chao2012

(1)

Research

Received 10/19/09 Revised 11/14/10 Accepted 01/11/11

Managing Perceived Stress Among College Students: The Roles of Social Support

and Dysfunctional Coping

Ruth Chu-Lien Chao

The author examined the conditions (i.e., social support and dysfunctional coping) under which perceived stress predicted psychological well-being in 459 college students. Hierarchical regression analyses indicated a significant 2-way interaction (Perceived Stress × Social Support) and a significant 3-way interaction (Perceived Stress × Social Support × Dysfunctional Coping) predicting well-being. Low social support deteriorated the association between stress and well-being. Only the frequent use of dysfunctional coping exacerbated the association between stress and well-being across high and low social support. Implications for counseling college students are discussed.

Keywords: stress, coping, well-being

F

or college students, life is full of stress (Roberti, Harrington, & Storch, 2006). Life often throws students off track with unexpected breakups, difficult tasks beyond their capabilities, and other situations that suddenly force them to make decisions about their future. All of these events can stress them enormously. College stressors have wide varieties, from academic work to uncertainty about the future, from difficulties in interpersonal relationships to dating problems, from self-doubt to family issues, and the list goes on.

Recent reports show that college students’ stresses have increased in sever- ity (Benton, Robertson, Tseng, Newton, & Benton, 2003). For students to manage their perceived stress, positive social support is as essential as good soil is to plants. Besides, useful coping is a tool to handle stress. Specifically, although students typically live under stress, some seem to manage stress better than their peers do. In the face of stress, they naturally seek support from families and friends and execute their coping to maintain well-being.

Although many researchers have reported that social support and coping are positively associated with well-being (Ben-Zur, 2009), there are two pitfalls in applying these associations to students. First, social support has been decreasing in the past decade (Arria et al., 2009); thus, it is inevitable to see how low social support may change the association between stress and well-being among students. Second, not all coping strategies are effective (Carver, Scheier, & Weintraub, 1989), because coping can be functional or dysfunctional. Almost all the previous literature considered only positive cop-

Ruth Chu-Lien Chao, Counseling Psychology Program, University of Denver. Correspondence concerning this article should be addressed to Ruth Chu-Lien Chao, Counseling Psychology Program, Morgridge College of Education, University of Denver, 1999 East Evans Avenue, Denver, CO 80208 (e-mail: [email protected]).

(2)

ing. Consequently, researchers’ one-sided consideration misses the fact that an increasing number of college students actually use dysfunctional coping (e.g., avoidance) to manage stress (Carver & Scheier, 1994). Thus, if low social support and dysfunctional coping are not taken into consideration, it is difficult to understand how some college students handle their perceived stress. Because of the necessity of adding the two concepts to more accurately describe students, the purpose of this study was to examine the moderating roles of social support and dysfunctional coping on the association between stress and well-being.

Social Support and Well-Being

If asked about how they manage stress, many college students would mention seeking support from people or the environment. Indeed, social support shows a long list of benefits (Ben-Zur, 2009; Lundberg, McIntire,

& Creasman, 2008) contributing to happiness and satisfaction with life.

Social support has been found to be negatively related to mental problems (Brown, Alpert, Lent, Hunt, & Brady, 1988), which are associated with stress. The relationship between social support and the well-being of college students has been well documented by numerous studies (e.g., Solberg

& Villareal, 1997; Wang & Castañeda-Sound, 2008). However, because social support has decreased in the past decade (Arria et al., 2009), what is missing is a consideration of how low social support affects the management of stress (Curran, Totenhagen, & Serido, 2010). In other words, college students with low support may also lack a buffer against stress. For college students, low social support is associated with various issues, such as parental divorce (Arria et al., 2009), a lack of friends (Curran et al., 2010), or an inadequate social life (Sirgy, Lee, & Bae, 2006). As a result of the rapid changes of today’s society, college students’ perceptions of social support have shifted from viewing their supports as stable to seeing them as variable and fluctuating when students need help most (Daniel, Evans, &

Scott, 2001). Perhaps available supports enable some students to maintain a relatively positive mental health despite stress. Still, students with low social support were found more likely to engage in less healthy activities, such as sedentary behavior, alcohol use, and too much or too little sleep (Thorsteinsson & Brown, 2008). Worse, college students’ lack of social support was found more likely to be related to life dissatisfaction or even suicidal behavior (Allgower, Wardle, & Steptoe, 2001). In an interview of 1,249 college students, 6% reported having suicidal ideation, and low social support was a predictor to suicidal ideation (Arria et al., 2009).

The aforementioned empirical evidence shows that, when perceiving insuf- ficient social support, college students would lack a buffer against life stress that deteriorates well-being. The level of social support could be the first moderator between stress and well-being. That is, when encountering life stress, the college students who have high social support may have a buffer

(3)

to moderate the association between stress and well-being, and those with low social support would lack the buffer against stress.

