Male Restrictive Emotionality and Evaluations of Online Versus Face-to-Face Counseling

Male Restrictive Emotionality and Evaluations of Online

Versus Face-to-Face Counseling

Aaron B. Rochlen, Lee N. Land, and Y. Joel Wong

The present study investigated men’s perceptions of online versus face-to-face counseling. One hundred ninety-one men with a range of restrictive emotionality and online counseling attitudes reviewed face-to-face or online counseling vignettes using a cognition- or emotion-based thera-peutic approach. Overall, participants’ evaluations of counseling approaches were similar across modality and theoretical approach. However, men with low as opposed to high levels of restric-tive emotionality expressed more favorable perceptions of face-to-face counseling. Significant improvements in online counseling attitudes were observed across the entire sample, with more pronounced improvements in online counseling value for men reviewing online counseling vi-gnettes. Results are discussed within the context of men’s underutilization of counseling services and the online counseling literature. Suggestions for further research are provided.

The challenges associated with men seeking and ultimately benefiting from counseling services have been well documented in the literature (Addis & Ma-halik, 2003; Brooks & Good, 2001; Rochlen, Bla-zina, & Raghunathan, 2002). In regard to men’s help-seeking behaviors, Addis and Mahalik (2003) recently concluded that overall, men of varied nation-alities, ethnicities, racial backgrounds, and ages seek professional help to a lesser extent than women. Studies on within-group differences have also added to our understanding and concern regarding men’s clinical treatment, with evidence suggesting that men most susceptible to psychological distress appear to be reluctant to seek available support services (Good & Wood, 1995).

A frequently recommended next step in addressing men’s underutilization of mental health services in-volves studying the appeal of alternative methods of providing counseling services to more counseling-resistant men (Addis & Mahalik, 2003; Hurst, 1997). To date, a few studies in this area have been initiated,

with efforts focusing on how men respond to alter-native methods of marketing therapy (Robertson & Fitzgerald, 1992; Rochlen et al., 2002) and different therapeutic styles (Wisch, Mahalik, Hayes, & Nutt, 1995; Rochlen & O’Brien, 2002). The current study intends to add to this research area through evaluat-ing men’s perceptions of counselevaluat-ing sessions as con-ducted using a face-to-face or online counseling mo-dality, with either an emotion- or cognition-focused therapeutic approach.

Theoretical Models and Within-Group Research Addressing Men’s

Help-Seeking Patterns

In considering the possible benefits of online coun-seling for men, it is first useful to consider the theo-retical accounts that aim to explain men’s overall reluctance to seeking professional help. In general, authors have frequently described the characteristics associated with interest and successful engagement in psychotherapy (e.g., being emotionally expressive) as being in contrast with the values of the male cul-ture and norms (Addis & Mahalik, 2003; Good, Gil-bert, & Scher, 1990; O’Neil, 1981a, 1981b; Robert-son & Fitzgerald, 1992; Wilcox & Forrest, 1992). This incongruence between the culture of masculinity and the counseling environment appears to be par-ticularly salient for men who adhere rigidly to tradi-tional male roles (Betz & Fitzgerald, 1993; Courte-nay, 2000).

Much of the empirical research that has incorpo-rated the above theoretical descriptions has been con-ducted with the Gender Role Conflict Scale (GRCS; O’Neil, Helms, Gable, David, & Wrightsman, 1986). Aaron B. Rochlen, Lee N. Land, and Y. Joel Wong,

De-partment of Educational Psychology, University of Texas at Austin.

We are grateful to Kevin Flahive, who served as a client in the audio vignette, and to Clara Hill and Shanna Smith for statistical consultation. Thanks also to Ken Trosky for his assistance with data collection and entry. Finally, gratitude is extended to Ryan Mckelly and Alissa Sherry for provid-ing thoughtful feedback on earlier versions of this article.

Correspondence concerning this article should be ad-dressed to Aaron B. Rochlen, Department of Educational Psychology, Counseling Psychology Program, University of Texas at Austin, 1 University Station D5800, Austin, TX 78712. E-mail: aaron.rochlen@mail.utexas.edu

O’Neil et al. (1986) proposed that at the core of male socialization is a fear and devaluation of femininity that leads to men’s unwillingness to be expressive, vulnerable, or emotional. As these characteristics are necessary for both men and women, O’Neil et al. speculated that men who rigidly adhere to these “rules” of masculinity are likely to suffer from a range of negative psychological consequences. Re-search has also demonstrated that men with high lev-els of gender role conflict report negative help-seeking attitudes (Blazina & Watkins, 1996; Good, Dell, & Mintz, 1989; Good & Wood, 1995).

A limited number of studies have used the GRCS measure in integrating personal and situational vari-ables in the study of men’s help-seeking patterns. In a frequently referenced study, Robertson and Fitzger-ald (1992) randomly assigned participants into two groups to evaluate different brochures offering ser-vices at a counseling center. Half the brochures de-scribed traditional counseling activities (individual and group counseling), whereas the remaining bro-chures described alternative services (classes, video-tapes, workshops, and structured interventions). Par-ticipants rated their likelihood of seeking help at the counseling center for a list of 17 potential issues. Men with more positive attitudes toward counseling expressed equal preferences for the two types of counseling brochures. Men with more negative atti-tudes toward help seeking and men with higher levels of gender role conflict had more interest in seeking professional help after viewing the nontraditional counseling services brochures.

In a study with particular relevance to the current project, Wisch et al. (1995) assessed the contribution of gender role conflict to men’s attitudes toward psy-chological help seeking after viewing two different therapeutic approaches. Participants were randomly assigned to watch a video segment of a counseling session involving interventions with an emotion-focused (e.g., “Let’s talk for a minute about your parents. What kind of feelings come up when you deal with them?”) or cognition-focused orientation (e.g., “Let’s talk for a minute about your parents. What kind of messages to yourself come up when you deal with them?”). Among men with high levels of gender role conflict, participants who viewed the cognition-focused video reported more positive help-seeking attitudes compared with participants who viewed the emotion-focused video.

