Evaluations of supportive messages provided by friends and romantic partners: an attachment...

In a questionnaire study of 270 college students we investigated relationships of adult attachment-style dimensions and perceptions of comforting messages received from friends and romantic partners. We found that comfort with closeness, preoccupied with relationships, and fear of intimacy all

associated with factors developed by Bippus (2001) on evaluations of comforting communication. These same three attachment-style dimensions were significantly related with perceptions of social support, further supporting the link between adult attachment orientations and feelings about others' supportiveness. Gender differences in evaluations of comforting messages were also examined. These findings will be discussed in terms of attachment theory.

Keywords: Attachment Styles; Supportive Communication Perceived Support

**********

The social support people receive from their close friends and romantic partners has positive effects on mental and physiological well being (e.g., Berkman, Glass, Brissette, & Seeman, 2000; Cunningham & Barbee, 2000). Scholars have shown that supportive communication between partners helps to maintain relationships and to promote commitment, liking, and satisfaction (e.g., Sprecher, Metts, Budeson, Hatfield, & Thompson, 1995; Stafford, Dainton, & Haas, 2000). Supportive communication has been conceptualized as comforting messages involving emotional support and instrumental support. Research suggests that personality may play a role in how people perceive specific types of supportive messages they receive from others (e.g., Sarason et al., 1991; Wallace & Vaux, 1993). In the current study we examine the associations between attachment-style dimensions and perceptions of supportive messages provided by friends and intimate partners.

Attachment theorists claim that attachment styles influence adults' ability to cope effectively with distressful situations (e.g., Bartholomew & Horowitz, 1991; Bowlby, 1973, 1980). For instance, Simpson, Rholes, and Nelligan (1992) found that women who were secure in attachment were more likely than those who were insecure to seek emotional support from their partners before engaging in an anxiety provoking experiment. Studies by Mikulincer and Florian (1995) and Sarason et al. (1991) show that security in attachment is linked with perceived social support--feelings that others will be available in times of need.

Supportive Communication

Researchers (Bippus, 2001; Burleson & Goldsmith, 1998; Cutrona & Russell, 1987) have proposed that supportive communication is a multidimensional construct involving emotional-focused messages (e.g., acceptance, expressions of comfort, concern) and information-focused messages (e.g., advice and problem solving). Applegate (1980) and Burleson (1994) argue that comforting messages vary in their level of sophistication or person-centeredness. Highly person-centered messages are emotion focused; they are sensitive to the distressed person's emotional state. They explicitly acknowledge, elaborate, legitimize, and encourage change in the distressed person's view of the distressful situation. Research indicates that people tend to evaluate highly person-centered messages more positively than less person-centered messages (Burleson & Goldsmith, 1998; Goldsmith, McDermott, & Alexander, 2000; Jones & Guerrero, 2002). In a study on how adults rate the effectiveness and quality of comforting interactions, Bippus (2001) showed that distressed persons identified both emotional and instrumental focused behaviors as important in the comforting process. Because we felt that friends and romantic partners would engage in emotional and instrumental comforting behaviors, we employed Bippus' (2001) five factor measure in this study. These factors include: General Negativity, consisting of counterproductive behaviors that might further upset the distressed person; Other Orientation, reflecting interpersonal involvement, caring, and acceptance; Problem Solving, involving the provision of advice for alleviating the substantive issues; Relating, involving comfort providers' disclosing their own similar experiences; and finally, Different Perspective, involving the provision of an alternate way of viewing problems or issues.

Attachment Theory and Comforting

Attachment theory should provide a good theoretical framework for understanding how people perceive the comforting messages they receive from friends and romantic partners. Attachment theorists (e.g., Ainsworth, Blehar, Waters, & Wall, 1978; Bowlby, 1973, 1980) argued that attachment styles are developed during early interactions with primary caregivers, serving as cognitive templates that regulate emotions and guide how people think about themselves in relation to others. When parents react to a child's distress by consistently providing affection and comfort, the child develops a secure attachment style, fostering feelings that others will continue to be supportive. If, however, parents are unreliable in providing consistent affection and emotional comfort to their children then children develop insecure attachment styles (Ainsworth et al., 1978; Bowlby, 1973, 1980).

