INTERRELATIONSHIPS OF ROLE CONFLICT, ROLE AMBIGUITY, ANDWORK-FAMILY CONFLICT WITH DIFFERENT FACETS OF JOBSATISFACTION AND THE MODERATING EFFECTS OF GENDER

James S. Boles, John Andy Wood, and Julie Johnson

The effects of role conflict, role ambiguity, and work-family conflict on overall salesperson job satisfaction have beenexamined in previous studies across work settings. Less attention has been given to the interrelationships between roleconflict, role ambiguity, and work-family conflict with various facets of job satisfaction. The moderating role of genderhas also received relatively little attention in the sales force management literature than it may warrant. In a study of 129business-to-business salespeople, it was found that the relationships of work-related role stress and work—family conflictwere different for the various facets of job satisfaction. Further, there were significant differences among these relation-ships between male and female salespeople. Sales management implications of these findings are presented and topics forfuture research are identified.

Role stress, as originally theorized by Katz and Kahn (1966),results from an employee's role conflict and role ambiguity.The existing literature on role stress is extensive and rangesacross almost all types of work environments (Fisher andGitelson 1983; Jackson and Schuler 1985; Johnston et al.1990). In particular, the area of professional selling has seenan extensive exploration of stresses resulting from work rolesand the work-related effects of stress (e.g., Behrman andPerreault 1984; Sager 1994). Over the past 20 years, researchhas also explored employee outcomes resulting from anothertype of stress. This additional type of stress occurs as a resultof trying to reconcile demands arising from an individualhaving obligations in both work and family roles.

There is evidence that stress arising from trying to simul-taneously resolve the conflicts between the duties of all ofthese roles can affect an employee's job satisfaction (e.g.,Higgins and Duxbury 1992; Kossek and Ozeki 1998). Fur-ther, the ambiguity that arises from being uncertain as to howmuch latitude in behavior a salesperson has in resolving theseconflicts can also reduce job satisfaction (Behrman andPerreault 1984; Fisher and Gitelson 1983). Findings havedemonstrated that work-family conflict (WFC)—a form of

James S. Boles (Ph.D., Louisiana State University), Associate Pro-fessor of Marketing, J. Mack Robinson College of Business, Geor-gia State University, mktjsb@langate.gsu.edu.

John Andy Wood (MBA, University of North Florida), DoctoralCandidate, J. Mack Robinson College of Business, Ceorgia StateUniversity, mktjawx@langate.gsu.edu.

Julie T. Johnson (Ph.D., Georgia State University), DeSoto FamilyFellow, Department of Marketing, Western Carolina University,jjohnson@email.wcu.edu.

inter-role conflict where the role pressures from work andfamily domains are mutually incompatible to some degree—negatively affects satisfaction within the family system(Netemeyer, Boles, and McMurrian 1996). In addition, otherstudies also indicate that WPC and its resultant stress canafl ect employees' attitudes toward work and their job satis-faction (Good, Sisler, and Gentry 1988; Greenhaus and Beutell1985).

However, there appear to be limitations within the exist-ing literature examining the interrelationships between roleconflict, role ambiguity, WFG, and job satisfaction. Mostearlier studies that include all three of these measures of stressuse measures of job satisfaction that are global, measuring anemployee's overall job satisfaction without considering thatthere may be mtiltiple facets of employee satisfaction that cotildbe differentially affected by work-related stress or WFG(Behrman and Perreault 1984; Boles, Johnston, and Hair1997; Sager 1994). These global job satisfaction measures failto account for the multiple dimensions of satisfaction such assatisfaction with coworkers, satisfaction with pay, satisfactionwith supervision, and so on, demonstrated by the scales suchas the INDSALES measure (Ghurchill, Ford, and Walker1974). Using an overall measure of job satisfaction can resultin job satisfaction scores, implying that all employees withsimilar scores are equally satisfied with all facets of the job.This limitation is particularly true for the more limited lit-erature examining the WFG —> job satisfaction relationship.

It appears that previous studies of the relationship betweenrole stress and job satisfaction for salespeople presume simi-larity of expectations and outcomes across individual per-sonality traits and characteristics. As noted by Schul and Wren(1992), many of these studies may have presumed that sales-people, regardless of gender, react similarly across all facets

Journal of Pmonai Selling & Sales Management, \o\. XXIII, no. 2 (spring 2003), pp. 99-113.© 2003 Pi Sigma Epsilon, Inc. All rights reserved.

ISSN 0885-3134 / 2003 $9.50 + 0.00.

100 Journal of Personal Selling & Sales Management

of job satisfaction. No clear evidence exists to support theassumption of homogeneity of role expectations and job sat-isfaction across gender. In fact, some existing research sug-gests that there may be differences between men and womenregarding the interrelationships between role conflict, roleambiguity, WFC, and job satisfaction (e.g., Babin and Boles1998; McNeilly and Goldsmith 1991).