A Three-Way Interaction Among Perceived Stress, Social Support,

and Dysfunctional Coping

A particularly useful lens for understanding the potential protective nature of social support and coping is Lazarus and Folkman’s (1984) paradigm on stress, appraisal, and coping. This paradigm consists of perceived stress (a relationship between the person and environment that is appraised as exceeding available resources), appraisal (one’s perception and assessment of social support in the situation), and coping (effortful or purposeful thoughts and actions to manage or overcome stressful situations; Frydenberg, 1997; Lazarus & Folkman, 1984). Social support, according to Lazarus and Folkman, is a resource in the social environment, and people appraise their resources when they encounter stress. Lazarus and Folkman specified that the effectiveness of a coping strategy depends on the extent to which it is appropriate to the internal and/or exter- nal demands of the situation. Carver et al. (1989) further explained that some responses to stress may be maladaptive. Specifically, the tendency to focus only on venting whatever distress one is experiencing may be less useful to resolve stress. Thus, Lazarus and Folkman’s conceptual framework seems appropriate for understanding college students’ stress, social support, and coping.

The college years are the period of emerging adulthood (18–25 years of age). These years are years of change and exploration, a crucial time of identity development (Arnett, 2000). Such change and exploration are embroiled in various stresses, such as leaving one’s family, making new friends, adjusting to a new academic life, dating, deciding on one’s major, and dealing with financial pressures. Thus, these students experience a variety of stresses in an entirely new social environment characterized by greater freedom and less adult super- vision than in their previous years (Lefkowitz, 2005), homesickness, a sense of isolation, and increased interpersonal conflict (Buote et al., 2007). These radical transitions during emerging adulthood may require them to appraise their resources of social support and coping strategies to manage stress. Yet, if asked how they cope with stress, some students would typically answer that they would play games or give up whatever they were pursuing (Brougham, Zail, Mendoza, & Miller, 2009). Is sleep a coping behavior? Yes, it is, although sleeping away stress is a dysfunctional coping behavior (Carver et al., 1989).

Coping is multidimensional and involves varied strategies ranging from overt behaviors to covert mental attitude. Some coping strategies are functional, whereas others are less so (Carver et al., 1989). Dysfunctional coping refers to the strategies with little or no effectiveness (Roth & Cohen, 1986).

Dysfunctional coping includes three aspects: (a) focusing on and venting of emotions, (b) behavioral disengagement, and (c) mental disengagement (Carver et al., 1989). Focusing on and venting of emotions indicates how

(4)

distress is central in emotions without adaptive behaviors. When individuals disengage behaviorally, they stop trying to deal with stress, and the stress still remains. Mental disengagement refers to the effort to put stress out of sight by engaging in various activities, as in the saying “out of sight, out of mind.”

Dysfunctional coping is indispensable to understanding college students. It is prevalent among today’s college students (Brougham et al., 2009). Unfor- tunately, despite the failure of dysfunctional coping strategies, these strategies still prevail among those students. For example, Sideridis (2008) found that the five most frequent coping strategies among students to reduce stress were browsing the Internet, sleeping and resting, using instant messaging, complaining, and watching TV or movies.

After engaging in dysfunctional coping, many students can still feel anxious because their stress is not resolved (Carver et al., 1989). Students may still not be satisfied, because such dysfunctional coping has not solved their stress but rather has merely relieved it for a while. In their study, Carver et al. (1989) found that high dysfunctional coping would exacerbate the association between stress and well-being and low dysfunctional coping would make no difference on the association because students did not have much dysfunctional coping.

The Present Study

The purpose of this study was to examine how college students handle stress and to explore whether dysfunctional coping exacerbates the association between stress and well-being for students with different levels of social support.

It is hoped that the findings of the present study will assist college students to deal with stress and help advance knowledge on dysfunctional coping in the face of life stress. Thus, the hypotheses were as follows:

Hypothesis 1: Social support moderates the relationship between students’

perceived stress and well-being. The greater social support students perceive, the more the support will buffer the effect of stress on well- being. However, the low social support perceived by students will exacerbate the negative association between stress and well-being.

Hypothesis 2: A three-way interaction of perceived stress, social support, and dysfunctional coping in predicting well-being exists in such a way that the association between perceived stress and well-being will be weakest when social support is low and the dysfunctional coping is high.

Method

Participants

A sample of 459 college students at a predominantly European American, large, public university in the Midwest completed an online survey. Of the participants, 390 (85%) were European American, 32 (7%) were African American, 23 (5%)

(5)

were Latina/o, and 14 (3%) were Asian American. The racial/ethnic ratio of this campus was 80% European American, 9% African American, 6% Latina/o, 3% Asian American, and 2% international students. Among the participants, there were 239 (52%) men and 220 (48%) women, with a mean age of 20.23 years (SD = 3.45; range = 18–35 years). With respect to academic status, 120 (26%) of the participants were freshmen, 115 (25%) were sophomores, 110 (24%) were juniors, and 114 (25%) were seniors. Twelve percent (n = 55) of the sample reported having had therapy or counseling.

