Sex Differences in Coping Behavior- A Meta Analytic Review and an Examination of Relative Coping

Sex Differences in Coping Behavior: A Meta-Analytic Reviewand an Examination of Relative Coping

Lisa K. TamresHealth and Community Systems

University of Pittsburgh School of Nursing

Denise Janicki and Vicki S. HelgesonDepartment of Psychology

Carnegie Mellon University

We used meta-analysis to examine recent studies of sex differences in coping. Womenwere more likely than men to engage in most coping strategies. The strongest effectsshowed that women were more likely to use strategies that involved verbal expressionsto others or the selfto seek emotional support, ruminate about problems, and usepositive self-talk. These sex differences were consistent across studies, supporting adispositional level hypothesis. Other sex differences were dependent on the nature ofthe stressor, supporting role constraint theory. We also examined whether stressor ap-praisal (i.e., womens tendencies to appraise stressors as more severe) accounted forsex differences in coping. We found some support for this idea. To circumvent this is-sue, we provide some data on relative coping. These data demonstrate that sex differ-ences in relative coping are more in line with our intuitions about the differences inthe ways men and women cope with distress.

Men and women are commonly thought to havedifferent styles of coping. Stereotypes of male copingbehavior typically revolve around two seemingly op-posite sets of behaviors. Men are believed to be morelikely to confront a problem head-on and also are as-sumed to be more likely to deny a problem exists.Women, on the other hand, are believed to exhibit amore emotional response to problems and are ex-pected to spend more time discussing problems withfriends or family. A number of older studies indicatethat men are more likely than women to exhibit di-rect, problem-confronting coping behaviors or alter-nately to avoid or deny the stressor (Pearlin &Schooler, 1978; Stone & Neale, 1984; Veroff, Kulka,& Douvan, 1981). Studies also show that women aremore likely than men to cope with emotion-orientedbehaviors and to seek social support (Billings &Moos, 1981; Folkman & Lazarus, 1980; Hamilton &Fagot 1988; Pearlin & Schooler, 1978; Stone &Neale, 1984). It has often been accepted that theproblem-oriented coping that men exhibit is adaptive,

whereas the emotion-oriented coping that women ex-hibit is maladaptive (Billings & Moos, 1981;Menaghan, 1982; Pearlin & Schooler, 1978).

In spite of these studies, sex differences in copinghave not yet been established conclusively (Porter &Stone 1995; Thoits, 1991). The literature on coping iscomplex when it comes to evaluating sex differences.First, investigators frequently group several distinctcoping behaviors under a single coping category in aneffort to simplify the findings either conceptually orempirically. The problem with this approach is that asingle coping category is composed of different behav-iors. For example, one study (Rosario, Shinn, Morch,& Huckabee, 1988) defined emotion-focused copingas resignation, denial, anger, controlling feelings, hu-mor, taking a break, physical activity, and socializingwith others, whereas another study (Folkman & Laza-rus, 1985) defined it as wishful thinking, distancing,emphasizing the positive, self-blame, tension reduc-tion, and self-isolation. Because these coping catego-ries are composed of different behaviors, it is difficultto generalize the results from one study to another. Inaddition, such broadly defined categories might bemasking a sex difference that is limited to only one ortwo of the specific behaviors. A second reason that it isdifficult to evaluate sex differences in coping is that thesame coping label has been applied to different copingbehaviors. This may explain why the findings for a spe-cific coping behavior are not always consistent across

Personality and Social Psychology Review2002, Vol. 6, No. 1, 230

Copyright 2002 byLawrence Erlbaum Associates, Inc.

2

A portion of this article fulfilled the requirements for a mastersthesis for Lisa K. Tamres. This work was partly supported by GrantR29 MI4866201A2 from the National Institutes of Health to VickiS. Helgeson.We are grateful to Michael Scheier and two anonymousreviewers for their helpful suggestions in regard to this work.

Requests for reprints should be sent to Vicki S. Helgeson, De-partment of Psychology, Carnegie Mellon University, Pittsburgh,PA 15213. E-mail: [email protected]

studies. For example, acceptance is a coping strategythat has been used to refer to positive acceptance, inwhich one accepts the problem so as to move on withones life (Carver, Scheier, & Weintraub, 1989), andresigned acceptance, in which the individual passivelyaccepts the problem because nothing can be doneabout it (Anderson & Leslie, 1991; Ebata & Moos,1991). Although the name for the behavior is the samein each case, the underlying motivation and attitude arequite different. Indeed, positive and resigned accep-tance are related to different outcomes (Anderson &Leslie, 1991; Carver et al., 1993). A third reason thatthe study of sex differences in coping is complicated isthat a single coping behavior might be given multiplelabels. For example, behavioral disengagement, dis-traction, and avoidance are all used to describe avoid-ing ones problems through involvement in competingactivities (Carver et al., 1989; Feldman, Fisher, Ran-som, & Dimiceli, 1995; Schwab, 1990).

This article evaluates the relatively recent literature(1990 to 2000) that compares mens and womens cop-ing strategies. We begin by providing a brief introduc-tion to coping theory and a theoretical discussion ofsex differences in coping. We then define coping be-haviors conceptually. We examine the literature to de-termine whether there are consistent sex differences inspecific coping behaviors through meta-analysis. Weexamine whether the nature of the stressor and stressorappraisal moderate these relations. Where sex differ-ences in coping can be identified, we explore possibleexplanations for these differences.

Coping Theory

Coping is often defined as cognitive and behavioralefforts made in response to a threat. A common modelof coping set forth by Lazarus and Folkman (1984)stresses that coping choices are dependent on both theappraisal of the threat (primary appraisal) and the ap-praisal of ones resources to address the threat (second-ary appraisal). In addition, Lazarus and Folkmansmodel stresses that coping is dynamic in nature (Cohen& Lazarus, 1973; Folkman & Lazarus, 1980; Lazarus& Folkman, 1984). Specifically, coping is a transactionbetween the threat, the appraisal, and the response.Therefore, coping behaviors will change over time asthese factors interact and also change over time. Thus,it is important to consider factors such as threat ap-praisal and the nature of the stressor when consideringsex differences in coping. Stressor appraisal could ex-plain sex differences in coping behavior if one sex en-gages in a coping strategy more than the other becauseone sex finds the event more stressful.

The nature of the stressor itself also might evokedifferent coping responses from men and women. Acomplete evaluation of how people typically cope with

stress should examine a range of stressors. In thisreview we take into consideration the nature of thestressor when we examine sex differences in coping.We examine a variety of different stressors. However,we only include studies that identify a specific stressor.We wanted to avoid studies in which participants re-port how they cope with hypothetical problems or stud-ies in which respondents report how they typicallycope with stress in general. Our focus on how peoplereported that they coped with a specific stressor en-ables us to understand how men and women cope withspecific problems and to see if the nature of the prob-lem influences the pattern of coping.

A wide variety of actions can be classified underthe definition of cognitive and behavioral effortsmade in response to threat. In practice, coping be-haviors have been placed into categories on either aconceptual or empirical basis. One study mightchoose several behaviors that conceptually fit a defi-nition of problem-oriented coping and then treat themanalytically as one coping behavior. Alternatively, in-vestigators might factor analyze an entire set of dis-crete coping behaviors and use the factors as copingdimensions. Each of these methods has producedfrom two to eleven coping factors (Ayers, Sandler,West, & Roosa, 1996; Billings & Moos, 1981; Carveret al., 1989; Folkman, Lazarus, Dunkel-Schetter,DeLongis, & Gruen, 1986; Miller, Kliewer,Hepworth, Sandler, 1994).

One popular classification system described byLazarus and Folkman, (1984) distinguishes betweenproblem-focused coping and emotion-focused coping.Problem-focused coping behaviors are aimed at alter-ing the stressor. Problem solving and planning are ex-amples. Emotion-focused behaviors are those that aredirected at altering the emotional response to thestressor. Examples are venting emotions, ruminating,avoidance, accepting the problem, interpreting theproblem in a positive light, and wallowing in blame.Although there are other classification systems of cop-ing (Billings & Moos, 1981; Epstein & Fenz, 1967;Overholser, Hemstreet, Spirito, & Vyse, 1989;Schwarzer & Schwarzer, 1996), the distinction be-tween problem-focused and emotion-focused coping isthe one that is most commonly cited when sex compar-isons are made.

As useful as the constructs of problem-focused andemotion-focused coping are in conceptualizing cop-ing, this dichotomy may not be useful in comparingmen and women. A variety of discrete behaviors are in-cluded under each of these two categories. Compari-sons of how men and women cope using these globalcategories may be misleading because sex differencesin coping might be confined to only some of the behav-iors within the categories rather than the overall cate-gories. Within these broad categories, men may bemore likely to exhibit one behavior and women may be

3

SEX DIFFERENCES IN COPING

more likely to exhibit a different behavior; these differ-ences would cancel each other out when all of the be-haviors are placed in the same category.

The assumption that problem-focused coping isadaptive and emotion-focused coping is maladaptivealso is problematic due to the broadness of these con-structs. First, it is possible that some problem-focusedand emotion-focused behaviors are adaptive more of-ten than others. For example, acceptance of a problemcouched in a positive light (I accept that the problemexists and go on with my life) and positive reappraisal(i.e., finding the good in the bad) are more likely thandenial and rumination to be adaptive emotion-focusedstrategies. Second, and more important, the adaptive-ness of a coping strategy is heavily dependent on thestressor. For example, problem-focused coping mightbe adaptive if the stressor is controllable, but emo-tion-focused coping might be more useful when thestressor is less controllable (Folkman, 1984). Even de-nial is adaptive in some situations but not others (Le-vine et al., 1987). Problem-focused coping also maynot be adaptive if one does not have the personal or so-cial resources required to alter the problem. In this re-view we examine discrete coping behaviors for differ-ent stressors.