The current study will extend the understanding of threekey elements of role stress—^WFC, role conflict, and role am-biguity—and their relationships with the multiple facets ofjob satisfaction. This research will explore if each dimensionof job satisfaction is differentially predicted by some or all ofthe stresses resulting from the salesperson filling multiple rolesincluding: internal employee of the firm, boundary-spannerbridging the gap between her or his firm and customers, andparticipant in nonwork relationships involving friends andfamily members (e.g.. Boles, Johnston, and Hair 1997; Singh1998). In addition, the current study will examine gender asa specific individual characteristic that differentially affectsreported levels of each of the dimensions of job satisfaction.

To summarize, this study extends the current sales litera-ture in two ways. First, the research examines the effects ofWFC and work-related role stress on various facets of jobsatisfaction. Second, it examines employee gender as a mod-erator of the relationships of role conflict, role ambiguity, andWFC with various facets of job satisfaction.

LITERATURE

Job Satisfaction

Job satisfaction is an important work-related attitude amongsalespeople for several reasons. First, satisfaction with the jobis directly related to organizational commitment (Brown andPeterson 1993). In addition, job satisfaction is either directly(Netemeyer, Johnston, and Burton 1990) or indirectly (Brownand Peterson 1994) related to a salespersons turnover inten-tions. Turnover intentions are a good indicator of future turn-over (Futrell and Parasuraman 1984).

Although some research has found a link between WFCand job satisfaction (Boles and Babin 1996; Good, Sisler, andGentry 1988), other studies have not found a direct relation-ship between those constructs (Bedeian, Burke, and Moffet1988). These earlier studies—both those reporting a signifi-cant relationship between WFC and job satisfaction and thosethat did not support that linkage—used a global measure ofsatisfaction as opposed to assessing employee job satisfactionacross its multiple facets. To accurately measure "job satisfac-tion," a number of characteristics of the job may need to beevaluated if one hopes to obtain a broad measure of employeebeliefs and attitudes about the job (Churchill, Ford, andWalker 1974). These characteristics or facets may not be of

equal importance to various individuals. Thus, it is possiblethat different types of role stress do not equally influence thesedifferent facets of an employee's job satisfaction.

This research explores rwo complementary explanationsof why outcomes of job satisfaction may differ across respon-dents. Evidence for the differential effects of role stress willcome from tests of the significance of the relationship of eachof the role stress constructs with each dimension of job satis-faction. In addition, individual characteristics of the sales-person may influence role expectations and thus the outcomeof the evaluation of each facet of job satisfaction. Whereasmany individual characteristics may influence the above evalu-ation process such as career-stage or job tenure, this studyuses the individual characteristic of gender. Gender is chosenbased on the extensive theoretical foundation on gender rolesas well as the body of empirical research supporting theseperspectives.

Work-Family Conflict

In modern, developed nations, most individuals participatein multiple roles in their lives. Increasingly, for botb men andwomen, the two primary roles as an adult are work and fam-ily (Zedeck 1992). Frequently, expectations from these twomajor life roles can be incompatible, resulting in a form ofinter-domain conflict called work-family conflict (Greenhausand Beutell, 1985; Netemeyer, Boles, and McMurrian 1996).For this reason, interest in WFC and its impact on employersas well as employees is growing, as evidenced by articles fromthe disciplines of sociology, psychology, and management(Conger and Rueter 1999; Livingston and Burley 1996; Mar-tins 2002; Phillips-Miller, Campbell, and Morrison 2000).Discussion concerning the importance of conflict betweenwork and family domains bas even expanded into the popu-lar press. An article in Business Week (Hammonds 1996) ratedbusinesses on how well they address issues in the work/familyrole system.

It appears that occupations involving some form of bound-ary-spanning activity are more likely to face work-related rolestress and to also have negative work-related outcomes asso-ciated with WFC (Bacharach, Bamberger, and Conley 1991).For example, salespeople, as boundary-spanners, are subjectedto a variety of demands from multiple sources, including in-ternal work roles and external expectations of customers, inaddition to their family roles (Boles, Johnston, and Hair 1997;Goolsby 1992). Previous studies suggest stress from role con-flicts affect the job satisfaction of salespeople (e.g., Behrmanand Perreault 1984; Netemeyer, Boles, and McMurrian 1996).

The failure of many previous sales force studies to con-sider tbe interrelationship between family as well as workdomains may have resulted in research findings that do not

spring 2003 101

fully account for the salesperson's perception of the existenceof conflict. Employee stress is a "perceptional phenomenonresulting from a comparison between the demand on a per-son and his ability to cope. An imbalance in this mechanism,when coping is important, gives rise to the experience of stress,and to the stress response" (Cox 1979, p. 25). This transac-tional view highlights the importance of perception and therelationship of the individual to the environment (i.e., worksetting, family setting). If there is an improper fit between anindividual and his or her environment, the individual experi-ences stress.