Bartholomew and Horowitz (1991) proposed a two-dimensional model that includes both a person's mental model of self and a mental model of others, thereby leading to four distinct attachment styles. Secures have positive views of self and others. They tend to have high self-esteem, he comfortable with intimacy, and to trust others. Preoccupieds have a negative view of self and a positive view of others. They tend to be preoccupied with relationships, to have low self-esteem, and to worry that others will abandon them. Dismissive-avoidant types have a positive view of themselves and negative views of others. They tend to have high self-esteem, avoid intimacy, and be overly independent. Fearful-avoidant types have a negative perception of themselves and others. They tend to fear intimacy and commitment, have low self-esteem, and low trust in others.

In the current study we examine five attachment-style dimensions that are based on these four attachment styles. We argue that adult attachment-style dimensions are related in predicable ways to perceptions of partners' skillfulness in providing comforting messages. Sarason et al. (1991) suggested that the difference between secure and insecure types is that secure persons see their social world as supportive. Florian, Mikulincer, and Bucholtz (1995) found that secure adults reported seeking and receiving more emotional and instrumental support from friends, romantic partners, and both parents than did three insecure types. The secure types also reported the highest level of perceived availability of support from friends, parents, and romantic partners.

Research suggests that people with different attachment-style orientations, as measured by attachment dimensions, should associate with evaluations of comforting messages provided by friends and romantic partners. Yet it is unclear how each of the attachment-style dimensions might associate with the five factors of comforting behaviors. A problem is that previous social support studies have used categorical measures to assess attachment styles (e.g., Mikulincer & Florian, 1998; Mikulincer, Florian, & Weller, 1993). Hence, our first research question explores associations between the five attachment-style dimensions and perceptions of the comforting behaviors provided by relational partners (friends and romantic partners).

RQ1: To what degree are the five attachment-style dimensions related to evaluations of the skillfulness of comforting behaviors provided by relational partners?

The second purpose of the current study is to understand the associations between attachment-style dimensions and perceived social support from relational partners. Perceived social support has been conceptualized as perceptions that others will be available to help in times of need (Sarason et al., 1991). Past studies indicate that the attachment-style dimensions should associate with perceived social support from friends and lovers in a fashion similar to evaluations of specific types of comforting behaviors. If this is the case, it would bolster our notion that how people evaluate comforting messages is partly due to their attachment-style orientation. This logic follows assumptions of attachment theory (Bowlby, 1980). People who are highly secure in attachment and have had supportive relationships in the past should continue to view their partners' comforting behavior as highly supportive. The following research question probes this issue:

RQ2: To what degree are the five attachment-style dimensions associated with perceived social support of relational partners?

Finally we investigated gender differences in perceptions of partners' comforting communication. Jones and Burleson (2003) point out that there has been much speculation as to what men and women consider effective supportive communication. One school of thought contends that men and women differ in emotional communication (e.g., Wood, 1997). Because men are less emotionally expressive and more instrumental, they should prefer messages focused on giving advice and helping solve problems. Being more expressive, women should prefer emotional focused comforting messages. Some studies have shown gender differences in perceptions of comforting (e.g., Kunkel, 2002; MacGeorge, 2003). Kunkel and Burleson (1999) found that women rate highly person-centered messages more favorably and rate less person-centered messages more negatively than do men. There are mixed results in this line of research. In most studies both men and women evaluate highly person-centered messages as the most effective (Jones & Burleson, 2003). It is unclear whether men will differ from women in terms of evaluating friends' and romantic partners' comforting behaviors, which leads to our last research question:

RQ3: Do men and women rate the skillfulness of friends' and romantic partners' comforting behaviors differently?

Methods

Participants and Procedures

Undergraduate students (N = 270; males n = 99; females n = 170, 1 missing data) in communication courses at a large university in southern California participated in a questionnaire study. Data from five people were dropped because they failed to complete portions of the questionnaire. The average age of participants was 23.5 years (SD = 5.40), and most classified themselves Euro-American (50%) followed by Mexican/Hispanic Americans (28.9%), African American (11.9%), and Asian American (7%). Five individuals did not identify their ethnicity. Respondents described a recent situation in which they spoke with a good friend or romantic partner about something that caused them to be upset. They also provided a brief 'script' of the conversation they had with their relational partner, an approach used in previous research on comforting messages (Bippus, 2001; Vangelisti & Crumley, 1998) to verify that participants understood the directions and reported on interactions that met the study requirements.