Theories of Sex Differences

One theory as to why women might engage in someof themoreemotion-focusedcopingstrategies thanmenis that women are thought to be the more emotionalsex. Although it is not clear whether women really expe-rience emotion more frequently than men (La France &Banaji, 1992;Larson&Pleck,1998),womenmayexpe-rience emotion more intensely (Fujita, Diener, &Sandvik, 1991), and women certainly express mostemotions more frequently than men (Brody & Hall,1993; Sprecher & Sedikides, 1993). Sex differences inemotional expression have been detected even among2-year-olds in which girls were observed to be moreemotionally expressive during play than boys(Malatesta, Culver, Tesman, & Shepard, 1989). Womencry more frequently than men, and women are morelikely than men to regard crying as a form of coping (DeFruyt, 1997). Even physiological data show greater fa-cial activity in the female than the male face when menand women express emotion (Grossman & Wood, 1993;La France & Banaji, 1992). This line of research sug-gests that women would be more likely than men to copewith distress by expressing emotions to others.

It is not clear, however, whether these sex differ-ences in emotional expression are innate or learned.Studies that find sex differences in emotional expres-sion among infants and small children would certainlyseem to suggest innate differences. However, a largebody of research shows parents of toddlers and pre-

school children behave in ways to elicit more emotionin girls than boys. For example, one study (Dunn,Bretherton, & Munn, 1987) found that mothers of18-month-olds were more likely to use words that de-scribed feeling states (e.g., sad, angry, happy), states ofconsciousness (e.g., awake, tired), and sensations (e.g.,hungry, cold) with girls than with boys. It is also truethat sex differences in the expression of emotion arelarger among men and women who have stronger ste-reotypes about gender and emotion (Grossman &Wood, 1993), and the female gender role (psychologi-cal femininity or communion) has been associatedwith the expression of emotion (Brody & Hall, 1993).These studies suggest that gender socialization mayplay a role in sex differences in emotional expression.

Another theory of sex differences in coping focuseson support-seeking behavior. Taylor and colleagues(Taylor et al., 2000) suggested that the fact that womenseek support more than men has biological underpin-nings. They suggest that the fight-or-flight responseto threat postulated by Cannon (1932) is more charac-teristic of men than women; instead, women are morelikely to tend-and-befriend when faced with threat.This theory is rooted in evidence from animal studiesthat have reported female animals demonstrating fewerstereotypical fear responses, such as fleeing, than maleanimals when subjected to an environmental threat(Klein, Popke, & Grunberg, 1998). This suggests thatthere is a biological basis for the sex differences in re-sponses to stress. However, gender socialization can-not be ruled out here, either. Women are encouraged toturn to others for support during times of stress,whereas help-seeking behavior is discouraged amongmen because it signifies weakness. For example, astudy (Derlega & Chaikin, 1976) that had college stu-dents read vignettes in which a person either did or didnot disclose a personal problem found that femaledisclosers were viewed as better adjusted than femalenondisclosers but that male disclosers were regarded asmore poorly adjusted than male nondisclosers. Ameta-analytic review of the literature shows thatself-disclosure increases liking of women but de-creases liking for men (Collins & Miller, 1994). Peoplein the social environment also provide more social sup-port to women than men, making it easier for women toturn to others for help (Eagly & Crowley, 1986). Thus,social forces may inhibit men from revealing emotionsin particular and seeking support in general.

There are other differences in mens and womenscoping that can easily be tied to gender socialization.Gender socialization theory would argue that menmight be more likely to cope with stress by denying theproblem or avoiding it because men are socialized toconceal their emotions. Because men stereotypicallyare considered to be action oriented, direct, and asser-tive, they also might be more likely to engage in prob-lem-focused coping.

4

TAMRES, JANICKI, & HELGESON

In sum, regardless of whether these differences incoping are innate or learned, the underlying theory isthat there are characteristic differences between menand women and these differences are reflected in theircoping choices. We refer to this as the dispositional hy-pothesis. Only behavioral genetic studies can deter-mine whether these sex differences are innate orlearned. Although we cannot distinguish biologicalfrom socialization reasons for these apparent sex dif-ferences in coping behaviors, the referenced literaturesuggests the character of the difference resides in gen-der rather than the environment.

By contrast, the situational hypothesis states thatsituations drive coping behaviors. Rosario et al. (1988)referred to this as role constraint theory. Sex differ-ences in coping behaviors are due less to underlyingpersonality differences between men and women thanto the different roles that men and women assume insociety and the different stressors men and womenface. If the nature of the stressor were held constant,sex differences in coping would disappear. For exam-ple, work stressors might be more conducive to prob-lem-focused coping than emotion-focused coping. Aproblem such as work overload or a difficult assign-ment might be best served by a direct, action-orientedapproach. To the extent that men spend more time inthe work environment than women, men might showgreater tendencies toward problem-focused coping.Perhaps family stressors are more conducive to emo-tion-focused coping, especially when the problem isinterpersonal. It might be difficult to change anotherpersons behavior and easier to manage ones ownemotions through acceptance or trying to find the posi-tive in the situation. To the extent that women facemore interpersonal stressors than men, women mightshow greater tendencies toward emotion-focused cop-ing. Thus, to the extent that previous research has notconsidered the nature of the stressor, sex differences incoping could be due to the fact men and women facedifferent sources of stress in their lives.

Although women have joined the workforce ingreater numbers than ever before, there are still a num-ber of role differences between men and women. Forexample, even in dual-career families, women spendmore time on housework than men and are more likelyto be the primary caretakers of children than men(Blair & Lichter, 1991). Discrepancies remain in theworkplace as well, with women having lower statusjobs than men and lower pay for comparable work(U.S. Bureau of Labor Statistics, 1998). Thus, workstressors may be perceived differently by men andwomen. Women may feel that they have little controlover their work situation and that problem-focusedcoping is risky, useless, or even impossible. Thus, therestill may be sex differences in coping within a givenstressor such as work because this stressor is perceiveddifferently by men and women. Men may engage in

more problem-focused coping than women with re-spect to work-related stressors because those stressorsare more amenable to personal control for men thanwomen. If this were the case, role constraint theoryrather than dispositional reasons would explain the sexdifference in coping. Relationship stressors constituteanother stressor domain that men and women mayperceive differently. When the stressor concerns an in-terpersonal issue, it is women who may confront theproblem and men who may withdraw either becauserelationships and relationship functioning are morecentral to the lives of women than men or because menand women hold unequal statuses in relationships withone another.

In this review, we attempt to evaluate the evidencefor the dispositional versus situational (role constrainttheory) hypothesis by (a) examining whether there aresex differences in coping behaviors averaged across allkinds of stressors and (b) examining whether sex dif-ferences in coping are influenced by the nature of thestressor. The dispositional hypothesis would predictthat if men and women were faced with the samestressor and had the same appraisals of that stressor,they would still behave in different ways. Role con-straint theory would predict that when men and womenconfront similar stressors and have similar threat ap-praisals, they would cope similarly.

There are at least two ways that coping might bedifferent for men and women. First, it is possible thatone sex might be more likely to exhibit a certain cop-ing behavior than the other sex. This has been thefocus of our discussion thus far. Second, the samecoping behavior might show different relations towell-being for men and women. For example, ventingthrough crying could leave a woman feeling relievedand less distressed but might leave a man feelinguncomfortable and more distressed. In this review weprimarily examine the former question, namelywhether the two sexes use the same coping behaviorsgiven the same stressor. The second question,whether coping is associated with differential out-comes for men and women, is also important to ad-dress because the outcome of a coping action couldaffect the likelihood of it being used again. Unfortu-nately, relatively few studies examine this issue. Weprovide an overview of the limited findings.

Questions to Be Addressed

Our goal in this article was to evaluate whether thereare sex differences in coping in the relatively recent lit-erature. We used meta-analysis to address this ques-tion. We examined sex differences in 17 specific cop-ing behaviors, each of which was evaluated by at leastsix studies. We also evaluated whether sex differencesin coping were moderated by the type of stressor and

5


stressor appraisal. If sex differences in coping holdacross stressors, the dispositional hypothesis is sup-ported. If sex differences in coping vary as a functionof the nature of the stressor, role constraint theorymight be supported. Mens and womens differentroles expose them to different stressors, which producesex differences in coping. However, it is also possiblethat sex differences in coping are influenced by the na-ture of the stressor because men and women appraisestressors differently. If sex differences in coping arelimited to those stressors appraised as more severe byone sex than the other, stressor appraisal rather thangender may be driving sex differences in coping.

Study 1: Meta-Analysis

Method

Reference articles were located through key-wordsearches using PsycINFO, PsycLIT, and Social Sci-ence Abstracts databases. The key word coping wasused in combination with the terms gender and sex, andwith the phrase human sex differences. In separatesearches, coping was replaced with the key phrasestress management. These searches were supple-mented by scanning reference lists from papers andthrough scanning titles from several pertinent journalsincluding Sex Roles, Journal of Personality and SocialPsychology, and Health Psychology.

Review Constraints

This review is limited to empirical studies reportedbetween 1990 and 2000. We chose to limit the studiesto English-speaking samples because there are indica-tions of cross-cultural sex differences in coping (Cope-land & Hess, 1995; Gerdes & Ping, 1994; Jung, 1995).Thus, studies from the United States, Canada (exceptFrench-Canadian), England, Australia, Ireland, andNew Zealand were included in this review. We exam-ined normal populations, meaning studies of copingin clinical or psychopathological populations were ex-cluded. For example, we excluded studies of popula-tions with clinical depression, alcoholism, and border-line personality disorder. Our interest is in describingmens and womens coping among a normal popula-tion. In this review we include studies that set out tocompare men and women as well as studies that com-pared men and women as a secondary analysis. Studieseither viewed coping as an outcome or tested coping asa mediator or moderator of the relation between an-other variable and an outcome.

