The measurement and prediction of conspiracy beliefs
A conspiracy theory or belief has typically been defined as an allegation of malevolent secrecy and plotting by a group of powerful actors, working in unison to fulfil sinister hidden goals at the expense of the general populace. Such beliefs tend to contradict common (and typically more benign) explanations for events and have the potential to reinforce or be used to ‘justify’ undesirable behaviours (e.g., discrimination, non-adherence to crucial healthcare practices, and environmental damage). However, the social psychological literature, specifically, concerning conspiracy beliefs is in its relative infancy. The overarching aim of this thesis is to provide greater coherency to future literature via a comprehensive examination of the measurement and prediction of conspiracy beliefs. A review of the existing research illustrates that, to date, the literature has tended to take a ‘fractionated’ approach to the study of conspiracy beliefs. That is, studies have tended to focus on scenario-specific conspiracies, and isolated predictors of conspiracy belief. Demonstrating that belief in real-world conspiracies and a generalised tendency to believe in conspiracies are equivalent has theoretical implications of understanding exactly what leads to these beliefs. To address this issue Study 1 examined the development, validation, and comparison of a Specific Conspiracy Belief Scale and a Generalised Conspiracy Belief Scale. A comparison of the relationships between various psychological predictor variables and both of these conspiracy belief scales was conducted in Study 2. These studies revealed that the Generalised Conspiracy Belief Scale was equivalent in performance in terms of its relationship to various predictor variables, and reliability and validity, to previously used specific conspiracy belief measures. The advantage of using the single generalised measure is its ability to be used consistently and comparatively across a range of different conspiracy scenarios. The review of the literature also revealed that although a number of predictor variables have been identified as being associated with conspiracy beliefs, studies have tended to only look at a relatively small subset of variables within a given study. Indeed, a critical analysis shows that the variables themselves may fall in to various (not necessarily independent) groupings or clusters: socio-political, personality, psychopathological, cognitive, and psychological control factors. Thus, the second goal of this thesis was to gain a better understanding of the relative contribution of the variety of variables that have been suggested as predicting conspiracy beliefs. A comprehensive analysis of the role played by a large number of potential predictor variables on their own and as part of domain groupings was performed within the context of a single population study. This issue formed a second aim of Study 2. The results showed that these variables can be reduced down to several common elements, which reveals there is no (as yet identified) single powerfully predictive psychological cause of conspiracy thinking. Rather, it is likely that psychopathological, socio-political, personality, and cognitive elements combine to explain individual differences in conspiracy belief. Finally, the validity of the Generalised Conspiracy Belief Scale and the relationships between a subset of key predictor variables (identified in Study 2) and conspiracy beliefs in the context of a wider population sample was a focus of Study 3. By using a large New Zealand-wide sample, Study 3 also broadened the scope of the thesis to examine the potential contribution of key demographic variables and psychological predictor variables in the prediction of conspiracy beliefs. Combining the demographic and psychological variables together in a hierarchical multiple regression revealed that nearly a quarter of variance in conspiracy belief was explained by these factors. However, when removing the shared variance of these predictors a number of demographic and psychological variables became non-significant or weakly predictive at best – a finding which again suggests that there are common elements that predict conspiracy belief. The remaining unique predictors of conspiracy thinking suggests that one of these common elements represent a hostile, suspicious, cynical, and threat-based worldview. Finally, although demographic variables do impact conspiracy beliefs, their unique effect is very small, and their effect works indirectly by impacting psychological predictors of conspiracy thinking. In conclusion, the current thesis has demonstrated that a single Conspiracy Belief Scale can serve as a useful and valid tool for future studies investigating conspiracy beliefs and that although individual psychological and demographic variables only weakly predict conspiracy beliefs on their own, they do cluster around potential themes which can aid in the development in a more comprehensive theoretical perspective on conspiracy.