Measuring interrogative suggestibility, questions of reliability and validity

Gudjonsson developed two scales to measure interrogative suggestibility: Gudjonsson Suggestibility Scales I and 2 (GSS I and GSS 2; Gudjonsson, 1984a; 1987c). The aims of the present thesis were to examine issues related to the reliability and validity of these scales. Three studies are presented. Study I assessed the effects of two interviewer styles on measures obtained on the GSS 1. The hypothesis was that a generally abrupt demeanour adopted by the interviewer would lead to higher scores than a friendly demeanour. Results showed that participants tested in the Abrupt condition gained higher scores on two of the post-feedback GSS measures than those tested in the Friendly condition. It was concluded that post-feedback scores may be more sensitive to social aspects of suggestibility than responses to leading questions. Study 2 assessed the effect of the same interviewer demeanours on a sample of adolescents, a more vulnerable population (e. g. Richardson, Gudjonsson, & Kelly, 1995). It was hypothesised that the abrupt demeanour would produce higher GSS I scores, than a friendly demeanour and that this difference would be more marked than that found for normal adults. Results did not support the hypothesis. Scores were lower in the Abrupt condition; this difference was significant for post-feedback responses to leading questions. It was concluded that results provided further evidence that GSS scores are not readily predictable. Study 3 aimed to investigate indicators of "faking bad" on the GSS. It was hypothesised that participants instructed to fake suggestibility would demonstrate a unique scoring pattern. Results supported the hypothesis. It was concluded that an elevated pre-feedback score in the absence of any other raised scores may indicate malingering on the GSS. Issues related to the reliability and validity of the scales are discussed.Gudjonsson developed two scales to measure interrogative suggestibility: Gudjonsson Suggestibility Scales I and 2 (GSS I and GSS 2; Gudjonsson, 1984a; 1987c). The aims of the present thesis were to examine issues related to the reliability and validity of these scales. Three studies are presented. Study I assessed the effects of two interviewer styles on measures obtained on the GSS 1. The hypothesis was that a generally abrupt demeanour adopted by the interviewer would lead to higher scores than a friendly demeanour. Results showed that participants tested in the Abrupt condition gained higher scores on two of the post-feedback GSS measures than those tested in the Friendly condition. It was concluded that post-feedback scores may be more sensitive to social aspects of suggestibility than responses to leading questions. Study 2 assessed the effect of the same interviewer demeanours on a sample of adolescents, a more vulnerable population (e. g. Richardson, Gudjonsson, & Kelly, 1995). It was hypothesised that the abrupt demeanour would produce higher GSS I scores, than a friendly demeanour and that this difference would be more marked than that found for normal adults. Results did not support the hypothesis. Scores were lower in the Abrupt condition; this difference was significant for post-feedback responses to leading questions. It was concluded that results provided further evidence that GSS scores are not readily predictable. Study 3 aimed to investigate indicators of "faking bad" on the GSS. It was hypothesised that participants instructed to fake suggestibility would demonstrate a unique scoring pattern. Results supported the hypothesis. It was concluded that an elevated pre-feedback score in the absence of any other raised scores may indicate malingering on the GSS. Issues related to the reliability and validity of the scales are discussed