We also required that studies evaluated responses tospecific stressors rather than hypothetical stressors or acombination of daily hassles that were undefined. Thiscriterion was important because (a) we wanted to ex-

amine the nature of the stressor as a moderator of sexdifferences or similarities in coping and (b) we believethat people are describing different behaviors whenthey report how they typically cope with stress ver-sus how they cope with a specific stressor. Inventoriesof daily hassles were used if there was a clearly definedstressor, such as caring for a family member with de-mentia. The stressors examined in this study wereplaced into four categories: (a) personal health (e.g.,injury or illness), (b) relationship (e.g., family con-flicts, marital problems), (c) achievement (e.g., workor school stress), and (d) others health (e.g., injury, ill-ness, or death).

Studies were excluded if they classified a con-glomeration of several different behaviors into onegeneral coping scale without (a) defining the distinctbehaviors that comprised the general scale or (b)comparing mens and womens responses to the dis-tinct behaviors. For example, studies that examined acomposite factor called problem-focused coping oremotion-focused coping without a clear explanationof what behaviors were included in these categorieswere excluded from this review (e.g., Brunswick,Lewis, & Messeri, 1992; Cummings, Davies, &Simpson, 1994). In cases in which studies used somebroad, ill-defined coping categories and some discretecoping behaviors, only the data from the discrete be-haviors were examined.

Finally, for inclusion in the meta-analysis, studiesneeded to provide some raw data on mens andwomens coping scores. We accepted either means andstandard deviations for men and women, statisticaltests of means differences (t tests or F statistics), or in afew cases statistical significance tests (p values) withsample sizes from which we inferred effect sizes.

In summary, studies were included if they werereported between 1990 and 2000, involved normalpopulations from the United States and other Eng-lish-speaking countries, examined a specific stressor,and provided analyzable data on specific coping be-haviors. We eliminated about half of the articles we lo-cated because they did not fit all of these criteria. Theprimary reasons for exclusion had to do with unaccept-able definitions or measures of coping behaviors or thefailure to compare men and women on adopted copingstrategies. We identified 50 eligible studies formeta-analysis. Each of these studies provided data on asingle sample. However, three studies evaluated cop-ing strategies with respect to two distinct stressors. Thestudies and their characteristics are shown in Table 1.The coping behaviors we evaluated are discussed in thefollowing section.

Definitions of Coping Behaviors

Because of the problems identified in the introduc-tionthe same label applying to different coping be-

6


7Table 1. Studies Included in Meta-Analysis

Author Sample Size StressorStressorCategory Sample Age Rangea Coping Variables

EffectSize r Measure

Abbey, Andrews &Helman (1991)

184 couples184 men184 women

Infertility PH Infertile couples 2244 Wishful thinkingGeneral problem focus

.56.42

WOCb

Abraham & Hansson(1996)

224120 men104 women

Work A Employed adults 4069 Active .00 PUB

Affleck et al. (1999) 147 Joint pain PH Volunteers with arthritis M = 62.8 Avoidance .22 PUBb48 men Seek SS-E .2199 women Religion .20

Active .16Positive reappraisal .18Venting .23

Anderson & Leslie (1991) 164 Family stress R Married couples with at-home child 2263 Seek SS-non .12 PUB82 men Seek SS-I .0982 women Religion .71

Positive reappraisal .09Archer, Keever, Gordon, &

Archer (1991)165 Demands of residency A Medical school residents 2345 Seek SS-non .11 WOC

101 men Positive reappraisal .1164 women Avoidance .00

Armstrong-Stassen (1998) 92 Company downsizing A Employees of large company ? Positive self-talk .04 LATACK53 men Active .1239 women Seek SS-non .13

Arthur & Hiebert (1996) 94 Educational transition A Students at polytechnical institute 1825+ Avoidance .00 COPE? men Denial .00? women Seek SS-I .00

Seek SS-E .14Religion .00Active .00Planning .00Positive reappraisal .00Venting .00

Borden & Berlin (1990) 61 Caregiving OH Spouses of persons with chronicdementia

3689 Seek SS-non .26 WOC27 men Positive reappraisal .00 nPUB34 women Wishful thinking .00

Self-blame .00Isolation .00General problem focus .00

(continued)

8Table 1. (Continued)



Broderick (1998) 174 Academics A 4th and 5th graders 912 Avoidance .04 nPUB74 boys Rumination .20100 girls

174 Relationship R 4th and 5th graders 912 Avoidance .12 nPUB74 boys Rumination .15100 girls

Butler & Nolen-Hoeksema(1994)

199 Dysphoric mood PH College participant pool 1822 Avoidance .01 PUB125 men Rumination .1474 women

Carroll & Shaefer (1994) 68 Death of child OH Couples bereaved by SIDS 2145 Seek SS-non .46 PUB34 men Seek SS-E .3434 women

Causey & Dubow (1992) 481 Academics A 4th6th graders ? Avoidance .12 nPUB248 boys Seek SS-non .17233 girls General problem focus .12

Venting .13Rumination .02

481 Relationship R 4th6th graders ? Avoidance .10248 boys Seek SS-non .22233 girls General problems focus .19


Choo, Levine, & Hatfield(1996)

250 Relationship break up R College students M = 23.1 Avoidance .18 PUB77 men Self-blame .00173 women

Christie & Shultz (1998) 181 Work A Employed adults attending nightschool

1858 Seek SS-I .06 LATACK56 men Seek SS-E .13125 women General problem focus .19

Exercise .14Avoidance .04

Crocker & Graham (1995) 235 Inability to reach sportsgoal

A Competitive athletes 1530 Seek SS-I .06 COPE123 male athletes Seek SS-E .24112 female athletes Active .01

Planning .04Self-blame .12Venting .01Wishful thinking .06

9DeVries, Hamilton, Lovett,& Gallagher-Thompson(1997)

170 Caregiving OH Adult caregivers of elderly relatives M = 59.4 Active .05 PUB31 men Planning .09139 women Seek SS-non .16

Seek SS-I .06Religion .21Positive reappraisal .09Positive self-talk .29Rumination .26Exercise .11Avoidance .14Venting .19

Encarnacion, Meyers,Ryan, & Pease (2000)

135 Pain PH Ballet dancers M = 19.2 Rumination .66 PUB21 men114 women

Feldman, Fisher, Ransom,& Dimiceli (1995)

166 Family demands duringadolescence

R Random sample of adults of semirural community

1926 Seek SS-E .30 A-COPE82 men Religion .0684 women Avoidance .06

Isolation .28Ferguson & Drotar (1994) 191 Demands of residency A Medical residents ? Avoidance .13 COPE

105 men86 women

Fife, Kennedy, & Robinson(1994)

250 Cancer PH Adults with cancer ? Active .23 WOC125 men Seek SS-non .14125 women Denial .10

Positive reappraisal .22Avoidance .15Religion .23

Franzini & Johnson (1991) 100 Dysphoric mood PH College participant pool M = 19.1 Seek SS-non .16 nPUB47 men Seek SS-I .1653 women Positive reappraisal .00

Exercise .00Gadzella, Ginther,

Tomcala, & Bryant(1991)

401 Work A Community men and women ? Seek SS-non .18 PUB158 men Exercise .17243 women Self-blame .13

Gianakos (2000) 176 Work A Employed college students M = 24.2 Avoidance .02 LATACK46 men Seek SS-non .05130 women Active .24

Positive self-talk .05Heppner, Cook, Strozier, &

Heppner (1991)79 Career transitions A Farmers attending career options

workshopM = 40.4 Seek SS-non .14 WOC

44 men General problem focus .0635 women Wishful thinking .07

Self-blame .09General problem focus .09

(continued)

10

Table 1. (Continued)



Hoffner (1995) 228 Scenes from horror PH 9th and 10th graders M = 15.1 Avoidance .11 PUBb134 boys Seek SS-E .1194 girls Positive reappraisal .00

Kahn & Cooper (1996) 225 Work A Foreign exchange dealers ? Active .00 PUB21 men Seek SS-non .29114 women

Klimes-Dougan & Bolger(1998)

63 Depressed mother OH Children M = 13.9 Active .04 PUBb29 boys Seek SS-I .2634 girls Denial .04


Kolt, Kirkby, & Lindner(1995)

115 Sports performance A Adolescent gymnasts 1320 Denial .09 WOCb32 male adolescents Seek SS-non .2583 female adolescents Isolation .04

Active .02Planning .01Positive reappraisal .05Venting .11Wishful thinking .04

Korabik & Van Kampen(1992)

35 Work A Managers M = 40.3 Avoidance .00 WOC17 men Seek SS-non .0018 women Seek SS-I .00

Seek SS-E .00Self-blame .24Wishful thinking .00

Kramen-Kahn & Hansen(1998)

208 Work A Psychotherapists M = 49.5 Active .29 PUBb73 men Seek SS-non .00135 women Seek SS-I .24

Avoidance .18Positive self-talk .37

Leana & Feldman (1991) 157 Laid off work A Employees of space industry 2160 Active .16 nPUB94 men Seek SS-I .0763 women Seek SS-E .25

Little & Hamby (1996) 501 Work A Therapists 3150 Active .10 nPUB207 men Seek SS-I .09294 women Seek SS-E .19

Exercise .00Avoidance .13

11

Lutzky & Knight (1994) 92 Caregiving OH Spouses of persons with chronicdementia

M = 69.9 Seek SS-non .27 WOC40 men Avoidance .3852 women General problem focus .00

