We believe that issues of methodology are often linked to problematic conceptual assumptions, and therefore important conceptual advances can derive from a clear methodological critique.  Methodological (and statistical) critiques underpin a number of important projects.  Similarly some conceptual advances come about with new or revitalised methods.

RWA, Fundamentalism, and Statistical Suppression

An apparently robust effect emerged in the literature on religiosity and prejudice, suggesting that the commonly found link between fundamentalist ideology and racial prejudice was reduced or even reversed when controlling for right-wing authoritarianism (RWA).  However, our analysis showed that this was a statistical artefact of the inflated correlation between the conventionalism component of RWA and religiosity.  After correcting for this inflated correlation, the “effect” disappeared and the expected relation between fundamentalism and racial prejudice was again found.  The uncomfortable link between religiosity and prejudice cannot be dismissed so easily.

A multi-component analysis of RWA

As part of the work on Religiosity and RWA, we also have contributed to the literature on the dimensionality of measures of RWA, showing in a novel way that at least 2, and more generally 3 factors can reliably be found in the RWA scale, and that this has important consequences for studies that focus on RWA, or control for RWA, in research on contested social attitudes. We have developed a convenient multi-component short form scale based on the 1996 version of the scale so that researchers with existing RWA datasets can introduce a multi-component analysis.

The Apples and Oranges bias

Our analysis of categorisation processes has highlighted a number of places in the literature where persons are assumed to have different qualities to groups, and these assumptions lead to methodological differences which in turn lead to confirmation of the original assumption.  We call these “Apples and Oranges” biases since persons are often assumed to be coherent and consistent with a clear core (Persons=Apples), whereas groups are assumed to be collections of people loosely held together by their group membership but lacking a clear core (Groups=Oranges).  We find these methodological biases in research on entitativity, cognitive load and face processing.

The Complexity of “Complexity”

We all have an intuitive notion about what it means for a person to be “complex” and the possible advantages and disadvantages of complexity or simplicity. The same might hold for groups.  Several  traditions have approached the notion of complexity differently leading to different measures and assumptions about what is advantageous.  One of our projects is to seek some integration of this diverse set of measures that can be used on similar types of self-representation and to help bring some coherence to these disparate approaches.