Criteria-based Content Analysis in True and Simulated Victims with Intellectual Disability

Abstract

The aims of the present study were to analyse people’s natural ability to discriminate between true and false statements provided by people with intellectual disability (IQTRUE = 62.00, SD = 10.07; IQFALSE = 58.41, SD = 8.42), and the differentiating characteristics of such people’s statements using criteria-based content analysis (CBCA). Thirty-three people assessed 16 true statements and 13 false statements using their normal abilities. Two other evaluators trained in CBCA evaluated the same statements. The natural evaluators differentiated between true and false statements with somewhat above-chance accuracy, even though error rate was high (38.19%). That lay participants could not effectively discriminate between false and true statements demonstrates that such assessments cannot be considered useful in a forensic context. The CBCA technique did discriminate at a better level than intuitive judgements. However, of the 19 criteria, only one significantly discriminated. More procedures specifically adapted to the abilities of people with intellectual disabilities are thus required. The presence of structured production, quantity of details, characteristics details and unexpected complications increased the probability that a statement would be considered true by non-expert evaluators. The classification made by the non-expert evaluators was independent of the participants’ IQ. A big data analysis is performed in search for better classification quality.