Martin et al. (1992) 165 Old age PH Older adults 60100+ Active .00 PUB63 men Planning .00102 women Isolation .00

165 Relationship R Older adults 60100+ Active .0063 men Planning .00102 women Isolation .00

McDougall (1998) 169 Old-age memory PH Elderly adults 5583 Seek SS-non .15 PUB41 men Religion .30128 women General problem focus .05

Positive reappraisal .09Self-blame .06Wishful thinking .00

McLeod, Kirkby, &Madden (1994)

206 Sports performance A Basketball players 1544 Denial .06 WOCb127 males Seek SS-non .2079 females General problem focus .10

Active .04Positive reappraisal .02Wishful thinking .06

Morrow, Thoreson, &Penney (1995)

206 Infertility PH Infertility clinic patients M = 32.8 Positive reappraisal .17 WOC86 men120 women

Nolen-Hoeksema, Parker,& Larson (1994)

253 Death of close other OH Bereaved adults 2086 Rumination .10 PUB73 men180 women

Parks & Pilisuk (1991) 176 Caregiving OH Adult caregivers of parents withAlzheimers disease

? Planning .00 PUB51 men Wishful thinking .22125 women Isolation .26

Prattke & Gass-Sternas(1992)

75 Artificial insemination PH Couples and women receivingartificial insemination

2355 Denial .03 WOC35 men Seek SS-non .1740 women Isolation .13

Active .23Positive reappraisal .21Self-blame .16Wishful thinking .28

Ptacek, Smith, & Dodge(1994)

114 Public speakinglaboratory task

A College participant pool 1739 Seek SS-non .23 nPUB55 men Avoidance .0059 women Positive reappraisal .00

Self-blame .00General problem focus .00

(continued)

12




Rokach & Brock (1998) 633 Loneliness R College students; community group 1379 Avoidance .11 nPUB295 males Seek SS-non .03338 females Religion .09

General problem focus .17Positive reappraisal .11

Rose, Strauss, Neundorfer,Smyth, & Stuckey(1997)

78 Caregiving OH Spouses of persons withAlzheimers disease

5388 Active .00 PUB29 men Wishful thinking .2449 women

Schnittger & Bird (1990) 329 Work/family R Full-time employed married adults M = 40 Avoidance .00 PUBb158 men Seek SS-non .27171 women Active .16

Positive reappraisal .19Schwab (1990) 50 Deceased child OH Married couples 2760 Active .54 nPUB

25 men Seek SS-I .2025 women Seek SS-E .29

Avoidance .03Isolation .31Venting .33Religion .13Exercise .07

Sigmon, Hotovy, & Trask(1996)

82 Laboratory-induceddysphoria

PH College participant pool ? Avoidance .00 COPE43 men Rumination .2839 women

13

Spirito, Stark, Gil, & Tyc(1995)

177 Chronic illness PH Children with chronic illnesses 718 Avoidance .00 PUB93 boys Seek SS-non .1284 girls Isolation .00

General problem focus .00Positive reappraisal .12Self-blame .12Wishful thinking .00

Stockdale (1998) 2,335 Sexual harassment R U.S. federal employees Mode = 3544 Active .19 nPUB553 men1,782 women

Sullivan, Tripp, & Santor(2000)

80 Induced pain PH College students 1928 Avoidance .00 nPUB38 men Positive reappraisal .0042 women Rumination .18

Positive self-talk .00Unruh, Ritchie, & Merskey

(1999)309 Pain PH Community volunteers 1865 Avoidance .02 PUB

152 men Seek SS-E .20157 women Planning .14

Active .17Venting .09Rumination .03Positive self-talk .17

Note: PH = personal health; A = achievement; R = relationship; OH = othershealth; SS-E = emotional social support; SS-non = nonspecific support; SS-I = instrumental social support; WOC = Ways of Coping Check-list (Lazarus & Folkman, 1984); PUB = other published measure; LATACK = Latack Coping scales (Latack, 1986); COPE = COPE (Carver, Scheier, & Weintraub, 1989); nPUB = nonpublished measure; A-COPE = Ad-olescent Coping Orientation for Problem Experience (Patterson & McCubbin, 1987).a? indicates no information on age is provided. bAdaptation of published measure.

haviors and the same coping behavior being assigneddifferent labelsit was important that we carefully de-fine the coping behaviors we examined. We relied onauthors descriptions of coping behaviors to determinethe appropriate category. We defined 17 specific cop-ing strategies, each of which was evaluated by a mini-mum of six studies. We had two independent ratersplace each of the coping styles into one of the follow-ing 17 categories. Interrater reliability was 85%. Dis-agreements were resolved by a third independent rater.A list of the definitions that guided our interpretationof the data follows.

Problem-focused behaviors. Problem-focusedcoping involves behaviors that are aimed at altering thestressor. It should be noted that this type of coping con-tains effectively two components: preparation (infor-mation seeking, planning) and action (problem solv-ing, active coping). The two components share theproperty of attempts to alter the stressor. Three individ-ual problem-focused behaviors, plus a fourth categoryof general problem-focused coping, are examined inthis meta-analysis:

1. Active coping. Active coping involves efforts tochange or remove the stressor. Taking action that is di-rected at the stressor is the key component.

2. Planning. This strategy includes gathering infor-mation, reviewing possible solutions to a problem, andother efforts that involve planning with the intent totake action about a problem.

3. Seek social support (instrumental). Seeking in-strumental social support refers to seeking specific,generally concrete help from friends and family. Thesupport is directed toward solving problems. Examplesof instrumental social support that people could re-ceive are money, baby-sitting services, advice, and of-fers to run errands.

4. Problem-focused coping (general). We includeda general problem-focused category of coping becausemany studies grouped several of the previously men-tioned behaviors into a single variable. These studiescombined more than one problem-focused copingstrategy into a general category that did not contain anyof the emotion-focused behaviors discussed following.

Emotion-focused behaviors. Emotion-focusedbehaviors are aimed at altering ones response to astressor. The behaviors classified under this categoryare quite different from one another, making a broademotion-focused category difficult to interpret. Forexample, one study (Sigmon, Hotovy, & Trask, 1996)classified emotion-focused coping as a combinationof seeking support and venting, whereas anotherstudy (Rosario et al., 1988) used a combination ofcognitive and emotional strategies, focusing on activ-ities outside of work and taking breaks. Instead of ex-

amining a broad emotion-focused category of copingper se, we evaluated 11 discrete emotion-focusedcoping behaviors:

1. Seek social support (emotional). Seeking socialsupport for emotional reasons refers to seeking outcomfort or emotional support from others.

2. Avoidance. Avoidance is defined as efforts todistract from or avoid the stressor. This includes bothbehavioral and mental efforts to keep ones mind offthe stressor. Individuals may keep busy with alternateactivities such as work, television, or leisure activitiesin an attempt to think of more pleasant things. The goalis to try not to think about the stressor, to avoid remind-ers of the stressor, and to avoid the stressor itself.

3. Denial. Denial includes denying the stressor ex-ists, distancing oneself cognitively from the stressor,and minimizing the importance of the stressor. Denialis distinct from avoidance in that the former entails thefailure to acknowledge a problem exists, whereas thelatter entails avoiding a known problem.

4. Positive reappraisal. Positive reappraisal in-volves trying to find the good in the situation, for ex-ample by trying to grow, learn, or derive some benefitfrom the stressor. The terms cognitive restructuring orcognitive reframing are classified as positive reap-praisalunless they involve denying the stressor ex-ists, in which case they are classified as denial (e.g.,Bird & Harris, 1990).

5. Isolation. Isolation is the removal of oneselffrom social activities. It includes not talking about thestressor and possibly avoiding other people in general.

6. Venting. Venting is the outward, sometimes pub-lic, release of emotions. Acting out, recklessness, andcrying are included here. Acting out can include behav-iors such as getting angry, yelling at others, swearing,making sarcastic comments, breaking things, or usingdrugs(Feldmanetal.,1995;Franzini&Johnson,1991).

7. Rumination. Rumination is focusing on onesproblems and their implications. Individuals who rumi-nate dwell on their problems. It may be viewed in oppo-sition to avoidance or other attempts to turn attentionaway from ones problems. Although people may be-lieve that thinking about a problem will lead to betterproblemsolving, there isevidence that rumination inter-feres with problem solving (Nolen-Hoeksema, 1991).

8. Wishful thinking. Wishful thinking refers towishing that the stressor were not there or imaginingthat the stressor will disappear on its own. Althoughmuch mental effort may be expended in thinkingabout these wishes, nothing is ultimately gained. Es-cape coping is included in this class of coping whenit refers to escapist strategies such as fantasizing,wishing, or hoping for miracles (e.g., Abbey, An-drews, & Halman, 1991).

9. Self-blame. When the individual focuses on hisor her own responsibility for the stressor, the coping

14


strategy used is self-blame. It is a self-critical, self-den-igrating way of thinking. Like rumination, self-blamelacks the solution-oriented aspect that self-analysismight otherwise imply.

10. Positive self-talk. Positive self-talk describesmaking self-statements that encourage oneself to feelbetter (e.g., reminding oneself of the good things inlife) or that reassure oneself that he or she is capable ofhandling the stressor.

11. Exercise. Exercise can be used to provide aphysical outlet for distress or as a way to distract one-self from the stressor. The underlying motivation is notclear. In either case, it is an emotion-focused activity.

Other coping behaviors. Two coping behaviorswere not readily classifiable as either problem-focusedor emotion-focused coping:

1. Seek social support (nonspecific). Several studiescombined items into one general support-seeking cate-gory that may have included seeking instrumental sup-port, emotional support, or both. Underlying each ofthese coping behaviors is the effort to seek help, com-fort, or assistance from others. This nonspecific sup-port-seeking category combines both problem-focusedand emotion-focused support-seeking behaviors.

