Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Anger coping styles and major personality dimensions: a closer look at the construct validity of the… Hogan, Brenda E. 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-0471.pdf [ 3.96MB ]
Metadata
JSON: 831-1.0088586.json
JSON-LD: 831-1.0088586-ld.json
RDF/XML (Pretty): 831-1.0088586-rdf.xml
RDF/JSON: 831-1.0088586-rdf.json
Turtle: 831-1.0088586-turtle.txt
N-Triples: 831-1.0088586-rdf-ntriples.txt
Original Record: 831-1.0088586-source.json
Full Text
831-1.0088586-fulltext.txt
Citation
831-1.0088586.ris

Full Text

ANGER COPING STYLES AND MAJOR PERSONALITY DIMENSIONS A CLOSER LOOK AT THE CONSTRUCT VALIDITY OF THE BEHAVIOURAL ANGER RESPONSE QUESTIONNAIRE (BARQ) by BRENDA E. HOGAN B.A., Honours, Queen's University, 1996 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES Department of Psychology We accept this thesis as corifomring to the required standard UNIVERSITY OF BRITISH COLUMBIA April, 1998 © Brenda E. Hogan, 1998 In p resen t ing this thesis in partial fu l f i lment of t h e requ i rements fo r an advanced degree at t h e Univers i ty o f Brit ish C o l u m b i a , I agree that the Library shall make it f reely available f o r re ference and study. I fu r ther agree that permiss ion f o r extens ive c o p y i n g o f th is thesis f o r scholar ly pu rposes may be g ran ted by the head o f m y d e p a r t m e n t o r by his o r her representat ives. It is u n d e r s t o o d that c o p y i n g o r p u b l i c a t i o n o f th is thesis f o r f inancial gain shall n o t b e a l l o w e d w i t h o u t m y w r i t t e n permiss ion . D e p a r t m e n t o f The Univers i ty o f Brit ish C o l u W b i a Vancouver , Canada Date (LA \«\to4 DE-6 (2/88) Abstract The purpose of the present study was to further examine the psychometric properties of the newly developed Behavioural Anger Response Questionnaire (BARQ) and investigate how anger coping styles are conceptually embedded in the Five Factor Model of Personality (FFM). Previous factor analyses have demonstrated that the BARQ captures 6 distinct anger coping dimensions: Aggressive Anger out, Avoidance, JJiffusion, Social Support Seeking, Rurnination, and Assertion. The BARQ and the NEO-FFI were adrriinistered to 400 participants (232 students, 168 community members; 146 men, 226 women; 203 Asian-Canadians, 185 Caucasian-Canadians). Internal consistencies for the BARQ subscales were moderately high, with the exception of Diffusion. Low iotercorrelations were observed between the subscales. Women scored significantly higher than men on Social Support Seeking and sigriificantly lower than men on Aggressive Anger Out. Asian-Canadians scored significantly higher on Avoidant Coping than Caucasian-Canadians. Factor analysis revealed a stable 6 factor solution that fairly represented the subscales of the BARQ; this factor structure was shown to be highly similar to that obtained with alternate samples. Factor analyses were conducted on the data from each subgroup; these factor structures were also shown to be highly similar. Highly significant correlations were observed between Rurnination and Neuroticism (r = .47, p < .001) and Aggression and Agreeableness (r = -.51, p < .001). A moderate negative correlation was observed between Rumination and Agreeableness (r = -.29, p < .001). These correlations remained stable across all subgroups. Avoidant coping did not correlate well with any of the NEO subscales. These results suggest that the BARQ is a psychometrically sound and original measure and that the factorial model of anger coping underlying the BARQ can only be partially placed within the structure of the FFM. Table of Contents Abstract ii List of Tables v List of Appendices vii Introduction 1 The Conceptualization of Anger 2 Gender and Cultural Differences in Anger Coping 4 The Development of the Behavioural Anger Response Questionnaire (BARQ) 6 Anger Coping Styles and Personality Dimensions 12 The Two Historical Paths to the Five Factor Model of Personality 13 The Five Factors and their Definitions 16 The Comprehensiveness and Validity of the Five Factor Model 18 The Universality of the Five Factor Model 19 Five Factors - Too Few or Too Many? 20 The Five Factors of Personality and their Relation to Anger and Hostility 21 Coping and the Five Factors of Personality 23 The Five Factor Model of Personality and the BARQ 26 Research Objectives 28 Method 29 Participants 29 Materials 30 Procedure 30 Statistical Analyses 30 Results 31 Further Evidence for the Validity of the BARQ 31 Age, Gender, and Cultural Comparisons of the BARQ Subscales 35 The Relationship between the BARQ and the NEO-FFI 35 iv Discussion 36 The Validity of the BARQ 37 Gender and Culture Differences in Anger Coping Styles 39 Anger Coping Styles and the Five Factors of Personality 40 Implications for Research on Cardiovascular Disease 44 Conclusions 47 References 48 Tables 60 Appendix 83 V List of Tables Table 1 Internal Consistencies for the BARQ Subscales for Combined Sample and Across Subgroups 60 Table 2 Intercorrelations between the BARQ Subscales for the Combined Sample (n=400) 61 Table 3 Comparison of Intercorrelations between the BARQ Subscales for the Student Sample (n=232)/Community Sample (n=168) 62 Table 4 Comparison of Intercorrelations between the BARQ Subscales for the Men(n=146)/Women(n=226) 63 Table 5 Comparison of Intercorrelations between the BARQ Subscales for the Asian-Canadians (n=203)/Caucasian-Canadians (n=185) 64 Table 6 ULS Factor Analysis with Varimax Rotation of BARQ Data from the Combined Sample 65 Table 7 ULS Factor Analysis with Varimax Rotation of BARQ Data from Rutledge and Linden (1996) 66 Table 8 Coefficients of Congruence for Comparison of BARQ Factors from the Present Study with BARQ factors from Rutledge and Linden (1996) 67 Table 9 ULS Factor Analysis with Varimax Rotation of BARQ Data from the Student Sample 68 Table 10 ULS Factor Analysis with Varimax Rotation of BARQ Data from the Community Sample 69 Table 11 Coefficients of Congruence for Comparison of BARQ Factors from the Student and Community Samples 70 Table 12 ULS Factor Analysis with Varimax Rotation of BARQ Data from the Asian-Canadian Sample 71 Table 13 ULS Factor Analysis with Varimax Rotation of BARQ Data from Caucasian-Canadians 72 Table 14 Coefficients of Congruence for Comparison of BARQ Factors from the Asian-Canadian and Other-Canadians Samples 73 vi Table 15 ULS Factor Analysis with Varimax Rotation of BARQ Data from Male Sample 74 Table 16 ULS Factor Analysis with Varimax Rotation of BARQ Data from Female Sample 75 Table 17 Coefficients of Congruence for Comparison of BARQ Factors from the Male and Female Samples 76 Table 18 Means, Standard Deviations, and t-test Results for Student and Cornmunity Samples, Men and Women, and Asian-Canadians and Caucasian-Canadians 77 Table 19 Correlation Coefficients for each of the BARQ Subscales with each of the NEO-FFI Subscales for the Combined Sample (n=400) 79 Table 20 Correlation Coefficients for each of the BARQ Subscales with each of theNEO-FFI Subscales for the Student Sample (n=232)/Community Sample (n=168) 80 Table 21 Correlation Coefficients for each of the BARQ Subscales with each of the NEO-FFI Subscales for the Asian-Canadian Sample (n=203)/ Caucasian-Canadian Sample 81 Table 22 Correlation Coefficients for each of the BARQ Subscales with each of the NEO-FFI Subscales for the Female Sample (n=226)/Male Sample (n=146) 82 List of Appendices Appendix A: Subscales and Items of the BARQ 1 Anger Coping Styles and Major Personality Dimensions: A Closer Look at the Construct Validity of the Behavioural Anger Response Questionnaire (BARQ) Chronic emotional distress is associated with poor health and premature death (Friedman et al., 1995; Shekelle et al, 1983), but it is not yet known precisely why such an association exists. High levels of anger/hostility and certain anger response styles have been singled out as potentially important psychosocial contributors to heart disease etiology and progression. Although this association is well established, the mechanisms linking the phenomena remain obscure. Additionally, there has been substantial controversy over the conceptualization and measurement of anger, as well as which components of anger (e.g., anger experience, anger expression or suppression) are related to the development of disease. Although many initial studies postulated that anger was a umdimensional construct (Chesney & Rosenman, 1985), it is becoming evident that a more complex conceptualization of anger is required. It is important to make a distinction between hostility and anger. Anger is defined as an emotional response which depends on the appraisal of events and the assignment of meaning to them (Arnold, 1960); it disrupts ongoing behaviour through agitation and cognitive interference (Novaco, 1975). Anger requires the discimination of an event as provocation that serves as a cue for active coping and is usually a transient experience. Hostility is understood as an attitude rather than an emotion (Buss, 1961) which is associated with disgust, indignation, contempt, and resentment (Plutchick, 1980). Hostility has been described as "a motivating force - a conscious or unconscious impulse, tendency, intent..." (Saul, 1976) that may not require provocation. Unlike anger, hostility is considered to be habitual. Although hostility may often be accompanied by anger, the two are considered to be distinct concepts. 2 The Conceptualization of Anger Spielberger et al. (1985) suggest that hostility can be viewed as enveloping two dimensions: state anger, which encompasses the emotion of anger provoked by environment stimulation, and trait anger, which involves habitual feelings of anger. This view of an interaction between situational and trait variables is widely accepted. Although situational variables in anger have also been investigated, the study of anger has been based predominantly on the trait approach (Ben-Zur & Breznitz, 1991; Lai & Linden, 1992; Rutledge & Linden, 1996). However, Spielberger's development of measures which differentiate trait anger from anger expression styles illustrates the need to isolate more specific dimensions of anger. More specifically, Spielberger differentiates anger level (in response to varying levels of provocation) from how individuals cope with their anger once aroused. Early conceptualizations of anger expression or anger coping proposed a dichotomy of "anger-in" and "anger-out" response styles. The anger-in response style was associated with passivity, self-deception, and suppression of anger, while pronounced anger-out behaviour was characterized as verbally or physically aggressive. It was believed that extreme anger suppression would be related to negative health outcomes, while anger expression would be linked to positive health outcomes (Alexander, 1939). This reflected what one could also call a hydraulic model; individuals who suppressed their anger were thought to experience a build up of internal "pressure", while those who released their anger were believed to avoid such a build up. The accumulation of pressure experienced by those who withheld their anger was predicted to cause negative health outcomes. However, the Social Conflict Model proposed by Linden and Lamensdorf (1990) suggests that the relationship between anger and health is much more complex; it proposes that meaningful styles of anger expression he more in the middle of a continuum between aggression and passivity. This model posits that the relationship of anger expression and blood pressure is best described as a U-shaped curve; individuals with extreme anger-out tendencies as well as those exhibiting excessive anger-in 3 behaviour are thought to generate social or intrapsychic conflict as a consequence, and were also noted as having the highest blood pressures (Linden & Lamensdorf, 1990). Interestingly, people with the lowest blood pressures tended to fall in between these extremes and were described as reflective anger copers (Gentry et al., 1982). Gentry et al.'s findings clearly support the usefulness of the Social Conflict Model. There is a growing body of research investigating the relationship between the expression of anger and cardiovascular disease (CVD). It has been repeatedly shown that expression style is important for the study of the link between anger and heart disease and essential hypertension (Friedman & Booth-Kewley, 1987; Julius, Harburg, Cottington, & Johnson, 1986; Linden, 1987; Schneider, Egan, Johnson, Drabny, & Julius, 1986). Further research has revealed that the expression of anger and the experience of anger have different cardiovascular consequences. The overt expression of anger has been shown to be related to increased cardiovascular reactivity in the laboratory, while the experience of anger or frustration without outward expression has not (Siegman et al., 1992; Suarez & Williams, 1990). Siegman and his colleages (1987) also demonstrated that the expression of anger was positively correlated with CVD, while the experience of anger without outward expression was actually negatively correlated with the severity of CVD. Siegman (1994) posits that it is "the expression of anger, not the mere experience of these feelings, that seems to be the toxic factor in CVD" (p. 182). However, if multiple anger expression styles do exist, then it is feasible that there are different forms of withholding anger, some which may have cardiovascular consequences and some which may not. As a consequence of the confusion over the components of anger, several recent studies have attempted to ascertain the dimensionality of a broad range of hostility/anger scales through a series of factor analyses. The work of Riley and Treiber (1989) suggests the existence of three distinct factors, namely anger experience/hostility, verbal/adaptive anger expression, and maladaptive/physical anger expression. In contrast, Miller et al. (1995) identified eight separate factors: hostile anger expression, perceived control over the 4 expression of one's anger, frequency of anger, ease of anger provocation, brooding, hostile outlook, cynicism, and sullenness. Friedman et al. (1995) administered a battery of self-report measures frequently used to assess the role of anger and hostility in physical health, and concluded that studies predicting health from chronic negative socioemotional patterns should include at least four measures: aggressive overt hostility, alienated bitterness, introversion, and anxiety/depression. These results suggest that multiple, observable anger expession styles may exist and that they are not adequately measured by existing measures of anger levels and anger expression styles. Gender and Cultural Differences in Anger Coping It has been speculated that whether a behaviour is considered socially acceptable is largely specific to gender and culture; hence, particular normative groups may manifest very different patterns of anger expression and no single style may be appropriate in all environments. One marhfestation of these differences can be observed in differences in expressing anger between men and women. Research has suggested that women are more likely than men to suppress anger or express it through somatic symptoms (Haynes et al., 1978). Thomas (1991) found that women tend to feel less free than men to directly express their frustrations. On the other hand, men report more physically or verbally aggressive responses to anger events than women (Harris, 1992a). Men and women also hold different beliefs about the effectiveness of different anger coping styles. While females anticipate that distracting themselves will improve their mood, men are more likely to report that behaving aggressively will improve their mood (Harris, 1992b). By suggesting that the effectiveness of different anger expression styles may vary across separate cultural subgroups, the aforementioned Social Conflict Model may offer a solution to such disparities. Lai and Linden (1992) demonstrated that women with anger-in tendencies did not show elevated resting blood pressures and displayed more rapid recovery from anger provocation (in contrast to the predicted slow recovery), whereas this was not true for men. A sociocultural learning explanation has been offered for this finding; women are traditionally raised to be 'peace-5 makers' and to remain non-aggressive (Hokanson, Willers, & Koropsak, 1968). Hence, the withholding of anger should not create dissonance with women's self-perception. The absence of resulting intrapsychic conflict may explain the observed cardiovascular benefits of anger suppression for women. Differences may also exist between the anger coping styles of different cultural groups. Triandis (1986) points out that Canada, the U.S., and other English speaking nations differ from East and Southeast Asian countries on the dimension of mdividualism-collectivism. Canada and other English-speaking countries tend to be ''individualistic" in nature, with individuals placing their personal goals above the goals of the collective, while East and Southeast Asian countries tend to be "collectivistic" in nature, with individuals subordinating their personal goals to the goals of the collective. Differences on this dimension have translated to differences in social behaviour (Mann, Redford, & Kanagawa, 1985). Interestingly, Matsumoto (1991) reported that individuals from Japan, a collectivist nation, reported much more anger targeting strangers than Americans or Europeans. In contrast, Americans and Europeans tended to report more episodes of anger toward known others with whom they had an ongoing relationship. Matsumoto suggested that the lower levels of reported anger toward in-group members by the Japanese was a consequence of the high value placed on harmonious relations within the in-group in that culture. Matsumoto also proposed that the anger toward out-group members reported by the Japanese might serve to build cohesion vvdthin the in-group and perhaps solidify one's place within the in-group. Other research has indicated that expression of anger from higher status persons toward lower status persons is more common in Japan than in the U.S. and may serve the purpose of mamtaining status differences (Matsumoto, 1996). However, Lebra (1984) argues that the norms of social harmony make Japanese especially sensitive to and aware of conflict. Although the Japanese may be urilikely to outwardly express anger in