Re-examining the issues: A response to Archer et al.

The purpose of this paper is to address and contest the issues brought forth by Archer et al. (2005) with regard to our original meta-analysis on the relationship between testosterone and aggression [Book, A. S., Starzyk, K. B., & Quinsey, V. L., (2001). The relationship between testosterone and aggression: A meta-analysis. Aggression and Violent Behavior.- A Review Journal, 6, 579 - 599]. We admit to some errors, such as the inclusion of multiple studies employing the same sample. Most of the discrepancies noted by Archer et al., however, were due to the inclusion of an erroneous table in the original article. We reanalyzed the data, again finding a small, positive relationship between testosterone and aggression. This relationship was moderated by age and time of day, confirming the results of our original analysis. Unlike the original analysis, we did find that the method of testosterone measurement was a moderator in the relationship. We disagree with Archer et al.'s claim that our definition of aggression was too broad. Not only are there theoretical reasons for the inclusion of studies on dominance and competition, but the exclusion of such studies did not significantly affect the correlation between testesterone and aggression. (c) 2005 Elsevier Ltd. All rights reserved.