Finally, in a recent study addressing men’s percep-tions of different theoretical approaches to career counseling, Rochlen and O’Brien (2002) found that overall, men preferred a more directive approach to career counseling over a more contextual,

emotion-oriented approach. In that study, preferences for therapeutic style were not modified by gender role conflict. Thus, men with varying degrees of gender role conflict preferred the more structured approach (identified as a person–environment fit approach) over the more emotion-oriented contextual approach (identified as a psychodynamic career counseling approach).

Online Counseling Overview and Potential Application for Male Clients

In the last 15 years, online counseling has been one of the most frequently discussed and debated prac-tices in the counseling literature (Alleman, 2002; Rochlen, Zack, & Speyer, 2004). Included in this debate are deliberations on the precise definition, ethical–legal considerations, regulatory and licensure issues, and the benefits and limitations of the practice (Alleman, 2002; Bloom, 1998; Heinlen, Welfel, Richmond, & Rak, 2003; Rochlen et al., 2004; Stofle, 2001). Although a full exploration of these issues is beyond the scope of this article, it is important to note one of the more frequently cited benefits of online counseling, namely the potential appeal to popula-tions who underutilize or stigmatize counseling ser-vices (Stofle, 2001). Cited advantages of online counseling for such populations include increased privacy and anonymity, decreased effort to begin therapy, and an atmosphere perceived as less threat-ening (Kraus, Zack, & Stricker, 2004; Rochlen et al., 2004). Despite these proposed benefits, we found no studies that have directly evaluated the appeal to populations who underutilize help-seeking services.

In regard to men and online counseling in particu-lar, one instrument development project did find that men and women expressed similar attitudes toward online counseling (Rochlen, Beretvas, & Zack, in press). For face-to-face counseling, women had more favorable attitudes than men, a finding consistent with the larger help-seeking literature (Addis & Ma-halik, 2003). In discussing these findings, Rochlen et al. (in press) suggested that future research investi-gate within-group differences that may provide more information on how online counseling might be per-ceived by different types of male help seekers.

Restrictive Emotionality

For the purposes of our current study, we decided to use the construct of restrictive emotionality to ex-plore within-group differences in regard to percep-tions and reacpercep-tions to online versus face-to-face

counseling. Restrictive Emotionality is one of four dimensions of gender role conflict and has been de-fined as having “difficulty and fears about expressing one’s feelings and difficulty finding words to express basic emotions” (O’Neil, Good, & Holmes, 1995, p. 176). Our decision to focus on Restrictive Emotion-ality was based on two central considerations. First, the Restrictive Emotionality dimension has yielded the strongest relationships with psychological dis-tress and negative help-seeking attitudes (e.g., Bla-zina & Watkins, 1996; Cournoyer, 1994; Good & Wood, 1995; Robertson & Fitzgerald, 1992; Sharpe & Heppner, 1991). Consequently, this construct seems particularly useful in detecting men that may be experiencing a range of psychological problems combined with a relative disinterest in utilizing coun-seling services. Second, men’s emotional inexpres-siveness has been a popular but controversial issue that has not been sufficiently researched in the con-text of therapy (Fischer & Good, 1997). Although many scholars (Brooks & Gilbert, 1995; Heppner & Gonzales, 1987; Levant, 2001; Scher, 1981) have suggested that men tend to be emotionally inexpres-sive because they are unaware of what they feel, others have posited that emotionally inexpressive men tend not to disclose their feelings because they choose not to rather than because of an inability to do so (Fischer & Good, 1997). The latter explanation raises the possibility that men who are inexpressive might be more willing to express their emotions in a setting that they perceive to be less psychologically threatening.

Objectives and Hypotheses

In consideration of the above literature, the current study had several objectives. The first objective, ex-ploratory in nature, was to generate descriptive in-formation regarding the relationships among men’s attitudes toward online counseling and the dimen-sions of male gender role conflict. The second ob-jective was to examine men’s evaluations of varied simulated counseling sessions, as conducted using different counseling modalities (online vs. face to face) and theoretical approaches (cognition vs. emo-tion oriented). On the basis of suggesemo-tions that online counseling may work well for populations who struggle with expressing emotions (Stofle, 2001; Zack, 2004), we hypothesized that men with high as opposed to low levels of Restrictive Emotionality would express more positive evaluations of the on-line counseling modality over a face-to-face condi-tion. We expected men with low levels of Restrictive Emotionality to evaluate the face-to-face counseling

conditions more positively. In addition, on the basis of previous literature (Robertson & Fitzgerald, 1992; Wisch et al., 1995), we hypothesized that men with high levels of Restrictive Emotionality would evalu-ate vignettes that displayed a cognitive-based (as op-posed to an emotion-based) theoretical approach more positively, regardless of the counseling modal-ity (online or face to face). Finally, we wanted to examine whether improvements in attitudes toward online counseling could be demonstrated on the basis of participation in the study. As has been demon-strated in several recent studies (Blazina & Marks, 2001; Gonzalez, Tinsley, & Kreuder, 2002; Rochlen et al., 2002), we anticipated improvements in atti-tudes toward online counseling as a result of being shown realistic portrayals of counseling scenarios carried out in a range of different modalities and approaches.

Method Participants

One hundred ninety-one undergraduates, recruited through undergraduate educational psychology courses at a large public university, served as partici-pants. All participants were given course credit for their involvement. The ethnic breakdown of the final participant pool was 108 Caucasian (56.5%), 1 Afri-can AmeriAfri-can (0.5%), 48 Asian AmeriAfri-can (25.1%), 25 Latino (13.1%), 3 biracial (1.6%), and 6 other (3.1%). Participants ranged in age from 18 to 43 years (M ⳱ 21.80 years, SD ⳱ 2.37). The over-whelming majority of the sample comprised juniors and seniors (87.40%).

Simulated Counseling Sessions

Our primary objective for creating the vignettes was to simulate in a realistic manner client–counselor interactions portrayed in either a face-to-face or on-line counseling context, with each scenario having primarily an emotion- or cognition-centered thera-peutic approach. We created these vignettes and gave them to several psychology doctoral students for feedback prior to having them reviewed in the ma-nipulation check.

