Belief in the paranormal: measurement development and evaluation

Drinkwater, Kenneth Graham (2017) Belief in the paranormal: measurement development and evaluation. Doctoral thesis (PhD), Manchester Metropolitan University.

Preview

Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (4MB) | Preview

Abstract

This thesis evaluated paranormal belief measurement. Particularly, it considered the weaknesses of existing published measures. An extensive literature review identified frequently used paranormal scales and common associates. Consideration of identified measures produced a comprehensive pool of items (see Dagnall et al., 2010a and 2010b). Analysis of these items assessed the factorial structure of paranormal belief. Research progressed through four discrete phases that evaluated measurement of paranormal belief. Phase I: Exploratory factor analysis: Respondents completed a 64-item scale. Analysis supported a conceptually coherent and internally reliable 8-factor solution (haunting/ghosts, extra-terrestrial, superstition, religious beliefs, psi [premonition/psychokinesis], extra-sensory perception, astrology, and witchcraft). Phase II: Confirmatory factor analysis: Tested the emergent structure (47-Items) further. Respondents completed items retained from phase I alongside additional items (indexing astrology and witchcraft) to create a 50-item scale. Following removal of items sharing excessive variance, analysis confirmed an 8-factor solution. The emergent measure demonstrated good internal reliability and validity (content and face). Phase III: Alongside the emergent scale, respondents completed established paranormal measures (Revised Paranormal Belief Scale and Australian Sheep-Goat Scale), a series of probabilistic reasoning tasks and a measure of proneness to reality test deficits. Results revealed the new measure was psychometrically sound, contained coherent subscales, assessed construct breadth and correlated positively with established measures. In addition, non-believers perform less well on perception of randomness reasoning tasks. Finally, belief in the paranormal correlated positively with proneness to reality testing deficits. Phase IV: Further examined the newly constructed measure alongside mental toughness to assess validity and reliability in a real world context. Findings were in line with previous phases, suggesting excellent levels of consistency, while correlational analysis produced ideas for additional development of paranormal items and subscales. Measurement of the current MMUpbs, psychometric performance and subscale coherence, reveal improvements for future item design.