By SHARON L. BROWN (1994)
Measurement & Evaluation in Counseling & Development, 27 (1), 308-315.

The study was conducted to explore the structure of a brief version of the ways of Coping (WOC) Questionnaire, and to promote better understanding of the coping strategies employed by students enrolled in stressful academic programs.

One of the more popular approaches that counselors can employ to understand clients' coping strategies invokes a contextual approach. This approach takes into account the interaction of situational and personal variables or the person/situation fit (Folkman, 1984; Folkman & Lazarus 1980, 1981, 1985; Folkman, Lazarus, Gruen & DeLongis, 1986; Lazarus & Folkman 1984; Lazarus & Launier, 1978; Parkes, 1984). From this perspective. coping is defined as "the person's cognitive and behavioral efforts to manage (reduce, minimize, master, or tolerate) the internal and external demands of the person-environment transaction that is appraised as taxing or exceeding the person's resources" (Folkman, Lazarus et al., 1986, p. 572). A chosen coping strategy is viewed as reflecting both the person's resources and situational demands.

This increased focus on the contextual process of coping has not adequately addressed the way that contextual variables influence coping processes. The stable elements in various person environment transactions have not been identified, although there is evidence suggesting that individuals' coping efforts vary across different situations (Folkman, Lazarus et al., 1986; Sherer, Weibe, Luther & Adams, 1988). Problem-focused coping strategies may be used in one situation, whereas emotion-focused strategies are used by the same person in another situation. Yet, as Folkman, Lazarus et al. (1986) point out, theoretically some stability should exist across situations, given the influence of personality characteristics. The Ways of Coping (WOC) questionnaire, developed and revised by Folkman and Lazarus (1980, 1981, 1985), has been used extensively in the investigation of coping strategies and person-situation interactions. Sherer et al. (1988) have argued that the biggest challenge to the contextual approach is its failure to demonstrate stability in its measures. Different factor structures as well as different coping strategies for different person-situation contexts have been found. Despite the argument that, in part, these varying effects across different contexts support the contextual approach, there are confounds of differences among people with differences among situations. This can be seen as undermining the comparability between situations and samples.

Sherer et al. (1988) reviewed research on coping that used the WOC questionnaire and found that all the studies differed "in at least two of the following characteristics: response format, factor analytic methods, sample populations, and coping situations" (p. 764). They then attempted to replicate the factor structure obtained with an undergraduate student sample to evaluate factor stability in reported coping strategies that were measured using the shortened version of the WOC. Their study (1988) provided some support for a convergence in factor structure for similar people in similar situations. Using the same 30 items as Folkman and Lazarus (1985), they isolated a similar structure involving five factors with no loadings under .33 on the primary factors and with congruence coefficients for the factors across studies of not less than .950.

The purpose of the current study was to explore further the construct validity of data collected when this shortened version of the WOC questionnaire is employed with student populations. An arguable limitation of the Sherer et al. (1988) study is that their student subjects were asked to report how they might cope in relation to a hypothetical vignette, whereas in the original study the coping that was reported on had actually occurred in response to an upcoming midterm examination. Also, in the original study data were collected with the 66-item WOC version, and it was in later analyses, reported in a private communication to Sherer et al. (1988), that the shortened version was identified.

In this study, actual, nonhypothetical program stressors provided the measurement context for evaluating the factor structure of the shortened instrument with a relatively homogeneous group of graduate students. This approach provided an opportunity to see if the factor structure generalizes to professional graduate populations characterized by high stress. Because less administration time would be required, a short form measure of ways of coping would make it feasible for the instrument to be used in more situations.

This study was conducted specifically to address three research questions. First, what is the structure underlying coping strategies that were employed by veterinary medicine students? Evaluating structure is one initial step in the process of establishing construct validity of scores from the measure. Second, what relative preferences do veterinary students have for various coping strategies? Third, viewed cross-sectionally, what are the differences, if any, in the coping strategy preferences of first-and third-year veterinary medicine students?
 

METHOD
Participants
The study sample was drawn from a large southwestern university veterinary medicine program. Students, participating on the basis of informed consent, were in the first (n = 105) and third (n = 102) years of the 4-year program. Most of the students (81%) ranged from 22 to 31 years of age and were White (92%). About two thirds were unmarried. Female-to-male ratios across years 1 and 3 were similar at 61/44 and 61/41, respectively. The percentage of students in the study was close to 96% of the combined classes from the two cohorts.