2. Religion. Praying, involvement in religious ac-tivities, speaking to religious authorities, and expres-sions of spirituality are classified as religious coping.This is not a clearly emotion-focused or problem-fo-cused form of coping. Individuals may view thismethod of coping as solution seeking (i.e., praying forguidance, praying for a solution) or as a way to managedistress (i.e., comfort from religion). Religion can be asource of instrumental as well as emotional support.

Overview of the Meta-Analysis

Meta-analytic procedures are statistical methods forcombining and comparing findings from multiple stud-ies. The procedure effectively treats each study as a re-spondent. We used meta-analysis to compare mensand womens use of the previously described copingbehaviors. When data were available, we also usedmeta-analysis to examine whether stressor type andstressor appraisal moderated sex differences in coping.

The productmoment correlation coefficient (r)was used as the effect size estimate. Effect sizes forindividual studies were calculated either from meansand standard deviations, if provided, or from t tests orF statistics. Fourteen studies reported that there wereno sex differences on some of the coping behaviorsbut did not provide the means or statistical tests. Inthese cases the effect sizes were assumed to be zero.Four studies reported significant sex differences incoping but did not provide means or statistical tests.In these cases, a significance level of p = .05 was as-

sumed, and an effect size was calculated accordingly.This is a conservative approach to detecting differ-ences as the significance levels were likely to besomewhat less than p = .05.

Because meta-analysis assumes independence ofeffect sizes, each study could only contribute one effectsize to a given analysis. In cases in which a study pro-vided data on two or more measures of the same copingstrategy, a combined effect size was computed bytransforming each r into a Fishers z coefficient, aver-aging the Fisher zs and converting the resulting z backinto an r (Rosenthal, 1991).

The meta-analysis results are the weighted mean ef-fect sizes based on the random effects model (Hedges& Olkin, 1985; Schwarzer, 1989). Each effect size isweighted by the number of participants in each study.This analysis is useful because the correlations becomemore stable as the sample size increases. We also ex-amined whether the effect was homogeneous acrossstudies. We used the chi-square statistic as a measureof sample homogeneity. If the studies have similar ef-fect sizes (homogeneous), the mean effect size is con-sidered to be robust. If the effect sizes are highly vari-able (heterogeneous), the mean effect size is lessinformative. In these cases, it is likely that other vari-ables moderate the effect size. We also examined themore conservative residual standard deviation measureof homogeneity. Cases in which the conclusions of thetwo tests of homogeneity differ are noted.

Results

Overview. First we use meta-analysis to deter-mine if there are sex differences in coping. Second, weexamine if sex differences in coping are moderated bythe type of stressor. Third, we explore whether stressorappraisal is associated with sex differences in coping.Finally, we review the few studies that test whethercoping behaviors are differentially related to outcomesfor men and women.

Sex differences in coping. Table 2 shows the re-sults of the meta-analyses for each of the 17 coping be-haviors. In general, women report greater use of themajority of coping behaviors compared to men. All ofthe effect sizes except for one have a negative sign, in-dicating that women use the strategy more than men. In11 of those cases, the effect was statistically signifi-cant. In no case did men engage in a coping strategymore than women.

Of the problem-focused coping strategies, womenwere significantly more likely than men to use activecoping, to seek social support for instrumental reasons,and to engage in general problem-focused coping.There also was a trend for women to engage in moreplanning than men. Of the emotion-focused copingstrategies, women were significantly more likely than

15


men to seek social support for emotional reasons, touse avoidance, to engage in positive reappraisal, to ru-minate, to engage in wishful thinking, and to employpositive self-talk. Women also were significantly morelikely than men to engage in the two coping strategiesthat were ambiguous with respect to the problem-fo-cused or emotion-focused distinction: to seek non-specific social support and to use religion. There wereno sex differences in denial, isolation, venting, orself-blame.

Although there were significant sex differences for11 of the 17 coping strategies (all in the direction ofwomen), the effect sizes were quite small. Only threeeffect sizes exceeded .15: seeking social support foremotional reasons, rumination, and positive self-talk.Notice that these 3 coping strategies all involve con-templation, expression of emotion, or botheither in-dependently or with others. The one sex difference thatshould be taken most seriously is seeking social sup-port for emotional reasons because these data are ho-mogenous across studies (see Table 2).

The effect sizes for the majority of the coping be-haviors were heterogeneous, suggesting that the effectsizes vary across the individual studies. It may be thatthe effect of a single study is inconsistent with the otherstudies and is an outlier. Alternatively, there may beother variables that discriminate different effect sizes,such as the nature of the stressor.

Each of the heterogeneous effects was examined foroutlier studies. Although outliers could occasionallybe identified, we could not detect any unusual charac-teristics of outlier studies that distinguished them from

the other studies. In addition, removal of outlier studiesdid not alter any conclusions. There was one exception.The effect size for religious coping increased from.07 to .16 (p < .001) when an outlier study was re-moved, and the effect became homogeneous. The au-thors (Rokach & Brock, 1998) of the outlier study thatshowed no sex difference in religious coping employedtheir own scale.

Stressor type. We examined whether the effectsizes for each of the coping strategies were influencedby the nature of the stressor. Interrater reliability fortwo persons classification of the stressor into one ofthe four categories was .90. Discrepancies were re-solved by discussions with a third independent rater.The results are presented in Table 3. We examined theeffect size for each of the four different kinds of stress-ors: personal health, relationship, achievement, andothers health. (Refer to Table 2 for the overall effectsize for each coping behavior.) Three of the four stress-ors were examined for all of the coping behaviors. Re-lationship stressors were not represented in studies ofdenial, wishful thinking, positive self-talk, or exercise.When only one study examined a stressor, we indicatedthe effect size for that study but placed it in parenthesesto reduce its impact from the overall conclusions.

A number of the coping strategies showed variabil-ity in effect sizes across the different stressors. Forthree emotion-focused coping strategiesventing,wishful thinking, and religionthe sex difference infavor of women was only significant for stressors thatreflected problems with personal health and others

16


Table 2. Meta-Analysis of Sex Differences in Coping Behaviors

Coping BehaviorsNo. of

Studies Total N

MeanEffectSize r SD p 95% CI Chi-Square Interpretation

Problem FocusedActive 22 6,036 .13 .11 .000 .16 to .11 Heterogeneous Women morePlanning 7 1,264 .04 .07 .087 .09 to .02 Homogeneousa Marginal womenSeek SS-I 12 1,958 .07 .10 .002 .11 to .02 Heterogeneous Women moreGeneral Problem Focus 11 2,378 .12 .13 .000 .16 to .08 Heterogeneous Women more

Emotion FocusedSeek SS-E 12 2,171 .20 .06 .000 .24 to .16 Homogeneous Women moreAvoidance 26 5,383 .03 .11 .005 .06 to .01 Heterogeneous Women moreDenial 6 635 .00 .06 .477 .08 to .08 Homogeneousa No sex differencePositive Reappraisal 19 3,315 .03 .11 .026 .07 to .00 Heterogeneous Women moreIsolation 8 985 .03 .15 .159 .09 to .03 Heterogeneous No sex differenceVenting 9 1,664 .03 .14 .103 .08 to .02 Heterogeneous Marginal womenRumination 10 2,014 .19 .15 .000 .23 to .15 Heterogeneous Women moreWishful Thinking 11 1,512 .13 .20 .000 .18 to .08 Heterogeneous Women moreSelf-Blame 9 1,517 .01 .11 .335 .06 to .04 Heterogeneous No sex differencePositive Self-Talk 6 1,035 .17 .15 .000 .23 to .11 Heterogeneous Women moreExercise 6 1,403 .04 .10 .059 .09 to .01 Heterogeneous Marginal women

OtherSeek SS-non 24 4,417 .10 .15 .000 .13 to .07 Heterogeneous Women moreReligion 9 1,675 .07 .14 .004 .11 to .02 Heterogeneous Women more

Note: CI = confidence interval; SS-I = instrumental social support; SS-E = emotional social support; SS-non = nonspecific support.aAlternate test of homogeneity (residual standard deviation) is heterogeneous.

17


Table 3. Meta-Analysis of Sex Differences in Coping Behaviors by Stressor Types

CopingBehavior Stressor

No. ofStudies Total N

MeanEffectSize r SD p 95% CI Chi-square Interpretation

ActiveaPersonal health 5 778 .15 .08 .000 .21 to .07 Homogeneousb Women moreRelationship 3 2,829 .18 .04 .000 .21 to .14 Heterogeneous Women moreAchievement 11 2,233 .06 .12 .002 .10 to .02 Heterogeneous Women moreOthers health 4 361 .11 .18 .015 .22 to .01 Heterogeneous Women more

PlanningaPersonal health 2 474 .09 .07 .023 .18 to .00 Homogeneousb Women moreRelationship 1 165 (.00) No sex differencesAchievement 3 444 .02 .02 .309 .07 to .12 Homogeneous No sex differencesOthers health 2 346 .04 .04 .206 .15 to .06 Homogeneousb No sex differences

Seek SS-IPersonal health 1 100 (.16) Women moreRelationship 1 164 (.09) No sex differenceAchievement 7 1,411 .04 .11 .055 .09 to .01 Heterogeneous Marginal womenOthers health 3 283 .13 .09 .014 .24 to .01 Homogeneousb Women more

General ProblemFocusa

Personal health 3 531 .17 .19 .000 .25 to .09 Heterogeneous Women moreRelationship 2 1,114 .17 .01 .000 .23 to .12 Homogeneousb Women moreAchievement 5 1,061 .06 .10 .019 .12 to .00 Heterogeneous Women moreOthers health 2 153 (.00) No sex differences