Each of the four conditions included a written overview of the client, a summary of the first coun-seling session, a media-based portrayal of the second session, and a written conclusion of the second ses-sion. For all conditions, the written client overview detailed a 26-year-old graduate student who had been experiencing increased stress at school and work

be-cause of difficulties with concentration and perfor-mance, as well as relationship concerns with his fi-ance´e. The first session summary illustrated some of the counselor’s interventions and general approach for working with the client. The media-based por-trayal of the second session was an audiotape for the face-to-face counseling participants and a computer monitor portraying an online counseling session for the online counseling participants.

For the face-to-face conditions, two 14-min audio-tapes were created to represent a segment of the sec-ond session of therapy in the simulated counseling sessions. The tapes used male actors to portray the counselor (32-year-old male psychologist) and the client (26-year-old masters student). In the emotion-focused audio recording, the counselor generally re-sponded to the client by focusing on the client’s emotions and inner world, for example, “I could defi-nitely tell that was an intense experience for you.” The client responded by reporting emotions, such as “Then I remembered that you told me to focus on my feelings, and I started to notice really how miserable I was feeling.” In the cognition-focused audio record-ing, the counselor intervened by focusing on the cli-ent’s cognitions, for example, “What exactly goes through your mind when you think of yourself as a failure?” The client responds by reporting thoughts, such as “I’m thinking it’s terrible that I’m the lousiest student in the class!”

For the online counseling conditions, participants reviewed simulations of online counseling sessions that were nearly identical to the original audiotape transcripts, with minor semantic changes introduced to emphasize the authenticity of the online therapy medium. For example, “We talked about” was modi-fied to “We chatted about” in the online conditions to mirror an actual online counseling session. These scenarios were designed to simulate the experience of participating in a real-time Internet text chat session, noted to be the most commonly used method of on-line counseling (Rochlen et al., 2004). In addition, participants worked on computers and controlled the progression of the online therapy dialogue in a man-ner designed to simulate actual participation in an online counseling session.

As the media-based portrayal of the second session was relatively brief, participants in all conditions also read a short passage describing the conclusion of the second session. These summaries also differed ac-cording to theoretical approach. In the emotion-centered condition, the client continued to explore his tendency to downplay and ignore his emotional re-actions, as well as how this related to his childhood experiences. In the cognition-centered condition, the

client proceeded to investigate the link between his cognitions and behaviors and was encouraged to no-tice how negative thought processes affected his life. This progression from the overview of the client’s initial problem through the first two counseling ses-sions was constructed to provide a detailed account of what might occur during a sequence of individual therapy and to reinforce the distinctions between the emotion-oriented and cognition-oriented approaches.

Measures

Manipulation check. To provide validation for the content and process of the audio and online coun-seling vignettes, we created a manipulation check with two subscales. The first scale was the Affective– Cognitive Content in Counseling Scale, developed by Wisch et al. (1995). The scale assessed the affective and cognitive content in the counseling vignettes. Raters used a 6-point scale to indicate how much the counselor in the vignettes focused on the client’s cog-nitions (1⳱ extremely cognition-focused) or emo-tions (6⳱extremely emotions-focused). The second scale was a three-item measure developed for this study called the Counselor Skills Scale. This scale assessed the perceived competence of the counselor in the vignette (e.g., whether the counselor demon-strated sound basic counseling skills) on a 5-point Likert scale (1 ⳱ strongly disagree, 5⳱ strongly agree). The counseling vignettes were assessed by 15 counseling psychology and 1 clinical psychology doctoral student. Each doctoral student was randomly assigned to evaluate one of the four counseling vi-gnettes. There was good preliminary evidence for the reliability of this measure, with an internal consis-tency estimate for all vignettes of .76. (See the “Re-sults” section for descriptive data and the results from the manipulation check analyses.)

Demographic questionnaire. A demographic questionnaire developed for this study included ques-tions about age, year in school, ethnicity, and aca-demic status.

Attitudes toward online counseling. To measure attitudes toward online counseling, we used the On-line Counseling Attitudes Scale (OCAS; Rochlen et al., in press). The OCAS is a 10-item questionnaire that assesses attitudes toward online counseling. This scale is comprised of two subscales: Value of Online Counseling (5 items) and Discomfort With Online Counseling (5 items). Responses are made on a 6-point Likert scale (1 ⳱ strongly disagree, 6 ⳱

strongly agree). Scale scores were calculated by sum-ming the responses to each item for the respective subscales, with high scores for the Value of Online

Counseling and Discomfort With Online Counseling subscales reflecting a strong sense of value toward online counseling and a high degree of discomfort associated with online counseling, respectively. In-ternal consistency estimates ranged from .77 to .90 over multiple studies for scores on both the Value of Online Counseling and Discomfort With Online Counseling subscales, and test–retest correlations of .70 to .88 were observed over a 3-week period on both subscales (Rochlen et al., in press). Evidence for convergent validity of the Value of Online Counsel-ing and Discomfort With Online CounselCounsel-ing sub-scales has been provided through correlations in the expected directions with general help-seeking atti-tudes, comfort in using e-mail, and self-reported like-lihood of using online counseling (Rochlen et al., in press). In the current study, internal consistency es-timates averaged between the pre- and postattitudes administrations were .93 and .91 for scores on the Value of Online Counseling and Discomfort With Online Counseling subscales, respectively.

