abstracts analyzed
1,727
describe qualitative work
43%
of method-naming abstracts use logistic regression
8%

Higher and adult education is the area in our analysis of 18,617 education dissertation and thesis abstracts that looks most like US doctoral research: it has the largest share of US-tagged work (17%) and the most qualitative designs.

What is in this sample

The most common topics are higher education research, employability, vocational and entrepreneurial education, service-learning and community engagement, sustainability in higher education, and adult and continuing education. Of the 1,727 abstracts, 43% describe qualitative work, 20% describe a survey, 12% are case studies, and 8% use mixed methods. Only 18% name a quantitative method.

The methods these dissertations name most

Bar chart of methods named in higher education abstracts: correlation 25%, linear regression 18%, SEM 13%, reliability 11%, mediation 11%, t test 8%, ANOVA family 8%, logistic regression 8%, moderation 7%, chi-square 6%, difference-in-differences or other causal designs 5%, factor analysis 4%. Correlation, linear regression, SEM, logistic regression, mediation, causal designs, and panel data are more common than in the rest of education; t tests and reliability are less common.
Share of the 315 method-naming abstracts. Circles show the rest of education.

Several methods are clearly more common here than in the other four areas:

  • Correlation (25% vs. 15%) and linear regression (18% vs. 7%).
  • Structural equation models (13% vs. 5%), often for models of student engagement or belonging.
  • Logistic regression (8% vs. 2%), for yes/no outcomes such as retention, graduation, or employment.
  • Mediation (11% vs. 7%).
  • Causal designs such as difference-in-differences or regression discontinuity (5% vs. 2%), and panel data models (4% vs. 1%), usually with institutional or administrative data.

t tests (8% vs. 26%) and reliability (11% vs. 19%) are much less common than in classroom-focused areas.

A note on change over time

Reliability rose from 0.5% to 4.4% of abstracts in this area, and structural equation models from 1.6% to 3.8%. OpenAlex added many repositories from outside the United States during this period, so shifts in education reflect who is in the index as much as how methods are changing. We don't chart them for that reason.

What these abstracts leave out

None of the 1,727 abstracts mentions a power analysis, and one mentions missing data. Institutional data often look complete but are not: transfer students, stop-outs, and missing survey responses all need a plan.

If you're planning a higher education dissertation or thesis

  • Match the model to the outcome. Retention and graduation are yes/no outcomes: use logistic regression, and report odds ratios or predicted probabilities. Time to degree is a survival outcome.
  • Be careful with causal language. Students choose programs, majors, and supports. Without a design such as difference-in-differences or matching, describe associations, not effects.
  • Check your structural model's measurement first. Before testing paths, show that each scale fits and is reliable in your sample.
  • Consider nesting. Students within institutions or programs share context; with multi-campus data, a multilevel model may be needed.

How we did this

We used OpenAlex, an open index of scholarly works, to collect every English-language dissertation and thesis in education with an abstract published from 2021 through September 2026, grouped them by OpenAlex research topic, and searched each title and abstract for about 40 statistical methods. The data include master's theses, probably some undergraduate theses, and many works from outside the United States, so “dissertations” here is short for all of them. Percentages describe how often abstracts mention a method, not how often it was used. Comparisons use Fisher's exact tests with a false-discovery-rate correction. The overview post has the full method and its limitations. The analysis code and result tables are on GitHub.