Proud to see Agora associate Maura Spiegelman contribute to this AEFP conversation on the future of federal education statistics. Her expertise reflects our commitment to building stronger data systems for researchers, educators, and communities. #AEFP2026

You are invited to join a webinar: Innovating For Whom We Measure in the AI Era. After registering, you will receive a confirmation email about joining the webinar. In an era of multi-modal AI, the opportunity exists to center the needs of learners, educators, families, and the wider community in ways unthought of previously. In addition to addressing how the sci...

Also, from the Study Group on Innovating for Whom We Measure in the AI Era, there's a February 24 webinar exploring how AI can support more equitable and useful measurement and improvement systems.

Register here: us06web.zoom.us/webinar/regi...

Accelerate Edtech Impact Summit • Stanford Accelerator for Learning How can educational technology better leverage learning science? This half-day summit draws together leaders in research, education, and technology to discuss, learn, and advance equitable and effecti...

Stanford’s 2026 AI and Education Summit is today, Wednesday, February 11. A strong lineup exploring how AI can support human relationships, trustworthy design, and evidence informed practice:
acceleratelearning.stanford.edu/conference/a...

#AIinEducation #ResponsibleAI #HumanCenteredAI

Congress Has Passed an Education Budget. See How Key Programs Are Affected Federal funding for low-income students and special education will remain level year over year.

Good news in the FY2026 education budget. Despite proposed cuts, Congress kept most ED programs funded at prior levels. This stability gives educators and researchers room to plan, innovate, and strengthen systems for kids and families. www.edweek.org/policy-polit...

OPRE is being significantly reduced under the new ACF restructuring. Independent research on programs like Head Start and child welfare is at risk. Read the joint statement from Data Foundation and Results for America. datafoundation.org/news/press-r...

Agora welcomes ED’s plans to strengthen IES. The Northern report reflects several recommendations we submitted last fall, from harmonized APIs to CEDS‑aligned research and coordinated TA, encouraging alignment for the field. www.ed.gov/about/news/p...

Are you smarter than an eighth grader in digital literacy and computational thinking? Test yourself with a real ICILS 2018 task and see how students reason through digital challenges. Read Linda’s new blog: edevaluator.org/explore-a-co...

One year ago, federal RIFs reshaped the Department of Education. We honor the public servants and contractors whose work built the nation’s education data systems. Agora remains committed to carrying that legacy forward with transparency, equity, and care.

Nurturing What We Invest In: Listening, Learning, and Evaluating During Universal Pre‑K and Child Care Expansion In Part 2 of our series on universal child care and PreK expansion, we look at how real‑time feedback from staff and families strengthens quality and keeps programs aligned with community needs.

Part 2 of our series on universal child care and PreK expansion is out. Jodi Jacobson Chernoff highlights why listening to staff and families during implementation is essential for protecting quality as programs grow.
edevaluator.org/nurturing-wh...

Proud to share that Agora’s Maura Spiegelman will join colleagues at AEFP on March 19. Our work pushes for evidence, equity, and systems change that strengthens opportunities for every learner. virtual.oxfordabstracts.com/event/public...

All told, this release of the BPS:20/22 data are a late but welcome bit of good news. I encourage postsecondary researchers to begin exploring the data on PowerStats:
nces.ed.gov/datalab/inde...

(19/19)

However, if the source variable list is an indicator, the other gender variables should survive in the source files and be made available for use with a restricted-use license.

nces.ed.gov/datalab/file...

(18/X)

The derived variable codebook appears to only have sex assigned at birth, with only two options, male and female.
nces.ed.gov/datalab/#SEX...

(17/X)

I want to conclude this thread by noting (without excessive editorialization, given the subject matter) another BPS variable that survived, albeit in diminished form.

The 2020 and 2022 BPS collections included data on gender identity, beyond the bounds of biological sex.

(16/X)

Natural Disasters in Longitudinal Education Data Without longitudinal education data, it may be harder for researchers to determine the immediate or long-term impacts many of these natural events will have on the persistence and attainment of U.S. p...

