Fairness of using different English accents: The effect of shared L1s in listening tasks of the Duolingo English test
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
This study aimed to answer an ongoing validity question related to the use of nonstandard English accents in international tests of English proficiency and associated issues of test fairness. More specifically, we examined (1) the extent to which different or shared English accents had an impact on listeners’ performances on the Duolingo listening tests and (2) the extent to which different English accents affected listeners’ performances on two different task types. Speakers from four interlanguage English accent varieties (Chinese, Spanish, Indian English [Hindi], and Korean) produced speech samples for “yes/no” vocabulary and dictation Duolingo listening tasks. Listeners who spoke with these same four English accents were then recruited to take the Duolingo listening test items. Results suggested that there is a shared first language (L1) benefit effect overall, with comparable test scores between shared-L1 and inner-circle L1 accents, and no significant differences in listeners’ listening performance scores across highly intelligible accent varieties. No task type effect was found. The findings provide guidance to better understand fairness, equality, and practicality of designing and administering high-stakes English tests targeting a diversity of accents.
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 | 0.001 | 0.009 |
| 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 | 0.001 | 0.000 |
| 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