The Elkins Hypnotizability Scale : validity, reliability, factor structure, and acceptability within a clinical sample.

Abstract

The current dissertation aimed to explore the reliability, validity, and factor structure of the Elkins Hypnotizability Scale (EHS) in a clinical sample of 173 post-menopausal women (Study 1). Study 1 included pre-existing data from a randomized-controlled trial on clinical hypnosis in the treatment of post-menopausal hot flashes. As hypothesized, the EHS demonstrated adequate reliability (α = .797) and high convergent validity with the current gold standard measure of hypnotizability, the Stanford Hypnotic Susceptibility Scale, Form C (SHSS: C) (ρ = .894). Results provided evidence that the EHS (M = 8.580) and SHSS: C (M = 8.415) were perceived to be equally pleasant. Additionally, Study 1 results indicated a high correlation between the SHSS: C and a shortened clinical version of the EHS (EHS: C) (ρ = .849), and a significant positive correlation between participants’ levels of dissociation on the EHS imagery task and total EHS score (ρ = .740). Study 1 confirmatory factor analysis results did not support the proposed four-factor structure of the EHS or SHSS: C. The results instead indicated that both the EHS (χ2 (9) = 8.412, p = .4932; CFI = 1.00; RMSEA < .001) and SHSS: C (χ2 (54) = 80.778, p = .0106; CFI = .971; RMSEA = .054) reflect a unidimensional structure in which the individual items on the scales are indicators for a general hypnotizability latent variable. Study 2 examined whether the model fit in Study 1 held in a multi-group sample consisting of the clinical sample in Study 1 and a collegiate sample of 225 Baylor University students. As predicted, measurement invariance test results suggested that the EHS (χ2 (18) = 38.575, p = .003, CFI = .986, RMSEA = .076) and the SHSS: C (χ2 (130) = 180.400, p = .002, CFI = .983, RMSEA = .044) demonstrate the same parsimonious single-factor structure, as demonstrated in Study 1. More specifically, the EHS reflected configural invariance across samples and the SHSS: C reflected strict invariance across samples. Implications for use of the EHS and EHS: C in clinical and research settings, as well as study limitations, and directions for future research were explored.