# Measuring Disparate Outcomes of Content Recommendation Algorithms with Distributional Inequality Metrics

arXiv: https://arxiv.org/abs/2202.01615 · HTML: https://mlsanity.com/paper/2202.01615/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 44 references that are in the map.
- First posted on arXiv 3 Feb 2022.

## Builds on (most cited first, 15 of 15)

- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness](https://mlsanity.com/paper/1711.05144/) (2017)
- [Fairness of Exposure in Rankings](https://mlsanity.com/paper/1802.07281/) (2018)
- [Fairness-Aware Ranking in Search & Recommendation Systems with Application to LinkedIn Talent Search](https://mlsanity.com/paper/1905.01989/) (2019)
- [FA*IR: A Fair Top-k Ranking Algorithm](https://mlsanity.com/paper/1706.06368/) (2017)
- [A Unified Approach to Quantifying Algorithmic Unfairness: Measuring Individual & Group Unfairness via Inequality Indices](https://mlsanity.com/paper/1807.00787/) (2018)
- [Measuring Fairness in Ranked Outputs](https://mlsanity.com/paper/1610.08559/) (2016)
- [Quantifying the Impact of User Attention on Fair Group Representation in Ranked Lists](https://mlsanity.com/paper/1901.10437/) (2019)
- [From Optimizing Engagement to Measuring Value](https://mlsanity.com/paper/2008.12623/) (2020)
- [Quantifying Search Bias: Investigating Sources of Bias for Political Searches in Social Media](https://mlsanity.com/paper/1704.01347/) (2017)
- [Comparing Fair Ranking Metrics](https://mlsanity.com/paper/2009.01311/) (2020)
- [Fairness through Experimentation: Inequality in A/B testing as an approach to responsible design](https://mlsanity.com/paper/2002.05819/) (2020)
- [Escaping the Impossibility of Fairness: From Formal to Substantive Algorithmic Fairness](https://mlsanity.com/paper/2107.04642/) (2021)
- [A simple method for measuring inequality](https://mlsanity.com/paper/2112.15284/) (2021)
- [Attention Inequality in Social Media](https://mlsanity.com/paper/1601.07200/) (2016)

## Built on by (most cited first, 8 of 8)

- [De-biasing "bias" measurement](https://mlsanity.com/paper/2205.05770/) (2022)
- [Optimizing generalized Gini indices for fairness in rankings](https://mlsanity.com/paper/2204.06521/) (2022)
- [Distribution-Free Statistical Dispersion Control for Societal Applications](https://mlsanity.com/paper/2309.13786/) (2023)
- [A Keyword Based Approach to Understanding the Overpenalization of Marginalized Groups by English Marginal Abuse Models on Twitter](https://mlsanity.com/paper/2210.06351/) (2022)
- [Bypassing the Popularity Bias: Repurposing Models for Better Long-Tail Recommendation](https://mlsanity.com/paper/2410.02776/) (2024)
- [Measuring Individual User Fairness with User Similarity and Effectiveness Disparity](https://mlsanity.com/paper/2602.02516/) (2026)
- [System Analysis for Responsible Design of Modern AI/ML Systems](https://mlsanity.com/paper/2204.08836/) (2022)
- [A Comparative Analysis of Bias Amplification in Graph Neural Network Approaches for Recommender Systems](https://mlsanity.com/paper/2301.07639/) (2023)

## Abstract (arXiv)

The harmful impacts of algorithmic decision systems have recently come into focus, with many examples of systems such as machine learning (ML) models amplifying existing societal biases. Most metrics attempting to quantify disparities resulting from ML algorithms focus on differences between groups, dividing users based on demographic identities and comparing model performance or overall outcomes between these groups. However, in industry settings, such information is often not available, and inferring these characteristics carries its own risks and biases. Moreover, typical metrics that focus on a single classifier's output ignore the complex network of systems that produce outcomes in real-world settings. In this paper, we evaluate a set of metrics originating from economics, distributional inequality metrics, and their ability to measure disparities in content exposure in a production recommendation system, the Twitter algorithmic timeline. We define desirable criteria for metrics to be used in an operational setting, specifically by ML practitioners. We characterize different types of engagement with content on Twitter using these metrics, and use these results to evaluate the metrics with respect to the desired criteria. We show that we can use these metrics to identify content suggestion algorithms that contribute more strongly to skewed outcomes between users. Overall, we conclude that these metrics can be useful tools for understanding disparate outcomes in online social networks.