Making sense of text: artificial intelligence-enabled content analysis
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
Purpose The purpose of this paper is to introduce, apply and compare how artificial intelligence (AI), and specifically the IBM Watson system, can be used for content analysis in marketing research relative to manual and computer-aided (non-AI) approaches to content analysis. Design/methodology/approach To illustrate the use of AI-enabled content analysis, this paper examines the text of leadership speeches, content related to organizational brand. The process and results of using AI are compared to manual and computer-aided approaches by using three performance factors for content analysis: reliability, validity and efficiency. Findings Relative to manual and computer-aided approaches, AI-enabled content analysis provides clear advantages with high reliability, high validity and moderate efficiency. Research limitations/implications This paper offers three contributions. First, it highlights the continued importance of the content analysis research method, particularly with the explosive growth of natural language-based user-generated content. Second, it provides a road map of how to use AI-enabled content analysis. Third, it applies and compares AI-enabled content analysis to manual and computer-aided, using leadership speeches. Practical implications For each of the three approaches, nine steps are outlined and described to allow for replicability of this study. The advantages and disadvantages of using AI for content analysis are discussed. Together these are intended to motivate and guide researchers to apply and develop AI-enabled content analysis for research in marketing and other disciplines. Originality/value To the best of the authors’ knowledge, this paper is among the first to introduce, apply and compare how AI can be used for content analysis.
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 | 0.004 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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