Instruments

Perceived Stress Scale (PSS; S. Cohen, Kamarck, & Mermelstein, 1983). The PSS is a 10-item measure designed to assess the degree to which individuals perceive their lives as stressful. Respondents rate each item on a 5-point Likert-type scale ranging from 1 (not at all satisfied) to 5 (very satisfied). Higher scores indicate greater perceptions of life stress; lower scores reflect lower perceptions of stress. The PSS showed adequate coefficient alphas (.84 and .85 for two college samples; S. Cohen et al., 1983); in this study, the coefficient alpha was .85. The PSS has been positively correlated with life-event scores, depressive and physical symptomatology, social anxiety, and maladaptive health-related behaviors (e.g., increased smoking; S. Cohen, Sherrod, & Clark, 1986).

These pieces of evidence indicate the construct and concurrent validity of the PSS. Furthermore, Kuiper, Olonger, and Lyons (1986) found the PSS to be associated with people’s greater vulnerability to depressive symptoms related to stressful life events.

Social Support Inventory (SSI; Brown et al., 1988). Social support was mea- sured with the SSI. It contains 39 statements on interpersonal needs (e.g., assistance in changing self-defeating behavior) for respondents to rate on two scales: (a) Need Strength (“How much of this type of help or support have you needed in the past month?”) and (b) Perceived Supply (“How much of this type of help or support have you received in the past month?”). The items for these two scales are scored on a 7-point Likert-type scale ranging from 1 (none) to 7 (very much). A Perceived Fit score is obtained by subtracting Perceived Supply ratings from Need Strength ratings and summing them across all 39 items. The SSI also contains a direct measure of satisfaction where respondents rate each Need Strength statement in terms of “How satisfied have you been with what you have received in terms of this type of help or support over the past month?” (1 = not at all to 7 = very much).

Higher scores indicate higher levels of perception and satisfaction with the support received. The focus of the present study was on college students’

levels of satisfaction with the social support they had received. The total Subjective Satisfaction score is obtained by summing ratings across all 39 items. The coefficient alpha was .95 in Brown et al.’s (1988) study and .90 in the current sample. The SSI has appropriate construct validity, being positively related to other social support measures and negatively related to depression measures (Brown et al., 1988).

(6)

COPE (Carver et al., 1989). The COPE is a 60-item self-report measure designed to assess different ways of responding to stress. The items are scored on 4-point Likert-type scale ranging from 1 (I usually don’t do this at all) to 4 (I usually do this a lot). The COPE assesses 13 patterns of coping, including dysfunctional coping, such as focus on and venting of emotions, denial, behavioral disengagement, and mental disengagement. High scores of particular patterns indicate a high tendency to use these coping patterns when experiencing stress. The focus of this study was on dysfunctional coping, which had a coefficient alpha of .77 as reported by Carver et al. (1989) and a coefficient alpha of .85 in the current sample. Convergent validity has been demonstrated with numerous personality measures, including optimism, control, self-esteem, internality, hardiness, self-monitoring, and anxiety. The COPE has also been used to assess strategies that individuals use when coping with specific life events (Carver & Scheier, 1994). Previous researchers have demonstrated that the COPE fitted the original factor structures well, with adequate convergent and discriminant validity. The COPE has been shown to correlate with various measures, including hassles and uplifts, physical symptoms, degrees of satisfaction with life, positive affectivity, and negative affectivity (Clark, Borman, Cropanzano, & James, 1995).

Mental Health Inventory (MHI; Veit & Ware, 1983). The MHI is a 38-item measure assessing both distress and well-being among adults. Ratings are made on a 5-point Likert-type scale. The MHI produces three indices: Psychological Distress, Psychological Well-Being, and Global Mental Health. The present study used the Psychological Well-Being index, which Veit and Ware (1983) reported as appropriate to use as an outcome measure. The MHI has been used extensively in studies of nonpsychiatric samples, and its psychometric adequacy has been well established (Siegel, Karus, Raveis, & Hagen, 1998;

Veit & Ware, 1983). The current sample was found to have a coefficient alpha of .89. Concurrent and convergent validity for the MHI Psychological Well-Being index have been established through positive correlations with measures of positive affect (Siegel et al., 1998).

Demographic questionnaire. The demographic questionnaire included ques- tions regarding race/ethnicity, age, gender, year in college, and whether the participant had previous experience with therapy or counseling.

Procedure

A research assistant announced the study project in class and then e-mailed reminders to 500 students in Introductory Psychology, History of Psychology, and Assessment courses. The participants were mostly psychology majors. All the participants were older than 18 years. Participants answered an online questionnaire package containing the PSS, SSI, COPE, MHI, and demographic questionnaire. Immediately after the completion of the questionnaire package, participant data were “dumped” to a password-protected data file on a network computer. Participants’ log-in details were immediately dumped

(7)

to a separate file; these files could not be combined to identify individual participants’ responses. However, the log-in data file was used to verify that individual student participated only once.

Of the 485 students who agreed to participate in the study, 479 completed the survey and an additional 20 responded to a validity item inaccurately. Thus, the completed data of 459 participants were used in the current analyses (a response rate of 92%). Those who agreed to participate received an invitation e-mail with a survey link. Participants were informed that the goal of the study was to identify factors that can change well-being among college students.

In the online survey, it was indicated to the participants that by completing the survey, they had consented to take part in the study. After the survey was completed, a debriefing form was provided. Also, participants could provide their contact information, which was stored in a separate data file, to enter a drawing for one of ten $20 cash prizes.