Seek SS-EPersonal health 3 684 .17 .04 .000 .24 to .10 Homogeneousb Women moreRelationship 1 240 (.30) Women moreAchievement 6 1,203 .16 .05 .000 .21 to .10 Homogeneousb Women moreOthers health 2 118 .32 .02 .000 .48 to .15 Homogeneousb Women more

AvoidanceaPersonal health 9 1,527 .04 .09 .044 .09 to .01 Homogeneousb Women moreRelationship 6 2,107 .03 .10 .099 .01 to .07 Heterogeneous Marginal menAchievement 11 2,320 .04 .10 .033 .08 to .00 Heterogeneous Women moreOthers health 3 312 .19 .13 .000 .30 to .08 Homogeneousb Women more

DenialPersonal health 2 157 .04 .06 .318 .20 to .12 Homogeneousb No sex differencesAchievement 3 415 .00 .06 .461 .09 to .10 Homogeneous No sex differencesOthers health 1 63 (.04) No sex differences

PositiveReappraisal

Personal health 9 1,264 .05 .12 .046 .10 to .01 Heterogeneous Women moreRelationship 3 1,126 .02 .13 .262 .04 to .08 Heterogeneous No sex differencesAchievement 5 694 .04 .04 .143 .12 to .03 Homogeneousb No sex differencesOthers health 2 231 .07 .04 .158 .19 to .06 Homogeneousb No sex differences

IsolationaPersonal health 3 417 .02 .05 .316 .12 to .07 Homogeneousb No sex differencesRelationship 2 405 .17 .14 .000 .26 to .07 Heterogeneous Women moreAchievement 1 115 (.04) No sex differencesOthers health 3 287 .11 .22 .035 .01 to .22 Heterogeneous Men more

VentingaPersonal health 2 456 .14 .07 .002 .23 to .04 Homogeneousb Women moreRelationship 1 481 (.14) Men moreAchievement 4 925 .06 .08 .042 .01 to .12 Homogeneousb Men moreOthers health 3 283 .16 .12 .003 .28 to .05 Homogeneousb Women more

RuminationaPersonal health 5 800 .22 .22 .000 .29 to .16 Heterogeneous Women moreRelationship 2 655 .19 .02 .000 .26 to .11 Homogeneousb Women moreAchievement 2 655 .07 .08 .040 .14 to .01 Heterogeneous Women moreOthers health 3 486 .17 .08 .000 .26 to .08 Homogeneousb Women more

Wishful ThinkingPersonal health 4 606 .22 .25 .000 .30 to .15 Heterogeneous Women moreAchievement 4 591 .01 .05 .397 .07 to .09 Homogeneousb No sex differencesOthers health 3 315 .18 .09 .000 .29 to .07 Homogeneousb Women more

(continued)

health. Most of these sex differences were homoge-neous. In the case of venting, men were more likelythan women to use this strategy to cope with achieve-ment and relationship stressors. Avoidance partly fitthis pattern of findings. The largest sex difference(women more than men) in avoidance occurred for oth-ers health stressors; there was a trend for men to useavoidance more than women to cope with relationshipstressors. For two coping strategies, planning and posi-tive reappraisal, the sex difference was only significantfor personal health stressors, but these effects werevery small.

One coping strategy revealed no sex differences av-eraging across all of the studies but showed sex differ-ences for particular stressors. Women were more likelythan men to use isolation for relationship stressors,whereas men were more likely than women to use iso-lation to cope with others health stressors.

The one significant effect that was homogenous inthe overall meta-analysis remained consistent whenthe nature of the stressor was examined. Women weremore likely than men to seek social support for emo-tional reasons for each of the four stressors. Womenalso were more likely than men to seek instrumentalsupport for three of the four stressors and to seek non-specific support for all four stressors. (Note the effectsizes for nonspecific support were larger and homoge-neous for personal health and others health stressors.)

Other effects that were not homogenous in the overallmeta-analysis still appear to hold across the differentstressors. Women were more likely than men to engagein active coping, to ruminate, and to engage in positiveself-talk for each of the four stressors. These are thecoping strategies that revealed the largest effect sizes inthe overall meta-analysis. Women also were morelikely than men to use general problem-focused copingfor three of the four stressors (all but others health).The nonsignificant effects for denial, self-blame, andexercise also held across the four different stressors.

One claim of previous researchers is that men usemore problem-focused coping and more avoidancethan women. There is no evidence that men use moreproblem-focused coping than women for any of thestressors. However, there is some evidence that menuse more avoidant or withdrawal strategies for rela-tionship and others health stressors. Both of thesestressors involve other people and may be perceived asless personally controllable than personal health andachievement stressors. Perhaps men are more likely touse avoidant and withdrawal behavior when they per-ceive the stressor as uncontrollable. These ideas arehighly speculative.

Another way to examine the influence of stressortype on sex differences in coping is to examine thepattern of sex differences across the coping strategiesfor each of the four stressors separately. For personal

18



CopingBehavior Stressor

No. ofStudies Total N


Self-BlamePersonal health 3 421 .05 .10 .173 .05 to .14 Homogeneous No sex differencesRelationship 1 250 (.00) No sex differencesAchievement 4 785 .05 .13 .100 .12 to .02 Heterogeneous Marginal womenOthers health 1 61 .00 No sex differences

Positive Self-TalkPersonal health 2 389 .14 .07 .004 .23 to .04 Homogeneousb Women moreAchievement 3 476 .16 .20 .000 .24 to .07 Heterogeneous Women moreOthers health 1 170 (.29) Women more

ExercisePersonal health 1 100 (.00) No sex differencesAchievement 3 1,083 .04 .11 .094 .10 to .02 Heterogeneous Marginal womenOthers health 2 220 .07 .08 .153 .20 to .06 Homogeneousb No sex differences

Seek SS-nonaPersonal health 5 603 .14 .02 .000 .22 to .06 Homogeneousb Women moreRelationship 4 1,607 .10 .15 .000 .15 to .05 Heterogeneous Women moreAchievement 12 2,297 .08 .16 .000 .12 to .04 Heterogeneous Women moreOthers health 4 391 .26 .11 .000 .35 to .16 Homogeneousb Women more

ReligionPersonal health 3 398 .25 .05 .000 .34 to .15 Homogeneousb Women moreRelationship 3 1,037 .03 .08 .180 .03 to .09 Heterogeneous No sex differencesAchievement 1 94 (.00) No sex differencesOthers health 2 220 .19 .03 .002 .32 to .06 Homogeneousb Women more

Note: CI = confidence interval; SS-I = instrumental social support; SS-E = emotional social support; SS-non = nonspecific support.aThese coping styles were included in studies that assessed more than one stressor. Thus, sum of total Ns for individual stressors is not equal to thetotal N for all stressors. bAlternate test of homogeneity (residual standard deviation) is heterogeneous.

health stressors, women were more likely than men toengage in 13 of the 17 coping strategies, and the ma-jority of these effects were homogeneous across stud-ies. Seven of the effects were .15 or greater. There isno strategy that men used more than women in thecase of personal health stressors. One wonders ifwomen engage in more coping strategies than menbecause women are more distressed about personalhealth problems than men. We know that women, onaverage, are less healthy than men as indicated bywomens reports of poorer general health, more ill-ness, more physician visits, and greater use of medi-cation compared to men (Macintyre, Hunt, & Sweet-ing, 1996; Verbrugge, 1989).

We found the fewest sex differences in coping forrelationship stressors. For only 6 of the 13 strategiesexamined did women engage in the behavior more thanmen (active coping, general problem-focused coping,seek emotional support, isolation, rumination, andseek nonspecific support). For relationship stressors,men were more likely than women to use avoidanceand venting. For achievement stressors, 9 of 17 effectswere significant in the direction of women, but nearlyall of these effects were quite small. Only two effectsizes exceeded .15: seek emotional support and posi-tive self-talk. For others health, 9 of 17 effect sizeswere significant in the direction of women. Nearly allof these (7) were more substantive, greater than .15.Thus, there are two stressors, personal and othershealth, that clearly indicate more coping on the part ofwomen compared to men.

Stressor appraisal. When comparing mens andwomens coping behavior, it is also important to deter-mine whether men and women perceive the samestressor as equally severe. If men and women appraisea stressor differently, sex differences in coping may bedue to appraisal rather than to preferred coping strate-gies. Research has shown that women use a wider vari-ety of coping strategies or expend more coping effortsthan men (Thoits, 1991, 1994). The implication is thatwomen perceive greater threat than men and conse-quently expend more effort by using more coping strat-egies. Our review has shown that women are morelikely than men to engage in most coping strategies. Itis important to determine whether women are engagingin more coping than men because they view the stress-ors as more severe.

The best way to determine if stressor appraisal ac-counts for sex differences in coping is to see if sexdifferences disappear when stressor appraisal is con-trolled in a statistical analysis. In other words, doesstressor severity mediate the relation between sex andcoping? Only two of the reviewed studies have em-ployed this analytic strategy. Ptacek, Smith, andDodge (1994) examined college students ways ofcoping with an achievement-oriented stressordeliv-

ering a lecture. In that study, there were no sex differ-ences in stressor appraisal, and stressor appraisal didnot account for women seeking social support morethan men or men using problem-focused coping morethan women. However, in a study of caregivers ofspouses with Alzheimers disease (Rose, Strauss,Neundorfer, Symth, & Stuckey, 1997), women ap-praised the stressor as more severe than men, andwomen were more likely than men to engage in wish-ful thinking. The sex difference in wishful thinkingbecame nonsignificant when distress was statisticallycontrolled in a series of regression analyses. Rose etal. tested and showed that stressor appraisal ac-counted for womens greater use of a coping strategycompared to men.