Gender role conflict. The GRCS (O’Neil et al., 1986) is a 37-item questionnaire that assesses men’s thoughts, feelings, and attitudes concerning gender role behaviors. The measure consists of four patterns of gender role conflict, namely, Success, Power, and Competition; Restrictive Emotionality; Restrictive Affectionate Behavior Between Men; and Conflict Between Work and Family. Responses are made on a 6-point Likert scale (1 ⳱ strongly disagree, 6 ⳱

strongly agree). The scales are calculated by sum-ming the individual items, with high scores reflecting a greater expression of gender role conflict within that particular domain. A total GRCS score, assessing the overall level of gender role conflict and adher-ence to traditional models of masculinity, can be computed by summing the total scores of the four subscales (O’Neil et al., 1995). Test–retest reliability over a 4-week period, for scores on all subscales, ranged from .72 to .86 (O’Neil et al., 1986). The GRCS subscales have been shown to be negatively related to psychological well-being (Blazina & Wat-kins, 1996; Cournoyer & Mahalik, 1995) and atti-tudes toward help seeking (Good & Wood, 1995). In our study, internal consistency for scores on the total GRCS was .91, with alphas ranging from .86 to .91 for subscale scores. In the current study, our primary interest was in the 10-item Restrictive Emotionality subscale, which we used as an independent variable to analyze men’s preferences for different modalities and approaches to counseling.

Counselor rating. The Counseling Rating Form—Short Form (CRF–S; Corrigan & Schmidt, 1983) is a 12-item questionnaire that assesses

char-acteristics of a therapist on three domains: Attractive-ness, ExpertAttractive-ness, and Trustworthiness. This version was a modified scale of the original 36-item Coun-selor Rating Form, developed by Barak and LaCrosse (1977). Respondents rated the extent to which a counselor demonstrates the characteristic of a posi-tive adjecposi-tive on a 7-point Likert scale (1⳱not very, 7 ⳱ very). Using this shortened version, Corrigan and Schmidt (1983) demonstrated an equivalent fac-tor structure, along with adequate levels of reliability and internal consistency estimates (above .80 for all scales), with the original measure. High scores on each of the respective scales reflect a high degree of perceived attractiveness, expertness, and trustworthi-ness. In the present study, internal consistency esti-mates for the current sample on the Attractiveness, Expertness, and Trustworthiness scales were .94, .85, and .89, respectively. In addition to the three con-structs of Expertness, Attractiveness, and Trustwor-thiness, the CRF–S has been found to measure pri-marily a general construct—a “good counselor factor” (Tracey, Glidden, & Kokotovic, 1988). As has been done in past studies (e.g., see Guinee, Tracey, & Terence, 1997), we created a total coun-selor evaluation index by combining the subscales. The internal consistency estimate for the total scale score was .91.

Counseling approach evaluation. The Counsel-ing Approach Evaluation Form (CAEF; Lyddon, 1989) is a six-item questionnaire that assesses pref-erence for styles of counseling. The CAEF is com-posed of two scales, each with three items. The first scale measures evaluations of the counseling ap-proaches with regard to the likelihood of benefiting from and using the particular approach. The second scale assesses preferences for the counseling ap-proach in relation to others’ opinions. The correlation between the two scales was .94. Responses were made on a 7-point scale. Scores for each scale were based on the mean scores of the three items. High scores on the first scale indicate greater personal in-terest in the counseling approach, whereas high scores on the second scale reflect perceptions that peers would be attracted to the approach. Lyddon and Adamson (1992) reported internal consistency esti-mates of .96 and .93 for their sample for the first and second scales, respectively, and a correlation of .92 between the two scales. Test–retest reliabilities over a 1-week time period ranged from .87 to .91. The CAEF was used to show preferences for different styles of counseling based on personal epistemology (Lyddon, 1989) and worldview (Lyddon & Adam-son, 1992). As has been done in past studies (Lyddon, 1989; Lyddon & Adamson, 1992), the two scales

were summed to create a total counseling evaluation score, with an internal consistency estimate of .90.

Procedure

The study consisted of three stages. The first stage involved randomly assigning each participant to one of the four conditions (online counseling or face-to-face counseling and emotion or cognition centered) and having each participant complete the following questionnaires: (a) a demographic questionnaire, (b) the GRCS, and (c) the OCAS. In the second stage, all participants first read a written summary of the client and a description of the first counseling session. Par-ticipants in the face-to-face conditions then listened to a 14-min audio recording of the client’s second counseling session. Participants assigned to one of the two online counseling conditions were given an individual laptop and instructions for how to proceed through the simulated online counseling session. Af-ter listening to or reading through the segment of the second counseling session, all participants were then asked to review a summary of the conclusion of the second counseling session. The final stage, posttest-ing, involved having all participants complete the CAEF (evaluating the counseling approach), the CRF–S (evaluating the counselor), and the OCAS (online counseling attitudes). At the conclusion of the study, participants were given a debriefing statement with an explanation of the purpose of the study and a description of various on-campus counseling and support services.

Results Manipulation Check

Among the 16 doctoral students who participated in the manipulation check, 5 were men and 11 were

women. The ethnic breakdown of the respondents was 11 Caucasians (68.8%), 1 African American (6.3%), 1 Asian American (6.3%), 1 Latino (6.3%), 1 biracial (6.3%), and 1 whose ethnicity was not indi-cated (6.3%). Their average age was 27.19 years (SD

⳱2.91).

A multivariate analysis of variance (MANOVA) was conducted with counseling modality (online vs. face to face) and therapeutic approach (emotion vs. cognition centered) as the independent variables and scores on the Affective–Cognitive Content in Coun-seling Scale and the Counselor Skills Scale as the dependent variables. No main effects or interactions emerged for counselor skills, indicating that the counselor was perceived as being equally skillful and competent across all conditions. The relatively high mean (13.38 of 15,SD⳱1.54) suggests that across all four conditions, the counselor was perceived as having a high level of counseling skill.

As expected, a main effect emerged for approach,

F(1, 12) ⳱ 37.07, p < .001, on the Affective– Cognitive Content in Counseling Scale, indicating the raters regarded the counselors in the two emotion-centered vignettes as focusing primarily on the cli-ent’s emotions. In contrast, the counselor in the cog-nition-oriented vignettes was perceived as focusing on the client’s cognitions.