Results

Preliminary Analyses and Descriptive Statistics

Table 1 shows descriptive statistics, internal consistency reliability estimates, and correlations among the variables of interest. The zero-order correlations among the variables indicated that both perceived stress and dysfunctional coping were related negatively to psychological well-being, but social support was related positively to psychological well-being. A series of analyses of variance was then conducted on psychological well-being to examine whether the dependent variable was a function of participants’ race/ethnicity, gender, year in college, and previous experience with therapy or counseling.

None of these demographic variables had a significant effect on psychological well-being: race/ethnicity, F(4, 455) = 0.78, p > .05; gender, F(1, 458) = 0.69, p > .05; year in college, F(5, 454) = 0.34, p > .05; and previous experience with therapy or counseling, F(1, 458) = 1.26, p > .05. Because people of different races/ethnicities may or may not have different levels of well-being (Keith & Brown, 2010), a series of six pairs of t tests was conducted for the

*p < .05. **p < .01. ***p < .001.

Table 1

Means, Standard Deviations, and Correlations for the Study Variables

Variable 1. Age

2. Perceived stress 3. Social support 4. Dysfunctional coping 5. Psychological well-

being

2

1 3 4

— .08 –.09 .02 .02

–.12*— .41**

–.19**

— –.01 .52***

— –.23**

M

5 SD a

—

20.23 1.59 3.81 2.50 4.20

3.45 0.45 0.71 1.63 1.02

.85 .90 .85

.89

(8)

four racial/ethnic groups. All the effect sizes of mean differences ranged from .03 to .09; these effect sizes are below the minimum effect size (J. Cohen, 1988). These two pieces of evidence indicate that race/ethnicity has no significant relationship to psychological well-being. As another preliminary check, J. Cohen’s (1992) power analysis was used to ensure that the present study had adequate power for its analysis. It was found that, in order to reach a power of .80 when α = .01, the sample size should be at least 388 to have a small population sample effect (see J. Cohen, 1992, p. 157). This study had a sample size of 459, which exceeded J. Cohen’s (1992) criterion.

Additionally, a correlation analysis between age and psychological well-being was conducted and no significant relationship between age and psychological well-being was found (r = .02, p = .52). Thus, in general, the dependent variable did not differ significantly in terms of these demographic variables (all ps > .05). Therefore, none of these variables were included in the multiple regression analyses.

Another issue with multiple regression analysis is normality, because substantial departures from normality can adversely affect the analyses; thus, the data need to meet the regression assumptions of normality (J. Cohen, Cohen, West, & Aiken, 2003). Results of the multiple regression analyses indicated that the skewness of the residuals ranged from –.17 (Z = –1.33, p > .05) to –.19 (Z = –1.54, p > .05) and the kurtosis of the residuals ranged from .54 (Z

= 1.65, p > .05) to .57 (Z = 1.68, p > .05). The results indicated no statistical departure from normality.

Moderation Analyses

To test the hypotheses on the moderating roles of social support (Hypothesis 1) and dysfunctional coping (Hypothesis 2) in the link between perceived stress and psychological well-being, I followed Dawson and Richter’s (2006) recommendation and used hierarchical multiple regression analyses to test moderator effects. I followed Aiken and West’s (1991) suggestion and used centered variables (i.e., mean deviation scores) to reduce multicollinearity between interaction terms and main effects when testing for moderator effects. In the analyses, I entered main effects (Step 1), two-way interaction terms (Step 2), and three-way interaction terms (Step 3).

Hypothesis 1: Perceived stress, social support, and Perceived Stress × Social Support.

For Hypothesis 1, Step 1 indicated that perceived stress, social support, and dysfunctional coping accounted for 40% of psychological well-being. In Step 2, the two-way interactions significantly predicted psychological well-being (ΔR² = .01, p < .05). In addition, the regression coefficient for the two-way interaction of Perceived Stress × Social Support was statistically significant (see Table 2). Champoux and Peters (1987) and Chaplin (1991) compre- hensively reviewed the social science literature and reported that interaction terms typically account for approximately 1% to 3% of the variance, although J. Cohen (1992) indicated that an R² value of .01 indicates a small effect

(9)

size. In Step 3, a three-way interaction was found to significantly contribute to the variance of psychological well-being (ΔR² = .01, p < .05). Again, this result is consistent with what is typically accounted for by interaction terms (i.e., approximately 1% to 3% of the variance), according to Champoux and Peters’s and Chaplin’s reviews of the social science literature.

After finding a significant two-way interaction effect, I then investigated the simple effect of the interaction by plotting SSI scores for PSS scores of one standard deviation above and below the mean (Aiken & West, 1991).

To check whether the slopes of the simple regression lines at high and low social support significantly differed from zero, I conducted a simple regression analysis as outlined by Aiken and West (1991). In this analysis, the criterion variable (i.e., psychological well-being) was regressed on the predictor (i.e., perceived stress), the moderator (i.e., social support) at a conditional value (e.g., high or low), and the Predictor × Moderator interaction. The t test for the regression coefficient of the predictor variable in this equation did reflect the significance of the simple slope (i.e., whether the slope is significantly different from zero).