A more indirect way to address whether stressor ap-praisal could account for sex differences in coping is toexamine stressor appraisal as a moderator variable inthe meta-analysis. If women engage in more of a cop-ing strategy than men only among studies in whichwomen appraise the stressor as more severe than men,one would certainly be concerned that appraisal ratherthan sex was driving coping. Testing moderation is notthe same as testing mediation, but we can at least usemoderation in meta-analysis to see if the results areconsistent with a mediational model. Just over half(26) of the 50 studies in the meta-analysis assessedstressor appraisal. None of the studies reported thatmen appraised the stressor as more severe than women.Of the 26 studies, 17 revealed that women reported thestressor as more severe, and 9 showed no sex differ-ences in stressor appraisal. These findings, in and ofthemselves, cause us to be concerned that stressor ap-praisal is driving sex differences in coping. Table 4 dis-plays the number of studies finding a sex difference instressor appraisal for each coping behavior. We usedmeta-analysis to determine if the effect size of studiesin which women appraised the stressor as more severediffered from the effect size of studies in which therewas no sex difference in stressor appraisal. To do this,we examined the seven coping behaviors in which atleast 2 studies showed women were more distressedthan men, and at least 2 studies showed no sex differ-ence in distress.

The results of the meta-analysis with stressor ap-praisal as a moderator variable are shown in Table 5. Anumber of findings suggest that stressor appraisal maybe responsible for the finding that women engage inmost of the coping strategies more than men. For fourof the seven coping strategies (active coping, avoid-ance, positive reappraisal, and self-blame), womenused the coping behavior more than men only in stud-ies in which women appraised the stressor as more se-vere; in studies in which there was no sex difference instressor appraisal, the effect size was not significant.

The other three coping strategies did not fit thispattern of findings. The sex difference in rumina-

19


tion held across studies in which women appraisedthe stressor as more severe and studies in whichthere was no sex difference in stressor appraisal; infact, the effect size was larger in the latter case. Forseeking nonspecific support, men were more likely

to use the strategy than women when women ap-praised the stressor as more severe, and womenwere more likely than men to use the strategy whenthere was no sex difference in stressor appraisal.Thus, it is unlikely that stressor appraisal is drivingthe sex difference in rumination or nonspecific sup-port seeking. For isolation, men were more likelythan women to use the strategy when women ap-praised the stressor as more severe, and there wasno sex difference in the use of isolation when menand women appraised the stressor as equally severe.We were unable to test stressor severity as a moder-ator of sex differences in seeking support for emo-tional or instrumental reasons because all of thestudies that assessed these coping strategies as wellas stressor appraisal found that women were moredistressed than men.

We also explored whether sex differences instressor appraisal were more likely to occur for certainstressor domains. Sex differences in stressor appraisalwere more likely to occur for personal health andachievement stressors. For personal health stressors, 7of the 10 studies showed that women appraised thestressor as more severe. This adds further support tothe idea that one reason women engaged in more cop-ing strategies compared to men for personal healthstressors is that women were more distressed than men.For achievement stressors, six of the eight studiesshowed that women appraised the stressor as more se-vere. However, this is the stressor domain that showedextremely small effect sizes for sex differences in cop-

20


Table 4. Number of Studies Evaluating Stressor Appraisaland Coping Behavior

Stressor Appraisal

Coping Behavior

WomenMore

StressedMen MoreStressed

No SexDifference

Active 7 0 2Planning 2 0 1Seek SS-I 3 0 0General Problem

Focus1 0 4

Seek SS-E 5 0 0Avoidance 8 0 5Denial 1 0 1Positive

Reappraisal7 0 3

Isolation 2 0 2Venting 2 0 0Rumination 3 0 2Wishful Thinking 5 0 1Self-Blame 3 0 2Positive Self-Talk 3 0 0Exercise 2 0 0Seek SS-Non 6 0 5Religion 2 0 1

Note: SS-I = instrumental social support; SS-E = emotional socialsupport; SS-non = nonspecific support.

Table 5. Meta-Analysis of Sex Differences in Coping Behaviors by Stressor Appraisal

Coping StressorNo. of

Studies Total N


ActiveWomen more 7 3,537 .17 .04 .000 .20 to .14 Homogeneousa Women moreNo sex difference 2 247 .08 .11 .112 .20 to .05 Homogeneousa No sex difference

AvoidanceWomen more 8 1,656 .09 .08 .000 .13 to .04 Homogeneousa Women moreNo sex difference 5 661 .02 .15 .291 .10 to .06 Heterogeneous No sex difference

Positive ReappraisalWomen more 7 1,065 .08 .10 .006 .14 to .02 Homogeneousa Women moreNo sex difference 3 257 .07 .10 .128 .19 to .05 Homogeneousa No sex difference

IsolationWomen more 2 251 .15 .18 .010 .02 to .27 Heterogeneous Men moreNo sex difference 2 226 (.00) No sex difference

RuminationWomen more 3 642 .08 .05 .026 .15 to .00 Homogeneousa Women moreNo sex difference 2 373 .15 .02 .001 .25 to .05 Homogeneousa Women more

Self-BlameWomen more 3 511 .15 .02 .000 .23 to .06 Homogeneousa Women moreNo sex difference 2 175 (.00) No sex difference

Seek SS-nonWomen more 6 932 .05 .14 .052 .01 to .12 Heterogeneous Men moreNo sex difference 5 428 .16 .15 .000 .25 to .06 Heterogeneous Women more

Note: CI = confidence interval; SS-non = nonspecific support.aAlternate test of homogeneity (residual standard deviation) is heterogeneous.

ing. Of the four studies that assessed stressor appraisalfor relationship stressors, two showed women ap-praised the stressor as more severe and two showed nosex difference. For othershealth stressors, two showedwomen appraised the stressor as more severe and threeshowed no sex difference.

Sex differences in the relation of coping to out-comes. As noted earlier, there are two ways in whichcoping might differ for men and women. The first isthat men or women might choose certain coping strate-gies over others. This was the subject of our meta-anal-ysis. The second is that certain coping strategies mightbe more or less adaptive for men or women. Ten of thestudies we reviewed examined whether there were sexdifferences in the relation of coping behaviors to out-comes. Seven of these found a sex difference in the re-lation of at least one coping strategy to an outcome.

In a study of college students dealing with schoolstress (Hovanitz & Kozora, 1989), men were bufferedfrom elevated dysfunction by problem-directed cop-ing, whereas women were buffered by seeking socialsupport. In a longitudinal study of adolescents copingwith family stress (Feldman et al., 1995), men whosought more social support had poorer adaptation 6years later, whereas women who sought more socialsupport had better adaptation. In Abraham andHanssons (1996) study of coping in the workplace,problem-directed coping was related to increased jobsatisfaction and decreased stress for men but not forwomen. These three studies suggest that problem-fo-cused coping is more adaptive for men than womenand seeking social support is more adaptive for womenthan men.

However, other studies do not support this conclu-sion. A study of farm families in career transitions(Heppner, Cook, Strozier, & Heppner, 1991) indicatedthat problem-directed coping was related to increasedprogress and increased perception of control for womenbut not for men. In a study of men and women copingwith organizational downsizing (Armstrong-Stassen,1998), active coping was associated with a lower per-ceived threat of job loss among women but not men. Onestudy (Holmbeck et al., 1997) found that positive rein-terpretation and growth was related to better outcomesfor men, and another study (Anderson & Leslie, 1991)found that positive reinterpretation and growth was re-lated to better outcomes for a subset of men who were intraditional, single wage-earning families. Thus, there issome evidence that problem-focused coping can benefitwomen and that at least one type of emotion-focusedcoping, in this case positive reinterpretation, can benefitmen. Clearly more research is needed in this area.

The results of these seven studies should be inter-pretedwithcautionbecausenoneof the investigatorsac-tually tested whether the relation of a coping strategy toan outcome significantly differed between men and

women. It also is possible that coping was differentiallyrelated to outcomes for men and women because therewere sex differences in stressor appraisal. This possibil-ity was not examined in any of these studies.

Discussion

We set out to review sex differences in coping withspecific stressors in studies published between 1990and 2000. First, we summarize the results from theoverall meta-analysis. Women were found to use morecoping strategies than men across a variety of behav-iors, including both problem-focused and emotion-fo-cused domains. The conventional wisdom in this areasuggests that men engage in more problem-focusedand avoidant coping than women. We found no evi-dence that men engage in more problem-focused cop-ing for any of the stressors. There were some hints thatmen may engage in more avoidant or withdrawal be-havior for some stressorsthe stressors having to dowith other people (relationships and others health).These are the two stressor domains that may be per-ceived as less controllable.

Most of the sex differences in coping were small.The largest effects had to do with behaviors that in-volved the contemplation or expression of feelings toothers (seeking emotional support) and the self (rumi-nation, positive self-talk). There was a single robust ef-fect (i.e., consistent across studies): seeking emotionalsupport. Thus, we are most confident that women aremore likely than men to seek emotional support acrossa range of stressors.

One problem typical of meta-analysis is thefile-drawer problem. Studies with significant sex dif-ferences may be more likely to be published than stud-ies that do not obtain significant differences. However,many of the studies in this review examined sex differ-ences in coping as a secondary rather than a primaryanalysis, which makes them somewhat less vulnerableto the file-drawer problem.