Preliminary Analyses

Descriptive data. Table 1 shows the means, stan-dard deviations, and intercorrelations for all measures used in the first stage of this study. Mean data for this sample were all within one standard deviation of those reported elsewhere for online counseling atti-tudes (Rochlen et al., in press) and gender role con-flict (Good & Wood, 1995). Regarding the correla-tions, a significant positive correlation emerged Table 1

Means, Standard Deviations, and Intercorrelations for Gender Role Conflict and Preintervention Online Counseling Attitudes Variables M SD 1 2 3 4 5 6 7 1. GRC-T 127.89 25.67 — 2. GRC-SPC 52.65 10.84 .82** — 3. GRC-RE 29.20 8.65 .76** .42** — 4. GRC-RABM 25.36 7.99 .78** .51** .56** — 5. GRC-CWF 20.51 6.05 .59** .43** .30** .23** — 6. OCAS(V) 15.64 5.94 .07 −.02 .15* .02 .08 — 7. OCAS(D) 16.51 5.21 .15* .13 .06 .10 .17* −.04 —

Note. N⳱191. GRC-T ⳱Gender Role Conflict total scale score; GRC-SPC⳱Success, Power, and Competition subscale; GRC-RE⳱Restrictive Emotionality subscale; GRC-RABM⳱Restrictive Affectionate Behavior Between Men subscale; GRC-CWF⳱Conflict Between Work and Family subscale; OCAS(V)⳱Online Counseling Attitudes Scale (Value of Online Counseling subscale); OCAS(D)⳱Online Counseling Attitudes Scale (Discomfort With Online Coun-seling subscale).

between Restrictive Emotionality and online counsel-ing value, indicatcounsel-ing more positive perceptions of the value of online counseling for men with difficulty with emotional expression. In contrast, online coun-seling discomfort was positively correlated with the Conflict Between Work and Family subscale and the total gender role conflict score.

Classification into restrictive emotionality group.

Participants were categorized as having high or low levels of restrictive emotionality based on their re-spective scores on the Restrictive Emotionality sub-scale being above or below the sample median (M⳱

29.20,SD⳱ 8.64). Using this criterion, 94 and 97 participants were categorized as belonging to the low and high restrictive emotionality conditions, respec-tively. A t test conducted between participants as-signed to the low versus high restrictive emotionality levels indicated significant differences in the ex-pected direction (i.e., high with greater restrictive emotionality),t(189)⳱–19.95,p< .001.

Differences between conditions. To address whether differences emerged on the demographic variables (age and ethnicity) and preliminary mea-sures (attitudes toward online counseling and restric-tive emotionality) among participants in the four con-ditions, a MANOVA (for continuous variables) and chi-square analysis (for categorical variables) were conducted. The chi-square analysis for ethnicity was not significant. The overall MANOVA for differ-ences in age, restrictive emotionality, and attitudes toward online counseling was significant,F(4, 186)

⳱3.27,p< .05. However, at the univariate level, no significant results emerged, suggesting roughly

equivalent scores on these indices among participants in the four different conditions.

Primary Data Analyses

Counselor and counseling approach evaluations.

Table 2 presents the means and standard deviations for posttesting counseling approach evaluation, coun-selor evaluation, and posttesting online counseling attitudes by modality, approach, and restrictive emotionality.

To examine the perceptions of the different vi-gnettes, we conducted a MANOVA with modality (online vs. face to face), approach (emotion vs. cog-nition based), and restrictive emotionality (high vs. low) as the independent variables. Dependent vari-ables were overall counseling approach evaluation (using the CAEF) and counselor evaluation (using the composite index on the CRF–S). Main effects will be interpreted first, followed by interactions.

Regarding main effects, no significant findings emerged for counseling approach or restrictive emo-tionality. This suggests that overall, participants with high and low levels of restrictive emotionality ex-pressed roughly equivalent evaluations of the coun-seling approach and counselor characteristics. Simi-larly, no differences emerged between participants assigned to the emotion- versus cognition-focused therapeutic approaches. A significant interaction emerged for Modality × Restrictive Emotionality,

F(2, 182)⳱3.41,p< .05. In examining the interac-tion at the univariate level, a significant effect was found for counseling approach,F(1, 183)⳱6.65,p

Table 2

Means and Standard Deviations for Postintervention Data by Modality, Restrictive Emotionality, and Counseling Approach

Measure

Face-to-face Online

Low RE High RE Low RE High RE

Emotion (n⳱21) M(SD) Cognitive (n⳱23) M(SD) Emotion (n⳱28) M(SD) Cognitive (n⳱24) M(SD) Emotion (n⳱30) M(SD) Cognitive (n⳱20) M(SD) Emotion (n⳱19) M(SD) Cognitive (n⳱26) M(SD) CAEF 27.33 (5.69) 30.35 (4.37) 26.04 (7.17) 25.71 (6.99) 25.00 (8.38) 25.75 (6.89) 27.63 (8.53) 27.77 (6.83) CRF 57.67 (6.58) 57.00 (6.40) 56.02 (7.95) 54.79 (7.48) 56.00 (11.08) 57.20 (6.99) 57.47 (9.73) 55.88 (8.17) OCAS(V) Pretest 16.52 (6.07) 16.61 (5.19) 15.21 (5.94) 14.79 (7.00) 14.67 (5.24) 15.00 (5.66) 15.47 (7.18) 17.04 (5.73) Posttest 17.10 (6.31) 18.85 (5.17) 16.25 (6.40) 15.52 (7.01) 16.50 (6.13) 17.40 (5.19) 18.21 (8.16) 19.12 (5.57) OCAS(D) Pretest 15.76 (4.71) 14.35 (4.66) 18.30 (5.93) 17.71 (4.93) 15.08 (4.59) 17.40 (4.69) 16.53 (6.67) 16.96 (4.80) Posttest 15.24 (4.97) 15.22 (3.97) 18.07 (6.41) 17.38 (5.67) 14.00 (5.38) 16.60 (3.99) 15.53 (6.64) 13.73 (4.54)

Note. Low RE⳱ low scores on the Restrictive Emotionality subscale; High RE⳱ high scores on the Restrictive Emotionality subscale; CAEF⳱Counseling Approach Evaluation Form; CRF⳱Counselor Rating Form—Short Form; OCAS(V)⳱Online Counseling Attitudes Scale (Value of Online Counseling subscale); OCAS(D)⳱Online Counseling Attitudes Scale (Discomfort With Online Counseling subscale).