Figure 1 shows that the relationship between perceived stress and psychological well-being was significant and negative for the low social support slope (b

= –2.80, p < .001) but nonsignificant for the high social support slope. These results indicate that college students who perceived low social support were more vulnerable to stress than were those who perceived high social support.

Low social support was one standard deviation (–1 SD) below the mean, and high social support was one standard deviation (+1 SD) above the mean. The difference between these two regression lines was also significant, as indicated by the significant regression coefficients found for the interaction terms in

Table 2

Hierarchical Multiple Regression analysis Predicting Psychological Well-being From Perceived Stress, Social Support, Dysfunctional

Coping, and Their Interaction (N = 459)

Step and Variable Step 1

Perceived stress Social support Dysfunctional coping Step 2

Perceived Stress × Social Support Perceived Stress × Dysfunctional Coping

Social Support × Dysfunctional Coping

Step 3

Perceived Stress × Social Support × Dysfunctional Coping

B SE B b R² DR² –1.32

6.98 –0.96 –0.01 0.23 0.21

0.79 .47 .44 .47 .00 .40 .44

.38 –.12**

.60***

–.08*

–.57**

.02 .02

.09*

.41

.42

.43 .40

.01

.01 DF 95.58***

2.88*

4.32*

df (3, 423)

(3, 420)

(1, 419)

*p < .05. **p < .01. ***p < .001.

(10)

the tests of the moderator effects, according to the procedure recommended by Aiken and West (1991).

Hypothesis 2: Perceived stress, social support, dysfunctional coping, and their interactions. Table 2 shows that, in Step 3, a three-way interaction was found to significantly contribute to the variance of psychological well-being (ΔR² = .01, p < .05). After seeing that Step 3 showed a significant three-way interaction, I then used the same procedure described earlier to plot the three-way interaction. As illustrated in Figure 2, among students who reported high levels of social support, the association between perceived stress and psychological well-being was significantly different from zero for those who tended to use high levels of dysfunctional coping (b = –1.99, p < .05) but was not different for those who used low levels of dysfunctional coping.

In contrast, the patterns of interaction among those with low levels of social support tell quite a different story (see Figure 3). The association between perceived stress and psychological well-being was significantly different from zero both among those who tended to use high levels of dysfunctional coping (b = –3.82, p < .001) and among those who used low levels of dysfunctional coping (b = –1.86, p < .05).

These results indicate that college students who reported low social support and frequent use of dysfunctional coping were most vulnerable to stress.

Discussion

In this study, the conditions (i.e., social support and dysfunctional coping) under which perceived stress predicts college students’ psychological well-being were

FIguRe 1

Interaction effects of Perceived Stress and Social Support on Psychological Well-being

***p < .001.

49 — 48 — 47 — 46 — 45 — 44 — 43 — 42 —

Low (1 SD Below) High (1 SD Above) Perceived Stress

—

Psychological Well-Being —

= High Social Support = Low Social Support Two-Way Interaction: Perceived Stress × Social Support

b = –0.52

b = –2.80***

(11)

examined. The current results supported both Hypotheses 1 and 2. Specifically, the results for Hypothesis 1 showed a significant two-way interaction between perceived stress and social support. When experiencing stress, college students need assurance that others (e.g., parents, friends) are willing to listen and talk.

= Low Dysfunctional Coping

= High Dysfunctional Coping 48 —

47 — 46 — 45 — 44 — 43 — 42 — 41 —

—

Three-Way Interaction: High Social Support

b = 0.33

b = –1.99*

FIguRe 2

at High Social Support, the Interaction effects of Perceived Stress and Dysfunctional Coping on Psychological Well-being

*p < .05.

= Low Dysfunctional Coping

= High Dysfunctional Coping 50 —

49 — 48 — 47 — 46 — 45 — 44 — 43 — 42 — 41 — 40 —

—

Three-Way Interaction: Low Social Support

b = –1.86*

b = –3.82***

FIguRe 3

at low Social Support, the Interaction effects of Perceived Stress and Dysfunctional Coping on Psychological Well-being

*p < .05. ***p < .001.

(12)

Friends, family, and significant others can provide instrumental, informational, or emotional assistance. This assistance is commonly referred to as social support and is considered a resource that positively affects individuals’ well-being (Brougham et al., 2009; Lazarus & Folkman, 1984). Unfortunately, some students do not feel satisfied with the support they receive, and this dissatisfaction with their support may cause them to lack a buffer against the effect of stress on their well-being. Social support of college students includes (a) support from college integration (e.g., interactions with faculty, fellow students, and friends) and (b) support from family (Allgower et al., 2001). Not surprisingly, students perceiving low acceptance in social environments could be quite vulnerable to stress and show low levels of well-being. Moreover, many college students experience multiple stressors, including struggling for independence, problems with roommates or friends, worries about dating, and concerns about grades (Arnett, 2000). Rather than detailing stressors, counselors must emphasize that low levels of social support can exacerbate well-being (Lefkowitz, 2005), and this study specified particular components of support that moderated the association between stress and well-being.