an in-group setting, this does not mean that they do not experience anger. Rather, it seems plausible that they employ different anger coping techniques. Lebra contends that the Japanese are more likely to choose 6 to accept a situation rather than enter into conflict and become angry. Some of this anger may be redirected at the self. If this research is generalized from Japan to other collectrvist societies, then it seems likely that individuals from East or Southeast Asia will most frequently employ anger coping strategies that involve withholding anger and frustration. In contrast, individuals from North America and other individualist societies would be less likely to utilize such strategies since less emphasis is placed on group cohesion and they tend to more freely express anger to friends and family. It seems evident that gender and cultural differences exist in styles of anger coping. These differences may explain discrepancies in cardiovascular variables between men and women and individuals of different ethnic identity. Additionally, the ability to identify differences in anger expression between genders and cultures would be quite valuable to the field of social psychology. However, without a measure that captures the full range of anger coping styles it is impossible to determine exactly where these differences exist. The Development of the Behavioural Anger Response Questionnaire (BARQ) The existing confusion over the conceptualization and measurement of anger has made it difficult to address the question of whether or not certain dimensions of anger can predict cardiovascular health. The need for a reliable, multidimensional, and parsimonious tool designed to measure anger response styles led to the development of the Behavioural Anger Response Questionnaire (BARQ) (Rutledge & Linden, 1996). The major objective for this project was to create a tool that captured the curvilinear nature of the Social Conflict Model and incorporated a wide range of anger coping styles. Using factor-analytic methods, the BARQ was found to represent 6 distinct, non-overlapping anger coping styles, which were labelled as follows: Aggressive Anger Out, Anger Diffusion, Assertive Anger Coping, Social Support Seeking, Avoidant Coping, and Rumination. Aggressive Anger Out is characterized by outward displays of anger, either verbal or physical (two representative test items are: "I swear, use foul language, or curse at the person who annoyed me" and "I hit or push the 7 person who angered me"). Anger Diffusion is defined as a direct, yet passive and non-violent, method of dealing with anger, and involves occupying oneself with unrelated activities, such as doing housework or engaging in creative efforts (two representative test items are: "I work off my frustration by exercising or taking a brisk walk" and "I simply put more energy than usual into my work or studies"). Assertive Anger Coping involves a direct but non-aggressive interaction with the person responsible for causing the angering event (two representative test items are: "I let things cool off a little and then talk to the angering person about what happened" and "I stay in the situation, collect my thoughts, and then express how I feel rather than verbally attacking the person who angered me"). Social Support Seeking is characterized by behaviors in which the person actively or indirectly seeks out a person external to the event with whom to discuss his/her feelings (two representative test items are: "I leave the situation and look for someone (like a co-worker or a friend) who likely agrees with me when I describe what happened" and "I think about he problem for a while; later - for example in the evening - I discuss the incident with my spouse, partner, or a friend"). Anger Avoidance involves efforts to forget about or ignore feelings of anger (two representative test items are: "I convince myself that this is not worth getting upset about" and "I convince myself that the angering event is not important"). Finally, Rumination is an internally mediated way of venting feelings of anger, involving repetitive thought patterns of replaying the frustrating event that lack the outward physical manifestations of anger (two representative test items are: "T tMnk repeatedly about what I really would have liked to have done but did not" and "I develop a revenge fantasy where I 'get even' with the person who angered me"). The items reflecting the different subscales of the BARQ are displayed in Appendix A. The term 'anger coping style' is defined as a characteristic response (which may be behavioural, cognitive, or both) to situations perceived as irritating, frustrating, or infiiriating. It is thought that any anger coping style is an attempt to reduce the psychological distress caused by anger, however, some are considered to be more adaptive than others. Basic assumptions behind the BARQ are that different people employ different types of anger 8 coping behaviours, and that different anger coping reactions may ultimately lead to predictable health outcomes. The development and validation of the BARQ was a multi-step process which is summarized below (for a more detailed description see Rutledge and Linden, 1996). The belief that a more ideal measure of anger coping was needed to capture the curvilinear model of anger expression and health outcome proposed by the Social Conflict Model was discussed during informal meetings of Wolfgang Linden's research group. Semi-structured discussion of personal experiences with anger-provoking situations and response behaviours suggested that there may be at least four distinct response styles that can easily be incorporated within the Social Conflict Model. The 'working labels' applied at this time were "Direct Anger-Out", "Diplomacy", "Complain (to another person)", and "Shut-up". These labels were believed to describe behaviour covering a continuum from aggression to passivity. The research group then generated a large set of items descriptive of anger coping styles that had either been directly used or personally observed in others and that represented different points along the continuum suggested by the Social Conflict Model. The items were written as gender-neutral for universal apphcability. This process resulted in a set of 29 items which was labelled Version 1 of the Behavioural Anger Response Questionnaire (BARQ 1). A pilot study was then conducted using a small sample of convenience. Each of the 29 items was typed on individual index cards. The index cards were placed visibly in front of each participant, who was then asked to group the cards in a manner reflecting which items he/she believed belonged together. The resulting information was subjected to a smallest space analysis (a form of multidimensional scaling) which displays similarity of items by clustering them together on a scatterplot. Items receiving highly variable categorical ratings end up being "lost in space" while others are visually grouped together. The results of this analysis suggested that four, five, or six clusters might exist. For the expected four-factor solution, the distribution of items was: 7 "Direct Anger Out" items, 5 "Diplomacy" items, 9 "Complain" items, and 8 "Shut-up" items. At this point, it was agreed that at least four, and possibly 5 or 6, subscales might be required to fully capture the dimensionality of anger 9 coping responses. New items were added while the original subscales were revised by discarding items too distant from the clusters and rewriting others. This version became known as BARQ 2, which again consisted of 29 items. The next phase of development involved recruiting subjects from the undergraduate subject pool. A total of 199 students completed the questionnaire package. In order to receive full credit points, the student participants were asked to give an identical questionnaire package to a non-student family member or a family friend. One hundred and eighty eight (188) community participants were thus recruited. The test package consisted of 4 measures: the BARQ 2, the Spielberger Trait Anger Scale (STAS; Spielberger et al., 1983), the Spielberger Anger Expression Scale (SAES; Spielberger et al., 1985), and the Balanced Inventory of Desirable Responding (BIDR-6; Paulhus, 1991). The student and community samples were analyzed separately, although similar results were ultimately observed for both samples. Factor analyses were conducted, combining the student and corrimunity samples. Factor solutions with 4, 5, and 6 factors were examined by objectively grouping the items loading on the factors and assessing their content. The six factor solution was deterrrrined to be the most meaningful, both theoretically and empirically. The six subscales were named Aggressive Anger Out, Avoidant Coping, Support Seeking, Assertion, Diffusion, and Rumination. Low factor intercorrelations between the BARQ subscales were observed, indicating orthogonality of the six anger response styles. As expected, the BARQ subscale of Aggressive Anger Out significantly correlated with the Anger-out scale of the SAES (r = .72, p < .001) and the STAS (r = .25, p < .001). Spielberger's Anger-in dimension seemed to have been broken down into the Rumination and Avoidant Coping items. The Spielberger scales correlated weakly or nonsignificantly with the other identified factors (i.e., diffusion, support seeking, and assertion). The BARQ subscales generally showed low correlations with the Impression Management (IM) and Self-Deception Enhancement (SDE) scales of the BIDR-6, indicating that BARQ scores are not critically confounded by response styles. Internal consistencies for each of the subscales were moderate, ranging from .55 (Diffusion) to .76 10 (Aggressive Anger Out) in the student sample and from .57 (Assertion) to .75 (Social Support Seeking) in the conimunity sample. The data from this study were also used to examine gender differences in anger expression styles. In concordance with previous research findings, women showed a greater reliance on the more diplomatic styles of anger expresssion. More specifically, nonsignificant differences appeared between men and women on the Aggressive Anger Coping and Avoidant Coping dimensions, while women reported using Social Support Seeking and Diffusion anger response styles significantly more frequently than men. The next step (Study 4) was to have the 29 existing BARQ items rated independently by 7 raters with some trairiing in the meaning of the 6 factor solution. Interrater reHability was high. However, there remained concern over the inadequate level of internal consistency for some of the subscales. The Aggressive Anger Out, Rumination, and Social Support Seeking subscales showed acceptable levels of internal consistency, while Avoidant Coping, Diffusion, and Assertive Anger Coping appeared to be subpar (alpha levels below the .70 to .75 range). The small number of items included in each of the latter subscales was in part responsible for the observed moderate values. More items were generated so that there were at least 6 items per factor. This process resulted in a total of 37 items with 6 items per subscale, with the exception of the Aggression subscale which had 7 items. This version became known as BARQ 3. The objective of the next study was to determine the internal consistency of the BARQ 3. A test package consisting of the BARQ 3, SAES, STAS, STEAM-S (Standardized Experience of Anger Measure, Linden et al., 1997), and a measure of social support was completed by 107 student participants. Although improved, the internal consistency of the Avoidant Coping and Diffusion subscales remained below the acceptable range of .7 to .75. One month later, 71 of the original student subjects completed a second questionnaire package which consisted of the BARQ 3 and the STEAM-S. The one month test-retest correlations demonstrated good test-retest reHabihty for the six subscales of the BARQ. 11 In an effort to further improve the internal consistency and inter-item correlations of the subscales, the items were revised a final time. Items were removed if they consistently showed low inter-item correlations and were replaced with items which were anticipated to correlate better. The order of the items was altered to ensure that items from each subscale were randomly distributed. These refinements, which ultimately resulted in the BARQ 4,did not change the number of items per subscale. The next step was to examine the factor structure, internal consistency, inter-item correlations, and construct validity of this revised version of the BARQ. It was hypothesized that the increased number of items and the item revisions would lead to improved internal consistency. Once again, a battery of questionnaires, including the BARQ 4, STAS, SAES, and BIDR-6, was administered to a student (N=T02) and community sample (N=107). As expected, the internal consistency of some of the subscales improved markedly. All six BARQ subscales were determined to now have moderate to very good internal consistency (in the range of .70 to .85). Inspection of the correlational results of the BARQ 4 with the other measures revealed that the Aggression subscale of the BARQ was significantly correlated with both the STAS and the Anger-out scales of the SAES,while the Rumination subscale was significantly correlated with the STAS and the Anger-in scale of the SAES. Nonsignificant or negative correlations were observed between the six BARQ subscales and the IM and SDE scales of the BIDR-6, confirming earlier findings that the BARQ is not critically confounded by response styles. Orthogonality of the six factors was supported by the low subscale intercorrelations. When the student and community samples were combined, factor analyses again suggested that a six factor structure provided the best solution. This series of studies resulted in the current version of the BARQ, which consists of 37 items and 6 subscales. The six subscales are titled Aggressive Anger Out, Anger Diffusion, Assertive Anger Coping, Social Support Seeking, Avoidant Coping, and Rumination. Each item is scored on a 5-point scale according to the frequency of the behaviour described, ranging from 1 (rarely) to 5 (very frequently). With the exception of Aggressive Anger Out, 12 each subscale consists of 6 items, so each subscale score can range from a minimum of 6 to a maximum of 30. Because the Aggressive Anger Out subscale has 7 items, it has a minimum score of 6 and a maximum score of 35. The BARQ has good test-retest reliability, moderate to high internal consistencies, a stable factor structure, good construct validity with convergent and divergent validity components, and good concurrent validity. The BARQ 4 is believed to be a psychometrically sound, multi-dimensional tool for measuring anger coping which incorporates the curAdlinear concept of the Social Conflict Model. The present study provided further evidence for the psychometric value of the BARQ. It allowed an opportunity to examine the internal consistencies of the BARQ subscales and inter-scale correlations of the BARQ with a different sample. It also permitted an assessment of the stability of the factor structure of the BARQ. Because of the nature of the sample, comparions between different subgroups are also permitted: men and women, students and non-student community members, and Asian-Canadians and Caucasian-Canadians. Anger Coping Styles and Personality Dimensions Although it is well accepted that situational variables are important mediators of anger expression, the high test-retest correlations of the BARQ indicate that a given individual's choice among anger coping responses is relatively stable and is likely a function of more general personality attributes. The development of a tool designed to measure dimensions of anger coping requires an understanding of their relations to the other, well-established basic dimensions of personality. Ideally, the underlying conceptualization of the tool should be anchored in prevailing models of personality. The research which led to the development of the BARQ demonstrated the existence of six independent anger coping styles. As well, the BARQ's moderate to good test-retest correlations described in the previous section suggest that people may display consistent preference for a particular type of anger coping, which may indicate that general dispositions play a meaningful role in determining choice of anger response style. However, rather than isolating a single personality characteristic as the sole deteraunant of the type of anger coping response employed, it seems more reasonable to 13 assume that certain personality characteristics relate to some types of anger coping mechanisms, but not to others. It is also possible that some anger coping styles are more related to basic personality dimensions than others. Research in the realm of personality has repeatedly demonstrated the existence of a Five Factor Model of Personality (FFM), comprised of extraversion or surgency, agreeableness, conscientiousness, emotional stability vs neuroticism, and culture or openness to experience (Digman & Takemoto-Chock, 1981; McCrae & Costa, 1982; McCrae & Costa, 1985a; Norman, 1963; Ostendorf & Angleitner, 1992). The FFM is currently considered to be a robust and comprehensive conceptualization of the dimensions of personality. The primary objective of the present study is to examine how the anger response styles measured by the BARQ correspond with the "Big Five". The Two Historical Paths to the Five Factor Model of Personality It is well known that the Five Factor Model (FFM) emerged from studies of natural language trait terms (John, Angleitner, & Ostendorf, 1988). Beginning the search for personality dimensions in the natural language is based on the premise that "all important individual differences will have been noted by speakers of a natural language at some point in the evolution of the language and encoded in trait terms; by decoding these terms, we can discover the basic dimensions of personality.. .analyses of language will provide a comprehensive taxonomy of personality traits" (McCrae & John, 1992). Following this logic, Cattell (1946) formed a previously developed list of trait terms into synonym clusters and then created rating scales contrasting groups of adjectives. After several decades of research, Cattell contended that there existed at least 16 factors in the personality domain (CattelL 1973). However, Fiske (1949) used a version of Cattell's rating scales and identified only five factors, four of which can be recognized as the contemporary Extraversion, Agreeableness, Emotional Stability, and Intellect, although the fifth scale was not as clear-cut (Digman, 1996). The first distinct appearance of the "Big Five" was observed by Tupes and Christal (1961; as cited by McCrae & John, 1992). This study, which borrowed rating scales from 14 Cattell, found five factors that were stable across replications and in their reanalyses of Cattell's (1947, 1948) and Fiske's (1949) correlations. Norman (1963) replicated this finding, presenting a five factor solution based on the correlations of 20 peer rating scales. It was at this early date that Norman suggested that a "relatively orthogonal and highly stable structure of personal characteristics" had been identified. Hence, the lexical perspective spawned a great deal of research searching for the foundations of personality. However, as a consequence of controversy over the essential dimensionality of personality constructs, further research in this area was scarce until the 1980's. Historically, personality research has involved questionnaires with scales designed to measure constructs derived from personality theory, rather than from the lexical perspective (McCrae & John, 1992). For example, Eysenck (1947) proposed that personality essentially begins with two broad dimensions, Extraversion and Neuroticism. A third dimension of Psychoticism was subsequently added (Eysenck, 1955). These factors were assumed to exist at a high level of abstraction, organizing more specific traits such as rigidity and accuracy. Eysenck was careful to distinguish these dimensions from intelligence (Digman, 1996). Esyenck provided useful measures based on his conception of personality (Esyenck & Esyenck, 1964, 1975). It eventually became clear, however, that Esyenck's (1955) dimensions did not consume the full range of personality attributes. Costa and McCrae (1976) proposed another broad domain of personality traits in addition to the original dimensions of Extraversion and Neuroticism, labeled "Openness to Experience". It was at this point that the lexical and questionnaire traditions merged, an occurrence which eventually led to the birth of the contemporary FFM. The NEO model, proposed by Costa and McCrae (1980), focused on three major personality dimensions: Neuroticism, Extraversion, and Openness to Experience. This model was intended to summarize broad clusters of personality traits. However, its development began with personality scales widely used in the field of personality research, rather than with 15 trait names in a natural language (McCrae & Costa, 1985b). The NEO Inventory was designed by Costa and McCrae (1980) to measure the three dimensions of Neuroticism, Extraversion, and Openness to Experience, based on their 1976 findings from cluster analyses of the 16PF. In 1987, questionnaire measures of agreeableness-antagonism and conscientiousness-undirectedness were developed to represent Norman's lexically-based final two factors of agreeableness and conscientiousness, and the NEO was revised to include the new scales (McCrae & Costa, 1987). Convergence for all five factors was observed across both observers and instruments when they examined adjective scales and questionnaire measures in an adult sample on whom peer ratings on parallel instruments were available (McCrae & Costa, 1987). Similar findings have subsequently been reported (Trapnell & Wiggins, 1990). There has been some criticism of Costa and McCrae's development of the NEO-PI. Block (1995) purports that the original NEO was "carefully tailored" to fit their decision to focus on Neuroticism, Extraversion, and the lesser known Openness to Experience. He points out that the facet distinctions offered for each broader construct were derived from their own convictions on how each domain should be articulated rather than through theoretical or empirical means. Additionally, he suggests that by surrounding the adjectives defining the lexical "Culture" factor with their personally selected openness adjectives, their revised adjective factor of "Openness to Experience" became a better match for the "Openness to Experience" dimension they proposed in 1976. Of equivalent concern was the fact that NEO items selected to constitute the Agreeableness and Concientiousness scales were selected mainly on the basis of their correlations with adjective factor scores formulated to mark Agreeableness and Concientiousness. Thus, the subsequent correspondence with the adjective based Agreeableness and Conscientiousness scores was not surprising. Block proposes that the above methodological flaws may help explain the relation of the lexical and questionnaire measures. 16 The Five Factors and their Definitions There has been some debate over the definitions of the five factors (McCrae & John, 1992). Johnson and Ostendorf (1993) point out that some of the confusion stems from the fact that certain personality terms are included by different research groups in different factors. This is partly due to the nature of factor analyses: interpretation of factors depends on the secondary loadings of items, which vary from study to study and depend on the content of the variables entered into the analysis. Hence, these authors suggest that some personality terms are "blends"; attempting to locate these 'blends' on only one of the five factors will always be problematic. Other problems are conceptual or stem from disagreements between different research groups. There is probably the least controversy over the definition of Neuroticism. This factor is accepted to represent individual differences in the tendency to experience distress, and the associated cognitive and behavioural styles. High Neuroticism scorers experience nervous tension, frustration, depression, and self-consciousness which are frequently associated with irrational thinking, low self-esteem, poor impulse control, and ineffective coping (Costa & McCrae, 1987). Individuals low on Neuroticism are more likely to be calm, relaxed, and even-tempered. Although Extraversion is perhaps the most well-known factor of the FFM, there has been some confusion about what it represents. McCrae and John (1992) suggest that the discrepancy stems from the fact that Extraversion and Agreeableness mutually define the Interpersonal Circumplex, around which interpersonal terms are evenly distributed. The major uncertainty about Extraversion concerns whether it aligns with the Dominance or Affiliation axis of the circumplex. Goldberg (1990) identified this factor with Dominance, however others have suggested that it exists at some point between Dominance and Warmth (McCrae and Costa, 1989). Consequently, there is not complete consensus on which elements are central and which are peripheral to Extraversion. Many researchers agree that sociability, affiliation, cheerfulness, activity level, assertiveness, ambition, and sensation seeking are 17 characteristics of individuals who are high scorers on Extraversion (McCrae & Costa, 1987; McCrae & John, 1992). However, it has been suggested that affiliation should be divided into warmth and gregariousness, and that ambition should be considered an element of Conscientiousness (McCrae and John, 1992). Low scorers on Extraversion are considered to be quiet, reserved, shy, silent, and withdrawn (McCrae & John, 1992). The bulk of the controversy concerns the definition of Openness to Experience. It is well accepted that this factor is defined by such terms as imaginative and perceptive (McCrae & John, 1992). However, McCrae and John point out that many traits related to Openness to Experience are not represented among English trait adjectives. Research using questionnaires have found evidence of a much broader range of defining terms, including creativity, differentiated emotions, aesthetic sensitivity, and unconventional values. McCrae and John posit that this factor is not equivalent to measured intelligence; it is a dimension of personality, not intellectual ability. Alternatively, Hogan (1986) and Goldberg (1990) propose that this factor should be defined by terms such as "intelligent" and "perceptive" and should be perceived as an intellect factor. The final two factors are far less familiar than Neuroticism, Extraversion, or Openness to Experience. Agreeableness is often associated with characteristics such as nurturance, caring and emotional support (McCrae & John, 1992), while its opposite pole, antagonism, is described by traits such as skepticism, callousness, stubbornness, and irritability (McCrae & Costa, 1987; Ostendorf and Angleitner, 1992). Costa et al. (1991) contend that terms denoting conformity and compliance also belong to this factor, although conformity has been linked to the negative pole of Openness to Experience and Conscientiousness by other researchers (McCrae & Costa, 1985a, Johnson, 1983). Concientiousness is described by adjectives such as hardworking, ambitious, energetic, and persevering (McCrae & Costa, 1987). An individual low on Conscientiousness is considered to be undirected and lazy. 18 The Comprehensiveness and Validity of the Five Factor Model In order to deterrriine the structural validity of the FFM, it is useful to examine studies which simultaneously analyze the items from a large set of personality questionnaires. Noller, Law, and Comrey (1987) employed the Comrey Personality Scales (CPS; Comrey, 1970), the 16PF scales (Cattell et al, 1970), and the EPI scales (Eysenck & Eysenck, 1968) in an effort to explore the common factor structure. The most important part of the common variance of the questionnaire scales could be explained by four of the five Norman factors. A factor representing Openness to Experience (or Culture) could not be identified. Zuckerman, Kuhlman, and Camac (1988) factor analyzed a set of 46 questionnaire scales designed to measure constructs of biologically oriented temperament. Extraversion and Neuroticism were clearly identifiable. A third factor was identified, which Zuckerman et al. (1988) labeled "impulsive-unsocialized sensation seeking", which was believed to include aspects of both Openness and Conscientiousness. A fourth factor was also identified which was thought to possibly represent the negative pole of Agreeableness. Costa and McCrae (1988a) examined the relation of the Personality Research Form (PRF; Jackson, 1984), which was formed on the basis of the influential list of needs described by Murray (1938), to the NEO Personality Inventory (NEO-PI; Costa & McCrae, 1985). Although the PRF and NEO-PI have very different conceptual origins and measure different aspects of personality, the joint factor analysis demonstrated that the needs measured by the PRF can be meaningfully organized within the framework of the FFM. Ostendorf and Angleitner (1992) explored the structural validity of the FFM on the basis of 576 items extracted from inventories of different personality theories (PRF, Freiburg Personality Inventory (FRP-R; Fahrenberg et al., 1984), EPI, and NEO-PI). Although Extraversion and Neuroticism were the most robust factors, factor analysis revealed that the first five factors which emerged were well ahead of the others in terms of percentage of explained variance. These factors were found to be clear variants of the Big Five. 19 It is apparent that at least variants of the "Big Five" have been found in numerous studies. As Digman and Inouye (1986) stated: "if a large number of rating scales is used and if the scope of the scales is very broad, the domain of personality descriptors is almost completely accounted for by five robust factors". However, Ostendorf and Angleitner (1992) point out that the frequent emergence of Neuroticism and Extraversion as the most salient factors can be attributed to the overrepresentation of both Extraversion and Neuroticism items included in most prominent personality questionnaires. Alternatively, questionnaire items measuring Openness, Agreeableness, and Conscientiousness are generally underrepresented in questionnaire data. Hence, even in comprehensive samples of questionnaire items Extraversion and Neuroticism emerge as the most robust factors. Of these two factors, Neuroticism is most frequently found to be the most dominant factor.. This may be due to the fact that many prominent personality questionnaires are designed to measure clinical aspects of personality. The Universality of the Five Factor Model If personality traits are formed primarily by culture, then very different personality styles and trait structures should exist in different cultures. Alternatively, if personality traits represent variations in basic human tendencies, then personality structure may be universal. The FFM suggests a dispositionalist explanatory hypothesis that the five factors correspond to biological tendencies. Studies addressing this question have produced mixed results. The FFM is clearly identifiable in a study involving German subjects (Ostendorf, 1990). However, a study using Chinese trait terms did not show a clear correspondence with the dimensions of the FFM (Yang & Bond, 1990). This discrepancy may be explained as evidence that personality dimensions varying across cultures, or conversely that the lexical approach has certain limitations, such as confounding differences in personality structure with differences in personality language (McCrae & Costa, 1997). More recently, Costa and McCrae (1997) assessed the cross-cultural generalizability of the FFM, using translated versions of the Revised NEO Personality Inventory (NEO-PI-R; 20 Costa & McCrae, 1992) on German, Portugese, Hebrew, Chinese, Korean, and Japanese samples, and compared the resulting structure with the American factor structure. It was concluded that a very similar structure of personality can be found in these highly diverse cultures with languages from 5 distinct language families. Costa and McCrae suggested that these findings imply that personality trait structure is indeed universal. Five Factors - Too Few or Too Many? It has been suggested that fewer than five factors are necessary to encompass the dimensions of personality. Zuckerman et al. (1988) argued that three factors which corresponded to Eysenck's Extraversion, Neuroticism and Psychoticism account for the majority of the variance. Other researchers have also suggested three factor models (Gough, 1987) . However, McCrae and Costa (1987) demonstrated that when fewer factors were extracted from 80 adjective pairs, they could not be matched across two samples (one sample of self-reports and one of peer ratings), but an almost exact match was found with five factors. Similar results have been reported by Goldberg (1990). Alternatively, it has been argued that five factors are insufficient to adequately summarize all that is known about individual differences in personality (Mershon & Gorsuch, 1988) . However, advocates of the FFM contend that the five factors do not exhaust all possible descriptions of personality traits, but that they represent the highest hierarchical level of trait description (McCrae & John, 1992). A more serious question concerns whether there are additional major factors which are not included in the "Big Five". Digman and Takemoto-Chock (1981) found some evidence of a Culture factor in their analyses. Peabody and Goldberg (1989) found the existence of a Values factor in the structure of perceived relations among traits, although this was not confirmed in studies involving actual subjects. Although analyses of specific personality instruments frequently show evidence of potential additional factors, McCrae and John point out that this is probably due to method artifacts, sampling variability, or the selection of particular variables. Lexical studies have repeatedly demonstrated that additional factors are not replicable (Goldberg, 1990; Ostendorf, 1990). It 21 seems that if the "Big Five" are understood to represent the highest hierarchical level of trait description, the FFM can be considered an adequate and comprehensive model of personality. The Five Factors of Personality and their Relation to Anger and Hostility Fueled by evidence of a link between anger/hostility and cardiovascular disease, researchers have begun to exaniine the role of particular personality domains in the experience and expression of anger/hostility. Hostility and anger have emerged as facets of the factors of Neuroticism and Agreeableness (Costa, McCrae, & Dembroski, 1988). Costa et al. suggest the existence of two forms of hostility: neurotic and antagonistic. Neurotic hostility is thought to be characterized by the frequent and intense experience of anger, while antagonistic hostility is characterized by cynicism, callousness, and uncooperativeness. Costa et al. report that across two replicating samples who were administered the NEO-PI and a battery of items pulled from several anger and hostility measures, anger experience (i.e. neurotic hostility) was related chiefly to Neuroticism while anger expression (i.e. antagonistic hostility) was related almost exclusively to low Agreeableness. Correlations between Neuroticism, Agreeableness and various forms of hostility as measured by Rosenman's (1978) Structured Interview (SI) reveal that Potential for Hostility and Hostile Style are strongly related to low Agreeableness, while Potential for Hostility, Hostile Content, and Hostile Intensity had moderately large correlations with Neuroticism. Consequently, it appears that hostility as the tendency to experience anger and frustration but abstain from outward expression of anger is related to Neuroticism, while hostility as a facet of Agreeableness is interpreted as an interpersonal orientation characterized as hostile and irritable, or alternatively cold, contemptuous, and callous. This would lead to the expectation that Neuroticism will likely be related to an anger coping style characterized by an intense anger experience but a suppression of an external anger response, such as Rumination, while low Agreeableness will likely be related to strong anger-out tendencies, or the Aggressive Anger Out coping style. Interestingly, it has been demonstrated that Neuroticism is not a robust predictor of actual physical health outcomes, although it does reliably predict somatic complaints (Costa & 22 McCrae, 1987; Watson & Pennebaker, 1989). In fact, Siegman et al. (1989) reported a significant