Clinicians and researchers are often faced with the difficult task of assessing individuals with anger problems. Norm-referenced measures provide a useful adjunct to interview methods but do not routinely assess the consequences of a client’s anger. The present study was conducted to refine the Anger Consequences Questionnaire (ACQ), a promising instrument developed through cluster analysis. After common factor analysis suggested a 6-factor solution, the construct validity of the revised ACQ was supported through relationships with trait anger, anger expression/control, aggression, hostility, stress, depression, and anxiety. Regression analyses showed that aggressive anger consequences (e.g., physical and verbal aggression, property damage, etc.) and damaged relationships were predicted by trait anger and aggressive anger expression. In contrast, the experience of negative emotions (e.g., depression, anxiety, guilt, etc.) and self-harm as a result of one’s anger were predicted by trait anger and anger suppression.
The cost of maladaptive anger is well known. Clients seeking treatment often report angry feelings as part of their presenting concerns (Edmondson & Conger, 1996). In both clinical and non-clinical settings, excessive anger and dysfunctional anger expression have been linked to aggressive behavior, reduced social support, occupational stress and burnout, substance abuse, anxiety and depression, and a variety of physical and mental health problems (Brondolo et al., 1998, Deffenbacher, 1992, Deffenbacher, 1993, Diamond, 1982, Goldstein and Niaura, 1992, Matthews and Haynes, 1986, Schum et al., 2003, Sirois and Burg, 2003 and Tescher et al., 1999).
Unfortunately, the task of classifying an individual’s experience and/or expression of anger as maladaptive is often difficult. The study of anger has received considerably less attention than violent behavior or other negative emotions such as anxiety or depression (DiGiuseppe, Eckhardt, Tafrate, & Robin, 1994). Moreover, the absence of diagnostic categories for which dysfunctional anger is the defining feature from the Diagnostic and Statistical Manual of Mental Disorders—4th Edition Text Revision (DSM-IV-TR; American Psychiatric Association, 2000) hinders effective assessment ( Lachmund, DiGiuseppe, & Fuller, 2005). In addition, the experience and expression of anger are guided by cultural norms, complicating the identification of problematic anger and increasing the possibility of misunderstanding ( Tanaka-Matsumi, 1995). Finally, the fact that anger is a common emotion ( Averill, 1983) means that it is important for researchers and clinicians to use norm-referenced measures to facilitate consideration of base rate information. While instruments such as the State-Trait Anger Expression Inventory-2 (STAXI-2; Spielberger, 1999) are a useful starting point for making normative comparisons of the frequency and intensity of anger and its expression, deviation from the norm is not sufficient to classify an emotion or a behavior as dysfunctional ( Maddux, Gosselin, & Winstead, 2005).
Recognizing a need to gather information about the adverse consequences associated with elevated trait anger and/or maladaptive anger expression, Deffenbacher, Oetting, Lynch, & Morris (1996) developed the Anger Consequences Questionnaire (ACQ) to assess the frequency of negative anger-related consequences (i.e., adverse outcomes of an individual’s manner of anger expression). Items were developed through interviews with highly angry college students and scaled so that respondents reported the frequency with which they experienced each consequence during the past month. The 33-item ACQ was developed through TRYSYS key cluster variable analysis. Eight clusters were identified (αs = .63–.92) and replicated during a second assessment with the same participants: “Physical Fights”, “Verbal Fights”, “Damaged Friendships”, “Property Damage”, “Hurt Self-Physically”, “Alcohol Use”, “Negative Emotions”, and “Legal/Vocational Difficulties”. After expanding the ACQ to 42-items, Lynch, Morris, Deffenbacher, & Oetting (1998) reported the addition of “Tense/Uptight” and “Reckless Driving” clusters and the deletion of the “Legal/Vocational Difficulties” cluster.
The present study was conducted in order to determine whether the ACQ could be improved through factor analytic methods. Cluster analysis suffers from several limitations (Gorsuch, 1997), and factor analysis is preferable for grouping items or variables (Hair, Anderson, Tatham, & Black, 1998). In addition, the ACQ contains 9-items that did not cluster, complicating interpretation. Thus, principal axis factoring was used to explore the latent structure of the ACQ to determine whether an empirical basis for placing items on scales would yield an improved instrument. Following the factor analytic revisions to the ACQ, the revised version was compared with measures of trait anger, anger expression/control, aggression, stress, anxiety, and depression in order to assess its construct validity. It was expected that anger consequences would be positively correlated with trait anger, maladaptive anger expression, and aggression. Inverse relationships between anger consequences and adaptive anger control were also predicted.