< .01, but not counselor evaluation. Post hoc pairwise comparisons for counseling approach revealed sig-nificant differences for participants in the face-to-face conditions,F(1, 183)⳱4.26,p< .05, but not the online conditions, F(1, 183) ⳱ 2.52, p < .11. These results suggest that men with low levels of restrictive emotionality reported more favorable re-views of the face-to-face counseling vignettes than men with high levels of restrictive emotionality.

Posttesting attitudes toward online counseling.

To examine improvements in online counseling atti-tudes from participation in the study, we conducted a repeated measures MANOVA with three groups and one within-groups factors. The between-groups factors were modality (online vs. face to face), approach (emotion vs. cognition oriented), and restrictive emotionality (high vs. low). The within-group factor, time, was comprised of the pre- and postadministrations of the Value of Online Counsel-ing and Discomfort With Online CounselCounsel-ing subs-cales. No main effects emerged for the between-groups factors of mode, approach, or restrictive emotionality. This suggested no overall differences in the extent of improvement in attitudes based on assigned condition, approach of the therapist, or level of restrictive emotionality.

A main effect did emerge for time,F(2, 182) ⳱ 19.89, p < .001. At the univariate level, improve-ments in attitudes were observed via an increase in online counseling value,F(1, 183)⳱37.76,p< .001, and a decrease in online counseling discomfort,F(1, 183)⳱ 5.73,p< .05, across the entire sample. In addition, a Mode × Time interaction emerged,F(2, 181)⳱ 3.91,p< .05. Univariate analysis indicated significant findings for online counseling value,F(1, 183)⳱4.07,p< .05, and discomfort,F(1, 183)⳱ 4.96, p< .05. Post hoc pairwise comparisons sug-gested that men assigned to review one of the online counseling conditions reported less discomfort with online counseling at posttesting than participants in the face-to-face counseling conditions,F(1, 183)⳱ 3.80,p< .05. Post hoc analyses indicated no signifi-cant differences between participants assigned to the online versus face-to-face counseling conditions in the posttesting perceptions of online counseling value.

Discussion

The purpose of this project was to evaluate men’s perceptions of online versus face-to-face counseling as conducted using an emotion- versus cognition-based therapeutic approach. Collectively, the results suggest few differences in perceptions between

on-line and face-to-face counseling by the sample of men in the current study. There was some evidence of the appeal of online counseling as compared with face-to-face counseling, particularly for men with self-described discomfort expressing their emotions. Reviewed in further detail below, this evidence emerged from the correlational data, perceptions of the counseling vignettes, and the improvements in online counseling attitudes.

A review of the correlations prior to the interven-tion reveals a mixed relainterven-tionship between several di-mensions of gender role conflict and online counsel-ing attitudes. Men who had higher levels of restrictive emotionality reported a greater apprecia-tion for the value of online counseling than men with lower levels of restrictive emotionality. Importantly, this relationship stands in contrast to the robust pat-terns suggesting that men with various dimensions of gender role conflict, including Restrictive Emotion-ality, report negative help-seeking attitudes (Blazina & Watkins, 1996; Good et al., 1989; Good & Wood, 1995; Wisch et al., 1995). Evidence of this relation-ship does support one commonly cited potential ben-efit of online counseling, namely its appeal to popu-lations who are uncomfortable with verbal expression of feelings (Alleman, 2002; Zack, 2004). However, some of the correlations between gender role conflict and online counseling emerged in the opposite direc-tion, with heightened online counseling stigma re-ported by men with conflict between work and family relations and overall gender role conflict (based on total score). In consideration of the mixed patterns of responses, more research is needed to clarify what may be important disparities between expressed value and discomfort with online counseling by dif-ferent types of men.

Additional important information regarding men’s perceptions toward online counseling emerged from participants’ ratings of the various counseling vi-gnettes. Overall, the participants’ evaluations were not significantly different across modality (online vs. face to face), approach (emotion vs. cognition cen-tered), or restrictive emotionality (high vs. low lev-els). These findings parallel recent research demon-strating similar and positive evaluations of the process and outcome of online as compared with face-to-face counseling (Cook & Doyle, 2002; Day & Schneider, 2002). Contrary to our prediction, there were no significant differences between the evalua-tions of the cognition- and the emotion-centered vi-gnettes, regardless of the level of restrictive emotion-ality. These findings do not support research that has suggested the appeal of cognition-focused orienta-tions for men with traditional gender role orientaorienta-tions

(Hurst, 1997; Wisch et al., 1995). In interpreting this discrepancy, one possible explanation is that the use of audio recordings in our study was less effective in accentuating the differences between the two thera-peutic approaches (as contrasted with video record-ings). An alternative interpretation is that men with differing levels of restrictive emotionality appear to be equally comfortable with cognition- and emotion-centered approaches to counseling. This interpreta-tion is supported by a distincinterpreta-tion between the current study and past research designs. In the current study, participants directly evaluated the counselor and the counseling approach used in the vignettes, whereas past studies utilized help-seeking measures as the de-pendent variables. Further research is needed to clarify the types of therapeutic approaches to coun-seling resistant men might prefer.

Our hypothesis regarding evaluations of online counseling by men with high levels of restrictive emotionality received mixed support. We predicted that men with high levels of restrictive emotionality would evaluate the online counseling conditions (both approach and counselor) more favorably than the face-to-face counseling conditions. Men who had more discomfort expressing their emotions had less favorable reviews of the counseling approaches in the face-to-face conditions than men with less discomfort expressing their feelings. However, men with high and low levels of restrictive emotionality had ap-proximately equal ratings of the counselor approach in the online conditions. In other words, men with high as opposed to low levels of restrictive emotion-ality reported less favorable evaluations of face-to-face counseling and roughly equal evaluations of on-line counseling. Considering the call for validation of a greater range of nonverbal counseling options (Fischer & Good, 1997; Heesacker & Prichard, 1992), these preliminary findings are encouraging.

