Chapter 19 Kelly: Psychology of Personal Constructs 589ways. This work represents truly exciting ways that Kelly’s personality theory
may be applied to enable healing in those who suffer from the internalization of
culturally prejudiced construals within themselves.
Reducing the Threat to Feminist Identification
A puzzling phenomenon in social justice research is the widespread tendency of
many people to agree with feminist values but not to identify as feminists. Often this
is referred to as the “I’m not a feminist but” phenomenon, wherein individuals deny
feminist identification but follow up immediately with stated agreement with many
specific feminist values such as the belief that men and women and boys and girls
should have equal opportunities and choices (e.g., Zucker, 2004). Why should this
matter? Studies show that feminist identification is associated with a host of psycho-
social benefits. Those who call themselves feminists, compared with nonfeminist
self-labelers, are less socially dominant, have less hostile and benevolent sexist atti-
tudes, have greater self-efficacy, and have a better capacity to reject sexist and body
objectifying attitudes (e.g., Leaper & Arias, 2011; Zucker & Bay-Cheng, 2010).
Moradi, Martin, and Brewster (2012) sought to use personal construct theory’s
notion of threat to predict who does and does not identify with being a feminist.
To understand this, the authors provided the example that if a person considers
being assertive as a desirable character trait, views him- or herself as assertive, and
construes feminists as assertive, then integrating feminist identity into the self-
concept would not be threatening. If, on the other hand, the person thinks of
assertiveness as undesirable and incongruent with her or his ideal self and construes
feminists as assertive, then integrating feminist identity into the self-concept would
introduce Kelly’s notion of threat because it would be perceived as an intolerable
challenge to that person’s existing core construct of self.
In their first study, Moradi and colleagues (2012) used the same Rep test
method as in their previous study of internalized homophobia. In this case, 91 college
students rated the 30 bipolar constructs three times: first, in terms of their actual self,
second in terms of their preferred or ideal self, and third, “if you were a feminist.”
So, imagine that one pole is “selfish versus unselfish.” Students first circled whether
they associated themselves more with the term selfish or unselfish. Next, they circled
whether they would prefer to see themselves as associated with the term selfish or
unselfish. Finally they circled whether, if they were a feminist, they would associate
themselves with the term selfish or unselfish. As in the other study, threat scores
were calculated by counting the number of times a student’s actual-self matched their
ideal-self but differed from their feminist-self. The results showed, as hypothesized,
that greater discrepancy between actual-self, ideal-self, and feminist-self (greater
“threat” to the self-construct), resulted in lower feminist identification. In other
words, the more someone’s actual and ideal selves are removed from their view of
a feminist self, the lower the likelihood of identifying as a feminist.
Next, Moradi and colleagues intervened to reduce threat by changing students’
construals of the threatening construct of “feminist.” In Kelly’s terms, they sought to
increase the range of convenience and permeability of the threatening construct. To
do so, 115 college students received either an intervention or not (a control). Both
groups received pre- and post- measures of feminist threat and feminist identification.
The intervention was an interaction, in the context of a class period, with a group of