Instrumentation
The Ways of Coping questionnaire was developed to study coping processes and the effects of contextual processes on them (Folkman & Lazarus, 1980, 1981, 1985). The latest long version (Folkman & Lazarus, 1985) with 66 items covers a wide variety of cognitive and behavioral strategies that people report they use to deal with either internal or external demands in stressful situations. Folkman, Lazarus, Dunkel-Schetter, DeLongis and Gruen (1986) have noted that the strategies were derived from both the current literature and from their conceptual framework, as exemplified in Lazarus and Launier (1978).

Folkman and Lazarus (1985) analyzed a total of 324 questionnaires collected from 108 undergraduates over three separate occasions, and they isolated a six-factor solution. Eight subscales were produced: a Planful Problem Solving scale, a Seeking Social Support Scale, and six emotion-focused scales. Subsequent studies produced similar numbers of factors but different subscales (Endler & Parker, 1990), thus supporting the need for caution in generalizing results across samples.

The WOC version used in this study consisted of the 30 items reported in the Sherer et al. (1988) study in their evaluation of factor structure congruency with Folkman and Lazarus' (1985) results. Five factors accounting for 42.9% of the variance were extracted in Sherer's study, but eigen-values were unreported. A 4-point Likert-scale response format has been use] consistently across all studies 0 = not used; 1 = used somewhat; 2 = used quite a bit, 3 = used a great deal. Examples of items include: "Talk to someone about how I am feeling"; "Have fantasies or wishes about how things might turn out": "I make a plan of action and follow it": "Accept it since nothing can be done": "I rediscover what is important in life": and "Go on as if nothing is happening."

In the current study, as in the Folkman and Lazarus (1985) study with students, the respondents were asked to rate the extent to which they used the coping strategy in relation to actual stressful situation(s) in the veterinary program, (e.g., exams, schedules, finances). We thought that the broadly defined scope of actual stressors would avoid some of the assumptions made in choosing one stressor and assuming that it is equally stressful for all persons. Both the proposed study and the original study (Folkman & Lazarus, 1985) differ from Sherer et al.'s (1988) study, because the situational demands for coping involved personally relevant stressors and frequency ratings of the related coping strategies used. This format is probably more consistent with the contextual/situational approach and may be more realistic than collecting responses reflecting projected coping in relation to hypothetical vignettes.

Procedure
Students completed the WOC anonymously at the end of scheduled class time. There were no student examinations being administered at or near the time of the data collection that might have affected the results.

RESULTS
We conducted a principal components analysis to address the study's first research question. Analysts differ quite heatedly over the utility of principal components as opposed to common or principal factor analysis. For example, an entire special issue regarding this controversy was recently published in Multivariate Behavioral Research, (Mulaik, 1992). The difference between the two approaches involves the entries used on the diagonal of the correlation matrix that is analyzed. Principal components analysis uses ones on the diagonal, whereas common factor analysis uses estimates of reliability, usually estimated through an iterative process.

The two methods yield increasingly more equivalent results as either (a) the factored variables are more reliable or (b) the number of variables being factored is increased. Snook and Gorsuch (1989, p. 149) explained this second point, noting that "As the number of variables decreases, the ratio of diagonal to off-diagonal elements also decreases, and therefore the value of the communality has an increasing effect on the analysis." For example, with decreases, and therefore the value of the communality has an increasing effect on the analysis." For example, with 10 variables the 10 diagonal entries in the correlation matrix represent 10% (10% 100) of the 100 entries in the matrix, but with 100 variables the diagonal entries represent only 1% (100 / 10,000) of the 10,000 matrix entries. Gorsuch (1983) suggested that with 30 or more variables the differences between solutions from the two methods are likely to be small and lead to similar interpretations. Previous research indicated that between five and eight factor solutions are typically found in coping research of this type. In the current study, the first 10 eigenvalues of the intervariable correlation matrix associated with the components prior to rotation (Thompson, 1989) were: 5.57, 3.15, 2.87, 1.48, 1.38, 1.13, 1.08, 1.01, 0.99, and .91, respectively. In the current study, based on prior findings and application of Cattell's (1966) "scree" test, five- and six-component solutions were extracted.