negative correlation between severity of CVD and neuroticism, as well as between severity of CVD and the experience of hostility without outward expression of anger. It is possible that these negative correlations are an artifact of how patients are selected for coronary angiography. Costa (1986) suggests that neurotic individuals may be prone to hypochondriacal angina-like pains despite good health, leading to their selection for coronary angiography. This may consequently result in an artificial inverse relationship between Neuroticism and severity of CVD. However, Seigman (1991) also reported that a negative correlation also exists between neuroticism and cardiovascular reactivity. This observation makes it more difficult to dismiss this finding as an artifact of subject selection. Felsten (1996) suggests that the lack of association between neurotic hostility and objective health problems may be due, at least partially, to an absence of exaggerated behavioural and physiological responses to stressors. Hence, although neurotic individuals are expected to experience frequent and exaggerated anger and frustration, they tend to refrain from outward expression of anger and consequently avoid the accompanying physiological arousal. Alternatively, low Agreeableness, or Antagonism, appears to be closely related to CVD. It has been suggested that the objective or expressive aspects of hostility, such as aggression, associated with antagonism may be responsible for this relationship (Dembroski et al., 1989; Siegman et al., 1987). Expressive hostility has also been associated with high cholesterol and low density lipoprotein levels (Dujovne & Houston, 1991) and heart rate and blood pressure reactivity (Felsten & Leitten, 1993). Felsten (1996) points out that the expressively hostile person is quick to react aggressively in situations perceived to be angering or frustrating. These outward behavioural responses are likely accompanied by intense physiological arousal, which together may account for the relationship between expressive hostility and CVD. Other research has also examined the relationship between personality dimensions and health behaviour and outcomes. Marshall et al. (1994) conducted a factor analysis of health 23 related personality scales, revealing 3 conceptually meaningful domains: Optimistic Control, Anger Expression, and Inhibition (of anger). Although these dimensions were correlated with the NEO-FFI subscales, none of the factors were clearly subsumed by a single personality domain. Anger Expression was linked with low Agreeableness and high Neuroticism, while Inhibition was associated with high Neuroticism and low Openness to Experience. These results are similar to those described previously. However, Optimistic Control was defined by instruments tapping positive health behaviours, such as optimism, hope, internal control, self-esteem, faith in one's abilities, and the capacity to derive meaning from life. This factor was associated with high Extraversion and Conscientiousness, and low Neuroticism. Booth-Kewley and Vickers (1994) reported that Extraversion, Conscientiousness, and Agreeableness were associated with more wellness behaviours and increased accident control behaviour, while Neuroticism was related to fewer wellness behaviours and less accident control behaviour. These findings imply that some personality dimensions are associated with better adjustment and more advantageous health-related behaviours, which may be related to the use of more adaptive anger coping techniques. Coping and the Five Factors of Personality As the FFM has become accepted as the prevailing model of personality, researchers have begun to examine the association between the dimensions of personality derived from the FFM and certain individual response styles. This stream of research has the potential to help investigators better ascertain the role of personality in certain cognitions and behaviours. In particular, the relationship between coping and the major dimensions of personality has received a great deal of attention. According to Fleishman (1984), coping can be defined as "both overt and covert behaviours that are taken to reduce or eliminate psychological distress or stressful conditions". Different people tend to employ different coping tactics, and these tactics may vary in their effectiveness at decreasing distress. Some controversy exists over whether coping strategies are related to personality, or whether personality variables are merely artifacts or nuisance factors. Although past research suggests that a comprehensive 24 model of coping must consider both situational and personality factors, it is generally accepted that there exists a relationship between general personality factors and specific coping behaviours (Fleishman, 1984; O'Brien & DeLongis, 1996). However, other authors consider some types of coping mechanisms to be independent of personality variables, despite that fact that they are temporally and situationally constant and independent of the type of stress reaction (Janke, Erdmann, & Kallus, 1985 as cited by Krohne, 1996; Lazarus, 1993). Despite the fact that there is a growing body of literature dedicated to the investigation of the relationship between coping styles and personality, there are no studies of which we are aware that examine the relationship between anger coping styles and the FFM. Anger coping responses differ from more general coping responses in that while coping responses occur in response to stressful events such as losses, threats, or challenges (McCrae & Costa, 1986), anger coping is a more specific concept that is considered to be a response to a situation perceived to be irritating, frustrating, or infuriating. Like coping responses in general, attempted anger coping responses may or may not be adaptive. Despite their differences, coping and anger coping are both responses to stressful situations. Consequently, it may be useful to examine the relationship between coping and the five dimensions of personality to help describe the role of personality in anger coping. Similar to anger coping, more general coping is not a unidimensional phenomenon. Folkman and Lazarus (1980) differentiate between problem-focused and emotion-focused coping. Problem-focused coping refers to acts taken to remove or mitigate the source of stress, while emotion-focused coping refers to attempts to reduce the psychological distress caused by the stressful situation. Pearlin and Schooler (1978) proposed three categories of coping dimensions: responses that change the situation, responses that alter the meaning or appraisal of the stress, and responses intended to control the emotional consequences of problems. More detailed taxonomies also exist. Factor analytic studies of the most widely used coping measure, the Ways of Coping Inventory (Folkman & Lazarus, 1980) revealed six main types of coping: problem-focused coping, seeking support, focusing on the positive, 2 5 distancing (i.e., minimizing threat by becoming psychologically detached from the perceived stress), wishful thinking (i.e., engaging in fantasies about escaping or avoiding the stressor), and self-blame (Marshall & Dunkel-Schetter, 1987, as cited in Bolger, 1990). Keeping in mind that not all coping mechanisms may be related to personality variables, recent studies examining the relationship between coping and the FFM have produced interesting results. Individuals high on Neuroticism tend to be less likely to engage in problem-focused coping (Hooker, Frazier, & Monahan, 1994), and instead are more likely to engage in forms of emotion-focused coping, such as escape-avoidance and self-blame (Bolger, 1990; Endler & Parker, 1990). McCrae and Costa (1986) found that people high on Neuroticism likely demonstrate increased hostility, withdrawal, wishful thinking, escapist fantasies, self-blame, sedation, passivity, and indecisiveness as coping mechanisms. Neuroticism has been found to be related to increased neurotic coping and decreased mature coping, suggesting that neuroticism may not only enhance maladaptive coping strategies but also promote fewer adaptive coping strategies (Costa & McCrae, 1990). O'Brien and Delongis (1996) reported that individuals high on Neuroticism use less planful problem solving and are more likely to use confrontation. The findings of these studies indicate that individuals high on Neuroticism tend to dwell on self-depreciating thoughts when faced with stressful situations, which suggests that people who use Rumination as an anger coping mechanism may also score high on Neuroticism. The observed association between withdrawal, escapist fantasy, and Neuroticism should also he noted, as it suggests that a relationship between Avoidant Coping and Neuroticism may exist. Individuals who score high on Extraversion appear to be more likely to engage in problem-focused coping than those low on this factor (Hooker et al., 1994). They are more likely to cope via seeking social support and are less likely to rely on emotion-focused strategies (O'Brien & Delongis, 1996; Hooker et al, 1994), including less avoidance, self-blame, and wishful thinking (McCrae & Costa, 1986). Their coping strategies may include positive thiriking, substitution, restraint, and increased use of rational action (McCrae & 26 Costa, 1986). Hence, it seems plausible that individuals who are extraverted may employ Assertion and Social Support Seeking as anger coping mechanisms, and be unlikely to use strategies such as Avoidant Coping. McCrae and Costa (1986) reported that individuals high on Openness to Experience are more likely to employ humour as a coping mechanism, while those low on Openness to Experience were more likely to rely on faith when confronted with a stressful situation. Those high on Openness to Experience also appear to use positive reappraisal more frequently than those low on Openness to Experience (O'Brien & DeLongis, 1996). These authors also found that individuals high on Openness to Experience tended to take a broader, more creative view of stressful situations. However, Hooker et al. (1994) found no relationship between coping mechanisms and the Openness to Experience factor. The inconsistency of findings for this factor makes it difficult to predict any potential relationship between anger coping mechanisms and Openness to Experience. The Agreeableness factor of the FFM has been associated with such coping responses as social support seeking and infrequent use of self-blame, avoidance, and confrontation (Hooker et al, 1994; O'Brien & Delongis, 1996. Therefore, it seems probable that individuals high on Agreeableness would be unlikely to utilize Aggressive Anger Out strategies, while those low on Agreeableness may frequently demonstrate aggressive tendencies. Individuals high on Agreeableness may turn to techniques such as Social Support Seeking. Individuals high on Conscientiousness have been found to employ more problem-focused coping techniques (e.g., planful problem solving) than emotion-focused techniques (e.g., escape and avoidance) (O'Brien & Delongis, 1996; Hooker et al., 1994). Hence, it may be the case that these individuals are likely to use Assertion as an anger coping mechanism. The Five Factor Model of Personality and BARQ The creation of the BARQ stemmed from a general dissatisfaction in the field of behavioural medicine with inventories designed to measure anger. Anger has repeatedly been associated with coronary heart disease and essential hypertension (Alexander, 1939; Diamond, 27 1982; Jorgensen et al., 1996). Of the many different dimensions of anger, it appears that expression style is particularly important for the study of the link between anger and heart disease and essential hypertension (Friedman & Booth-Kewley, 1987; Julius, Harburg, Cottington, & Johnson, 1986; Linden, 1987; Schneider, Egan, Johnson, Drabny, & Julius, 1986). Recent research suggests that multiple, observable anger expressions styles likely exist and are simply not captured by existing measures of anger levels and anger expression styles. The lack of a reliable and parsimonious tool designed to measure anger expression styles unquestionably hinders research examining the relationship between anger and heart disease. The need for such a tool led to the development of the BARQ. Research focusing on anger and hostility has been based on the assumption that the experience of anger involves an interaction between trait and situation variables (Ben-Zur & Breznitz, 1991; Lai & Linden, 1992; Spielberger et al., 1985). Personality has been understood to be important in determining how anger is expressed. Hence, the development of a tool designed to measure anger expression should involve exarnining how the tool relates to personality. This process should increase understanding of the role of personality in deternnning how anger is expressed as well as provide evidence for construct validity of the BARQ. The present study proposes to examine the relation of the 6 anger dimensions measured by the BARQ to the five factors of personality measured by the NEO-FFI. Despite some criticism, the FFM has been accepted by the personality psychology community as a comprehensive model of the dimensions of personality. If the FFM indeed covers all overlapping aspects of personality, then the six anger response styles measured by the BARQ should also be fully encompassed by the factors of Neuroticism, Extraversion, Openness to Experience, Agreeableness, and Conscientiousness. If the anger response styles measured by the BARQ are fully captured by the five factors of personality then it appears that these 6 response styles are only a specific case of the "Big Five" and can be fully explained as a function of general personality traits. An incomplete overlap between the BARQ and the NEO-FFI would suggest that the five factors of personality do not adequately 28 capture the full dimensionality of anger coping styles. This may indicate that the FFM is incomplete, or alternatively that anger coping styles cannot be fully explained by personality. Given existing findings in the Big Five and coping literatures, we predict that the factorial model of anger coping underlying the BARQ can only be partially placed vvithin the structure of the FFM. More specifically, it is expected that the anger response style of Rumination as measured by the BARQ will correlate positively with the NEO measure of Neuroticism, and that the anger response style labeled Aggressive Anger Out will be related to the opposite pole of Agreeableness, consequently correlating negatively with the NEO measure of Agreeableness. It is also anticipated that the BARQ measures of Social Support Seeking and Assertive Anger Coping will correlate positively with the NEO's Extraversion and Conscientiousness, and negatively with Neuroticism. However, it is predicted that the BARQ constructs of Anger Diffusion and Avoidant Coping will not correlate well with any of the five factors measured by the NEO. Research Objectives This study was designed with two main objectives. First of all, the adnrinistration of the BARQ to a large sample of individuals will allow further examination of the psychometric properties of this instrument. Furthermore, the composition of the sample in this study allowed for the comparisons of men and women, older community members and students, and Asian-Canadians and Caucasian-Canadians. This was expected to provide evidence for the generalizability of the BARQ as well as allowing exploration of potential differences between subgroups in styles of anger coping. Secondly, this study will examined the relationship between anger coping styles and the five factors of personality. It was intended to determine whether the six anger coping styles measured by the BARQ are completely or only partially anchored within the structure of the FFM. To address these matters, correlational techniques were employed to examine the relationship between the BARQ subscales and the NEO factors. As well, through examination of the strength of correlations between certain anger coping styles and personality 29 factors, it was hoped that this study would provide valuable information about the relative importance of the role of certain personality traits in detemrifiing anger coping styles. Method Participants Students taking first- or second-year psychology courses were recruited from the departmental subject pool. Three hundred (300) questionnaire packages were distributed. A total of 232 students completed and returned the questionnaire package for credit points. There were 77 males and 144 females (11 students did not specify their gender), with ages ranging from 17 to 21 years and an average age of 20.1. In order for the participants to receive full credit points, they were recruited as secondary research assistants by having them give an identical test package to a non-student family member or family friend. Once completed, these questionnaires were to be sealed by the respondent before returning it to the student. These sealed envelopes were either directly returned to the laboratory or collected by a research assistant during class time. This method has multiple advantages: it provides relatively easy access to a large community sample and it also results in a sample that is matched to the student sample in terms of ethnic composition and social class. A total of 170 community packages were returned. Two participants had to be excluded due to missing data, leaving a total of 168 community participants. There were 70 males and 82 females (16 corrmiunity participants did not specify their gender). The age range was 14 to 70 years with an average age of 35.6 years. . In total, there were 400 participants. The age range was 14 to 70 years with an average age of 28 years. There were 246 women and 147 men. The sample consisted primarily of two major cultural groups, Asian-Canadians and Caucasian-Canadians. Asian-Canadian was defined as any individual who was born in South or Southeast Asia or had one or both parent(s) born in East or Southeast Asia. The Caucasian-Canadian group included individuals who were born in Canada, the U.S. or Europe and had both parents born in Canada, the U.S., or Europe. There were 203 Asian-Canadians and 185 Caucasian-30 Canadians. Twelve individuals were excluded from this categorization because they did not meet the criteria for either group. Materials The questionnaire package consisted of the Behavioural Anger