Attachment-style dimensions were operationalized with a multi-item measure used in previous research (e.g., Bachman & Guerrero, 2001; Guerrero, 1998) developed by Griffin and Bartholomew (1994). The five subscales were rated on 7-point Likert-type scales: comfort with closeness (e.g., 'I find it easy to get emotionally close to others,' [alpha] = .79); trust in others (e.g., 'I think most people are trustworthy,' [alpha] = .78); preoccupied with relationships (e.g., 'I worry that others don't value me as much as I value them,' [alpha] = .87); general avoidance (e.g., 'I find it difficult to depend on others,' [alpha] = .78); and fear of intimacy (e.g., 'Romantic partners want to be closer than I feel comfortable being,' [alpha] = .73).

Bippus' (2001) measures for rating the skillfulness of comforting behavior were used to assess general negativity (e.g., 'Seemed to be judging me,' [alpha] = .93); other orientation (e.g., 'Was focused on my feelings,' [alpha] = .94); problem solving (e.g., 'Helped me to solve my problem,' [alpha] = .88); relating (e.g., 'Showed me that they could relate to my issue,' [alpha] = .82); and different perspective (e.g., 'Showed me another way of viewing my problem,' [alpha] = .86).

Perceived social support was measured with an instrument employed by Wallace and Vaux (1993). The measure consisted of 12 statements rated on 7-point Likert-type scales. The stems were slightly modified to assess perceptions about friends and romantic partners (e.g., 'It is easy to ask for help from my friends when I need it,' 'I can't rely on my romantic partner for support,' [alpha] = .85).

Results

A canonical correlation was computed to test the first research question to determine if the set of attachment-style dimension variables associated with the set of variables representing ratings of relational partners' comforting behaviors. (1) A significant canonical correlation, Rc = .23, F(25, 956.21) = 4.21 p < .001, reflected a congruence between comfort with closeness, preoccupied with relationships, and fear of intimacy, and ratings of the comforting behaviors (Table 1). Comfort with doseness was inversely related with negativity and positively related with other orientation and problem solving. A reverse pattern was found for both the preoccupied with relationships and fear of intimacy dimensions. Trust in others and general avoidance did not figure significantly as correlates of perceptions to comforting attempts. Although both the skillfulness variables of relating and different perspective also loaded on the canonical variate, their standardized coefficients indicated that they contributed little unique variance. A second canonical correlation was significant, but because the sample size-to-variable ratio was too small and the correlation was weak it was not interpreted. (2)

To address the second research question, a linear regression was computed for the five attachment-style dimension variables on the measure of perceived social support. The model was significant, F(5,263) = 101.03, p < .001, adjusted [R.sup.2] = .65, with comfort with closeness ([beta] = .40), preoccupied with relationships ([beta] = -.43), and fear of intimacy ([beta] = -.24) emerging as significant predictors. Again, trust in others and general avoidance were not significant.

For the third research question, a MANOVA was calculated to test the effect of sex and relationship type on perceptions of skillfulness evaluations. No significant interaction emerged, but there were main effects for sex of recipients, F(5, 247) = 4.38, p < .002, partial [[eta].sup.2] = .08, and type of relationship, F(5, 247) = 4.74, p < .001, [[eta].sup.2] = .09. Follow-up univariate yielded significant gender differences for two of the skillfulness measures: general negativity, F(1, 251) = 16.39, p < .001, [[eta].sup.2] = .06, and other orientation, F(1, 251) = 9.40, p < .003, [[eta].sup.2] = .04. Males rated their comfort providers higher on general negativity (M = 2.84, SD = 1.14) than did females (M = 2.25, SD = 1.14), but females rated their friends and intimate partners higher on other orientation (M = 5.52, SD = 1.31) than did males (M = 5.05, SD = 1.05). For relationship type, a significant a univariate test, F(1, 251) = 7.40, p < .01, [[eta].sup.2] = .03, revealed that friends were rated higher on other orientation (M = 4.86, SD = 1.54) than romantic partners (M = 4.32, SD = 1.55).

The results from the MANOVA suggested that if sex and relational type were added to the set of independent variables in the canonical correlation they could alter the derived canonical variates. We computed a post hoc analysis with sex and relational type and the attachment-style dimensions in the first set variables representing ratings of relational partners' comforting behaviors. The attachment-style dimensions obtained the largest canonical correlation, Rc = 28, F(35, 1024.64) = 3.64, p < .001, and accounted for slightly more variance, and sex after dummy coded emerged as a defining variable (Table 2).