I made a case for the utility of these FEMA data in a blog post for Agora last September: edevaluator.org/natural-disa...

So I'm very happy that these data are now available as part of BPS.

(15/X)

These FEMA data were incorporated to permit researchers to better analyze the relationships natural disasters and extreme weather events may have on postsecondary students' longitudinal outcomes.

(14/X)

You can find more information about the FEMA source data in the BPS:20/22 documentation, here: ies.ed.gov/sites/defaul...

The FEMA file "Contains data from FEMA for the institutions BPS:20/22 survey respondents attended. There is a separate record for each institution."

(13/X)

Another feature of BPS:20/22 I'm excited to see available to researchers: FEMA data.

If you examine the source variables list, you will find a category for a BPS22FEMA file, which includes FEMA data.

nces.ed.gov/datalab/file...

(12/X)

However, with the 3-year data available in BPS:20/22, we will have to rely on the three-year outcomes. So, some longitudinal analyses of this cohort are still possible.
(11/X)

If the BPS 2025 collection had proceeded to the field, it would have provided six-year longitudinal data on students who began their postsecondary educations around the time of the COVID-19 pandemic. That represents the loss of an irreproducible opportunity for priceless postsecondary data. (10/X)

Now for some bad news. As designed, BPS is supposed to have two collections after the base year. For this round of BPS:
* base year: 2020
* first follow-up: 2022
* second follow-up: 2025
In February 2025, we were preparing to field the 2025 collection when the BPS contract was cancelled.
(9/X)

Therefore, BPS:20/22 is uniquely positioned to allow researchers to examine the long-term, longitudinal outcomes (including persistence, attainment, and labor) of college students who began college during the pandemic. (8/X)

Now, as for the variables themselves, and why this particular round of BPS is so exciting...
As implied by the study's name, the BPS:20/22 cohort began college in AY 2019-20. That is to say, these students entered college for the first time around the start of the COVID:19 pandemic. (7/X)

While PowerStats cannot analyze these source variables, the inclusion of this source information can help researchers to:
1) Better understand the derived variables that rely on those source data.
2) Decide whether to apply for the restricted-use datafile (which includes the source variables). (6/X)

Beginning Postsecondary Students (BPS) BPS collection: 2020/2022 About BPS This study of undergraduate students who are first-time, beginning students collects information from students at at the end of their first year, and then three and six years after starting in postsecondary education. Use this study to learn about student persistence, transfer, degree attainment, and student demographics and experiences. Visit Study Homepage Population Beginning students who were surveyed at the end of their first year, and then three and six years after first starting in postsecondary education. Codebooks Variables by Subject Variables by Name Source Variables Source Variables List Keywords Students’ persistence, progress and attainment of a degree, Labor force experiences ANALYSIS TYPES AVAILABLE Percentage Distribution Averages, Medians, & Percents Percentiles Linear Regression Logistic Regression Correlation Matrix

For experienced users of PowerStats ( nces.ed.gov/datalab/inde... ), if you review the BPS:20/22 information tile (pictured here), you'll see a new link to a "Source Variables List" that provides a list of variables in the source datafiles. (5/X)

As for the data release itself, I'm pleased to see some useful features survive for researchers to use in the BPS:20/22 data. Here are a few of these... (4/X)

While I'm pleased these data are finally seeing the light of day, it's unfortunate that these data were delayed an additional year when NPSAS and BPS data releases are already lengthy processes that mitigate the freshness of the data. (3/X)

First, this data release was being prepared early last year but were delayed in February 2025 by the wide-scale cancellation of many IES contracts by DOGE. (2/X)

#IES and #NCES have announced that data for the 2020-22 Beginning Postsecondary Students Longitudinal Study (BPS:20/22) are available for analysis on DataLab, here: nces.ed.gov/datalab/inde...

I'm pleased to see these data available and want to share some observations, below: (1/X)