ABSTRACT
I read the Williams, Wormith, Bonta, and Sitarenios (2017) critique of the Singh, Grann, and Fazel (Citation2011) meta-analysis as someone who has struggled with conceptual and methodological issues when conducting risk assessment meta-analyses. I offer four lessons for risk assessment meta-analysts, based on my experiences as a risk assessment meta-analyst and consumer of risk assessment research. Lesson 1: We should accept the Dodo bird verdict in risk assessment instrument research and then move forward. Lesson 2: If you must compare, compare within samples. Lesson 3: Meta-analyses are often more useful for showing how much we don't know than how much we do know. Lesson 4: There are not enough risk assessment instrument studies examining the predictive validity of commonly used score interpretations.
Acknowledgment
I would like to thank Daniel C. Murrie for his comments on an earlier draft of this commentary.