Discussion

Our findings indicate support that attachment-style orientations are linked to perceptions of comforting behaviors of friends and romantic partners. We found three of the five attachment-style dimensions were significantly associated with general negativity, other orientation, and problem solving. Comfort with closeness, a distinct component of the secure attachment style, was inversely associated with general negativity and positively associated with other orientation and problem solving. These results indicate that individuals who reported higher scores for comfort with closeness were unlikely to think that their friends and lovers judged them when they were distressed. They viewed their partners as highly interested in their problems; their partners encouraged them to talk about the distressful situation, expressed empathy, and focused on their thoughts and feelings. Respondents who are highly comfortable with closeness rated their friends and romantic partners highest on providing advice and helping to solve problems.

Our findings for the preoccupied with relationships and fear of intimacy dimensions (both distinct components of the insecure attachment styles) revealed an opposite pattern. The findings reveal that these types are more likely to feel that their friends and romantic partners judge them when they become distressed. Respondents who scored higher on these two dimensions reported higher scores in general negativity, suggesting that people who are highly preoccupied and fearful of intimacy might view their partners as being cold and uninterested in their problems. Overall, respondents who scored higher on the two insecure dimensions did not view their comforters as effective in helping solve their problems.

The results reveal clear distinctions between individuals who reported themselves as comfortable with closeness and those who reported themselves as highly preoccupied with relationships or fear of intimacy. People who feel comfortable with close relationships report their friends and romantic partners to be caring and highly skilled comfort providers. The findings for preoccupied with relationships and fear of intimacy are probably similar because both attachment styles represent individuals who experience high levels of anxiety, need self-validation from others, and have difficulty coping with distress (Bartholomew & Horowitz, 1991). Perhaps when these types become distressed, they have unrealistic expectations of the support people can offer. Whatever the partner does to help is simply judged poorly. A study of college students during the SCUD missile crisis by Mikulincer et al. (1993) revealed similar results: Preoccupied types were just as likely as secure types to receive social support from others, but they reported the support as not helpful.

Our second research question involved the associations between the attachment-style dimensions and perceived social support. Similar to our findings for evaluations of specific comforting behaviors, comfort with closeness was positively related to perceived social support, whereas the preoccupied and fear of intimacy dimensions were negatively related. This provides further support for attachment theory and our argument regarding attachment-style dimensions and perceptions of others' support. Attachment-style dimensions are not only related to perceptions of others' willingness to provide help in time of need, but are also related in the same way to evaluations of specific comforting messages they have received from friends and lovers.

Our results also demonstrate gender differences in evaluations of partners' comforting behavior. Male respondents rated their friends and lovers higher on general negativity than did females, suggesting that men are more likely to perceive that their relational partners are less interested in their problems and to make judgments about them when they become distressed. It may be that male respondents are drawing on gender stereotypes, feeling that their partners view them as 'weak' for needing help. In sharp contrast female respondents rated their friends and romantic partners higher on the other orientation than did males, suggesting that women view their partners as more emotionally involved, focused on their feelings, and interested in helping them with their problems.

Interestingly, both males and females rated their friends higher on the other orientation than they rated their romantic partners. This suggests that college age adults are more apt to turn to friends for emotional support than to other people. Samter, Whaley, Mortenson, and Burleson (1997) noted that young adults view their friends as their main source of emotional support and are more likely to seek advice, comfort, and reassurance from friends than from family members and romantic partners. Perhaps young adults try to keep their interactions with romantic partners on a positive tone, preferring to share their daily hassles and problems with friends.

In summary this study offers further evidence that attachment styles play a role in how people perceive and interpret the support provided by close friends and romantic partners. The more secure and comfortable people are with close relationships the more positive their evaluations of the comfort provided by friends and lovers. The more preoccupied and fearful of intimacy people are in attachment, the more negative they evaluate comfort providers. Male respondents evaluated their partners as more critical of them when they became distressed and less emotionally supportive than did female respondents. The greatest limitation of this study is that respondents were all college students enrolled in communication courses. Repeating the sampling with nonstudent populations might provide more insight. Another limitation is that the results are based on recalling a comforting interaction. It is limited by not being able to verify how accurate the respondents were when remembering these interactions.