In contrast to the above results regarding evalua-tions of the counseling approaches, there were no significant findings relevant to counselor evaluations. One possible explanation for the difference in results between the assessment of the counseling approach and that of the counselor is that the participants were able to draw a distinction between the personal at-tributes (e.g., level of trustworthiness) of the coun-selor and the approach to counseling (which focuses more on his theoretical orientation in counseling). Hence, some participants might have had positive impressions of the counselor as a person and yet not have liked the approach to counseling. This might account for why the participants with high and low levels of restrictive emotionality had similar evalua-tions of the counselor, but men with greater

restric-tive emotionality had poorer evaluations of the coun-seling approach in the face-to-face condition compared with men with low levels of restrictive emotionality.

A promising conclusion from the study was the generally positive impact of participation in the study on attitudes toward online counseling. Our data sug-gest that overall, participation in the project led to increases in reported levels of value of online coun-seling and decreases in online councoun-seling discomfort. These findings are consistent with past studies dem-onstrating that providing men with informative por-trayals of psychotherapy can positively influence their opinions, attitudes, and expectations about vari-ous help-seeking services (Blazina & Marks, 2001; Gonzalez et al., 2002; Rochlen et al., 2002).

However, we had expected that participants as-signed to the online condition would evidence more substantial improvements in both value and discom-fort with online counseling. The findings did indicate that participants in the online counseling conditions experienced greater decreases in discomfort with on-line counseling but equivalent levels of improvement in the perceived value of online counseling than par-ticipants in the face-to-face counseling conditions. These results suggest that the primary benefit of our intervention could have been the exposure of partici-pants to the process of how an online counseling session might unfold, thus normalizing the experi-ence of online counseling for them. This exposure seemed to have had the effect of decreasing the par-ticipants’ discomfort with online counseling but not yielding substantially higher levels of perceived value. Perhaps more experiential interventions would yield greater increases in expressed level of value with the process of online counseling.

Of importance, there are several limitations in this study that suggest the need for further research. For one, the participants in this study were exposed to portrayals of online and face-to-face counseling and were not involved in the counseling process. Future research should evaluate whether and how different populations benefit from actual participation in coun-seling using varied therapeutic techniques and/or mo-dalities. Such research may want to include further evaluation of online counseling as well as a range of therapeutic interventions and modalities that may be more appealing and/or effective for counseling resis-tant men (e.g., coaching, workshops, self-help manu-als). Second, the current study is limited by its focus on the general dichotomy between cognition- versus emotion-centered approaches to counseling. More commonly practiced in the field are a wide range of theoretical approaches that can be empirically

evalu-ated in studies using similar methodologies. Third, our study did not examine other types of distance therapy apart from a chat-based approach to online counseling. Future research should compare and evaluate varied forms of distance therapy, such as the use of video conferencing and e-mail. Fourth, it is possible that a range of other variables not included in the current study influenced perceptions of online versus face-to-face counseling. Variables that should be considered in future research include expectations and/or knowledge of online counseling, familiarity with using technology, use of the Internet, and even typing speed. Finally, the generalizability of our study is further limited by the homogenous nature of the sample, consisting of primarily Caucasian, col-lege-aged men with moderate levels of gender role conflict and restrictive emotionality. Future research should study more diverse populations in regard to age, ethnicity, restrictive emotionality, and online counseling attitudes.

References

Addis, M. E., & Mahalik, J. R. (2003). Men, masculinity, and the contexts of help seeking.American Psychologist, 58, 5–14.

Alleman, J. R. (2002). Online counseling: The Internet and mental health treatment.Psychotherapy, 39, 199–209. Barak, A., & LaCrosse, M. B. (1977). Comparative

percep-tion of practicum counselor behavior. A process and methodological investigation.Counselor Education and Supervision, 16, 202–208.

Betz, N. E., & Fitzgerald, L. F. (1993). Individuality and diversity: Theory and research in counseling psychology. Annual Review of Psychology, 44, 343–381.

Blazina, C., & Marks, L. I. (2001). College men’s affective reactions to individual therapy, psychoeducational work-shops, and men’s support group brochures: The influence of gender-role conflict and power dynamics upon help-seeking attitudes. Psychotherapy: Theory, Research, Practice, Training, 38, 297–305.

Blazina, C., & Watkins, C. E. (1996). Masculine gender role conflict: Effects on college men’s psychological well-being, chemical study usage, and attitude toward help-seeking.Journal of Counseling Psychology, 43, 461–465. Bloom, J. W. (1998). The ethical practice of Web counsel-ing. British Journal of Guidance and Counseling, 26, 53–59.

Brooks, G. B., & Gilbert, L. A. (1995). Men in families: Old constraints, new possibilities. In R. Levant & W. Pollack (Eds.),The new psychology of men(pp. 252–279). New York: Basic Books.

Brooks, G. R., & Good, G. E. (Eds.). (2001). The new handbook of psychotherapy and counseling with men. San Francisco: Jossey-Bass.

Cook, J. E., & Doyle, C. (2002). Working alliance in online

therapy as compared to face-to-face therapy: Preliminary results.Cyber Psychology and Behavior, 5(2), 95–105. Corrigan, J. D., & Schmidt, L. D. (1983). Development and

validation of revisions in the Counselor Rating Form. Journal of Counseling Psychology, 30, 64–75. Cournoyer, R. J. (1994). A developmental study of gender

role conflict in men and its changing relationship to psy-chological well-being (Doctoral dissertation, Boston Col-lege, 1994). Dissertation Abstracts International, 54, 6476.

Cournoyer, R. J., & Mahalik, J. R. (1995). Cross-sectional study of gender role conflict examining college-aged and middle aged men.Journal of Counseling Psychology, 42, 11–19.

Courtenay, W. H. (2000).Constructions of masculinity and their influence on men’s well-being.Social Science and Medicine, 50, 1385–1401.

Day, S. X., & Schneider, P. L. (2002). Psychotherapy using distance technology: A comparison of face-to-face, video, and audio treatment.Journal of Counseling Psy-chology, 49, 499–503.

