1. Self-Esteem Instability and Perceived Aggression Ashton C
Self-Esteem Instability and Perceived Aggression Ashton C. Southard, Brian Enjaian, Christopher J. Holden, & Virgil Zeigler-Hill Oakland University
Introduction Recent debate has considered the association between self-esteem and aggression. One view suggests that low self-esteem may sometimes cause aggression (Donnellan, Trzesniewski, Robins, Moffitt, & Caspi, 2005). However, research has revealed that this relationship is weak and inconsistent. A rival hypothesis suggests that it is actually high self-esteem that leads to aggressive behavior (e.g., Bushman & Baumeister, 1998). We believe some of this confusion may be explained by the large focus on self-esteem level and the relative lack of attention to other aspects of self-esteem such as temporal stability (i.e., fluctuations in moment to moment self-esteem over time). Self-esteem instability is operationalized as the magnitude of change in state self-esteem across repeated measurements (Kernis, 2005). It is important to consider self-esteem level (i.e., whether self-esteem is low or high) in conjunction with self-esteem instability because previous research has found that high levels of unstable self-esteem are related to a number of important outcomes including defensiveness (Kernis, Lakey, & Heppner, 2008), psychological adjustment (Zeigler-Hill & Wallace, 2012), and interpersonal style (Zeigler-Hill, Clark, & Beckman, 2011). Purpose The present study attempted to clarify the association between aggression and self-esteem by examining the possibility that self-esteem instability moderates the association between self-esteem level and aggression.
Method
Participants, Procedure, and Measures. A total of 234 (34 men, 200 women) undergraduates enrolled in introductory psychology courses at a university in southern region of the United States participated in exchange for partial fulfillment of class required research credit. All measures were completed via a secure website. Participants completed measures of self-esteem level (Rosenberg Self-esteem Scale; RSES, Rosenberg, 1965), completed a measure of state self-esteem (i.e., a modified version of RSES) each day for up to seven consecutive days, and recruited up to five friends or family members (1,078 perceivers participated in the study) to complete questionnaires concerning the perceived aggression (Forms and Functions of Aggression Scale; Little, Henrich, Jones, & Hawley, 2003) of the participant (i.e., the target).
Results
Results of the multilevel modeling analysis predicting perceived aggression are displayed in Table 1. Results revealed significant main effects for self-esteem level and self-esteem instability, which were qualified by their interaction. The predicted values for this interaction are displayed in Figure 1. Simple slopes analyses found that the slop of the line representing the association between self-esteem instability and pure overt aggression was significant for those with high self-esteem level, but not for those with low self-esteem. This suggests that those with stable high self-esteem were perceived as less aggressive than those with unstable high self-esteem or low self-esteem.
Table 1
Results of multilevel analysis predicting perceived aggression from self-esteem level and self-esteem instability.
Coefficient SE t d
Intercept
1.46
0.03
43.75***
Self-esteem level (SEL)
-0.09
0.04
-2.41*
.32
Self-esteem instability (SEI)
0.10
2.13*
.28
SEL x SEI
0.07
1.99*
.26
Figure 1
Predicted values for perceived aggression illustrating the interaction of self-esteem level and self-esteem instability at values one standard deviation above and below their respective means.