A panel for lemons? Positivity bias, reputation systems and data quality on MTurk
The DOI matches its Crossref record; 24 of the 24 references it cites were found in a registry or an index.
Abstract
Purpose: The purpose of this paper is to investigate how the effectiveness of systems for ensuring cooperation in online transactions is impacted by a positivity bias in the evaluation of the work that is produced. The presence of this bias can reduce the informativeness of the reputation system and negatively impact its ability to ensure quality. Design/methodology/approach: This research combines survey and experimental methods, collecting data from 1,875 Mechanical Turk (MTurk) workers in five studies designed to investigate the informativeness of the MTurk reputation system. Findings: The findings demonstrate the presence of a positivity bias in evaluations of workers on MTurk, which leaves them undifferentiated, except at the extremity of the reputation system and by status markers. Research limitations/implications: Because MTurk workers self-select tasks, the findings are limited in that they may only be generalizable to those who are interested in research-related work. Further, the tasks used in this research are largely subjective in nature, which may decrease their sensitivity to differences in quality. Practical implications: For researchers, the results suggest that requiring 99 per cent approval rates (rather than the previously advised 95 per cent) should be used to identify high-quality workers on MTurk. Originality/value: The research provides insights into the design and use of reputation systems and demonstrates how design decisions can exacerbate the effect of naturally occurring biases in evaluations to reduce the utility of these systems.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Regression Analysis of Count Data
Cameron, A. C., & Trivedi, P. K. (2013) · doi:10.1017/cbo9781139013567 · Cited by 6,640
The Effect of Word of Mouth on Sales: Online Book Reviews
Chevalier, J. A., & Mayzlin, D. (2006) · Journal of Marketing Research · doi:10.1509/jmkr.43.3.345 · Cited by 6,003
Evaluating Online Labor Markets for Experimental Research: Amazon.com's Mechanical Turk
Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012) · Political Analysis · doi:10.1093/pan/mpr057 · Cited by 4,008
Identifying careless responses in survey data.
Meade, A. W., & Craig, S. B. (2012) · Psychological Methods · doi:10.1037/a0028085 · Cited by 3,865
Running experiments on Amazon Mechanical Turk
Paolacci, G., Chandler, J., & Ipeirotis, P. G. (2010) · Judgment and Decision Making · doi:10.1017/s1930297500002205 · Cited by 3,767
Conducting behavioral research on Amazon’s Mechanical Turk
Mason, W., & Suri, S. (2012) · Behavior Research Methods · doi:10.3758/s13428-011-0124-6 · Cited by 3,172
Beyond the Turk: Alternative platforms for crowdsourcing behavioral research
Peer, E., Brandimarte, L., Samat, S., & Acquisti, A. (2017) · Journal of Experimental Social Psychology · doi:10.1016/j.jesp.2017.01.006 · Cited by 2,957
The Efficient Assessment of Need for Cognition
Cacioppo, J. T., Petty, R. E., & Feng Kao, C. (1984) · Journal of Personality Assessment · doi:10.1207/s15327752jpa4803_13 · Cited by 2,913
Pavlou, P. A., Liang, H., & Xue, Y. (2007) · MIS Quarterly · doi:10.2307/25148783 · Cited by 2,646
Building Effective Online Marketplaces with Institution-Based Trust
Pavlou, P. A., & Gefen, D. (2004) · Information Systems Research · doi:10.1287/isre.1040.0015 · Cited by 2,516
Cited by
Data Collection via Online Platforms: Challenges and Recommendations for Future Research
Newman, A., Bavik, Y. L., Mount, M., & Shao, B. (2021) · Applied Psychology · doi:10.1111/apps.12302 · Cited by 306
An innovation resistance theory perspective on purchase of eco-friendly cosmetics
Sadiq, M., Adil, M., & Paul, J. (2021) · Journal of Retailing and Consumer Services · doi:10.1016/j.jretconser.2020.102369 · Cited by 262
The Perils of Using Mechanical Turk to Evaluate Open-Ended Text Generation
Karpinska, M., Akoury, N., & Iyyer, M. (2021) · Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2021.emnlp-main.97 · Cited by 56
“Reluctance of requesters to reject HITs leads to positive bias in workers’ qualifications (Matherly, 2019).”
BackgroundMethodInterplay of rationality and morality in using ChatGPT for academic misconduct
Zhang, L., Amos, C., & Pentina, I. (2025) · Behaviour & Information Technology · doi:10.1080/0144929x.2024.2325023 · Cited by 48
“To enhance data quality, we implemented guidelines in accordance with recommendations from seminal work regarding online panels (Goodman and Paolacci 2017; Matherly 2019).”
A four-dimensional model of Asian American stereotypes
Bu, W., & Borgida, E. (2021) · Group Processes & Intergroup Relations · doi:10.1177/1368430220936360 · Cited by 9
“As Matherly (2019) noted, data quality on MTurk has generally been high despite inflated performance ratings.”
Background
Metadata
- Authors
- Matherly, T.
- Venue
- European Journal of Marketing
- Publisher
- Emerald
- Published
- 2019-02-11
- Volume
- 53
- Issue
- 2
- Pages
- 195-223
- ISSN
- 0309-0566 (ISSN-L)
- OpenAlex
- W2908862202
- Cited by
- 75
- References
- 24
- FWCI
- 11.16
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: A panel for lemons? Positivity bias, reputation systems and data quality on MTurk. https://app.cochat.ai/papers/10.1108/ejm-07-2017-0491
- The paper
- Matherly, T. (2019). A panel for lemons? Positivity bias, reputation systems and data quality on MTurk. European Journal of Marketing. https://doi.org/10.1108/ejm-07-2017-0491
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.