The solutions were rotated to the varimax criterion. Previously, some researchers using this measure have rotated the factors obliquely. Orthogonal solutions are preferable to oblique solutions, however, even if the oblique solutions are slightly more interpretable. This is because invoking an oblique rotation requires the estimation of a whole new class of parameters, the interfactor correlation coefficients. As more parameters are estimated, there is more capitalization on sampling error, so there is a concomitant decrease in the likelihood that results will replicate. Furthermore, all oblique rotations imply the existence of higher-order factors that should be extracted (Gorsuch, 1983), and it has been recommended that higher-order factoring should be continued until the solution is orthogonal or a single higher-order factor is isolated (Gorsuch, 1983). Thus, from this point of view, an oblique solution in exploratory factor analysis is not entirely acceptable.

The five factor solution provided the most interpretable and parsimonious solution and accounted for 48.2% of the covariance among responses to the items. This solution was accepted as the solution that warranted interpretation. These factors were labelled: (a) Problem Solving; (b) Social/Emotional Support Seeking; (c) Wishful Thinking; (d) Acquiescence, and (e) Reactive Openness to Experience.

The factor structure coefficients for the five factors are presented in Table 1. Item 20 ("I am inspired to do something creative.") had a disproportionately low communality coefficient (h2 = .14), and therefore was not deemed to be particularly salient in defining any of the five factors. The next lowest communality coefficient was .29 for Items 48 and 35. The interpretability of this solution, and the simple structure reflected in the Table 1 results, both suggest that an orthogonal solution was reasonable for these data.

Conventional factor scores (e.g., the regression estimates produced by default in many statistics packages when factor scores are requested) can be computed for a principal components analysis. The computational scoring algorithm is:

Z[sub NXV] R[sup -1 sub VXV] P[sub VXF] = F[sub NXF']

where Z is the data matrix standardized such that the variables' means are all zero and the variables' standard deviations are all one, R is the intervariable correlation matrix, and P is the factor pattern coefficient matrix. This scoring algorithm yields factors scores that are themselves in z-score form; that is, they have a mean of zero and a standard deviation of one. Such factor scores are not useful when the researcher wishes to compare mean scores across factors, be standardize the raw data by dividing scores on each variable by the standard deviation of the variable, without subtracting the mean of each variable from each score on a given variable. This procedure yields factor scores that each have a standard deviation of one, but the mean factor score on each factor is now solely a function of the original data. Factor scores produced in this manner are perfectly correlated with the factor scores computed in the conventional manner; they are simply not all forced to have means of zero.

The procedure can be readily implemented with widely available computer packages. For example, in Statistical Package for the Social Sciences (SPSS), the first step is to invoke a series of COMPUTE statements that divide the scores by the variables standard deviations, such as:

COMPUTE QUES1 =QUES1 /.92467844

Second, one COMPUTE statement per factor is used to compute the factor scores. The SPSS output labels one scores by the variables' standard deviations, such as: COMPUTE QUES1 =QUES1 /.92467844

Second, one COMPUTE statement per factor is used to compute the factor scores. The SPSS output labels one matrix that it prints, the factor score matrix; this W matrix is the rightmost portion of the conventional scoring algorithm:

R[sup -1 sub VXV] P[sub VXF] = W[sub VXF]

The values from this output matrix are then used in a series of COMPUTE statements; for example:

COMPUTE FSCORE1=(QUES1 * -0.10796) + (QUES2 * +0.04009) etc.

With respect to the study's second research question, (What relative preferences do veterinary medical students have for various coping strategies?) the means (each with SD for the total sample = 1.00) on the five factors were 4.86 (Problem Solving), 2.32 (Social/Emotional Support Seeking), 3.09 (Wishful Thinking), 3.84 (Acquiescence), and - 0.76 (Reactive Openness to Experience). Thus, the students perceived themselves as invoking a variety of coping strategies, including, in order: Problem Solving, Acquiescence, Wishful Thinking, and Social/Emotional Support Seeking. The students perceived themselves as making relatively limited use of Reactive Openness to Experience.

With respect to the study's third research question, (Viewed cross-sectionally, what are the differences, if any, in the coping strategy and preferences of first- and third-year veterinary medical students?) one-way two-level ANOVAs were calculated. Statistically significant differences (F = 4.61, df = 1/ 192, p = .0331) were noted only on Factor III, Wishful Thinking. The third-year students placed more emphasis (M = 3.24, SD = 1.06) on this strategy than did the first-year students (M= 2.94, SD= 0.91). This difference involved a standardized effect size of roughly one-third of a standard deviation. Furthermore, it should be noted that this difference would not have been statistically significant if a Bonferroni correction had been invoked.
 