References

Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: Apsychological study of the strange situation. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Applegate, J. L. (1980). Adaptive communication in educational contexts: A study of teachers' communicative strategies. Communication Education, 29, 158-170.

Bachman, G., & Guerrero, L. (2001, November). Attachment-style dimensions as predictors of relational maintenance behavior. Paper presented to the Interpersonal Communication Division of the National Communication Association, Atlanta, GA.

Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61 (2), 226-244.

Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration to health: Durkheim in the new millennium. Social Science and Medicine, 51, 843-857.

Bippus, A. M. (2001). Recipients' criteria for evaluating the skillfulness of comforting communication and the outcomes of comforting interactions. Communication Monographs, 68, 301-313.

Bowlby, J. (1973). Attachment and loss: Vol. II. Separation: Anxiety and anger. New York: Basic Books.

Bowlby, J. (1980). Attachment and loss (Vol. 3). New York: Basic Books

Burleson, B. R. (1994). Comforting messages: Features, functions, and outcomes. In J. A. Daly & J. M. Wiemann (Eds.), Strategic interpersonal communication (pp. 135-161). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Burleson, B. R., & Goldsmith, D. J. (1998). How the comforting process works: Alleviating emotional distress through conversationally induced reappraisals. In P. A. Andersen & L. K. Guerrero (Eds.), Handbook of communication and emotion: Research, theory, applications, and contexts (pp. 245-280). San Diego, CA: Academic Press.

Cunningham, M. R., & Barbee, A. P. (2000). Social support. In C. Hendrick & S. S. Hendrick (Eds.), Close relationships: A sourcebook (pp. 272-285). Thousand Oaks, CA: Sage.

Cutrona, C. E., & Russell, D. W (1987). The provisions of social relationships and adaptations to stress. In W. H. Jones & D. Perlman (Eds.), Advances in personal relationships (Vol. 1, pp. 37-67). Greenwich, CT: JAI Press.

Florian, V., Mikulincer, M., & Bucholtz, I. (1995). Effects of adult attachment style on the perception and search for social support. Journal of Psychology, 129, 665-676.

Goldsmith, D. J., McDermott, V. M., & Alexander, S. C. (2000). Helpful, supportive and sensitive: Measuring the evaluation of enacted social support in personal relationships. Journal of Social and Personal Relationships, 17, 369-391.

Griffin, D. W., & Bartholomew, K. (1994). Models of self and other: Fundamental dimensions underlying measures of adult attachment. Journal of Personality and Social Psychology, 67, 430-445.

Guerrero, L. K. (1998). Attachment-style differences in the experience and expression of romantic jealousy. Personal Relationships, 5, 273-291.

Jones, S. M., & Burleson, B. R. (2003). Effects of helper and recipient sex on the experience and outcomes of comforting messages: an experimental investigation. Sex Roles: A Journal of Research, 48, 1-18.

Jones, S. M., & Guerrero, L. A. (2002). The effects of nonverbal immediacy and verbal person centeredness in the emotional support process. Human Communication Research, 27, 567-596. Kunkel, A. (2002). Explaining sex differences in the evaluation of comforting messages: The mediating role of interaction goals. Communication Reports, 15, 29-42.

Kunkel, A. W., & Burleson, B. R. (1999). Assessing explanations for sex differences in emotional support: A test of the different cultures and skill specialization accounts. Human Communication Research, 25, 307-340.

Lewis-Beck, M. S. (1980). Applied regression: An introduction. Beverly Hills, CA: Sage. MacGeorge, E. L. (2003). Gender differences in attributions and emotions in helping contexts. Sex Roles, 48 (3-4), 175-182.

Mikulincer, M., & Florian, V. (1995). Appraisal and coping with a real-life stressful situation: The contribution of attachment styles. Personality and Social Psychology Bulletin, 21, 408-416.

Mikulincer, M., & Florian, V. (1998). The relationship between adult attachment styles and emotional and cognitive reactions to stressful events. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships (pp. 143-165). New York: Guilford Press.

Mikulincer, M., Florian, V., & Weller, A. (1993). Attachment styles, coping strategies and posttraumatic psychological distress: Impact of the Gulf War in Israel Journal of Personality and Social Psychology, 64, 817-826.