Response Questionnaire (BARQ; Linden et al., 1998) and the shorter version of the NEO, the NEO Five Factor Inventory (NEO-FFI; Costa & McCrae, 1992). Information about ethnic background and acculturation was collected via a modified version of the Suinn-Lew Asian Self-Identity Acculturation Scale (Suinn, Richard-Figueroa, Lew, & VigiL 1987). Information about social support was collected, but was not analyzed for the purpose of this study. Procedure Student and cornmunity participants were permitted to complete the questionnaire packages at home. They were asked to complete the questionnaires as honestly and accurately as possible, and were assured of their anonymity. Participants were asked to return the completed questionnaire package in a sealed envelope which indicated whether they belonged to the student or corrrmunity sample. Community participants returned their completed test package to the students, who in turn returned the packages during class time or directly brought them to the laboratory. Statistical Analyses Internal consistencies and inter-scale correlations were calculated for each of the BARQ subscales. In order to detennine whether the correlational structure of the subscales differed between these subgroups, subscale inter-correlations for each of the subgroups were also calculated and compared. An absence of meaningful differences allowed the subgroups to be combined, thereby increasing statistical power. The combined data were subjected to exploratory factor analyses. In order to determine whether the factor structure of this data was similar to the factor structure of the data obtained by Rutledge and Linden (1996), coefficients of congruence were calculated for each of the factors (Harman, 1976). The data were then broken down into the subgroups 31 (student and community, men and women, and Asian-Canadians and Caucasian-Canadians). Factor analyses were conducted on each subgroup. In order to determine whether the factor structures of the student and cornmunity samples, men and women, or Asian-Canadians and Caucasian-Canadians differed, coefficients of congruence were calculated for each set of factors. Means and standard deviations were calculated for the student and corjomunity samples, men and women, and Asian-Canadians and Caucasian-Canadians for each of the subscales . T-tests were performed to detennine whether any of these subgroups exhibited differences in anger coping styles. Correlational techniques were employed to investigate the relationship between the five personality factors measure by the NEO-FFI and the six anger coping styles measured by the BARQ. Correlation coefficients between each anger coping style and each personality dimension were calculated for the combined sample and each of the subgroups. Results Further Evidence for the VaUdity of the BARQ The present study confirmed that the BARQ is a sound psychometric instrument, demonstrating acceptable levels of validity. The internal consistencies for most of the BARQ subscales in the combined sample and in each subgroup (student and community samples; men and women; Asian-Canadians and Caucasian-Canadians) ranged from moderate to high (see Table 1). Only Diffusion fell below the acceptable standard of .70. Low intercorrelations between the BARQ subscales were observed for the combined sample (see Table 2). Intercorrelations between the BARQ subscales for each of the above subgroups are provided in Tables 3-5 respectively. The only subscale intercorrelation that remained at statistically significant levels across all the subgroups was between Rumination and Aggressive Anger Out, ranging from .29 (student sample) to .46 (cornmunity sample). No significant differences between the inter-scale correlations of students and community members, men and women, or 32 Asian-Canadians and Caucasian-Canadians were observed (see Tables 3, 4, 5). An alpha level of .001 was employed to maintain an overall .05 risk of type I error. The absence of meaningful correlational differences between the subgroups allowed the combination of the subgroups (Harman, 1976). The combined data were subjected to exploratory factor analyses. Several criteria were used to determine the number of factors, including the scree plot of the eigenvalues, the Kaiser rule for eigenvalues greater than 1, and the use of the maximum likelihood ratio test. For the purpose of our analyes, loadings greater than .3 were accepted as nontrivial. Results of the maximum likelihood ratio test suggested 8 factors. This statistic was used to provide an upper end estimate, as it almost invariably overestimates the number of factors (Hakstian, Rogers, & Cattell, 1982). The Kaiser rule for eigenvalues greater than 1 suggested 8 factors, while the scree plot indicated breaks at 5 factors. Examination of the five, six, seven, and eight factor rotated solutions showed that the six factor model provided a substantially clearer solution with an acceptable number of loadings per factor. As a final criterion, a communality threshold of. 1 was set for the items (Hakstian, Suedfeld, Ballard, & Rank, 1986). Any item with a communality value beneath . 1 was ekminated from the solution. As in previous factor analyses using the BARQ (Rutledge & Linden, 1996), an unweighted least squares (ULS) method of extraction was employed. Given the generally low communality values found with individual items, along with the expectation that unequal number of items would load on the anger coping style subscales, weighting by item communalities would potentially unbalance the solution in favour of the items with higher communalities and those factors with a greater number of loadings. The ULS method was therefore chosen to assign equal weighting to items. A total of 3 oblique and 2 orthogonal solutions were examined. A standard varimax rotation was compared to 3 oblimin rotations and an equamax rotation. Solutions were rated according to the number of hyperplane values (loadings < .1), number of complex items (items loading greater than .3 on two or more factors), and for interpretive clarity. When these 33 criteria were weighed in combination, the varimax solution provided the clearest solution. The low inter-correlations observed between the subscales provided additional support for the use of an orthogonal solution. This process resulted in the use of a ULS factor analysis with a varimax rotation. Items with loadings of .3 or greater on the six factors are presented in Table 6. These items all possessed communality values greater than the . 1 threshold. The item clusters were then scratinized in an attempt to identify underlying constructs. The items loading on each factor were very closely aligned with the items from each subscale. Four items had complex loadings, but each had it's highest loading on the factor with which it was perceived to fit best. Factor 1 clearly represented the Assertive Anger Coping Subscale, consisting of all six of the items from this subscale. An item from the Rumination subscale, item 7, also loaded on this factor, although it's loading was relatively low. All 7 of the items from the Aggressive Anger Out subscale loaded on Factor 2. Item 2 was complex, with a low negative loading on Factor 5. Two items from other subscales loaded on this factor; item 11, from the Avoidant Coping subscale, had a low negative loading, and item 30, from the Rumination subscale, had a relatively low positive loading. Factor 3 consisted of all 6 items from the Social Support Seeking subscale. Factor 4 appeared to represent the construct of Rumination, consisting of five out of the six Rumination subscale items; item 7 was missing. Item 30 was complex, also loading on Factor 2. Factor 5 clearly aligned with Avoidant Coping. All items from the subscale were represented, although item 12 (Diffusion) and item 2 (Aggression; negative loading) appeared as complex loadings. Items 12 and 2 had relatively low loadings on this factor. Factor 6 appeared to represent the Diffusion subscale; all items were represented. Item 12 overlapped with Factor 5. Although not perfectly clean cut, the six factors clearly showed underlying similarity and appeared to fairly represent the subscales of the BARQ. In order to determine whether the factor structure of the BARQ obtained in the present study differed from the 6 factor structure obtained by Rutledge and Linden (1996), the 34 factor structures from the two samples were compared. The factor loadings for the six factors obtained by Rutledge and Linden are provided in Table 7. Coefficients of congruence were calculated for factors believed to represent the same underlying construct. Factors with coefficients of congruence over .80 were interpreted as highly similar (Harman, 1976). This process revealed that the factor structures of the two samples were analogous (see Table 9). Overall, it appeared that the factor structure of the BARQ remained very stable across two independent samples. The combined sample was then broken down into student and corrimunity subgroups. A ULS factor analysis with varimax rotation was performed on each sample (see Tables 9 and 10 for the factor loadings). Based on the criteria outlined previously, a six factor solution was found to provide the clearest solution. In each case, the six factors showed underlying similarity and were representative of the subscales of the BARQ. The factor structure of each sample was compared by calculating coefficients of congruence for the factors from each sample representing the same anger coping style (see Table 11). The only factor that did not reach the acceptable standard of .80 or higher was Avoidant Coping. The other five factors were Mghly similar, indicating that the rest of the factor structure was stable for the student and community samples. A second comparison was made between the Asian-Canadians and the Caucasian-Canadians. A ULS factor analysis with varimax rotation was conducted on each subgroup. Again, a six factor solution was found to be the best for both the Asian-Canadian and the Caucasian-Canadian samples. Factor loadings for each sample are provided in Tables 12 and 13. The factor structures of the samples were compared using coefficients of congruence (see Table 14). The coefficient of congruence calculated for the factors representing Social Support Seeking fell slightly below the acceptable standard of .80, with a value of .79. However, the rest of the factor structure was very similar for the Asian-Canadian and the Caucasian-Canadian subgroups. 35 The same procedure was followed for the male and female subgroups in order to determine whether the factor structures displayed any substantial differences. A six factor solution provided interpretive clarity; the factors showed underlying similarity and appeared to be closely related to the anger coping subscales of the BARQ. The factor loadings for each sample are provided in Tables 15 and 16. The coefficients of congruence ranged from .82 to .95 (see Table 17), once again indicating that the factor structure was stable across the two samples. Age. Gender, and Cultural Comparisons of the BARQ Subscales Table 18 provides means, standard deviations, and statistical test results for the student and community samples, men and women, and Asian-Canadians and Caucasian-Canadians. Alpha levels for these tests were set at .001 in order to maintain an overall .05 risk of type I error. No significant differences were observed between the student and community samples on any of the BARQ subscales. Men were found to be significantly higher than women on Aggressive Anger Coping (t = 4.24, p < .001), with an effect size of .45. Women were found to be significantly higher than men on Social Support Seeking (t = -3.38, p < .001), with an effect size of .36. Asian-Canadians scored significantly higher on Avoidant Coping than Caucasian-Canadians (t = -4.67, p < .001), with an effect size of .48. The Relationship between the BARQ and the NEO-FFI In order to determine the extent to which the BARQ subscales overlapped with the subscales of the NEO-FFI, correlations between each of the BARQ subscales and each of the NEO subscales were calculated and are provided in Table 191. Because of the large number of correlation coefficients calculated, an alpha level of .001 was adopted to maintain an overall .05 risk of Type I error. As expected, a high correlation was observed between the BARQ subscale of Rumination and the NEO subscale of Neuroticism (r = .47, p < .001). A high because of some moderate inter-scale correlations between the NEO subscales, a correlation matrix using the NEO factors and the BARQ subscales was also constructed. This correlation matrix was highly similar to the matrix reported in the text, indicating that any correlations existing between the NEO subscales did not influence the results reported. 36 negative correlation existed between the BARQ subscale of Aggressive Anger Out and the NEO subscale of Agreeableness (r = -.51, p < .005). Moderate positive correlations were observed between the BARQ subscale of Diffusion and the NEO subscale of Neuroticism (r = .21, p < .001), the Assertion subscale of the BARQ and the Conscientiousness, Extraversion, and Openness subscales of the NEO (r = .20, p < .001; r = .25, p < .001; r = .21, p < .001 respectively). Moderate negative correlations were found between the BARQ subscale of Rumination and the NEO subscales of Agreeableness and Conscientiousness (r = -.29, p< .001; r = -.23, p < .001), the Aggressive Anger Out subscale of the BARQ and the NEO's Conscientiousness (r = -.21, p < .001), and the BARQ subscale of Assertion and the NEO's Neuroticism (r = -.20, p < .001). A correlation matrix of the BARQ and NEO-FFI subscales was calculated for each of the subgroups (student and community samples; men and women; Asian-Canadians and Caucasian-Canadians). These correlation matrices are provided in Tables 19 to 22. In each of these subgroups, the correlation between Rumination and Neuroticism remained high and significant, as did the negative correlation between Aggressive Anger Out and Agreeableness. The moderate correlation between Rumination and Agreeableness remained stable, but other correlations tended to vary between the subsamples. It should be noted that the correlations that did not remain stable across the subgroups also tended to be quite low (r < .31) and failure to find statistical significance is most likely due to less power associated with the smaller samples used for the subgroup analyses. Discussion There were two objectives to the present study. The primary intent was to provide further evidence of the construct validity of the BARQ by comparing it with a well established measure of major personality dimensions, the NEO-FFI. This approach also allowed exploration of the relationship between personality and anger expression. However, before investigating the relationship between the BARQ and the FFM, it was important to demonstrate the stability of the factor structure of the BARQ to ensure that this principal 37 objective had strong footing. Thus, the secondary objective of this study was to affirm that the BARQ had sound psychometric properties, including a stable factor structure and acceptable levels of reliablity and validity. The Validity of the BARQ The present study provided further evidence that the BARQ is a psychometrically sound measure of anger coping styles. The internal consistencies for most of the BARQ subscales ranged from moderate to high. This held true in the combined sample and in each of the subgroups. The subscale that appeared to have a lower than acceptable level of internal consistency was Diffusion. Throughout the development of the BARQ, Diffusion has persisted as the weakest subscale. Although a previous study (Rutledge & Linden, 1996) found the most recent version of the Diffusion subscale to have an acceptable level of internal consistency, the current results suggest that this subscale may warrant further development or re-conceptualization. There are several possible explanations for the low reliability of the Diffusion subscale. First of all, it may require a high level of psychological sophistication to perceive Diffusion as a form of coping with anger. Consider the subscale item "I simply put more energy into my work or studies". Not all individuals may realize that their increased work or school involvement was an indirect expression of their anger or frustration. Secondly, the Diffusion subscale items refer to behaviours, such as writing poetry, exercising, and organizing the office, that may have a low base rate, at least for some individuals. Both of these conditions may contribute to the low level of internal consistency for the Diffusion subscale. However, with this noted exception, the items within the BARQ subscales showed a high degree of homogeneity. Low intercorrelations between the BARQ subscales were observed, indicating orthogonality of the six anger coping styles. This finding was also reported by Rutledge and Linden (1996). The low intercorrelations suggest very little overlap between the different anger coping styles measured by the BARQ. Thus, although individuals may utilize more than one form of anger coping, the BARQ anger response styles do not appear to regularly co-38 occur within individuals. However, it should be noted that the correlation between Rumination and Aggressive Anger Coping, although not very strong, was consistent across each of the subgroups. This may be a result of the nature of aggressive anger coping; this form of anger expression is the most visible, and verbally or physically aggressive behaviour tends to be stigmatized in North American society. Thus, individuals who rely on aggression as an anger coping technique may end up having mixed feelings about their behaviour, and consequently engage in raminative behaviour after the fact. In the series of studies described by Rutledge and Linden (1996), factor analyses were conducted in order to determine the factor structure of the BARQ. A six factor solution was repeatedly shown to be the most meaningful, both theoretically and empirically. This finding was replicated in the present study. Although there was some overlap of subscale items, the six factors showed definite underlying similarity and clearly represented the subscales of the BARQ. The factor solution obtained in the present study was shown to be highly similar to that reported by Rutledge and Linden, demonstrating the stability of the