Highly Accurate, But Still Discriminatory: A Fairness Evaluation of Algorithmic Video Analysis in the Recruitment Context
Open accessThe DOI matches its Crossref record; 35 of the 35 references it cites were found in a registry or an index.
Abstract
The study aims to identify whether algorithmic decision making leads to unfair (i.e., unequal) treatment of certain protected groups in the recruitment context. Firms increasingly implement algorithmic decision making to save costs and increase efficiency. Moreover, algorithmic decision making is considered to be fairer than human decisions due to social prejudices. Recent publications, however, imply that the fairness of algorithmic decision making is not necessarily given. Therefore, to investigate this further, highly accurate algorithms were used to analyze a pre-existing data set of 10,000 video clips of individuals in self-presentation settings. The analysis shows that the under-representation concerning gender and ethnicity in the training data set leads to an unpredictable overestimation and/or underestimation of the likelihood of inviting representatives of these groups to a job interview. Furthermore, algorithms replicate the existing inequalities in the data set. Firms have to be careful when implementing algorithmic video analysis during recruitment as biases occur if the underlying training data set is unbalanced.
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
A Survey on Bias and Fairness in Machine Learning
Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., & Galstyan, A. (2019) · arXiv · doi:10.48550/arxiv.1908.09635 · Cited by 5,812
A Survey on Bias and Fairness in Machine Learning
Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., & Galstyan, A. (2022) · ACM Computing Surveys · doi:10.1145/3457607 · Cited by 5,812
Bertrand, M., & Mullainathan, S. (2004) · American Economic Review · doi:10.1257/0002828042002561 · Cited by 4,504
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 · cited 1×
“However, since a large number of videos have been evaluated and the AMT workers represent a cross-section of American society (Paolacci et al. 2010), and several experiments showed that AMT is an excellent opportunity to gain a representative sample of participants, e.g., Thomas and Clifford…”
BackgroundDwork, C., Hardt, M., Pitassi, T., Reingold, O., & Zemel, R. (2012) · Proceedings of the 3rd Innovations in Theoretical Computer Science Conference · doi:10.1145/2090236.2090255 · Cited by 3,501
Barocas, S., & Selbst, A. D. (2014) · doi:10.2139/ssrn.2477899 · Cited by 3,048 · cited 5×
“These types of algorithms have been considered to be biased because human biases were transferred to the algorithm (Barocas and Selbst 2016), thereby making them ‘‘unfair’’ (Mehrabi et al. 2019).”
BackgroundMethodInfluentialWhat Should Be Done with Equity Theory?
Leventhal, G. S. (1980) · Social Exchange · doi:10.1007/978-1-4613-3087-5_2 · Cited by 3,010 · cited 1×
“For example, Leventhal (1980) describes fairness as equal treatment based on people’s performance and needs.”
BackgroundOpensmile: the munich versatile and fast open-source audio feature extractor
Eyben, F., Wöllmer, M., & Schuller, B. (2010) · Proceedings of the 18th ACM international conference on Multimedia · doi:10.1145/1873951.1874246 · Cited by 2,642 · cited 6×
“Regarding the job interview score, the ROCHCI predictions were significantly different for Asians and African-Americans, with effect sizes of - 0.3 and - 0.4, respectively.”
BackgroundInfluentialRank Analysis of Incomplete Block Designs: I. The Method of Paired Comparisons
Bradley, R. A., & Terry, M. E. (1952) · Biometrika · doi:10.2307/2334029 · Cited by 2,412 · cited 1×
“calculated utilizing the Bradley-Terry-Luce model (Bradley and Terry 1952).”
Barrick, M. R., Mount, M. K., & Judge, T. A. (2001) · International Journal of Selection and Assessment · doi:10.1111/1468-2389.00160 · Cited by 2,356
Cited by
Ethics of AI-enabled recruiting and selection: a review and research agenda
Hunkenschroer, A. L., & Luetge, C. (2022) · Journal of Business Ethics · doi:10.1007/s10551-022-05049-6 · Cited by 447
Fairness and Bias in Algorithmic Hiring: A Multidisciplinary Survey
Fabris, A., Baranowska, N., Dennis, M. J., Graus, D., Hacker, P., Saldivar, J., Zuiderveen Borgesius, F., & Biega, A. J. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3696457 · Cited by 117
Is AI recruiting (un) ethical? A human rights perspective on the use of AI for hiring
Hunkenschroer, A. L., & Kriebitz, A. (2023) · AI and Ethics · doi:10.1007/s43681-022-00166-4 · Cited by 111
“In the assessment stage, face recognition software can be used to analyze video interviews, evaluate applicants’ responses, and provide insight into certain personality traits and competencies [28].”
Method
Metadata
- Authors
- Köchling, A., Riazy, S., Wehner, M. C., & Simbeck, K.
- Venue
- Business & Information Systems Engineering
- Published
- 2020-11-24
- Volume
- 63
- Issue
- 1
- Pages
- 39-54
- ISSN
- 1867-0202 (ISSN-L)
- 2363-7005
- OpenAlex
- W3108750004
- Cited by
- 80
- References
- 35
- FWCI
- 9.48
- Type
- journal article
- Language
- en
- Open access
- hybrid
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: Highly Accurate, But Still Discriminatory: A Fairness Evaluation of Algorithmic Video Analysis in the Recruitment Context. https://app.cochat.ai/papers/10.1007/s12599-020-00673-w
- The paper
- Köchling, A., Riazy, S., Wehner, M. C., & Simbeck, K. (2021). Highly Accurate, But Still Discriminatory: A Fairness Evaluation of Algorithmic Video Analysis in the Recruitment Context. Business & Information Systems Engineering. https://doi.org/10.1007/s12599-020-00673-w
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.