Regarding Hypothesis 2, it was found that a significant three-way interaction (Perceived Stress × Social Support × Dysfunctional Coping) significantly contributed to the variance of college students’ well-being. High levels of dysfunctional coping deteriorated the association between stress and well- being at both high and low social support (see Figures 2 and 3); however, low levels of dysfunctional coping also exacerbated the association between stress and well-being at low social support. Thus, taken together, the current findings highlight the significance of dysfunctional coping (Carver et al., 1989;

Clark et al., 1995). Specifically, even as students perceived high levels of social support, continuing to use dysfunctional coping strategies would lower their well-being. In other words, dysfunctional coping may deteriorate the buffer of social support to decrease well-being. Worse, in recent years, dysfunctional coping is becoming more popular among college students (Brougham et al., 2009), and, to date, not too much attention has been paid to this problem.

The findings of this study have advanced knowledge of and awareness for social support and coping of college students in two ways. First, Hypothesis 1 shows that satisfaction with social support is a buffer, and Hypothesis 2 shows that dysfunctional coping can deteriorate well-being. For instance, the timely acceptance by a professor after a student fails an exam may help the student by acting as a buffer against academic stress, thereby helping the student maintain his or her well-being (O’Connor & O’Connor, 2003). Unstable friendships, family issues, and low support from college are indices of low social support that fails to soothe students’ stress at school, in relationships, and so on (Bulduc, Caron, & Logue, 2007; Buote et al., 2007). In order to understand college students, counselors must understand how low social support worsens the association between stress and well-being. This study contributes to the understanding of college life by articulating the exacerbat- ing effect of low social support on stress to deteriorate students’ well-being.

(13)

Second, this study demonstrated that not all coping is functional and help- ful (Carver & Scheier, 1994; Carver et al., 1989). Dysfunctional coping may deteriorate well-being, even though students are satisfied with their social support. Actually, although no one is immune from daily stress, college students do live with varied social supports (Ben-Zur, 2009). With various coping strategies, some students can thrive whereas others drop out of school in distress despite being satisfied with their social support (Bulduc et al., 2007). In this study, it was found that in addition to low social support, dysfunctional coping strategies (e.g., constant TV watching or habitually venting problems) further decreased well-being; however, in similar situations, less dysfunctional coping strategies resulted in more well-being (Carver &

Scheier, 1994; Carver et al., 1989).

Limitations

This study has four limitations. First, because students’ stress, social support, coping, and well-being were assessed only with self-reported measures and college environment and context were not assessed, the results cannot be generalized to apply to other methods of investigation. For instance, inter- viewing students who participate in social groups may show different results from those in the present study. Besides, the current findings may explain only the perceived social support and dysfunctional coping of students in a university in the Midwest, because the findings are not based on a nationwide sample. Second, only the general stress of college students was assessed, not their specific stressors. The findings cannot be generalized to cover particular stressors, such as professor–student conflicts in classrooms or on-campus violence. Third, this study includes very few racial/ethnic minority students;

therefore, the results cannot explain the stress, social support, coping, and well-being of students from racial/ethnic minority cultures. In the present study, it was not demonstrated that dysfunctional coping is cross-culturally universal in deteriorating the well-being of all students. Finally, because this study used a correlational research design, the findings do not necessarily indicate the causal relationships of the predictors (or independent variables) on the criterion variables (or dependent variables).

Future Research Directions

The following four recommendations are suggested for future research. First, it is suggested that future studies examine social support and dysfunctional coping via other methodologies (such as qualitative research) than the ones used in the present study. For example, students’ subjective reflections may prove or disprove the current results, offer information that was not considered, or extend the results to better understand how college students cope with stress.

Next, it is recommended that researchers study the moderation effect of social support among students of varied cultural backgrounds. It was found that social support was critical for students, but students of different cultures may need

(14)

other types of support, such as financial help (Olson, Garriott, Rigali-Oiler, &

Chao, 2009). Appreciating the role of a specific component of social support in a specific cultural framework is essential for counselors who attempt to provide effective help to students of various cultures. Moreover, it may beneficial to counselors’ understanding of the well-being of college students of different cultures if researchers examine how students variously use dysfunctional coping, because students of different cultures may use dysfunctional coping in different ways. It is necessary to know how dysfunctional coping interweaves cultures. Although racial/ethnic differences on these variables were not found in the present study, a substantial amount of the literature supports differences in stress, social support, coping, and well-being across groups. Thus, future studies need to be conducted to further explore cross-cultural comparison on these variables. Finally, it is recommended that future researchers examine how functional coping (such as problem-focused coping) interacts with dysfunctional coping (Carver et al., 1989). For instance, students could use both functional and dysfunctional coping at the same time, with both forms of coping interact- ing to affect well-being.

Implications for Counseling

This study has three implications for counseling. First, when counseling college students, counselors need to know students’ perceptions of the social support they receive. Therefore, it is beneficial that counselors invite students to talk about their social environment rather than assume that most students enjoy a highly supportive life. If counselors know that some students have few social supports, they can explore with them which specific components of social support were so low as to render them quite vulnerable to stress.