factor structure of the BARQ. Further evidence for the stability of the factor structure of the BARQ was obtained by conducting factor analyses on the data from each of the subgroups (men and women; students and community members; Asian-Canadians and Caucasian-Canadians). Based on empirical and theoretical criteria outlined previously, a six factor solution was found to provide the clearest solution for each subgroup. In each case, the factors clearly showed underlying similarity and were representative of the subscales of the BARQ. The factor structure of the female and male samples were shown to be highly similar. Comparable results were found when the factor structure of the Asian-Canadian and Caucasian-Canadian samples were compared; a slight difference existed between these subgroups on the factor representing Social Support Seeking. Five of the six factors were largely similar when student and cornmunity members were compared. Only the factor which represented the Avoidant Coping subscale of the BARQ differed between the students and community members. Overall, it can 39 be concluded that the BARQ has a stable factor structure across age groups, gender, and different cultural groups. A six factor solution which can be easily interpreted as representing the six anger coping styles measured by the BARQ has repeatedly been confirmed. Gender and Culture Differences in Anger Coping Styles The data from this study were also used to examine gender differences in anger coping styles. Women showed a greater reliance on social support seeking as a self-reported anger coping mechanism. This finding was in concordance with previous studies, including previous work on the BARQ, which suggested that women tend to discuss their anger experience more often than men (Haynes et al., 1978; Rutledge & Linden, 1998; Thomas, 1991). However, some of this research has also suggested that women are more likely than men to suppress their anger. This study and the results reported by Rutledge and Linden contradict this finding: women did not report using Avoidant Coping more frequently than men. Although Harris (1992) found that men are more likely to report that behaving aggressively would elevate their mood, previous examination of gender differences in anger response styles measured by the BARQ (Rutledge & Linden) did not find this difference. However, in the present study, men were more likely than women to utilize the style of Aggressive Anger Coping. These findings suggest that although women are less likely than men to react aggressively to angering or frustrating situations, they are not necessarily more likely to report suppression of their anger. Asian-Canadians were found to use Avoidant Coping significantly more frequently than Caucasian-Canadians. This finding was in concordance with the observations of Lebra (1984). Lebra stated that in collectivist societies such as Japan, individuals place high value on group solidarity. In order to preserve harmonious relationships, individuals are likely to quietly choose to accept a frustrating situation rather than enter into conflict and become openly angry. This emphasis on acceptance of difficulties and avoidance of confrontation is in stark contrast to the finding that mdividuals from individualist societies, such as the U.S., report more episodes of anger toward known others with whom they share a personal 40 relationship than toward strangers. The findings of the present study confirm that Asian-Canadians report more Avoidant Coping techniques than other individuals in Canada. Presumably these differences would be stronger if comparisons had been made with individuals currently living in a "pure" collectivist society. Additionally, it would be expected that these differences would dissipate as Asian-Canadians become more acculturated within Canadian society (Feldman et al, 1992). Although previous research has indicated that individuals living in North America feel less inhibited in expressing anger toward in-group members than East or Southeast Asians (Matsumoto, 1991), differences were not observed between the Caucasian-Canadians and Asian-Canadians on any of the BARQ subscales measuring outward expression of anger. It is possible that this may be due to an increased level of acculturation in the Asian-Canadian sample used in this study. Anger Coping Styles and the Five Factors of Personality The moderate to good test-retest correlations reported for all BARQ subscales by Rutledge and Linden (1996) indicate that people display a consistent preference for a particular type of anger coping. This suggests that personal dispositions may play an important role in determining individual choice of anger coping response. The most original part of the present study examined the relationship between anger coping styles and personality by comparing the BARQ subscales with the subscales of the NEO-FFI, a well-known and validated measure of the five factors of personality. It was hypothesized that some, but not all, of the anger coping styles measured by the BARQ could be explained as a function of personality. In other words, it was reasoned that the five factors of personality would not adequately capture the full dimensionality of anger coping styles. This expectation was confirmed. Specifically, individuals who were high on Neuroticism were more likely to rely on the anger coping style of Rumination, and the correlation coefficient was sufficiently large to suggest that "ramination" is a proxy for Neuroticism. mdividuals who were low on Agreeableness (i.e., antagonism, which is conceptually similar to hostility) were more likely to utilize the anger coping style of Aggression. The correlations between these variables were 41 strong and persisted when each of the subgroups was examined separately. Other correlations were statistically significant in the combined sample, but were quite low and tended to vary across the subgroups. Avoidant Coping did not correlate significantly with any of the personality dimensions, indicating that its utilization may be unrelated to personality. These findings suggest that the anger coping styles measured by the BARQ cannot be adequately explained as a function of the "Big Five" personality factors. It should also be noted that a highly similar pattern of results was observed when the NEO factors were correlated with the BARQ subscales. The moderate correlations between some of the NEO subscales could have potentially influenced the size of the correlations between the BARQ and NEO subscales; the observation that the essentially orthogonal NEO factors showed a very similar pattern of correlations with the BARQ subscales adds strength to the conclusions of the current study. Past research investigating the role of personality in anger and hostility has implicated Neuroticism and Agreeableness as personality dimensions mediating the experience and expression of anger. Neurotic hostility, characterized by the frequent and intense experience of anger without outward expression, was found to be related to Neuroticism, while antagonistic hostility, described as a stable, hostile and irritable interpersonal orientation often accompanied by verbal and physical aggressive outbursts, was related chiefly to low Agreeableness (Costa et al., 1989). Interestingly, the anger coping style of Rumination, which is described as a very active but internally mediated way of dealing with feelings of anger lacks the outward physical or verbal manifestations of anger and was found to be strongly related to Neuroticism. Likewise, Aggressive Anger Coping, characterized by outward physical or verbal displays of anger, was found to be highly correlated with low Agreeableness. It is possible that Rurnination is similar to Costa et al.'s description of neurotic hostility, while frequent use of Aggressive Anger Coping is a manifestation of antagonistic hostility. Given that Rumination and Aggressive Anger Coping are strongly and consistently related to NEO factors of Neuroticism and Agreeableness, the question arises as to how this relationship should be conceptualized. Buss and Plomin's (1984) theory of temperament 42 suggests that there are some inherited personality traits present in early childhood that are predictive of late development; these traits show continuity or at least have residuals for later personality. Included in these defined temperaments is emotionality, which is depicted as the frequency, duration, and intensity of crying and temper tantrums. Also of importance in the conceptualization of emotionality is the direction of the response: whether the child turns toward the anger-arousing stimulus in anger, or away from the threatening stimulus in fear. Finally, Buss and Plomin emphasize that the strength of the verbal or physical threat required to elicit fear and the degree of frustration or restriction required to elicit anger are important aspects of emotionality that can influence personality development. This theory posits that how young children cope with frustrating or angering situations can influence later personality. If true, it may be the case that children who respond with physical and verbal aggression to threatening or frustrating stimuli are more likely to develop an antagonistic personality, while those who tend to turn inward or respond with fear may be more likely to form a neurotic personality. This would imply that the anger coping styles mtimately related to personality, Rurriination and Aggression, may actually be inherited dispositions that can lead to the development of specific personality types, namely Neuroticism and Antagonism (low Agreeableness). Furthermore, these anger coping styles should persist, at least to some degree, in later life. This appears likely because the NEO factors have been shown to be stable longitudinally (Costa & McCrae, 1980;1988b; Costa & Metter, 1994), which suggests that Rumination and Aggression should endure over the lifespan as well, due to their close relationship with Neuroticism and Agreeableness. Although Buss and Plornin's (1984) theory of temperament provides an interesting perspective on the potential relationship between anger and personality, it is also possible that the opposite is true; that is, personality type is a deterrriinant of anger coping style. This line of thinking is intuitively logical; an individual with a certain "personality" or set of defining traits may be more likely to rely on some coping strategies and patterns of behaviour more than others. If this is true, based on the findings of this study, individuals high on Neuroticism 43 should consistently rely on Rurnination as an anger coping style, and individuals low on Agreeableness should consistently rely on Aggressive Anger Coping. The personality dimension of Neuroticism is associated with nervous tension, irrational tliinking, frustration, depression, self-consciousness, and ineffective coping. It seems quite plausible that individuals with these traits might rely on an anger coping mechanism that is internally mediated yet very active, involving ruminative patterns of replaying the frustrating event. The opposite pole of Agreeableness, that is, antagonism, is described by traits such as skepticism, callousness, stubbornness, and irritability; individuals possessing these traits may be more likely to utilize outward displays of anger as a way of coping with frustration. It should be noted that although these two possibilities seem oppositional, they may not be mutually exclusive. There are a range of anger coping styles and personality dimensions that may be related in varying ways; the possibility that a more complex, interactional relationship between anger coping and personality exists should not be discounted. As mentioned previously, Rutledge and Linden (1996) reported moderately high test-retest correlations for the BARQ subscales. If the observation that individuals display a consistent preference for a particular anger coping style is interpreted as an indication that personal dispositions may play an important role in determining choice of anger coping response, then among the BARQ subscales, Rumination and Aggression should show the highest test-retest correlations. Indeed, this was found to be true. But what about the other BARQ subscales? Individuals who reported relying on other anger coping mechanisms did show some degree of consistency at different points in time. However, the relationship was not as robust, and the correlational analyses revealed that some of other anger coping styles were only moderately related to personality dimensions, while still others appeared to be unrelated. The personality dimension of Agreeableness was found to show a moderate negative correlation with Rumination. This relationship persisted across the combined sample and the subgroups. Hence, individuals who are more antagonistic in nature tend to rely, at least at 44 times, on the anger coping style of Rurrrination. However, this association was not as strong as the relationship between low Agreeableness and Aggressive Anger Coping. This might imply that the most probable response to an angering or frustrating event for a person who is low on Agreeableness is an outward physical or verbal reaction, but at times these individuals may utilize a more internal, mrninative response. This could indicate that although an antagonistic personality is a significant determinant of the use of Rumination, situational variables play an important role in deteirnining when this anger coping response is actually employed. Though other correlations between personality dimensions and anger coping styles emerged as statistically significant, these correlations tended to be quite low and varied substantially across the subgroups, indicating that they are not very stable. For these reasons, it was decided that it would be misleading to interpret these relationships or the differences between the subgroups. However, it is interesting to note that Assertive Anger Coping did appear to be related to the personality dimensions of Conscientiousness, Extraversion, and (low) Neuroticism in the combined sample, Conscientiousness and Extraversion in the female sample, and Extraversion in the Caucasian-Canadian sample. High Conscientiousness, high Extraversion, and low Neuroticism are associated with well-adjusted individuals and have been shown to be related to positive health behaviours (Marshall et al., 1994; Booth-Kewley &Vickers, 1994). Individuals who score high on these dimensions may be more likely, at least in some situations, to rely on Assertion as an anger coping technique. If this association does exist, it may be that Assertive Anger Coping is a "good" form of anger expression that may be linked with positive health outcomes. This possibility will be discussed in the next section. Imphcations for Research on Cardiovascular Disease The experience and expression of anger have emerged over recent years as important psychosocial variables which may have cardiovascular consequences. The most severe of these consequences is, of course, CVD. It has been shown that the physical or verbal expression of anger, rather than the experience of anger, is the more "toxic" factor in the 45 development of CVD (Siegman et al, 1987; Helmig et al., 1991). It has been demonstrated that expression of anger is also a risk factor for women despite the fact that men tend to physically or verbally express their anger more than women (Siegman et al., 1992). Also of interest is the fact that a positive relationship has been observed between the outward expression of anger (when subjects are actively provoked or harassed) and cardiovascular reactivity in the laboratory (Siegman et al., 1992; Suarez and Williams, 1990). In each of these studies no parallel association was observed between the experience of anger/hostility (without outward expression) and cardiovascular reactivity. Not surprisingly, given what is known about the relationship between hostility and personality dimensions, low Agreeableness (antagonism) has been found to be related to CVD and cardiovascular reactivity, while a negative relationship has been found to exist between Neuroticism and CVD and cardiovascular reactivity. It has been postulated that the negative relationship between Neuroticism and CVD is actually artifactual and a consequence of neurotic individuals complaining of somatic pain in the absence of any evidence of CVD, resulting in their selection for coronary angiography (Costa, 1986). However, more recently Seigman (1994) has pointed out that the negative correlations found between Neuroticism and resting BP and HR measurements (Watson & Pennebaker, 1989) also exists between neuroticism and cardiovascular reactivity (Siegman et al.,1991) and consequently cannot be so readily dismissed as artifacts of subject selection. The relationship between these variables remains to be clarified. These findings, in combination with the results of the present study, suggest a potentially complex relationship between anger coping styles and cardiovascular health. In the past, research has focused primarily on the experience and the dichotomous "out" versus "in" expression of anger. The present study suggests that these anger response styles may be parallel to the BARQ anger coping styles of Rumination and Aggressive Anger Coping. If this were true, then it would be expected that Rumination would be negatively correlated with cardiovascular reactivity and possibly be unrelated to risk of CVD, while Aggressive Anger 46 Out would be positively correlated with cardiovascular reactivity and risk of CVD. Alternatively, Rumination and Aggression may be forms of anger coping that are not completely synonymous with the "neurotic hostility" and "antagonistic hostility" described by Costa et al.