Rejoinder to Goldberg, Lee and Ashton (2008): explaining counterintuitive findings

Abstract

In their critique of our erratum and addendum (Anderson & Ones, 2008), Goldberg, Lee, and Ashton (2008) allege clerical errors (1) where participants' HPI scores were incorrectly matched to other inventory scale scores and (2) in scoring of the OPQ and BPI scales. The first point was fully addressed by our erratum and addendum and Goldberg et al. do not present any new evidence that was not already considered. In this response, we further demonstrate that Goldberg et al. do not and cannot prove clerical errors in scoring OPQ and BPI scales. The circumstantial arguments presented by Goldberg et al. are shown to be fallacious. We respond to all conjectures and criticisms raised by their paper. We acknowledge, as we did in our previous publication (Anderson & Ones, 2003) that some of the specific convergent correlations and findings reported are not normative, but nonetheless explainable given psychometric and statistical artefacts that influence individual primary study results. Even basing our analyses upon the ‘reconstituted’ dataset generated and preferred by Goldberg et al., we demonstrate that our original conclusions do not change substantially. Responsible science requires the publication of all results, not only those confirming extant hypotheses and expectations. We continue to stand by the findings and conclusions reported in our previous publications (Anderson & Ones, 2003; Ones & Anderson, 2002). Copyright © 2008 John Wiley & Sons, Ltd.