Fischer, A. R., & Good, G. E. (1997). Men and psycho-therapy: An investigation of alexithymia, intimacy, and masculine gender roles.Psychotherapy, 34, 160–170. Gonzalez, J. M., Tinsley, H. E., & Kreuder, K. R. (2002).

Effects of psychoeducational interventions on opinions of mental illness, attitudes towards help seeking, and expec-tations about psychotherapy in college students.Journal of College Student Development, 43, 51–63.

Good, G. E., Dell, D. M., & Mintz, L. B. (1989). Male role and gender role conflict: Relations to help seeking in men.Journal of Counseling Psychology, 36, 295–300. Good, G. E., Gilbert, L. A., & Scher, M. (1990). Gender

aware therapy: A synthesis of feminist therapy and knowledge about gender.Journal of Counseling and De-velopment, 68, 376–380.

Good, G. E., & Wood, P. K. (1995). Male gender role conflict, depression, and help-seeking: Do college men face double jeopardy?Journal of Counseling and Devel-opment, 74, 70–75.

Guinee, J. P., Tracey, T. J., & Terence, J. G. (1997). Effects of religiosity and problem type on counselor.Journal of Counseling and Development, 76, 65–73.

Heesacker, M., & Prichard, S. (1992). In a different voice, revisited: Men, women, and emotion.Journal of Mental Health Counseling, 14, 274–290.

Heinlen, K. T., Welfel, E. R., Richmond, E. N., & Rak, C. F. (2003). The scope of Web counseling: A survey of services and compliance with NBCC standards for the ethical practice of Web counseling.Journal of Counsel-ing and Development, 81, 61–69.

Heppner, P. P., & Gonzales, D. S. (1987). Men counseling men. In M. Scher, M. Stevens, G. Good, & G. A. Eichen-field (Eds.),Handbook of counseling and psychotherapy with men(pp. 30–38). Newbury Park, CA: Sage. Hurst, M. A. (1997).The best fit in counseling men: Are

there solutions to treating men as the problem? Unpub-lished doctoral dissertation, Ball State University, Mun-cie, IN.

Kraus, R., Zack, J. S., & Stricker, G. (2004).Online coun-seling: A handbook for mental health professionals.San Diego, CA: Academic Press.

Levant, R. F. (2001). Desperately seeking language. In G. R. Brooks & G. E. Good (Eds.),The new handbook of psychotherapy and counseling with men(pp. 424–443). San Francisco: Jossey-Bass.

Lyddon, W. J. (1989). Personal epistemology and prefer-ence for counseling.Journal of Counseling Psychology, 35, 423–429.

Lyddon, W. J., & Adamson, L. A. (1992). Worldview and counseling preference: An analogue study. Journal of Counseling and Development, 71, 41–47.

O’Neil, J. M. (1981a). Male sex role conflicts, sexism, mas-culinity: Psychological implications for men, women, and the counseling psychologist.The Counseling Psycholo-gist, 9, 61–81.

O’Neil, J. M. (1981b). Patterns of gender role conflict and strain: Sexism and fear of femininity in men’s lives. Per-sonnel and Guidance Journal, 60, 203–210.

O’Neil, J. M., Good, G. E., & Holmes, S. (1995). Fifteen years of theory and research on men’s gender role con-flict: New paradigms for empirical research. In R. Levant & W. Pollack (Eds.),The new psychology of men(pp. 164–206). New York: Basic Books.

O’Neil, J. M., Helms, B. J., Gable, R. K., David, L., & Wrightsman, L. S. (1986). Gender Role Conflict Scale: College men’s fear of femininity.Sex Roles, 14, 335–350. Robertson, J. M., & Fitzgerald, L. F. (1992). Overcoming the masculine mystique: Preferences for alternative forms of assistance among men who avoid counseling.Journal of Counseling Psychology,39, 240–246.

Rochlen, A. B., Beretvas, S. N., & Zack, J. S. (in press). Development and validation of the on-line and face-to-face Counseling Attitude Scales. Measurement and Evaluation in Counseling and Development.

Rochlen, A. B., Blazina, C., & Raghunathan, R. (2002). Gender role conflict, attitudes toward career counseling,

career decision-making, and perceptions of career coun-seling advertising brochures.Journal of Men and Mas-culinity, 3, 127–137.

Rochlen, A. B., & O’Brien, K. M. (2002). The relation of male gender role conflict and attitudes toward career counseling to interest in and preferences for career coun-seling styles. Psychology of Men and Masculinity, 3, 9–21.

Rochlen, A. B., Zack, J. S., & Speyer, C. (2004). Online therapy: Review of relevant definitions, debates, and cur-rent empirical support.Journal of Clinical Psychology, 60, 269–283.

Scher, M. (1981). Men in hiding: A challenge for the coun-selor. The Personnel and Guidance Journal, 60, 199– 202.

Sharpe, M. J., & Heppner, P. P. (1991). Gender role, gender role conflict, and psychological well-being in men. Jour-nal of Counseling Psychology, 38, 323–330.

Stofle, G. S. (2001).Choosing an online therapist. Harris-burg, PA: White Hat Communications.

Tracey, T. J., Glidden, C. E., & Kokotovic, A. M. (1988). Factor structure of the Counselor Rating Form—Short. Journal of Counseling Psychology, 35, 330–335. Wilcox, D. W., & Forrest, L. (1992). The problems of men

and counseling: Gender bias or gender truth?Journal of Mental Health Counseling, 14, 291–303.

Wisch, A. F., Mahalik, J. R., Hayes, J. A., & Nutt, E. A. (1995). The impact of gender role conflict and counseling technique on psychological help-seeking in men. Sex Roles, 33, 77–89.

Zack, J. S. (2004). The technology of online counseling. In R. Kraus, J. Zack, & G. Stricker (Eds.),Online counsel-ing: A handbook for mental health professionals(pp. 93– 121). San Diego, CA: Elsevier Academic Press.

Received November 3, 2003 Revision received January 28, 2004