A systematic discourse analysis of how U.S. political leaders frame disability: Implications for students with disabilities (ICPSR 307656)
This study examined how U.S. political leaders publicly framed disability during calendar year 2025, using a systematic discourse analysis of public statements and federal legislation. It also examined how this discourse was reflected in federal education policy, including statements by the Secretary of Education and education-related bills.
The study analyzed 121 public statements about people with disabilities made by White House and Cabinet officials, and 32 federal bills introduced in 2025 that could affect the rights, services, or educational opportunities of individuals with disabilities. Statements were identified through a two-stage process that combined a custom Python-based web-scraping tool, which extracted verbatim, attributed quotations from news articles, press releases, interview transcripts, official speeches, and social media posts using the OpenAI GPT-4 API, with manual verification searches conducted in ChatGPT Plus and Perplexity AI Pro. Federal bills were identified through Congress.gov. All statements were reviewed by the authors to confirm accuracy, attribution, and date. Each statement and bill was independently scored by two human coders and by GPT-4 using an author-developed four-point rubric grounded in the social and human rights models of disability, ranging from 1 (dehumanizing) to 4 (affirming).
The data contain one record per statement, including the speaker's name and title, date, verbatim quotation, source, context, an analysis of the framing, a score from 1 (dehumanizing) to 4 (affirming), and the rationale for the score. The bills data are available on the project website and are not included in this deposit.
The study was approved by the university Institutional Review Board (March 2025) and was preregistered on the Open Science Framework (https://osf.io/hbfe5/).
From Hollywood to classroom: How Film and Television Can Be Used to Teach Civil Rights. (ICPSR 208702)
Congressional Record for 104th-110th Congresses: Text and Phrase Counts (ICPSR 33501)
Please note that inconsistencies have been identified in some of the data accompanying this collection related to the variable "speechID." Potential data users are advised that the files in DS15-DS20 may be compromised and should be used with caution.
This qualitative data collection contains original and processed text from the United States Congressional Record for the 104th-110th Congresses. The Congressional Record includes text from both chambers, the United States House of Representatives and the United States Senate. For each Congress the archive includes the original tagged text files, parsed files that separate the text into individual speeches, speaker metadata that can be linked to the parsed files, and counts of two-word phrases (bigrams) by speaker, party, and date.