Feist−Feist: Theories of
Personality, Seventh
EditionV. Learning Theories 18. Kelly: Psychology of
Personal Constructs(^574) © The McGraw−Hill
Companies, 2009
procedure, participants rated role titles on whether they were more descriptive of
women than of men, men than of women, or neither/both. Gendered ratings were
scored 1 and nongendered ratings (either of both or neither) were scored 0, with
possible scores ranging from 0 to 20. In addition to the Rep test, participants com-
pleted a questionnaire concerning gender stereotyping and whether they applied
gender stereotypes to strangers in social situations and a questionnaire on sexist
gender attitudes.
Results showed that gender was a basic category for many participants, with
no one scoring 0, and the mean was slightly less than 10 out of 20. Additionally,
those who used gender most as a way of categorizing people on the Rep also were
more likely to apply gender stereotypes to strangers in social situations. Harper and
Schoeman (2003) concluded that “participants who frequently engaged in gender
stereotyping also organized their person schemas in terms of gender. This suggests
that participants who use gender stereotypes in perceiving strangers also tend to cir-
cumscribe their perceptions of friends, family members, and acquaintances along
gendered lines” (p. 523).
Smoking and Self-Concept
Previous research on self-concept and adolescent smoking has tended to find rela-
tively negative self-concepts of smokers compared with nonsmokers. More specifi-
cally, smokers have greater disparity between real and ideal self-concepts as well as
lower self-esteem (Burton, Sussman, Hansen, Johnson, & Flay, 1989; Webster,
Hunter, & Keats, 1994). Because different smokers smoke for different reasons,
however, an idiographic approach such as the Rep test should be better than the con-
ventional methods at measuring these differences.
The idiographic Rep test was used by Peter Weiss, Neill Watson, and Howard
Mcguire (2003) in two groups of college students, which included both smokers and
nonsmokers. More specifically, these researchers assessed participants’ views of smok-
ers’ and nonsmokers’ personalities using the Rep test. They predicted that smokers
would identify with and rate their own personalities more similar to the personality
descriptions they had of other smokers than of nonsmokers. They also predicted
lower self-concept (e.g., greater real versus ideal self disparity) for smokers than for
nonsmokers.
Participants first gave initials of three smokers and three nonsmokers they
knew, and the researcher later presented the participants with 18 different triads of
two smokers and one nonsmoker and 18 triads of two nonsmokers and one smoker.
Participants then used their own words to describe how the two were similar to each
other and different from the third. Next, they judged how the two were different from
each other and similar to the third. This resulted in 18 smoker personality trait terms
and 18 nonsmoker personality trait terms unique to each participant. The 36 person-
ality terms were then used in a self-concept assessment in which participants rated
four distinct forms of self-concept using a standard 7-point Likert scale: real self,
ideal self, real social self, and ideal social self. The real self-rating was assessed by
participants’ rating themselves as they really are, whereas the ideal self was the rat-
ing as they would ideally be. The social self-ratings were assessed by asking partic-
ipants to rate themselves as others their age would see them.
568 Part V Learning Theories