Corrigendum to “A meta-analysis of criterion effects for the Implicit Relational Assessment Procedure (IRAP) in the clinical domain” [Journal of Behavior Therapy and Experimental psychiatry 48 (2015) 59–65]

The authors regret that due to transposing errors, identified by Hussey (2025) and by Vahey, Nicholson, and Barnes-Holmes (2025), the following corrections should be made to Vahey et al. (2015). Firstly, the Forest plot should be corrected in the following ways. The r value displayed beside the study row listed as “Vahey et al. (2009)” should be corrected from .53 to .579; which also entails updating the corresponding 95% confidence interval (CI) from (.27, .72) to (.34, .75). Likewise, the r value displayed beside “Nicholson, McCourt et al. (2013)” in that same Forest plot should be corrected from .48 to .46; thus, entailing updates of the corresponding 95% CI from (.27, .79) to (.10, .72). Additionally, the study sample size stated for “Timko et al. (2011, Study 1)” should be adjusted from 41 to 50, thus entailing an update in the overall N size for the Forest plot from 494 to 503. Secondly, we determined that one of the Funnel plot datapoints, at coordinates (.49, .16), should be moved to coordinates (.68, .20). This correction did not change the overall profile of the original Funnel plot. Critically, the above corrections collectively resulted in only a marginally increased meta-analytic effect of r̅ = .46 (from the original estimate of r̅ = .45) without a 95% credibility interval (CR). Indeed, those corrections also did not meaningfully affect any of the remaining supplementary analyses regarding publication bias or statistical power contained within Vahey et al. (2015). For example, Kendell's τ(15) in the discussion section of the original article should be updated from .09 to .15, and its p-value should change from .66 to .43. Likewise, the findings from the relevant Vevea and Woods (2005) sensitivity analyses remain virtually unchanged, with extremely slight respective adjustments from −.007, −.007, −.016, −.016 to −.007, −.007, −.015, −.016. Nonetheless, the authors would like to apologise for any inconvenience caused. References.