Samter, W., Whaley, B. B., Mortenson, S. R., & Burleson, B. R. (1997). Ethnicity and emotional support in same-sex friendship: A comparison of Asian-Americans, African-Americans, and Euro-Americans. Personal Relationships, 4, 413-430.

Sarason, I. G., Pierce, G. R., Sherin, E. N., Sarason, I. G., Waltz, J. A., & Poppe, L. (1991). Perceived social support and working models of self and actual others. Journal of Personality and Social Psychology, 60, 272-287.

Simpson, J. A., Rholes, W. S., & Nelligan, J. S. (1992). Support-seeking and support-giving within couple members in an anxiety-provoking situation. Journal of Personality and Social Psychology, 62, 434-446.

Sprecher, S., Metts, S., Burleson, B., Hatfield, E., & Thompson, A. (1995). Domains of expressive interaction in intimate relationships: Associations with satisfaction and commitment. Family Relations, 44, 203-210.

Stafford, L., Dainton, M., & Haas, S. M. (2000). Measuring routine and strategic relational maintenance: Scale revision, sex versus gender roles, and the prediction of relational characteristics. Communication Monographs, 3, 306-323.

Stevens, J. (1996). Applied multivariate statistics for the social sciences. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Vangelisti, A. L., & Crumley, L. P. (1998). Reactions to messages that hurt: The influence of relational contexts. Communication Monographs, 65, 173-196.

Wallace, J. L., & Vaux, A. (1993). Social support network orientation: The role of adult attachment style. Journal of Social and Clinical Psychology, 12, 354-365.

Wood, J. T. (1997). Gendered lives: Communication, gender, and culture (2nd ed.). Belmont, CA: Wadsworth.

Notes

[1] This analysis considers relationships among two composite variables which are the weighted sum of two variables sets--in this case, the attachment style-dimensions and the skillfulness variables. To guard against multicollinearity in a set of independent variables, Lewis-Beck (1980) recommends regressing each independent variable against all others to guard against values close to 1.0. The highest adjusted [R.sup.2] yielded by this approach was an acceptably low .40.

[2] Though significant at the .01 level, the second canonical correlation was not interpreted for two reasons. First, Stevens (1996) recommends a sample size-to-variable ratio of 42/1 if interpreting the first two canonical pairs, but 20/1 if interpreting only the first. The ratio in this study was 27/1. Second, the correlation was a weak .30, which Stevens cautions against attempting to interpret.

Guy Bachman (PHD, Arizona State University, 2002) and Amy Bippus (PHD, University of Texas at Austin, 2000) are Assistant Professors in the Communication Studies Department at California State University, Long Beach. Correspondence to: Guy Foster Bachman, Department of Communication Studies MHB-717, California State University, Long Beach, 1250 Bellflower Blvd, Long Beach, CA 90840, USA. Email: gfbachman@csulb.edu. A portion of this paper was presented at the 2004 National Communication Association Conference. The authors wish to thank the two anonymous reviewers and Walter Zakahi for their helpful comments.

Table 1 Canonical Loadings and Coefficients

                                       Canonical       Standardized
                                    variate-variable   coefficients
                                      correlations
Attachment style variables
  Comfort with closeness                  .86              .84
  Preoccupied with relationships         -.50             -.24
  General avoidance                      -.39             -.02
  Trust in others                         .25             -.44
  Fear of intimacy                       -.67             -.38
Comforting skillfulness variables
  General negativity                     -.81             -.25
  Other orientation                       .95              .87
  Problem solving                         .70              .25
  Relating                                .43             -.23
  Different perspective                   .45             -.23

Table 2 Canonical Loadings and Coefficients with Sex and
Relationship Type

                                       Canonical       Standardized
                                    variate-variable   coefficients
                                      correlations

Attachment style variables
  Comfort with closeness                  .80              .81
  Preoccupied with relationships         -.44             -.18
  General avoidance                      -.33             -.O1
  Trust in others                         .19             -.44
  Fear of intimacy                       -.66             -.33
  Sex (dummy coded for male)             -.45             -.26
  Relationship type                       .23              .07
Comforting skillfulness variables
  General negativity                     -.81             -.30
  Other orientation                       .93              .87
  Problem solving                         .66              .21
  Relating                                .33             -.34
  Different perspective                   .43             -.17