Second, intervention of some components of social support could crucially help students with low social support. For example, counselors can offer psychoeducation to help them cultivate satisfactory social support, such as finding role models or mentors to guide them in their academic work. Learning how to find role models to support them during times of stress could very possibly increase their well-being. Or else, students can learn to cultivate self-assurance to develop hope about their ability to persevere, thereby buffering themselves against stress. A counseling center can help students develop skills to create social support in school and within students’ families (Rayle, Kurpius, & Ar- redondo, 2006). If students are satisfied with the social support they receive from school and in their family environments, students could maintain or even enhance their well-being.

Third, counselors can allow college students to describe what, why, how, and when they use dysfunctional coping, even though they are familiar with the characteristics of dysfunctional coping. Counselors can help students be aware of many sorts of dysfunctional coping (e.g., venting emotions, behavioral disengagement, mental disengagement), how dysfunctional coping deteriorates well-being, and strategies to reduce dysfunctional coping such as

(15)

resorting to listening to music, complaining, or giving up. Cultivating useful strategies to manage stress could reduce dysfunctional coping. When students admit that they cannot complete a school project, counselors may promote students’ motivation and persistence. In helping students deal with various dysfunctional coping strategies, counselors need to know what specific roles each dysfunctional coping strategy plays. For example, when students feel frustrated, they may tend to vent their distress; when students feel annoyed, they may choose behavioral disengagement to stay away from the source of annoyance. To further help students reduce the use of dysfunctional coping, counselors can conduct psychoeducation sessions on how to manage frus- tration and stay focused while performing tasks. As indicated in the present study, low levels of dysfunctional coping, especially when combined with high levels of social support, will not exacerbate the association between stress and well-being. In other words, students’ low dysfunctional coping crucially helps maintain their well-being and makes them less vulnerable to stress (see Figure 2).

Finally, counselors can provide outreach workshops with high social support (Schwitzer, 2005) to highlight (a) the positive roles of high social support and (b) the negative roles of dysfunctional coping. Social support involves families, friends, and even classmates, and helping students cultivate their social support network is one of the top priorities in counseling. Such outreach workshops should be designed to educate students on these benefits to reduce dysfunctional coping.

References

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions.

Thousand Oaks, CA: Sage.

Allgower, A., Wardle, J., & Steptoe, A. (2001). Depressive symptoms, social support, and personal health behaviors in young men and women. Health Psychology, 20, 223–227. doi:10.1037/0278- 6133.20.3.223

Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the early twenties. American Psychologist, 55, 469–480. doi:10.1037/0003-066X.55.5.469 Arria, A. M., O’Grady, K. E., Caldeira, K. M., Vincent, K. B., Wilcox, H. C., & Wish, E. D.

(2009). Suicide ideation among college students: A multivariate analysis. Archives of Suicide Research, 13, 230–246. doi:10.1080/13811110903044351

Benton, S. A., Robertson, J. M., Tseng, W.-C., Newton, F. B., & Benton, S. L. (2003). Changes in counseling client problems over 13 years. Professional Psychology: Research and Practice, 34, 66–72. doi:10.1037/0735-7028.34.1.66

Ben-Zur, H. (2009). Coping styles and affect. International Journal of Stress Management, 16, 87–101. doi:10.1037/a0015731

Brougham, R., Zail, C., Mendoza, C., & Miller, J. (2009). Stress, sex differences, and coping strategies. Current Psychology, 28, 85–97. doi:10.1007/s12144-009-9047-0

Brown, S. D., Alpert, D., Lent, R. W., Hunt, G., & Brady, T. (1988). Perceived social support among college students: Factor structure of the Social Support Inventory. Journal of Counsel- ing Psychology, 35, 472–478.

Bulduc, J. L., Caron, S. L., & Logue, M. E. (2007). The effects of parental divorce on college students. Journal of Divorce & Marriage, 46, 83–104.

(16)

Buote, V. M., Pancer, S. M., Pratt, M. W., Adams, G., Birnie-Lefcovitch, S., Polivy, J., & Wintre, M. G. (2007). The importance of friends: Friendship and adjustment among 1st-year university students. Journal of Adolescent Research, 22, 665–689. doi:10.1177/0743558407306344 Carver, C. S., & Scheier, M. F. (1994). Situational coping and coping dispositions in a stressful

transaction. Journal of Personality and Social Psychology, 56, 267–283. doi:10.1037/0022- 3514.56.2.267

Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies: A theoretically based approach. Journal of Personality and Social Psychology, 56, 267–283. doi:10.1037/0022- 3514.56.2.267

Champoux, J. E., & Peters, W. S. (1987). Form, effect size, and power in moderated regression analysis. Journal of Occupational Psychology, 60, 243–255.

Chaplin, W. F. (1991). The next generation in moderation research in personality psychology.