(1989). At this time, there has not yet been any research investigating the relationship between the BARQ subscales and cardiovascular variables. There are currently studies underway in our laboratory that will help clarify how the BARQ subscales of Rumination and Neuroticism (and/or other subscales) are related to cardiovascular health. In addition to Rumination and Aggressive Anger Out, the BARQ provides measures of four other independent styles of anger coping: Assertion, Social Support Seeking, Diffusion, and Avoidant Coping. It remains to be seen how these anger coping styles are related to CVD. It is quite possible that some of these ways of dealing with anger may be unrelated to CVD or may actually serve a preventative function. Any discussion of how anger coping may be related to blood pressure and cardiovascular health is speculative, as no physiological data were collected in this study. Current theories and recent research may lead to different predictions regarding the nature of the link between anger coping and cardiovascular variables. The Social Conflict Model proposed by Linden and Lamensdorf (1990) proposes that meaningful styles of anger expression lie on a continuum between aggression and passivity. In this model, the relationship between anger expression and blood pressure is described as a U-shaped curve. Individuals with extreme anger-out tendencies as well as those exhibiting excessive anger-in behaviour are predicted to have the highest blood pressures, while those falling between the extremes are predicted to have the lowest blood pressures. These expectations were confirmed by Linden and Lamensdorf (1990). Siegman's (1989; 1994) research suggests an alternative possibility. Given the high correlation between Neuroticism and Rumination, and Siegman's finding that Neuroticism and "neurotic hostility" were unrelated to cardiovascular health, it is possible that no meaningful relationship between Rumination and cardiovascular health exists. A similar line of reasoning would lead to the prediction that a strong positive relationship exists between Aggressive Anger Coping and 47 cardiovascular health. The relationship between the range of anger coping styles measured by the BARQ and cardiovascular variables, such as blood pressure, reactivity, and risk of CVD, will be examined in the near future in an effort to clear these muddy waters. In the present study, Assertive Anger Coping was found to be moderately correlated with Conscientiousness, Extraversion, and low Neuroticism in the combined sample and in some of the subgroups. Some research has demonstrated that these personality dimensions were related to better adjustment and positive health behaviours (Marshall et al., 1994; Booth-Kewley and Vickers, 1994). This may suggest that Assertive Anger Coping may be associated with lower CVD severity. If this is the case, training individuals to deal with their anger assertively might decrease the risk of CVD. The finding that Assertive Anger Coping is only moderately related to personality would make the possibility of'retraining'' individuals to cope with anger and frustration more feasible. These issues can be addressed in future research examining the relationship between anger and cardiovascular health. Conclusions There is an important limitation to this study that deserves mention. All relationships examined in this study were self-reports, which have inherent limits. We do not yet know how BARQ self-reports relate to actual behaviour. One way of assessing this is through peer validation and this project is currently in the planning stages. However, the current study has shown that the BARQ is a comprehensive measurement tool that captures the multiple dimensions of anger coping and that BARQ scores correlate closely and the predicted directions with other measures that have undergone behavioural validation. It provides a measure of six independent anger coping styles. Although some anger coping styles are rooted in personality, the BARQ captures a wider range of response styles that are less a function of personality and are likely determined by situational factors as well. It is believed that the BARQ can facilitate the study of the role of anger coping in CVD. 48 References Alexander, F. (1939). Emotional factors in essential hypertension. Psychosomatic Medicine. 1. 175-179. Arnold, M . B. (1960) Emotion and Personality. New York: Columbia University Press. Ben-Zur, H. & Breznitz, S. (1991). What makes people angry: Dimensions of anger-provoking events. Journal of Research in Personality. 25, 1-22. Block, J. (1995). A contrarian view of the five-factor approach to personality description. Psychological Bulletin. 117. 187-215. Bolger, N . (1990). Coping as a personality process: A prospective study. Journal of Personality and Social Psychology. 59. 525-537. Booth-Kewley, S. & Vickers, R. R. Jr. (1994). Associations between major domains of pesonality and health behavior. Journal of Personality. 62, 281-298. Buss, A. (1961) The Psychology of Aggression. New York: Wiley. Buss, A. & Plomin, R. (1984). Temperament: Early Developing Personality Traits. : London: Lawrence Erlbaum Associates. Cattell, R. B. (1946). The description and measurement of personality. Yonkers, NY: World Book. Cattell, R. B. (1947). Confirmation and clarification of primary per sonahty factors. Psychometrika. 12. 197-220. Cattell, R. B. (1948). The primary personality factors in women compared with those in men. British Journal of Psychology, 1. 114-130. Cartel, R. B., Eber, H. W., & Tatsuaka, M . M. (1970). The handbook for the Sixteen Personality Factor Ouestionnairre. Champaign, JL: Institute for Personality and Ability Testing. Cattell, R. B. (1973). Personality and mood by questionnaire. San Francisco: Jossey-Bass. 4 9 Chesney, M . & Rosenhan, R. (Eds.) (1985). Anger and Hostility in Cardiovascular and Behavioral Disorder. New York: McGraw Hill. Comrey, A. L. (1970). Manual for the Comrey Personality Scales. San Diego, CA: EdITS. Costa, P. T., Jr. (1986). Is Neuroticism a risk factor for CAD? Is Type A a measure of Neuroticism? In T. Schmidt, T. Dembroski, & G. Blumchen (Eds.), Biological and psychological factors in cardiovascular disease (pp. 85-95). New York: Springer-Verlag. Costa, P. T., Jr. & McCrae, R. R. (1976). Age differences in personality structure: A cluster analytic approach. Journal of Gerontology. 31. 564-570. Costa, P. T, Jr. & McCrae, R. R. (1980). Still stable after all these years: Personality as a key to some issues in adulthood and old age. In P. B. Baltes & O. G. Brim, Jr. (Eds.), Life Span Development and Behavior (Vol. 3, pp.65-102). New York: Academic Press. Costa, P. T., Jr. & McCrae, R. R. (1985). The NEO Personality Inventory manual. Odessa, FL: Psychological Assessment Resources. Costa, P. T., Jr. & McCrae, R. T. (1987). Neuroticism, somatic complaints, and disease: Is the bark worse than the bite? Journal of Personality. 55. 301-316. Costa, P. T., Jr. & McCrae, R. R. (1988a). From catalog to classification: Murray's needs and the five-factor model. Journal of Personality and Social Psychology. 55. 258-265. Costa, P. T., Jr. & McCrae, R. R. (1988b). Personality in adulthood: A six year study of self reports and spouse ratings on the NEO Personality Inventory. Journal of Personality and Social Psychology, 54. 853-866. Costa, P.T., Jr. & McCrae, R. R. (1990). Personality: Another hidden factor in stress research. Psychological mquiry. 1, 22-24. Costa, P. T., Jr. & McCrae, R. R. (1991). Facet scales for Agreeableness an d Digman, J. M. (1990). Personality structure: Emergence of the five-factor model. Annual Review of Psychology. 41. 417-440. 5 0 Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO PI-RTM) and NEO Five-Factor Inventory (NEOFFI) professional manual. Odessa, FL: Psychological Assessment Resources. Costa, P. T., Jr. & McCrae, R. R. (1995). Solid ground in the wetlands of personality: A reply to Block. Psychological Bulletin. 117. 216-220. Costa, P. T., Jr., McCrae, R. R., & Dye, D. A. (1991). Facet scales for Agreeableness and Conscientiousness: A revision of the NEO Personality Inventory. Personality and Individual Differences. 12. 887-898. Costa, P. T., Jr., McCrae, R. R., & Dembroski, T. M . (1989). Agreeableness versus Antagonism: Explication of a potential risk factor for CHD. In A. Sieman & T. M. Dembroski (Eds.), In Search of Coronary-Prone Behaviour (pp. 41-63). BGllsdale, NJ: Erlbaum. Costa, P. T., Jr. & Metter, E. J. (1994). Personality stability and its contribution to successful aging. Journal of Geriatric Psychology, 27, 41-59. Dembroski, T. M. , MacDougall, J. M. , Costa, P. T., Jr., & Grandits, G. A. (1989). Components of hostility as predictors of sudden death and myocardial infarction in the Multiple Risk Factor Intervention Trial. Psychosomatic Medicine, 47. 219-233. Diamond, E. (1982). The role of anger and hostility in essential hypertension and coronary heart disease. Psychological Review. 92. 410-433. Digman, J. M . (1996). The curious history of the five-factor model. In J. S. Wiggin (Ed.) The Five-Factor Model of Personality: Theoretical Perspectives. New York: The Gnxilford Press. Digman, J. M. & Inouye, J. (1986). Further specification of the five robust factors of personality. Journal of Personality and Social Psychology. 50. 116-123. Digman, J. M. , & Takemoto-Chock, N. K. (1981). Factors in the natural language of personality: Re-analysis, comparison, and interpretation of six major studies. Multivariate Behavioral Research. 16. 149-170. 51 Dujovne, V. F. & Houston, B. K. (1991). Hostility-related variables and plasma lipid levels. Journal of Behavioral Medicine. 14. 555-565. Endler, N. & Parker, J. D. A. (1990). Multidimensional assessment of coping: A critical evaluation. Journal of Personality and Social Psychology. 58. 844-854. Eysenck, H. J. (1947). Dimensions of Personality. London: Routledge & Kegan Paul. Eysenck, H. J. (1955). Psychiatric diagnosis as a psychological and statistical problem. Psychological Reports. 1, 3-17. Eysenck, H. J., & Eysenck, S. B. G. (1964). Manual of the Eysenck Personality Inventory. London: University Press. Eysenck, H. J., & Eysenck, S. B. G. (1968). Manual of the Eysenck Personality Inventory. Sand Diego, CA: Educational and Industrial Testing Service. Eysenck, H. J., & Eysenck, S. B. G. (1975). Manual of the Eysenck Personality Questionnaire. San Diego: EdiTS. Fahrenberg, J., Hampel, R., & Selg, H. (1984). Das Freiberger Personhchkeitsinventar (FPI-R) [The Frieberg Personality Inventory]. Gottingern, FRG: Hogrefe. Feldman, S. S., Mont-Reynaud, R., & Rosenthal, D. A. (1992). When east moves west: The acculturation of values of Chinese adolescents in the U.S. and Australia. Journal of Research on Adolescence, 2, 147-173. Felsten, G. (1996). Five-factor analysis of Buss-Durkee Hostility Inventory neurotic hostility and expressive hostility factors: Implications for health psychology. Journal of Personality Assessment. 67. 179-194. Felsten, G , & Leitten, C. L. (1993). Expressive, but not neurotic hostility is related to cardiovascular reactivity during a hostile competitive task. Personality and Individual Differences. 14. 805-813. Fiske, D. W. (1949). Consistency of the factorial structures of personality ratings from different sources. Journal of Abnormal and Social Psychology. 44. 329-344. 52 Fleishman, J. A. (1984). Personality characteristics and coping patterns. Journal of Health and Social Behavior. 25. 229-244. Folkman, S. & Lazarus, R. S. (1980). An analysis of coping in a middle-aged community sample. Journal of Health and Social Behavior. 21. 219-239. Friedman, H. & Booth-Kewley, S. (1987). Personality, Type A behavior, and coronary heart disease: The role of emotional expression. Journal of Personality and Social Psychology. 48, 1299-1315. Friedman, H. S., Tucker, J. S., &Reise, S. P. (1995). Personality dimensions and measures potentially relevant to health: A focus on hostility. Annals of Behavioral Medicine. 17. 245-253. Gentry, W., Chesney, A., Gary, H., Harburg, E. (1982). Habitual anger-coping styles. I. Effect on mean blood pressure and risk for essential hypertension. Psychosomatic Medicine. 44, 195-201. Goldberg, L. R. (1990). An alternative "description of personality": the Big-Five factor structure. Journal of Personality and Social Psychology. 59. 1216-1229. Gough, H. G. (1987). California Psychological Inventory Administrator's Guide. Palo Alto, CA: Consulting Psychologists Press. Hakstian, A. R., Rogers, W. T., & Cattell, R. B. (1982). The behaviour of number-of-factors rules with simulated data. Multivariate Behavioral Research, 17. 193-219. Hakstian, A. R., Suedfeld, P., Ballard, E. J., & Rank, D. S. (1986). The Ascription of Responsibility Cniestionnaire: Development and empirical extensions. Journal of Personality Assessment. 50. 229-247. Harman, H. H. (1976). Modern Factor Analysis. University of Chicago Press: Chicago. Harris, M. (1992a). Sex and ethnic differences in past aggressive behaviour. Journal of Family Violence. 7. 85-102. 53 Harris, M . (1992b). Beliefs about how to reduce anger. Psychological Reports, 70, 203-210. Haynes, S. G., Levine, S., Scotch, N. , Feinleib, M. , & Kannel, W. B. (1978). The relationship between psychosocial factors to coronary heart disease in the Framingham study: 1. Methods and risk factors. American Journal of Epidemiology. 107, 362-383. Helmig, L., Houston, B. K., Vavak, C. R., & Muffin, J. (1991). Hostility related variables, self-schemata and CHD. Paper presented at the meetings of the Society of Behavioural Medicine, Washington, DC. Hogan, R. (1986). Hogan Personality Inventory manual. Minneapolis: National Computer Systems. Hokanson, J., Willers, K., & Koropsak, E. (1968). The modification of autonomic responses during aggressive interchange. Journal of Personality. 36. 386-404. Hooker, K., Frazier, L. D., & Monahan, D. J. (1994). Personality and coping among caregivers of spouses with dementia. The Gerontologist. 34, 386-392. Jackson, D. N. (1984). Personality Research Form manual (3 r d ed.). Port Huron, MI: Research Psychologists Press. John, O. P., Angleitner, A., & Ostendorf, F. (1988). The lexical approach to personality: A historical review of trait taxonomic research. European Journal of Personality. 2, 171-203. Johnson, J. A. & Ostendorf, F. (1993). Clarification of the Five-Factor Model with the abridged Big Five dimensional circumplex. Journal of Personality and Social Psychology. 65. 563-576. Jorgensen, R.S., Johnson, B. T., Kolodziej, M . E., & Schreer, G. E. (1996). Elevated blood pressure and personality: A meta-analytic review. Psychological Bulletion. 120. 293-320. 54 Julius, M. , Harburg, E., Cottington, E., & Johnson, E. (1986). Anger-coping types, blood pressure, and all-cause mortality: A follow-up in Tecumseh, Michigan. American Journal of Epidemiology. 124. 220-233. Krohne, H. W. (1996). Individual differences in coping. In M. Zeidner & N. S. Endler (Eds.) Handbook of Coping: Theory. Research. Applications. New York: John Wiley & Sons, Inc. Lai, J. Y. & Linden, W. (1992). Gender, anger expression style, and opporiiinity for anger release determine cardiovascular reaction to and recovery from anger provocation. Psychosomatic Medicine. 54. 297-310. Lebra, T. S. (1984). Nonconfrontational strategies for management of interpersonal conflicts. In E. S. Kraus, T. P. Rohen, & P. G. Steinhoff(Eds), Conflict in Japan. Honolulu: University of Hawaii Press. Linden, W. (1987). On the impending death of the Type A construct: Or is there a phoenix rising from the ashes? Canadian Journal of Behavioural Science. 19. 177-190. Linden, W. &Feuerstein, M . (1981). Essential hypertension and social coping behavior. Journal of Human Stress. 7. 28-34. Linden, W. & Lamensdorf, L. (1990). Hostile affect and casual blood pressure. Psychology and Health. 4. 343-349. Linden, W., Leung, D., Chawla, A., Stossel, C , Rutledge, T., & Tanco. S. (1997). Social determinants of experienced anger. Journal of Behavioral Medicine. 20, 415-432. Linden, W., Rutledge, T., Earle, T. L, Lenz, J. W., & Chawla, A. (1998). There is more to anger coping than "in" and "out". Manuscript in preparation. Mann, L , Radford, M. & Kanagawa, C. (1985). Cross-cultural differences in children's use of decision rules: A comparison between Japan and Australia. Journal of Personality and Social Psychology. 49. 1557-1564. 55 Marshall G. N. , Wortman, C. B., Vickers, R. R., Jr., Kusulas, J. W., & Hervig, L. K. (1994). The Five-Factor Model of Personality as a framework for personality-health research. Journal of Personality and Social Psychology, 67, 278-286. Matsumoto, D. (1991). Cultural influences on facial expressions of emotion. Southern Communication Journal. 56. 128-137 Matsumoto, D. (1996). Unmasking Japan: Myths and Realities about the Emotions of the Japanese. Stanford: Stanford University Press. McCrae, R. R. & Costa, P. T., Jr. (1982). Consensual validation of personality traits: Evidence from self-reports and ratings. Journal of Personality and Social Psychology. 43, 293-303. McCrae, R. R. & Costa, P. T., Jr. (1985a). Openness to experience. In R. Hogan & W. H. Jones (Eds.) Perspectives in Personality (Vol. 1, pp. 145-172). Greenwich, CT: JAI Press. McCrae, R. R. & Costa, P. T., Jr. (1985b). Updating Norman's "adequate taxonomy": Intelligence and personality dimensions in natural language and in questionnaires. Journal of Personality and Social Psychology. 49. 710-721. McCrae, R. R. & Costa, P. T., Jr. (1986). Personality, coping, and coping effectiveness in an adult sample. Journal of Personality, 54. 385-405. McCrae, R. R. & Costa, P. T., Jr. (1987). Validation of the five-factor model across instruments and observers. Journal of Personality and Social Psychology. 52, 81-90. McCrae, R. R. & Costa, P. T., Jr. (1989). The structure of interpersonal traits: Wiggins' circumplex and the five-factor model. Journal of Personality and Social Psychology. 56, 586-595. McCrae, R. R. & Costa, P. T., Jr. (1997). Personality Trait Structure as a Human Universal. American Psychologist. 52. 509-516. McCrae, R. R, & John, O. P. (1992). An introduction to the five-factor model and its applications. Journal of Personality. 60. 175-215. 56 Mershon, B., & Gorsuch, R. L. (1988). Number of factors in the personality sphere: Does increase in factors increase predictabiHty of real-life criteria? Journal of Personality and Social Psychology. 55, 675-680. Miller, T. Q., Jenkins, C. D., Kaplan, G. A., & Salonen, J. T. (1995). Are an hostility scales alike? Factor structure and covariation among measures of hostility. Journal of Applied Social Psychology. 25. 1142-1168. Murray, H. A. (1938). Explorations in Personality. New York: Oxford University Press. Noller, P., Law, H., & Comrey, A. L. (1987). Catell, Comrey, and Eysenck personality factors compared: More evidence for the five robust factors? Journal of Personality and Social Psychology. 53. 775-782. Norman, W. T. (1963). Toward an adequate taxonomy of personality attributes: Replicated factor structure in peer nominaton personality ratings. Journal of Abnormal and Social Psychology. 66. 