Response pattern analysis: Assuring data integrity in extreme research settings
Bibliographic record
Abstract
Research summary : Strategy scholars increasingly conduct research in nontraditional contexts. Such efforts often require the assistance of third‐party intermediaries who understand local culture, norms, and language. This reliance on intermediation in primary or secondary data collection can elicit agency breakdowns that call into question the reliability, analyzability, and interpretability of responses. Herein, we investigate the causes and consequences of intermediary bias in the form of faked data and we offer Response Pattern Analysis as a statistical solution for identifying and removing such problematic data. By explicating the effect, illustrating how we detected it, and performing a controlled field experiment in a developing country to test the effectiveness of our methodological solution, we encourage researchers to continue to seek data and build theory from unique and understudied settings . Managerial summary : Any form of survey research contains the risk of interviewers faking data. This risk is particularly difficult to mitigate in Base‐of‐Pyramid or developing country contexts where researchers have to rely on intermediaries and forms of control are limited. We provide a statistical technique to identify a faking interviewer's ex post data collection, and remove the associated data prior to analysis. Using a field experiment where we instruct interviewers to fake the data, we demonstrate that the algorithm we employ achieves a 90 percent accuracy in terms of differentiating faking from nonfaking interviewers . Copyright © 2016 John Wiley & Sons, Ltd.
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 | 0.336 | 0.674 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.005 | 0.006 |
| Science and technology studies | 0.004 | 0.006 |
| Scholarly communication | 0.005 | 0.005 |
| Open science | 0.004 | 0.007 |
| Research integrity | 0.004 | 0.004 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".