# Oversampling Higher-Performing Minorities During Machine Learning Model Training Reduces Adverse Impact Slightly but Also Reduces Model Accuracy

arXiv: https://arxiv.org/abs/2304.13933 · HTML: https://mlsanity.com/paper/2304.13933/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 67 references that are in the map.
- First posted on arXiv 27 Apr 2023.

## Builds on (most cited first, 11 of 11)

- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimization](https://mlsanity.com/paper/1603.06560/) (2016)
- [Distributed Representations of Sentences and Documents](https://mlsanity.com/paper/1405.4053/) (2014)
- [Mitigating Unwanted Biases with Adversarial Learning](https://mlsanity.com/paper/1801.07593/) (2018)
- [Train faster, generalize better: Stability of stochastic gradient descent](https://mlsanity.com/paper/1509.01240/) (2015)
- [Non-stochastic Best Arm Identification and Hyperparameter Optimization](https://mlsanity.com/paper/1502.07943/) (2015)
- [On Hyperparameter Optimization of Machine Learning Algorithms: Theory and Practice](https://mlsanity.com/paper/2007.15745/) (2020)
- [Adversarial Removal of Demographic Attributes from Text Data](https://mlsanity.com/paper/1808.06640/) (2018)
- [Algorithmic Fairness](https://mlsanity.com/paper/2001.09784/) (2020)
- [Discrimination in the Age of Algorithms](https://mlsanity.com/paper/1902.03731/) (2019)
- [Mean Birds: Detecting Aggression and Bullying on Twitter](https://mlsanity.com/paper/1702.06877/) (2017)

## Abstract (arXiv)

Organizations are increasingly adopting machine learning (ML) for personnel assessment. However, concerns exist about fairness in designing and implementing ML assessments. Supervised ML models are trained to model patterns in data, meaning ML models tend to yield predictions that reflect subgroup differences in applicant attributes in the training data, regardless of the underlying cause of subgroup differences. In this study, we systematically under- and oversampled minority (Black and Hispanic) applicants to manipulate adverse impact ratios in training data and investigated how training data adverse impact ratios affect ML model adverse impact and accuracy. We used self-reports and interview transcripts from job applicants (N = 2,501) to train 9,702 ML models to predict screening decisions. We found that training data adverse impact related linearly to ML model adverse impact. However, removing adverse impact from training data only slightly reduced ML model adverse impact and tended to negatively affect ML model accuracy. We observed consistent effects across self-reports and interview transcripts, whether oversampling real (i.e., bootstrapping) or synthetic observations. As our study relied on limited predictor sets from one organization, the observed effects on adverse impact may be attenuated among more accurate ML models.