DISCUSSION
The results of this study lend support to the view that there are relatively stable underlying coping structures across samples or groups within specified coping situations, because the five factors isolated in the current study resemble those identified in two previous coping research studies with students (Folkman & Lazarus, 1985; Sherer et al., 1988). These studies, taken together, form a basis for comparative interpretation of the structure isolated here. Four of the five dimensions in the current study were similar to those reported by Folkman and Lazarus (1985) and Sherer et al. (1988), and named by these other researchers as Problem Solving, Seeking Emotional Support, Wishful Thinking, and Detachment.

The least similar factor across the studies was the fifth factor isolated in the current study. This factor is somewhat difficult to interpret. In the previous two studies, this factor was more vague than the first four factors, but it seemed interpretable as a factor named "Focus on the Positive." The item loading most highly on this factor suggests actively focusing on the positive: "Rediscover what is important in life." The negative loadings of the other two items ("I try to keep my feelings from interfering too much,"and "Go on as if nothing is happening") suggest the coping strategies of not repressing or ignoring the problem. In a sense, this is consistent with an active, as opposed to passive, coping stance. The diminished evidence for a coping factor describable as "Focusing on the Positive" may reflect the uniqueness of the veterinary medicine school experience and the types of students attracted to this highly competitive major.

The five factors differ in some respects from those identified in studies with student nurses (Parkes, 1984) or in studies with general members of a community (Folkman & Lazarus, 1980). The broad dimensions of coping represented in these previous studies were direct coping versus suppression, and emotion-focused versus problem-solving, respectively. The current study suggests that a multidimensional view of coping strategies may be required, although the structure underlying coping strategies may be somewhat different across populations or across situations.

The findings that (a) the first- and third-year veterinary medical students do not differ much cross-sectionally over a 2-year period with respect to reported coping strategies, and (b) first- and third-year veterinary medicine students use four of the five styles with relatively equal frequency suggest some potential implications for practitioner-researchers. Either or both of the following interpretations may be true: (a) the results may reflect the stress associated training in a highly competitive professional program, and (b) the results show the flexibility that bright people are capable of when coping with stress. It is noteworthy that wishful thinking strategies were used slightly more by the third-year students. Previous research suggests that this strategy, as compared with problem-solving coping, tends to be associated with less perceived potential for controlling outcomes of stressful situations (Folkman & Lazarus, 1980). Hypothetically, the third-year students may perceive their environment as less controllable and thus use wishful thinking and fantasy-based coping more frequently. Related to managing the stress of professional training, research in other areas has shown planful problem-solving to be associated with more positive than negative emotion compared with other coping strategies (Folkman & Lazarus, 1988).

More research is needed regarding coping strategies and coping outcomes in different situations of advanced professional programs. Longitudinal study of the change in student coping over the course of professional programs should be helpful in evaluating outcomes of any attempted program changes or coping or stress reduction interventions. Future research might investigate the relationship between perceived control, program flexibility, stress reduction interventions, psychological symptoms of stress, and coping variables.

One difficulty in the coping research is the reliance on self-report data and variations in defining the stressful environment (e.g., hypothetical vignettes, examinations, professional programs). Another problem, particularly in using the WOC questionnaires, is that researchers have tended to adapt the scales according to their purposes, thus decreasing the possibility of first evaluating the construct validity of scores from the instrument through comparable studies.

The current study attempted to define a relatively homogenous sample and context from which to evaluate factor structure in comparison to relatively similar prior studies. As Sherer et al. (1988) pointed out, if every aspect of coping is person- and situation-specific, there may be little meaning left in the coping concept as a generalized construct. What is required is a "search for convergence in dimensions and strategies for similar people/situation contexts as well as differences among dissimilar contexts" (p. 74). The current study represented one more step in that direction. A fruitful search for understanding of the dynamics of coping will ultimately provide counselors with valid theory and related useful measures that they can employ to assist clients who confront major life stresses.

                                                 Factor

Number        Item              I    II   III    IV    V  h[sup2]

26      plan of action         .76   .09   .04  -.05  -.02  .60
39      change something       .75   .07   .03  -.03  -.08  .57
52      different solutions    .64   .15  -.10   .08  -.01  .45
49      what's so to be done   .64   .07  -.11  -.11   .14  .45
2       analyze the problem    .63   .28  -.14  -.04  -.04  .50
62      go over in mind        .60   .10   .29  -.17   .08  .49
64      see other perspective  .56   .08  -.02   .23   .08  .38
23      change/good way        .50   .09  -.22   .12   .27  .40
35      not first hunch        .46   .09  -.02   .24  -.08  .29
48      past experience        .36   .23  -.18   .22  -.17  .29
20      do creative            .29   .08   .05   .10   .19  .14

