Jumpstarting the Justice Disciplines: A Computational-Qualitative Approach to Collecting and Analyzing Text and Image Data in Criminology and Criminal Justice Studies
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Computational methods are increasingly popular in criminal justice research. As more criminal justice data becomes available in big data and other digital formats, new means of embracing the computational turn are needed. In this article, we propose a framework for data collection and case sampling using computational methods, allowing researchers to conduct thick qualitative research – analyses concerned with the particularities of a social context or phenomenon – starting from big data, which is typically associated with thinner quantitative methods and the pursuit of generalizable findings. The approach begins by using open-source web scraping algorithms to collect content from a target website, online database, or comparable online source. Next, researchers use computational techniques from the field of natural language processing to explore themes and patterns in the larger data set. Based on these initial explorations, researchers algorithmically generate a subset of data for in-depth qualitative analysis. In this computationally driven process of data collection and case sampling, the larger corpus and subset are never entirely divorced, a feature we argue has implications for traditional qualitative research techniques and tenets. To illustrate this approach, we collect, subset, and analyze three years of news releases from the Royal Canadian Mounted Police website (N = 13,637) using a mix of web scraping, natural language processing, and visual discourse analysis. To enhance the pedagogical value of our intervention and facilitate replication and secondary analysis, we make all data and code available online in the form of a detailed, step-by-step tutorial.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.009 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.003 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it