources are also updated to improve access to the data.

• Summary Tables: Released with the provisional data files, Summary Tables provide a national-level look at the data. These tables show the operational status of schools and districts by type as well as the number of schools, students, and teachers by state.
   
• Locators and ElSi: There are two primary tools that can be used to access CCD data: the Locators (School Locator and District Locator) and the Elementary/Secondary Information System (ElSi). These tools are updated as the data files are released. The Locators are updated with each release, while ElSi is updated with the release of the provisional data files. Learn more about these tools.
   
• Online Documentation: The online documentation provides some general information about CCD. This information is not year specific, but it provides a detailed explanation about how the data are collected, processed, and reviewed.

• Reference Library: The reference library includes detailed documentation on various components of the CCD files that applies to multiple years, levels, and components of the data collection. The library includes crosswalks, documentation describing changes to the collection, and guidance for utilizing the data files, such as how to aggregate free or reduced-price lunch data.

Be sure to follow NCES on Twitter, Facebook, LinkedIn, and YouTube and subscribe to the NCES News Flash to stay up-to-date on future CCD releases and resources.

By Patrick Keaton, NCES

Last month, the National Center for Education Statistics (NCES) kicked off a major survey of adults (ages 16–74) across the nation to learn about their literacy skills, education, and work experience. Information collected through this survey—officially known as Cycle 2 of the Program for the International Assessment of Adult Competencies (PIAAC) in the United States—is used by local, state, and national organizations, government entities, and researchers to learn about adult skills at the state and local levels (explore these data in the PIAAC Skills Map, shown below).

Specifically, these data are used to support educational and training initiatives organized by local and state programs. For example, the Houston Mayor’s Office for Adult Literacy has used the PIAAC Skills Map data in developing the Adult Literacy Blueprint, a comprehensive plan for coordinated citywide change to address the systemic crisis of low literacy and numeracy in the city. In addition, the Kentucky Career and Technical College System developed a comprehensive data-driven app for workforce pipeline planning using the county-level PIAAC Skills Map data as one of the education pipeline indicators.

This is not the first time NCES is administering PIAAC. NCES collected PIAAC data three times between 2011 and 2017, when the first cycle of this international study was administered in 39 countries. Developed by the Organization for Economic Cooperation and Development (OECD), PIAAC measures fundamental cognitive and workplace skills needed for individuals to participate in society and for economies to prosper. Among these fundamental skills are literacy, numeracy, and digital problem-solving. Data from the first cycle of PIAAC (2011–17) provided insights into the relationships between adult skills and various economic, social, and health outcomes—both across the United States as a whole and for specific populations of interest (e.g., adults who are women, immigrants, older, employed, parents, or incarcerated). The OECD and NCES have published extensively using these data.

The current cycle (Cycle 2) of PIAAC will resemble the first cycle in that interviewers will visit people’s homes to ask if they are willing to answer background questionnaire and take a self-administered test of their skills. However, unlike the first cycle when respondents could respond to the survey on paper or on a laptop, this cycle will be conducted entirely on a tablet. PIAAC is completely voluntary, but each respondent is specifically selected to provide invaluable information that will help us learn about the state of adult skills in the country (participants can also receive an incentive payment for completing the survey).

PIAAC’s background questionnaire includes questions about an individual’s demographics, family, education, employment, skill use, and (new in Cycle 2 and unique to the United States) financial literacy. The PIAAC test or “direct assessment” measures literacy, numeracy, and (new in Cycle 2) adaptive problem-solving skills of adults.¹

Each sampled person’s response is not only kept confidential but also “anonymized” before the data are released (so that no one can ever definitively identify an individual from personal characteristics in the datafile).

The international report and data for PIAAC Cycle 2 is scheduled to be released by the OECD in December 2024.

Be sure to follow NCES on Twitter, Facebook, LinkedIn, and YouTube and subscribe to the NCES News Flash to stay up-to-date on PIAAC report and data releases and resources.