WHO DO WE THINK WE ARE? COMPARING INTERSECTIONAL IDENTITY TRENDS IN ASEE AND CEEA-ACEG USING NATURAL LANGUAGE PROCESSING AND REVIEW OF PROCEEDINGS
Bibliographic record
Abstract
In 2018, we overheard many CEEA-AGEC members stating that they have "found their people"; this led us to wonder what makes this evolving community unique. Using cultural historical activity theory to view the proceedings of CEEA-ACEG 2004-2018 in comparison with the geographically and intellectually adjacent ASEE, we used both machine-driven (Natural Language Processing, NLP) and human-driven (literature review of the proceedings) methods. Here, we hoped to build on surveys—most recently by Nelson and Brennan (2018)—to understand, beyond what members say about themselves, what makes the CEEA-AGEC community distinct, where it has come from, and where it is going. Engaging in the two methods of data collection quickly diverted our focus from an analysis of the data themselves to the characteristics of the data in terms of cultural historical activity theory. Our preliminary findings point to some unique characteristics of machine- and human-driven results, with the former, as might be expected, focusing on the micro-level (words and language patterns) and the latter on the macro-level (ideas and concepts). NLP generated data within the realms of "community" and "division of labour" while the review of proceedings centred on "subject" and "object"; both found "instruments," although NLP with greater granularity. With this new understanding of the relative strengths of each method, we have a revised framework for addressing our original question.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.022 | 0.088 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.035 | 0.037 |
| Science and technology studies | 0.004 | 0.006 |
| Scholarly communication | 0.014 | 0.017 |
| Open science | 0.002 | 0.005 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".