{"paper_id":"2408.02558","title":"Peer-induced Fairness: A Causal Approach for Algorithmic Fairness Auditing","authors":["Shiqi Fang","Zexun Chen","Jake Ansell"],"paper_date":"2024-08-05","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2408.02558v4","page_url":"https://thequant.space/flowcharts/peer-induced-fairness-a-causal-approach-for-algorithmic-fai/","scores":{"empirical_rigor":4,"math_complexity":7,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes a novel \"peer-induced fairness\" framework for auditing algorithmic fairness.","quote":"We propose a novel framework, ``peer-induced fairness'', which combines the strengths of counterfactual fairness and peer comparison strategy, creating a reliable and robust tool for auditing algorithmic fairness.","summary":"With the European Union’s Artificial Intelligence Act taking effect on 1 August 2024, high-risk AI applications must adhere to stringent transparency and fairness standards. This paper addresses a crucial question: how can we scientifically audit algorithmic fairness? Current methods typically…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The sample mean of peer probabilities follows a normal distribution (Central Limit Theorem).","type":"statistical","stated":"explicit","quote":"According to the Central Limit Theorem, the sample mean ¯T follows a normal distribution, and thus E[ ¯T], can be employed to estimate the overall predictive probabilities of favourable outcomes among peers"},{"assumption":"The hypothesis test uses a standard z‑test, requiring a large enough number of peers (N).","type":"statistical","stated":"explicit","quote":"Since ¯Ti follows a normal distribution and N is a large enough number, our hypothesis is consistent with the standard z-test, which is designed to evaluate the presence of algorithmic bias statistically."},{"assumption":"Logistic regression is used as the default predictive model.","type":"model","stated":"explicit","quote":"we use logistic regression as the default model for both prediction and fitting for simplicity"},{"assumption":"The peer‑identification threshold δ is set to 0.3 times the standard deviation of micro‑firms’ ICs.","type":"model","stated":"explicit","quote":"Without loss of generality, we set the default δ to 0.3 times the standard deviation of the micro-firms’ ICs."},{"assumption":"Micro‑firms are defined as the protected group (fewer than 10 employees and turnover < £2 million).","type":"data","stated":"explicit","quote":"Micro-firms are defined as those with fewer than 10 employees and an annual turnover of less than £2 million following the literature"},{"assumption":"The dataset does not exhibit a significant imbalance in the protected attribute.","type":"data","stated":"explicit","quote":"Our dataset does not exhibit a significant imbalance in the protected attributes, which is useful, as this would complicate testing our framework’s universality"}]},"datasets":{"status":"available","items":[{"name":"UK Archive Small and Medium‑Sized Enterprise Finance Monitor","provider":"BDRC Continental","kind":"other","access":"public","period":"2011Q1 to 2023Q4","universe":"UK SMEs (≤250 employees, ≤£25 million turnover)","quote":"The dataset used for this study is collected from the UK Archive Small and Medium‑Sized Enterprise Finance Monitor (BDRC Continental, 2023). The dataset compiles survey information on SMEs"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"weekly-monthly"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"Only 2.48% of micro‑firms are treated fairly while 41.51% face discrimination according to the audit.","kind":"statistical","quote":"our findings indicate that only 2.48% of micro-firms are treated fairly, while a staggering 41.51% face discrimination."},{"finding":"Audit results remain highly consistent when data imbalance is altered, demonstrating robustness of the framework.","kind":"theoretical","quote":"Even when we adjusted the data quality by altering the imbalance level, the audit results remained highly consistent with our original findings."}],"reported_performance":null,"out_of_sample":{"method":"unclear","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":8,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T18:04:51Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}