The Self-Appraisal Questionnaire (SAQ, Loza, 2005) is a self-report measure designed to predict violent and nonviolent recidivism. According to the authors of this study, published in the latest issue of the Journal of Interpersonal Violence, the SAQ has been shown, in several studies, to be valid in a variety of different “populations, settings, cultures, gender, and age groups” (p.672).
Despite this, some researchers wonder about the validity of the SAQ, because it is a self-report measure:
Such doubts arise from beliefs that self-report measures have inferior validity relative to professionally rated measures (Kroner & Loza, 2001) and a widespread belief that self-report questionnaires are more susceptible to lying and deception, especially when used in offending populations (p.673).
So the authors set out to investigate whether the SAQ is vulnerable to manipulation by offenders:
Two studies were conducted to investigate the vulnerability of the Self-Appraisal Questionnaire (SAQ) to deception and self-presentation biases. [...] In the first study, comparisons were made between 429 volunteer offenders who completed the SAQ for research purposes and 75 offenders who completed the SAQ as a part of the psychological assessments process required for consideration for early release. In the second study, 106 participants over two sessions completed the SAQ and the Balanced Inventory of Desirable Responding. Participants completed both measures under two separate sets of instructions: (a) Answers would be used for research purposes, and (b) answers would be used for making decisions about their release to the community [from the abstract].
Their results suggest that the SAQ is not vulnerable to manipulation:
The nonsignificant differences between the SAQ scores of the participants who completed the SAQ for research and those of the participants who completed the SAQ for decision on release purposes in two separate studies provide further support to the previously reported findings that the SAQ as a self-report measure not affected by deception and self-biases (p.679).
Reference:
- Wagdy Loza, Amel Loza-Fanous & Karen Heseltine (2007). The Myth of Offenders’ Deception on Self-Report Measure Predicting Recidivism: Example From the Self-Appraisal Questionnaire (SAQ). Journal of Interpersonal Violence 22(6):671-683
Photo credit: (fez), Creative Commons Licence
One Trackback/Pingback
[...] The Deception Blog has recently found research that provides evidence supporting that this might not be the case for all psychometrics: The Self-Appraisal Questionnaire (SAQ, Loza, 2005) is a self-report measure designed to predict violent and nonviolent recidivism. According to the authors of this study, published in the latest issue of the Journal of Interpersonal Violence, the SAQ has been shown, in several studies, to be valid in a variety of different “populations, settings, cultures, gender, and age groups” (p.672). [quote] [...]
Post a Comment