45      talk to someone        .06   .78   .11   .02   .24  .69
42      ask a friend           .28   .72   .06  -.03   .14  .62
18      seek out sympathy      .16   .65   .06  -.05   .02  .48
8       talk to who can do     .44   .59   .05  -.04  -.05  .54
28      let feelings out       .15   .58   .04   .00   .36  .49

57      daydream/imagine       .04   .07   .76   .13  -.08  .61
59      fantasies/wishes       .09   .08   .73   .17   .05  .58
55      wish can change        .08   .20   .70  -.08  -.07  .56
11      hope a miracle        -.11  -.21   .61   .11   .32  .55
58      wish would go away    -.22   .13   .60   .04  -.06  .43
21      try to forget         -.24  -.13   .38   .37  -.27  .44

53      accept it              .13  -.06   .18   .68   .03  .52
4       time/have to wait     -.03   .05   .11   .62  -.05  .41
12      go with fate           .09  -.38   .14   .60   .11  .55
24      wait and see           .00   .12  -.05   .57  -.35  .46
15      silver lining          .45   .00  -.11   .48   .08  .45

38      rediscover/life        .26   .18  -.11   .16   .69  .61
54      keep feelings from     .26  -.11  -.09   .16  -.49  .35
13      go on as if nothing   -.07  -.28   .06   .29  -.61  .55

 Post Rotation Trace           4.54  2.88  2.85  2.36  1.83

Endler, N. S., & Parker, J. (1990). Multidimensional assessment of coping: A critical evaluation. Journal of Personality and Social Psychology,58,844854.

Folkman, S. (1984). Personal control and stress and coping processes: A theoretical analysis. Journal of Personality and Social Psychology, 46, 839-852.

Folkman, S., & Lazarus. R. S. (1980). An analysis of coping in a middle-aged community sample. Journal of Health and Social Behavior. 21, 219-239.

Folkman, S., & Lazarus, R.S. (1981). Reply to Shinn and Krantz. Journal of Health and Social Behavior, 21, 219-239.

Folkman. S., & Lazarus, R. S. (1985). If it changes it must be process: Study of emotion and coping during three stages of a college examination. Journal o Personality and Social Psychology, 48, 150170.

Folkman. S., & Lazarus, R. S. (1988). Manual for the Ways of Coping Questionnaire. Palo Alto, CA: Consulting Psychologists Press.

Folkman, S., Lazarus, R. S., Dunkel-Schetter, C., DeLongis, A., & Gruen, R. J. (1986). Dynamics of a stressful encounter: Cognitive. appraisal, coping, and encounter outcomes. Journal of Personality and Social Psychology, 50, 992- 1003.

Folkman, S., Lazarus, R. S., Gruen, R. J., & DeLongis, A. (1986). Appraisal, coping, health status. and psychological symptoms. Journal of Personality and Social Psychology, 50, 571-579.

Gorsuch, R. L. (1983). Factor analysis (2nd ed.). Hillsdale, NJ: Erlbaum.

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

Lazarus, R. S., & Launier, R. (1978). Stress-related transactions between person and environment. In P. Plutchil & H. Kellerman (Eds.). Theories of emotion (pp. 189-217). New York: Academic Press.

Mulaik, S. (Ed.) (1992). Theme issues on principal components analysis. Multivariate Behavioral Research, 27 (3).

Parkes, K. R. (1984). Locus of control, cognitive appraisal, and coping in stressful episodes. Journal of Personality and Social Psychology, 46, 655-668.

Sherer, R. F., Weibe, F. A., Luther, D. C., & Adams, J. S. (1988). Dimensionality of coping: Factor stability using the Ways of Coping {Questionnaire. Psychological Reports, 62, 763-770.

Snook, S. C.. & Gorsuch. R. L. (1989). Component analysis versus common factor analysis: A Monte Carlo study. Psychological Bulletin, 106, 148-154.

Thompson, B. (1983. January). The calculation of factor scores: An alternative. Paper presented at the annual meeting of the Southwest Educational Research Association, Houston, Texas. (Order document #0450 1 from National Auxiliary Publication Service, P.O. Box 3513, Grand Central Station, NY, NY 10163)

Thompson. B. (1989). Prerotation and postrotation eigenvalues shouldn't be confused: A reminder. Measurement and Evaluation in Counseling and Development, 22(3), 114-116.