To evaluate whether sex differences in coping weremore supportive of the dispositional or situational hy-pothesis, we examined whether the nature of thestressor influenced sex differences in coping. Our gen-eral conclusion is that there is some evidence for boththeories. The dispositional hypothesis is supported forthe sex differences in coping behaviors that were ro-bust across stressor type: seeking emotional support,rumination, and positive self-talk. Women were morelikely than men to seek emotional support, to ruminate,and to engage in positive self-talk regardless of the na-ture of the stressor. Gender socialization may explainthese findings. Women may be socialized to seek outothers for emotional support. Women are encouragedbut men are discouraged from expressing feelings toothers, especially feelings about problems. Such ex-pressions are thought to foster connections among

21


women but may be viewed by men as revealing weak-nesses and exposing vulnerabilities. Rumination is acoping strategy that also appears to be encouraged inwomen. One study found that both male and femalecollege students were more likely to give women thanmen ruminative advice in response to depression; thatis, women were told to figure out why they were de-pressed (Ali & Toner, 1996). The sex difference in ac-tive coping also was consistent across stressor type, butthis finding does not support gender socialization the-ory because it is men rather than women who are ex-pected to engage in problem-focused coping.

The finding that women seek support from othersmore frequently than men also may be explained bytheories that focus on innate differences between menand women. That is, factors associated with the bio-logical category of sex rather than those associatedwith the social category of gender might influenceones preferred coping strategies. One such factor isthe pituitary hormone, oxytocin. Women possesshigher levels of oxytocin than men. Release ofoxytocin during times of stress is associated withdownregulation of the sympathetic nervous systemand facilitation of the parasympathetic nervous sys-tem. This neuroendocrine activity is expressed by apattern of tend-and-befriend rather than fight-or-flight(Taylor et al., 2000). According to this theory, womenwould be more likely than men to seek out the sup-port of others in times of stress. This theory alsocould be applied to the minor trend for men to engagein avoidant and withdrawal coping behaviors forsome stressor domains. Because the mitigating ef-fects of oxytocin on sympathetic activity are not aspronounced for men as they are for women, the flightpattern of behavior might provide the basis for mento engage in a stressor-avoiding coping strategy.

The nature of the stressor clearly influenced somecoping behaviors, partly supporting role constraint the-ory. Women were especially likely to engage in morecoping strategies than men for personal health and oth-ers health stressors. These are stressors to whichwomen may be more exposed compared to men. To theextent that women are more likely than men to facepersonal health problems and more likely than men tobe involved in the caretaking of others, it makes sensethat women report greater distress than men in re-sponse to these stressors and engage in more copingbehaviors compared to men.

We found the most mixed results for the domain ofrelationship stressors. Because women are socializedto focus more on relationships compared to men, onemight have expected men and women to cope quite dif-ferently with relationship stressors. This is the only do-main in which men were found to engage in two copingbehaviors more than women. Men were more likelythan women to use ventilation and avoidance, twoemotion-focused strategies, whereas women were

more likely than men to use active coping, use generalproblem-focused coping, seek social support for emo-tional and nonspecific reasons, use isolation, and userumination. Thus, women are using the strategies tocope with relationship stressors that they use to copewith all domains (seek emotional and nonspecific sup-port, rumination) but also using more problem-focusedstrategies. To the extent that relationships are morecentral to the lives of women than men, women may bemore motivated than men to engage in problem-fo-cused coping. These findings fit with the literature onthe demand and withdrawal pattern among couples thatshows that women are more likely to confront prob-lems (demand) in their relationships and men are morelikely to avoid (withdraw) them (Christensen &Heavey, 1993).

There is one case in which the nature of the stressorinfluenced coping, but the findings fit with thedispositional hypothesis rather than role constraint the-ory. Men were more likely than women to use avoidantand withdrawal strategies to cope with relationship andothershealth stressors. These are stressors that involveother people and may be less amenable to personalcontrol than personal health and achievement stressors.Men may be more likely to use avoidant strategies foruncontrollable stressors due to gender socialization(i.e., personal control is integral to the male genderrole) or due to biological underpinnings (i.e.,fight-or-flight response).

It is important to point out that our examination ofthe effect of stressor type on coping is limited in thatour stressor categories may not reflect distinct de-mands. Although we had two independent ratersagree that a stressor fell into one of the four catego-ries, the categories are broad. For example, work-re-lated stressors could be associated with the nature ofthe work itself or with interpersonal difficulties en-countered while at work. It is also possible that agiven category of stressor poses different problemsfor men and women. For example, women may con-ceive of personal health stressors as relationshipstressors if personal health problems interfere withcaretaking responsibilities. This could be an addi-tional reason why women appraised this stressor asmore severe than men.

It is notable that in all cases in the overall meta-anal-ysis where there was a sex difference, women used thecoping strategy more than men. It is also notable thatthe majority of studies that assessed stressor appraisalfound that women appraised the stressor as more se-vere than men. If stress is driving coping and womenare experiencing more stress than men, stress could beresponsible for sex differences in coping. This wouldsuggest that mens and womens coping patterns wouldbe more similar if they were equally stressed, support-ing role constraint theory. The situation (increasedstress) would be driving coping rather than an inherent

22


characteristic of men and women. Although we couldnot determine whether differences in stressor appraisalexplained women favoring specific coping behaviorsmore than men in terms of mediational analyses, wedid examine whether stressor appraisal moderated sexdifferences in coping. We found that for several of thecoping strategies, women were more likely to use thestrategy when they appraised the stressor as more se-vere. Thus, the concern that women are more likely touse most of the coping strategies compared to men dueto women being more distressed by the stressor thanmen is a viable explanation for these sex differences incoping. Future research in this area ought to testwhether stressor severity accounts for sex differencesin coping using mediational analyses.

An alternative way to evaluate sex differences incoping that would circumvent the stressor appraisal is-sue is to examine relative coping. Relative coping in-volves comparing how much an individual uses onestrategy compared to other strategies (Vitaliano,Maiuro, Russo, & Becker, 1987). Whereas raw copingscores do not take into consideration base rates of cop-ing, relative coping scores do (Vitaliano et al., 1987).Using raw coping scores, we found that women en-gaged in many coping strategies more frequently thanmen. However, men and women may still differentiallyfavor certain coping strategies, which would be re-flected in relative coping scores.

Four studies in our review examined sex differencesin relative coping. Grant and Compass (1995) study ofadolescents who had a parent diagnosed with cancerfound that girls were relatively more likely than boys touse rumination, but there were no sex differences in useof distraction. The findings for rumination are consis-tent with our review and the large body of work byNolen-Hoeksema (1987, 1990). The findings for dis-traction are in contrast to our finding that women en-gage in more avoidance than men, as well as previousresearch that has shown that men engage in moreavoidance than women (Nolen-Hoeksema, 1987,1990). Hurst and Hurst (1997) also used relative cop-ing scores in their study of occupational stress amongcorrectional officers. They found that women were rel-atively more likely than men to seek social support,and there was a trend for men to be relatively morelikely than women to use planful problem solving.There were no sex differences in avoidance, denial, orpositive reappraisal. A third study (Ptacek et al., 1994)reported both relative and raw coping scores. We usedthe raw coping scores in our analysis. The results usingrelative scores differed both from the results based onraw scores in Ptacek et al.s (1994) study as well as thismeta-analysis. Using raw scores, Ptacek et al. showedthat women seek more social support than men andthere was no sex difference in problem-focused cop-ing. Relative scores showed men reporting more prob-lem-focused coping than women and no sex difference

in seeking social support. A fourth study (Stanton,Tennen, Affleck & Mendola, 1992) examined relativecoping among infertile couples. Men were relativelymore likely than women to use problem-focused cop-ing and isolation, whereas women were relatively morelikely than men to seek social support and to use avoid-ance. There were no sex differences in self-blame orpositive reappraisal. Thus, sex comparisons in copingreveal quite different results depending on whether rawor relative scores are used.

Another issue to address with respect to sex differ-ences in coping, particularly as these differences relateto appraisal issues, is neuroticism. The tendency forwomen to appraise stressors as more severe than menmight not be due to sex per se but to differences in apersonality trait that is associated with sex: neuroti-cism. A study (Lynn & Martin, 1997) of 37 countriesfound that women tend to score higher on neuroticismthan men. Thus, it could be that womens higher levelsof neuroticism are responsible for womens tendenciesto appraise stressors as more severe and to engage inmore coping behavior. In fact, neuroticism has been as-sociated with high stress levels, even after controllingfor the effects of gender (Fontana & Abouserie, 1993).Moreover, because neuroticism has been associatedwith more emotion-focused coping (De Fruyt, 1997;Saklofske & Kelly, 1995), it is plausible that this per-sonality trait might differentially influence mens andwomens coping behavior preferences as well as theirstressor appraisals.

To further shed light on these issues, we comparedabsolute (raw scores) versus relative (ipsitized scores)coping in two samples. The first sample consisted ofcollege students coping with a problem at school andcoping with a relationship problem. The second sam-ple consisted of adult men and women coping with aheart problem. In the second sample, we measuredneuroticism and evaluated its relation to coping aswell as whether it accounted for any sex differencesin coping.

Study 2: Relative Coping

Method

Participants. Sample 1 consisted of 95 collegestudents (37 men, 58 women) who completed a ques-tionnaire during the first semester of their freshmanyear. The majority of these students (n = 93; 39 men, 54women) completed a second questionnaire during thefirst semester of their sophomore year. (Two men in thefirst wave did not return the initial questionnaire and 4women in the second wave did not return the secondquestionnaire.)

Sample 2 consisted of 262 cardiac patients (172men, 90 women) who were interviewed in the hospital

23


shortly after a successful angioplasty and then in theirhomes 6 months later.

Procedure. The college students completed theeight subscales from the COPE (Carver et al., 1989),shown in Table 6. They were asked to indicate howthey were coping with a school-related difficulty in thefirst semester of their freshman year. They were askedhow they were coping with a rel

Date post:	28-Sep-2015
Category:	Documents
Upload:	andres-archila-mateus
View:	36 times
Download:	2 times

Sex Differences in Coping Behavior- A Meta Analytic Review and an Examination of Relative Coping

Documents