Date created: | Last Updated:

Category: Project

Description: Outcome bias is the phenomenon whereby decisions which resulted in positive outcomes are rated more favorably than when the same decisions resulted in negative outcomes, ceteris paribus. We conducted a pre-registered replication of Gino, Moore, and Bazerman (2009) Study 1’s three scenarios (original’s: N = 120) with an extension adding the three scenarios from their Study 2. Our data was collected online with an Amazon Mechanical Turk sample recruited using CloudResearch (N = 402). We tested outcome bias by measuring participants’ ratings of how unethical, punishable, and blameworthy a decision maker’s behavior was in morally grey scenarios. We partially replicated outcome bias in ratings of punishment (original: η2G = .05 [.00, .14]; replication 1-3: η2G = .03 [.01, .11]; replication 4-6: η2G = .11 [.05, .18]) and blame (original: η2G = .12 [.03, .23]; replication 1-3: η2G = .06 [.05, .19]; replication 4-6: η2G = .16 [.08, .23]), but with support for outcome bias in ratings of unethicality in Scenarios 4-6 (η2G = .04 [.01, .08]) but not in Scenarios 1-3 (original: η2G = .059 [.004, .16]; replication: η2G = .00 [.00, .03]). Similarly, we only found support for the target’s finding that ratings of unethicality mediate the relationship between outcome and both perceptions of punishment and blame in Scenarios 4-6. We also added an extension of a control condition and found higher unethicality judgements when a decision resulted in a negative outcome relative to a control condition with no outcome information. Materials, data, and code are available on: https://osf.io/3bz2g/.