Construct validity, reliability and predictive validity of the dynamic risk assessment and management system (DRAMS) in offenders with intellectual disability

Abstract

Background. There have been a number of developments in the assessment of dynamic risk in the criminological literature. The dynamic risk assessment and management system (DRAMS) has been developed to facilitate the measurement of dynamic factors of risk for offenders with intellectual disability. Method. The study was designed to assess the construct validity, reliability and predictive utility of the DRAMS in 23 male forensic patients in a high secure setting. Predictions were made against independently collected incident data. Concurrent validity was assessed against the Ward Anger Rating Scale (WARS). Results. A reformulation of the individual section variables according to convergent and discriminant correlational analysis revealed Cronbach's alpha levels of >.8 for all sections apart from mood (alpha .750) and items retained to facilitate clinical information (alpha −.017). Two hundred pairs of WARS and DRAMS assessments revealed orderly sectional correlations. The sections of mood, antisocial behaviour, and intolerance/agreeableness predicted incidents with a medium effect size as did total DRAMS score. There were highly significant differences between assessments taken 1 or 2 days prior to an incident and control assessments conducted at least 7 days from an incident. Conclusions. The reformulated DRAMS has good construct and concurrent validity. It appears to be a reliable assessment tool and informs on risk levels relating to stable and acute proximal factors. It appears as an important addition to the utility of risk assessments for this population.

Evaluating the concurrent validity of the HCR-20 scales

“…Jung et al for the concurrent validity of these HCR-20 measures (e.g. Righthand et al 2005;Steptoe et al 2008), but this has not always been the case (see Nunes and Babchishin 2011). Our findings show some support for the Historical and Clinical scales of the HCR-20 consistent with the findings of other studies (e.g.…”

“…Juvenile Sex Offender Assessment Protocol' by Righthand et al 2005; 'Construct validity, reliability and predictive validity of the dynamic risk assessment and management system' by Steptoe et al 2008). Determining the construct validity of a test requires identification of the latent construct of interest and the integration of multifaceted evidence regarding the extent to which the test reflects that latent construct.…”

How do static and dynamic risk factors work together to predict violent behaviour among offenders with an intellectual disability?

“…; Steptoe et al . ). Dynamic measures also demonstrate moderate to high effect sizes in predicting sexual offending (e.g.…”

Why Do Some People with Intellectual Disability Engage in Offending Behaviour and What Can We Do About It? Editorial

“…() and Steptoe et al . () had already suggested that dynamic risk variables may be as good as or better than static variables in predicting violent and sexual incidents in offenders with intellectual disability. In this issue, Lofthouse et al .…”

“…Lofthous et al (2013) conducted an interesting analysis where they found that dynamic risk predictors may be act as a proxy for underlying static risk predictors. Lindsay et al (2004) and Steptoe et al (2008) had already suggested that dynamic risk variables may be as good as or better than static variables in predicting violent and sexual incidents in offenders with intellectual disability. In this issue, Lofthouse et al provide further evidence supporting these findings in relation to the perpetration of sexual offences.…”