Electronic Thesis and Dissertation Repository


Doctor of Philosophy




Dr. Richard D. Goffin

2nd Supervisor

Dr. Bertram Gawronski

Joint Supervisor


Process dissociation is introduced as a way to overcome methodological limitations currently hindering sexism research. Researchers have identified two main types of sexism in hiring contexts. Meta-analyses confirm that men are traditionally advantaged over women (Tosi & Einbender, 1985), and that both genders encounter discrimination when applying to a job typically associated with the other gender (Davison & Burke, 2000). One problem is that these two biases are often confounded. As a result, researchers have hitherto been limited to showing that the two biases exist, but are largely unable to quantify them.

A possible solution might be process dissociation. It provides a way of measuring processes without the need to isolate them (Jacoby, 1991). The purpose of the dissertation was to explore process dissociation within the context of hiring decisions.

The current dissertation consisted of three parts. A pretest developed materials for use in the main studies. Study 1 then explored how process dissociation estimates compare to existing tools commonly used to study sexism. The pro-male bias was found to relate to old-fashioned sexism. The gender-job fit bias related to benevolent sexism. Measures of bias appeared uncontaminated by internal and external motivation to respond without sexism. Biases did not relate as expected to other measures of sexism, including hostile, modern, and neo-sexism, and two Implicit Association Tests. Finding differential relationships with some expected correlates supports the validity of process dissociation parameters and helps elucidate how the parameters fit within existing sexism constructs.

Study 2 further investigated validity through independent manipulation. Participants were randomly assigned to receive one of four word-sorting tasks containing primes intended to selectively influence one of the two types of bias. These manipulations had the desired effect for only some participants. Though both biases were selectively affected, a full double dissociation was not achieved. Consequently, Study 2 results provide only partial support for the proposed causal mechanisms and independence of process dissociation parameters.

Overall, results illustrate that process dissociation may be a helpful tool for use in research on sexism in hiring decisions. Limitations of process dissociation and potential next steps are discussed.