Journal of Personality, 59, 143–178. doi:10.1111/j.1467-6494.1991.tb00772.x

Clark, K. K., Borman, C. A., Cropanzano, R. S., & James, K. (1995). Validation evidence for three coping measures. Journal of Personality Assessment, 65, 434–455. doi:10.1207/

s15327752jpa6503_5

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Mahwah, NJ:

Erlbaum.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159. doi:10.1037/0033- 2909.112.1.155

Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385–396. doi:10.2307/2136404

Cohen, S., Sherrod, D. R., & Clark, M. S. (1986). Social skills and the stress-protective role of social support. Journal of Personality and Social Psychology, 50, 963–973. doi:10.1037/0022- 3514.50.5.963

Curran, M., Totenhagen, C., & Serido, J. (2010). How resources (or lack thereof) influence advice seeking or psychological well-being and marital risk: Testing pathways of the lack of financial stability, support, and strain. Journal of Adult Development, 17, 44–56. doi:10.1007/

s10804-009-9077-8

Daniel, B. V., Evans, S. G., & Scott, B. R. (2001). Understanding family involvement in the college experience today. New Directions for Student Services, 94, 3–13. doi:10.1002/ss.7 Dawson, J. F., & Richter, A. W. (2006). Probing three-way interactions in moderated multiple

regression: Development and application of a slope difference test. Journal of Applied Psychol- ogy, 91, 917–926. doi:10.1037/0021-9010.91.4.917

Frydenberg, E. (1997). Adolescent coping: Theoretical and research perspectives. New York, NY:

Routledge.

Keith, V. M., & Brown, D. R. (2010). African American women and mental well-being: The triangulation of race, gender, and socioeconomic status. In T. L. Scheid & T. N. Brown (Eds.), A handbook for the study of mental health: Social contexts, theories, and systems (2nd ed., pp.

291–305). New York, NY: Cambridge University Press.

Kuiper, N. A., Olonger, J., & Lyons, L. M. (1986). Global perceived stress level as a moderator of the relationship between negative life events and depression. Journal of Human Stress, 12, 149–153.

Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York, NY: Springer.

Lefkowitz, E. S. (2005). “Things have gotten better”: Developmental changes among emerging adults after the transition to university. Journal of Adolescent Research, 20, 40–63.

doi:10.1177/0743558404271236

Lundberg, C. A., McIntire, D. D., & Creasman, C. T. (2008). Sources of social support and self-efficacy for adult students. Journal of College Counseling, 11, 58–72.

O’Connor, R. C., & O’Connor, D. B. (2003). Predicting hopelessness and psychological distress: The role of perfectionism and coping. Journal of Counseling Psychology, 50, 362–372.

doi:10.1037/0022-0167.50.3.362

(17)

Olson, A., Garriott, P. O., Rigali-Oiler, M. E., & Chao, R. (2009, August). Debt, class, and cop- ing: A model for counseling psychology trainees. Poster session presented at the annual convention of the American Psychological Association, Toronto, Ontario, Canada.

Rayle, A. D., Kurpius, S. E. R., & Arredondo, P. (2006). Relationship of self-beliefs, social support, and university comfort with the academic success of freshman college women. Journal of College Student Retention, 8, 325–343. doi:10.2190/R237-6634-4082-8Q18

Roberti, J. W., Harrington, L. N., & Storch, E. A. (2006). Further psychometric support for the 10-item version of the Perceived Stress Scale. Journal of College Counseling, 9, 135–147.

Roth, S., & Cohen, L. J. (1986). Approach, avoidance, and coping with stress. The American Psychologist, 41, 813–819.

Schwitzer, A. M. (2005). Self-development, social support, and student help-seeking: Research summary and implications for college psychotherapists. Journal of College Student Psychotherapy, 20, 25–52.

Sideridis, G. D. (2008). The regulation of affect, anxiety, and stressful arousal from adopting mastery-avoidance goal orientation. Stress & Health, 24, 55–69. doi:10.1002/smi.1160 Siegel, K., Karus, D., Raveis, V. H., & Hagen, D. (1998). Psychological adjustment of women

with HIV/AIDS: Racial and ethnic comparisons. Journal of Community Psychology, 26, 439–455.

doi:10.1002/(SICI)1520-6629(199809)26:5<439::AID-JCOP4>3.0.CO;2-M

Sirgy, M. J., Lee, D.-J., & Bae, J. (2006). Developing a measure of Internet well-being: No- mological (predictive) validation. Social Indicators Research, 78, 205–249. doi:10.1007/

s11205-005-8209-1

Solberg, V. S., & Villareal, P. (1997). Examination of self-efficacy, social support, and stress as predictors of psychological and physical distress among Hispanic college students. Hispanic Journal of Behavioral Sciences, 19, 182–201. doi:10.1177/07399863970192006

Thorsteinsson, E. B., & Brown, R. F. (2008). Mediators and moderators of the stressor–fatigue relationship in nonclinical samples. Journal of Psychosomatic Research, 66, 21–29.

Veit, C. T., & Ware, J. E. (1983). The structure of psychological distress and well-being in general populations. Journal of Consulting and Clinical Psychology, 51, 730–742. doi:10.1037/0022- 006X.51.5.730

Wang, C.-C. D., & Castañeda-Sound, C. (2008). The role of generational status, self-esteem, academic self-efficacy, and perceived social support in college students’ psychological well-being.

Journal of College Counseling, 11, 101–118.