Using Amazon Mechanical Turk for linguistic research
Bibliographic record
Abstract
Amazon?s Mechanical Turk service makes linguistic experimentation quick, easy, and inexpensive. However, researchers have not been certain about its reliability. In a series of experiments, this paper compares data collected via Mechanical Turk to those obtained using more traditional methods One set of experiments measured the predictability of words in sentences using the Cloze sentence completion task (Taylor, 1953). The correlation between traditional and Turk Cloze scores is high (rho=0.823) and both data sets perform similarly against alternative measures of contextual predictability. Five other experiments on the semantic relatedness of verbs and phrasal verbs (how much is ?lift? part of ?lift up?) manipulate the presence of the sentence context and the composition of the experimental list. The results indicate that Turk data correlate well between experiments and with data from traditional methods (rho up to 0.9), and they show high inter-rater consistency and agreement. We conclude that Mechanical Turk is a reliable source of data for complex linguistic tasks in heavy use by psycholinguists. The paper provides suggestions for best practices in data collection and scrubbing.
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 | 0.016 | 0.049 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.001 |
| Bibliometrics | 0.004 | 0.005 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.004 | 0.005 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.451 | 0.288 |
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".