574-583. Novaco, R. W. (1975) Anger Control. Lexington, Mass.: Heath. O'Brien, T. B. & DeLongis, A. (1996). The interactional context of problem-, emotion-, and relationship-focussed coping: The role of the big five personality factors. Journal of Personality. 64. 775-813. Ostendorf, F. (1990). Sprache mid Personfichkeitsstruktur: Zur Validitat des Funf-Faktoren-Modells der Personlichkeit [Language and personality structure: Toward the validation of the five-factor model of personality]. Regensburg, Germany: S. Roeder Verlag. Ostendorf F. & Angleitner, A. (1992). On the generality and comprehensiveness of the five-factor model of personality: Evidence for five robust factors in questionnaire data. Modern Personality Psychology. New York: Harvester Wheatsheaf. Paulhus, D. L. (1991). BIDR Reference manual for Version 6. Unpublished manuscript, University of British Columbia. 57 Peabody, D., & Goldberg, L. R. (1989). Some determinants of factor structures from personality-trait descriptors. Journal of Personality and Social Psychology. 57. 552-567. Pearlin, L. I. & Schooler, C. (1978). The structure of coping. Journal of Health and Social Behavior. 19. 2-21. Plutchik, R. (1980) Emotion: Apsychoevolutionary synthesis. New York: Harper & Row. Riley, W. T. & Treiber, F. A. (1989). The vahdity of multidimensional self-report anger and hostility measures. Journal of Clinical Psychology. 45. 397-404. Rosenman, R. (1978). The interview method of assessment of the coronary-prone behaviour pattern. In T. M . Dembroski, S. M. Weiss, J. L. Shields, S. G. Haynes, & M . Feinleib (Eds.), Biobehavioural bases of coronary heart disease. New York: Karger. Rutledge, T., & Linden, W. (1996). Development and validation of a new tool for the measurement of anger coping styles. Paper presented at the Annual Meeting of the Society of Behavioral Medicine, Washington, DC. Saul, L. J. (1976). The Psychodynamics of Hostility. New York: Jason Aronson. Schneider, R. H., Egan, B. M. , Johnson, E. H., Drabny, H., & Julius, S. Anger and anxiety in borderline hypertension. Psychosomatic Medicine. 48. 242-248. Shekelle, R. B., Gale, M. , Ostfeld, A. M. , & Paul, O. (1983). Hostility, risk of coronary heart disease, and mortality. Psychosomatic Medicine. 45. 109-114. Siegman, A. W. (1994). Cardiovascular consequences of expressing and repressing anger. In A.W. Siegman & T. W. Smith (Eds.), Anger. Hostility, and the Heart (pp. 173-197). Hillsdale NJ: Lawrence Erlbaum Associates, Inc. Siegman, A. W. (1989). The role of hostility, neuroticism, and speech style in Coronary-Artery Disease. In A. W. Siegman & T. M. Dembroski (Eds.) In Search of Coronary-Prone Behaviour: Beyond Type A (pp.41-63). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. 58 Siegman, A. W., Anderson, R. W., & Boyle, S. (1991). Repression, impression management, trait anxiety, and cardiovascular reactivity in men and women. Paper presented at the annual meeting of the Society for Behavioural Medicine, Washington, DC. Siegman, A. W., Anderson, R. A , Herbst, J., Boyle, S., & Wilkinson, J. (1992). Dimensions of anger-hostility and cardiovascular reactivity in provoked and angered men. Journal of Behavioral Medicine. 15. 257-272. Siegman, A. W., Dembroski, T. M. , & Ringel, N. (1987). Components of hostility and the severity of coronary heart disease. Psychosomatic Medicine. 49. 127-136. Spearman, C. (1927). The Abilities of Man. New York: Macmillan. Spielberger, C. D., Jacobs, G. A , Russell, S. F., & Crane, R. J. (1983). Assessment of anger: The State-Trait Anger Scale. In J. Butcher & C. D. Spielberger (Eds.), Advances in Personality Assessment (pp. 161-189). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Spielberger, C. D., Johnson, E. H , Russell, S. F., & Crane, R. J. (1985). The expression and experience of anger: Construction and validation of an anger expression scale. In Chesney, M . & Rosenhan, R. (Eds.) (pp.5-30). Anger and hostility in Cardiovascular and Behavioral Disorder. New York: McGraw Hill. Suarez, E. C , & Williams, R. B. (1990). The relationships between dimensions of hostility and cardiovascular reactivity as a function of task characteristics. Psychosomatic Medicine. 52. 558-570. Suinn, R. M. , Richard-Figueroa, K., Lew, S., & Vigil, P. (1987). The Suinn-Lew Asian Self-Identity Acculturation Scale: An initial report. Educational and Psychological Measurements. 47, 401-407. Thomas, S. P. (1991). Toward a new conceptualization of women's anger. Mental Health Nursing. 12. 31-49. Trapnell, P. D., & Wiggins, J. S. (1990). Extension of the Interpersonal Adjective Scales to include the big five dimensions of personality (IASR-B5). Journal of Personality and Social Psychology. 59. 781-790. 59 Triandis, H. C , Bontempo, R., Betancour, H., Bond, M. , Leung, K., Brenes, A., Georgas, J., Hui, H. C , Marin, G , Setiadi, B., Sinha, J. B. P., Verma, J., Spangenberg, I , Touzard, H., & de Montmoltin, G (1986). Individualism and collectivism across cultures. Australian Journal of Psychology. 38. 25-267. Watson, D. A., & Pennebaker, J. W. (1989). Health complaints, stress and distress. Exporing the central role of negative affectivity. Psychological Review. 96. 239-254. Yang, K , & Bond, M. H. (1990). Exploring implicit personality theories with indigenous or imported constructs: The Chinese case. Journal of Personality and Social Psychology. 58. 1087-1095. Zuckerman, M. , Kuhlman, D. M. , & Camac, C. (1988). What lies beyond E and N? Factor analyses of scales believed to measure basic dimensions of personality. Journal of Personality and Social Psychology. 54. 96-107. C O O o 3. P 3 5 3 CO c T3 O CO CD CD B" OQ O O o £l p B O o 5' OQ to N O <1 ON •o O O tO O O oo oo -0 to oo to oo O N 1-4 3 O N > OQ CD <*> <' CD CD o -j O N 1^1 oo o O N <J1 O N O N O N o O N oo -o N O 1^1 > CO CD g-5' 3 Cd O O to O O O O O N oo 1>J O O O N CO o O N O O 4^  O N to to O N II o 0 0 & ° 3 2 CD g " CD CD • 5 CO CO to cT 3. CD 3 O 3 CD 3 o > P V3 Pp" 3 P s 3 3 P o 3^  B g C/5 3 to 3" P r iCD < CD 3 H CD o CD CD Cd CO 3 o CD* 3* iCO p I CD a. > o CO I * CO CO CD O A CD O o 5' o CTO CO c T3 O a. 3 3 3 5 3 o > O < o 5' ol CTO P 3 5' 3 c/i CD g. o ' 3 OS * O o ON NO * os * £ > 3 cro cro cro -n CO _ tn O £ < ' o to to * o > > 3 cro cro erg CD S *-t CD _ </i O 22 £ < > CD a 5" 3 o 3 o > o < T3 o 5' ol cro P 3 3 3 O 3 CO CO CD o CD o FT siL" 5' cro CO c T3 •a o 3-CO p 3 a 1 Hi o 1 A SI i-I a 3 o n 0* CD rt fD fD 3 o o a ST G O oo fD O fD O 3 — CTQ G O 3 O 3. 3 5 si-s' 3 o > o < 12 2 5' ol CTQ g 3 a 5' 3 cn fD a 5" 3 3 O Q " fD O £ £. < o O J > > 3 cro CTQ erg >-t CD O 22 rt CD > cn cn fD 3. in o 3 te fD fD O 3 cn A o o o o oo d cn 5' 3 3-fD o 0 0 c o O S * N J OO * N J O O o oo o 0ft o os O J * to N O * 4^  O n * o > o < "2 2 5' 5l crq » 3 3 3 o 3 o fT cn 3> g* oo rt c a . fD £ 0 0 I fD* O J N J O S N J O N O O J o * N J N J O O O oo fD o fD o SC 5' CTQ 0 0 3 T J T3 O a N J O J N J o o 3 3 3 3 oo I O S oo Table 6 ULS Factor Analysis with Varimax Rotation of BARO Data from the Combined Samnle Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 BARQ01 .53 BARQ02 .44 -.30 BARQ03 .35 BARQ04 .65 BARQ05 .44 BARQ06 .68 BARQ07 .48 BARQ08 .66 BARQ09 .58 BARQ 10 .67 BARQ 11 -.35 .39 BARQ 12 .40 .54 BARQ 13 .67 BARQ 14 .73 BARQ 15 .71 BARQ 16 .65 BARQ 17 .50 BARQ 18 .53 BARQ 19 .37 BARQ20 .56 BARQ21 .51 BARQ22 .63 BARQ23 .58 BARQ24 .54 BARQ25 .56 BARQ26 .74 BARQ27 .73 BARQ28 .66 BARQ29 .55 BARQ30 .46 .51 BARQ31 .65 BARQ32 .70 BARQ33 .67 BARQ34 .68 BARQ35 .60 BARQ36 .73 BARQ37 .74 Note: loadings with absolute value < .30 are not shown. 66 Table 7 ULS Factor Analysis with Varimax Rotation of BARQ Data from Rutledge & Linden (1996) Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 BARQ01 .53 BARQ02 .38 -.38 BARQ03 BARQ04 .63 BARQ05 .59 BARQ06 .61 BARQ07 .32 .32 BARQ08 .57 BARQ09 .51 BARQ 10 .52 BARQ 11 -.37 .31 BARQ 12 .66 BARQ 13 .65 BARQ 14 .78 BARQ 15 .70 BARQ 16 .43 BARQ 17 .49 BARQ 18 .46 BARQ 19 BARQ20 .57 BARQ21 .64 BARQ22 .67 BARQ23 .68 BARQ24 .58 BARQ25 .60 BARQ26 .69 BARQ27 .81 BARQ28 .72 BARQ29 .52 BARQ30 .43 .38 BARQ31 .51 BARQ32 .70 BARQ33 .74 BARQ34 .33 .50 BARQ35 .56 BARQ36 .64 BARQ37 .59 Note: loadings with absolute value < .30 are not shown. Table 8 Coefficients of Congruence for Comparison of BARQ Factors from the Present Study with BARQ factors from Rutledge & Linden (1996) Rutledge & Coefficient of Anger Coping Style Present Study Linden (1996) Congruence Aggressive Anger Out Factor 2 Factor 2 .96 Assertion Factor 1 Factor 3 .95 Diffusion Factor 6 Factor 6 .86 Avoidant Coping Factor 5 Factor 4 .92 Rumination Factor 4 Factor 5 .92 Social Support Seeking Factor 3 Factor 3 .92 68 Table 9 ULS Factor Analysis with Varimax Rotation of BARQ Data from Student Sample Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 BARQ01 .56 BARQ02 .47 -.38 BARQ03 .38 BARQ04 .67 BARQ05 .39 BARQ06 .69 BARQ07 .41 BARQ08 .64 BARQ09 .61 BARQ 10 .73 BARQ 11 -.33 .42 BARQ 12 .40 .52 BARQ 13 .69 BARQ14 .70 BARQ 15 .72 BARQ 16 .65 BARQ 17 .52 BARQ 18 .46 BARQ 19 .38 BARQ20 .49 BARQ21 .49 BARQ22 .65 BARQ23 .62 BARQ24 .54 BARQ25 .60 BARQ26 .77 BARQ27 .75 BARQ28 .68 BARQ29 .57 BARQ30 .52 .43 BARQ31 .65 BARQ32 .81 BARQ33 .69 BARQ34 .73 BARQ35 .61 BARQ36 .77 BARQ37 .77 Note: loadings with absolute value < .30 are not shown. Table 10 ULS Factor Analysis Varimax Rotation of BARQ Data from Community Sample Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 BARQ01 .60 BARQ02 .46 BARQ03 .31 BARQ04 .51 BARQ05 .56 BARQ06 .71 BARQ07 .55 BARQ08 .68 BARQ09 .52 BARQ 10 .55 BARQ 11 -.40 BARQ 12 .68 BARQ 13 .64 BARQ 14 .69 BARQ 15 .67 BARQ 16 .62 BARQ 17 .40 BARQ 18 .52 BARQ 19 .32 BARQ20 .66 BARQ21 .53 BARQ22 .62 BARQ23 .56 BARQ24 .54 BARQ25 .52 BARQ26 .69 BARQ27 .68 BARQ28 .63 BARQ29 .62 BARQ30 .57 .41 BARQ31 .65 BARQ32 .54 BARQ33 .66 BARQ34 .63 BARQ35 .59 .31 BARQ36 .68 BARQ37 .67 Note: loadings with absolute value < .30 are not shown. Table 11 Coefficients of Congruence for Comparison of BARQ Factors from the Student and Community Samples Anger Coping Style Student Sample Community Sample Coefficient of Congruence Aggressive Anger Out Factor 4 Factor 1 .90 Assertion Factor 1 Factor 2 .95 Diffusion Factor 6 Factor 6 .90 Avoidant Coping Factor 5 Factor 5 .62 Rumination Factor 3 Factor 4 .94 Social Support Seeking Factor 2 Factor 3 .92 Table 12 ULS Factor Analysis Varimax Rotation of BARQ Data from Asian-Canadian Sample Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 BARQ01 .50 BARQ02 .52 BARQ03 .54 BARQ04 .72 BARQ05 .46 BARQ06 .66 BARQ07 .51 BARQ08 .74 BARQ09 .59 BARQ 10 .66 BARQ 11 -.30 .43 BARQ 12 .44 .54 BARQ 13 .68 BARQ 14 .77 BARQ 15 .74 BARQ 16 .56 BARQ 17 .44 BARQ 18 .59 BARQ 19 .38 BARQ20 .56 BARQ21 .44 BARQ22 .649 BARQ23 .63 BARQ24 .54 BARQ25 .61 BARQ26 .78 BARQ27 .77 BARQ28 .62 BARQ29 .52 BARQ30 .41 .59 BARQ31 .66 BARQ32 .73 BARQ33 .72 BARQ34 .68 BARQ35 .56 .30 BARQ36 .69 BARQ37 .78 Note: loadings with absolute value < .30 are not shown. 72 Table 13 ULS Factor Analysis Varimax Rotation of BARQ Data from Caucasian-Canadian Sample Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 BARQ01 .64 BARQ02 .35 .39 BARQ03 BARQ04 .42 BARQ05 .46 BARQ06 .71 BARQ07 .42 BARQ08 -.31 .50 BARQ09 .66 BARQ 10 .71 BARQ 11 -.40 BARQ 12 .60 BARQ 13 .65 BARQ 14 .63 BARQ 15 .64 BARQ 16 .74 BARQ 17 .39 BARQ 18 .33 BARQ 19 .32 BARQ20 .55 BARQ21 .62 BARQ22 .60 BARQ23 .53 BARQ24 .53 BARQ25 .51 BARQ26 .71 BARQ27 .65 BARQ28 .69 BARQ29 .64 BARQ30 .51 .31 BARQ31 -.35 .52 BARQ32 .77 BARQ33 .58 .33 BARQ34 .70 BARQ35 .63 BARQ36 .72 BARQ37 .67 Note: loadings with absolute value < .30 are not shown. Table 16. 73 Table 14 Coefficients of Congruence for Comparison of BARQ Factors from the Asian-Canadian and Caucasian-Canadian Samples Asian-Canadian Causacian- Coefficient of Anger Coping Style Sample Canadian Sample Congruence Aggressive Anger Out Factor 3 Assertion Factor 1 Diffusion Factor 6 Avoidant Coping Factor 5 Rumination Factor 4 Social Support Seeking Factor 2 Factor 2 Factor 1 Factor 6 Factor 5 Factor 4 Factor 3 .90 .91 .80 .87 .90 .79 Table 15 ULS Factor Analysis with Varimax Rotation of BARQ Data from the Male Sample Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 BARQ01 .57 BARQ02 .31 .44 BARQ03 BARQ04 .47 BARQ05 .46 BARQ06 .65 BARQ07 .45 BARQ08 .68 BARQ09 .45 BARQ 10 .58 BARQ 11 -.31 .38 BARQ 12 .613 BARQ 13 .53 .328 BARQ 14 .74 BARQ 15 .73 BARQ 16 .62 BARQ 17 .44 BARQ 18 .50 BARQ 19 .46 BARQ20 .54 BARQ21 .53 BARQ22 .64 BARQ23 .32 .49 BARQ24 .57 BARQ25 .58 BARQ26 .66 BARQ27 .78 BARQ28 .69 BARQ29 .65 BARQ30 .41 .67 BARQ31 .70 BARQ32 .63 BARQ33 .71 BARQ34 .62 BARQ35 .63 BARQ36 .68 BARQ37 .68 Note: loadings with absolute value < .30 are not shown. Table 16 ULS Factor Analysis with Varimax Rotation of BARQ Data from the Female Sample Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 BARQ01 .51 BARQ02 .51 -.37 BARQ03 . 3 9 BARQ04 .68 BARQ05 .42 BARQ06 .70 BARQ07 .51 BARQ08 -.30 .69 BARQ09 .64 BARQ 10 .72 BARQ11 -.40 .39 BARQ 12 .42 .50 BARQ 13 .71 BARQ 14 .66 BARQ 15 .69 BARQ 16 .67 BARQ 17 .51 BARQ 18 .560 BARQ 19 BARQ20 .51 BARQ21 .55 BARQ22 .64 BARQ23 .62 BARQ24 .49 BARQ25 .56 BARQ26 .81 BARQ27 .68 BARQ28 .64 BARQ29 .44 BARQ30 .52 .47 BARQ31 .61 BARQ32 .72 BARQ33 .64 .36 BARQ34 .72 BARQ35 .66 BARQ36 .75 BARQ37 .77 Note: loadings with absolute value < .30 are not shown. 76 Table 17 Coefficients of Congruence for Comparison of BARQ Factors from the Male and Female Samples Anger Coping Style Men Sample Women Sample Coefficient of Congruence Aggressive Anger Out Factor 2 Factor 2 .91 Assertion Factor 1 Factor 1 .95 Diffusion Factor 6 Factor 6 .82 Avoidant Coping Factor 5 Factor 5 .90 Rumination Factor 4 Factor 4 .93 Social Support Seeking Factor 3 Factor 3 .92 77 Table 18 Means. Standard Deviations, and t-test Results for Student and Community Samples. Men and Women, and Asian-Canadians and Caucasian-Canadians Student Community BARO Subscales Mean SD Mean SD df t value p Aggressive Anger Out 16.76 4.96 16.90 5.50 336.88 - .26 > .10 Assertion 16.26 5.00 16.77 4.63 398 -1.03 > .10 Diffusion 14.06 4.45 14.68 4.34 398 -1.41 > .10 Avoidant Coping 17.21 4.27 17.56 4.17 398 - .81 > .10 Rumination 18.35 5.00 17.20 4.68 398 2.35 > 10 Social Support Seeking 18.60 5.35 17.81 4.67 384.47 1.57 > .10 Men Women BARO Subscales Mean SD Mean SD df t value p Aggressive Anger Out 18.15 5.33 15.87 4.89 370 4.24 < .001 Assertion 16.21 4.78 16.46 4.95 370 - .49 > .10 Diffusion 14.14 4.39 14.47 4.43 370 - .72 > .10 Avoidant Coping 17.81 4.14 17.10 4.24 370 1.61 > .10 Rumination 17.96 4.75 17.77 4.94 370 .37 > .10 Social Support Seeking 17.25 4.82 19.05 5.15 370 -3.38 < .001 Caucasian- Asian-Canadians Canadians BARQ Subscales Mean SD Mean SD df t value p Aggressive Anger Out 16.41 4.94 17.24 5.36 386 -1.59 > .10 Assertion 16.64 4.97 16.28 4.72 386 .74 > .10 Diffusion 14.52 4.10 14.17 4.71 386 .77 > .10 Avoidant Coping 16.36 3.79 18.32 4.41 386 -4.67 < .001 Rumination 17.67 4.75 18.18 5.02 386 -1.02 > .01 Social Support Seeking 17.99 4.93 18.44 5.28 386 -.87 > .10 * T3 A © GO GO ro o ro o OP GO 5= T 3 T J o 3-*3 c 3 5 o 3 o > o < T J O 5' 51 CTQ P 3 rt) cn 5" 3 cn ro g. o' 3 £ > 3 era era era rt, ro _ cn O J£ 5. <' rt ^ ro H ro NJ NJ O N O NJ 4^  * * o 4^  NJ 4^ . NJ OJ O Oh 4^  Oh * Oh > ro ro P a ; CD* 3 ro cn cn O NJ O O N NJ NJ 4^  O OJ o Oh OJ o NJ O N * OJ OJ NJ N O O o 3 cn O ro' o 3 cn 3 ro o NJ o o NJ o o oo o o OJ © o NJ © o m X rt -1 ro o 3 NJ 4^  N O * o Oh NJ O NJ OJ ro 3 o O N O N * o N O NJ O N o. cn' 3 o o o 4^  o o o <1 NJ 4^  *^ NJ 4^  o ro 3 3 ro cn cn 83 Appendix A Subscales and items of the BARQ Subscales Items corresponding to subscales Aggressive Anger Out The person I am angry with gets a piece of my mind (that is, I tell him/her very directly how I feel). I make a sarcastic or critical remark to the person who annoyed me. I use vigorous gestures (for example: make a fist, wave my arms, or give a hand sign). I swear, use foul language, or curse at the person who annoyed me. I hit or push the person who angered me. I use an object to express my anger (for example: slam a door, hit a table, or drive too fast). I raise my voice. Anger Diffusion I express my feelings by playing music, writing a poem, or painting. I just keep busy hoping to work off my anger. I work off my frustration by exercising or taking a brisk walk. I write about the angering event in my diary or simply on a piece of paper. I simply put more energy than usual into my work or studies. I work off my frustration by doings things like cleaning the house, organizing the office, or by doing garden work. Assertive Anger Coping I let things cool off a little and then talk to the angering person about what happened. I stay in the situation, collect my thoughts, and then express how I feel rather than verbally attacking the person who angered me. In a calm voice, I tell the angering person how I honestly feel. I try to figure out what happened and plan to talk to the person who angered me once I calm down. While containing my frustration, I try to work on the problem with the person who angered me. I leave the situation, calm down, and then try to solve the problem by calmly talking to the person who annoyed me. 84 Social Support Seeking While in the situation, I stay calm on the surface. Later, I discuss the incident with a friend or family member over the phone. I leave the situation and look for someone (like a co-worker or a friend) who likely agrees with me when I describe what happened). I leave the situation, find a supportive person to listen to my story, and get his/her advice. I think about the problem for a while; later - for example in the evening - I discuss the incident with my spouse, partner, or friend. I leave the situation. Some time later I call a friend or family member to share my feelings. Even without planning it, I usually end up describing my feelings to a friend, partner, or family member. Avoidant Coping I convince myself that the angering event is not important. I leave the situation and try to forget the whole incident. I put the angering event out of mind. I avoid making a scene and keep my feeling to myself. I just wait to feel better. I convince myself that this is not worth getting upset about. Rurriination In my mind, I try to figure out why I really got upset. I develop a revenge fantasy where I oget eveno with the person who angered me. I think repeatedly about what I really would have liked to do but did not. I cannot easily stop thinking about the event. Annoying events bother me for a long time. In my mind, I re-play the frustrating event several times over. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088586/manifest

Comment

Related Items