For "WFC to occur, it appears that an individual must per-ceive role demands from both the work and family domainsas legitimate. In an employee's work/family roles, conflict isinterchangeable along both the work and family dimensions.Previous research has demonstrated that conflict between thework and family domain is correlated with attitudes aboutthe workplace and job (e.g., Bacharach, Bamberger, andConley 1991; Boles and Babin 1996; Netemeyer, Boles, andMcMurrian 1996). The role member must believe that be-cause of the existence of mutually exclusive role demands aresolution to the conflict does not exist. In the person's per-ception, the conflict must strain their values, desires, or goals.

The existing academic literature examining potential con-flicts between the work and family domains is based on re-search examining a wide range of occupations. Results of thesestudies consistently demonstrate that stress can result fromconflicts between work and family roles (e.g., Lynagh andMurphy 1996; Phillips-Miller, Campbell, and Morrison2000). Conflict between work roles and family roles has beenrelated directly or indirectly to job outcomes such as job dis-satisfaction, employee burnout, and turnover (Bacharach,Bamberger, and Conley 1991; Burke 1989; Frone, Russell,and Cooper 1992; Good, Sisler, and Gentry 1988). Thus,WFC is an issue that may provide some additional insightinto stress among members of a firm's sales force (Wilson1997).

Given that the perception of stress will only occur whenrole expectations are perceived to be incompatible, it is notsuggested that WFC will afifect all dimensions of job satisfac-tion. Rather, the nature of the role expectations will controlthe conflicts. For example, the employment role naturally cre-ates time constraints on nonwork or family roles. Present fi-nancial reward in the form of pay or potential rewards from apromotion may mitigate this temporal conflict given suffi-cient enumeration. However, if the reward is perceived as in-sufficient to compensate for time demands then thesatisfaction with pay will decrease. The above framework sug-gests the following:

Hla: Work-family conflict will be negatively related tosatisfaction with pay.

Hlb: Work—family conflict will be negatively related tosatisfaction with promotion.

Similarly, perceived conflict between family and work isprimarily the result of the structural requirements of the job.Policies and the individual that administers the policy areperceived as the arbitrators that can resolve the conflict. How-ever, as the stress from WFC increases, the flexibility of thepolicy and the ability of the supervisor to alleviate conflictsmay be limited. Unresolved stress results in a decrease in re-ported satisfaction with both policy and supervisor.

Hlc: Work—family conflict will be negatively related tosatisfaction with policy.

Hid: Work—family conflict will be negatively related tosatisfaction with supervisor.

Role Conflict and Role Ambiguity

The importance of role conflict and role ambiguity in bound-ary-spanning positions is evident from previous research (e.g.,Behrman and Perreault 1984; Sager 1994) and it is not thepurpose of this paper to provide an extensive review of thatbody of literature. Certainly, the existence and influences ofrole conflict among boundary-spanning positions is well docu-mented (e.g., Netemeyer, Johnston, and Burton 1990). Re-search indicates that employees in boundary-spanningpositions, whose jobs involve direct interaction with custom-ers or clients, will face conflicting demands from customers,supervisors, and coworkers (Churchill et al. 2000). These con-flicting role expectations appear to be a major component ofwork-related role stress (Behrman and Perreault 1984). Sinceprevious research suggests not all of the dimensions of jobsatisfaction will involve conflicting expectations the follow-ing hypotheses are tested.

H2a: Role conflict will be negatively related to satisfactionwith work.

H2b: Role conflict will be negatively related to satisfactionwith coworkers.

H2c: Role conflict will be negatively related to satisfactionwith supervisor.

H2d: Role conflict will be negatively related to satisfactionwith customers.

Role ambiguity, another aspect of work-related role stress,results when the duties and actions required of an employeeare unclear to the employee. Research indicates the strain as-sociated in dealing with issues of unclear responsibility isstrongly related to job stress (Behrman and Perreault 1984).A direct relationship between role conflict, role ambiguity,and job satisfaction has been widely reported in the organiza-

102 Journal of Personal Selling & Sales Management

tional literature (Fisher and Gitelson 1983; Jackson andSchuler 1985) as well as the sales literature (e.g., Netemeyer,Johnston, and Burton 1990). For role ambiguity to exist, aperceived lack of clarity must exist for the salespersons. Thisis likely to occur in situations involving interpersonal rela-tions, which indicate the following hypotheses. In addition,the nature of boundary-spanning positions requires salespeopleto satisfy multiple expectations in an uncertain environment.

H3a: Role ambiguity will be negatively related to satisfac-tion with work.

H3b: Role ambiguity will be negatively related to satisfac-tion with coworkers.

H3c: Role ambiguity will be negatively related to satisfac-tion with supervisor.

H3d: Role ambiguity will be negatively related to satisfac-tion with the customer.

Employee Differences Based on Gender

Previous work in nonmarketing settings finds that gender dif-ferences exist in job attribute preferences (e.g.. Mason 1995;Wiersma 1990). These studies indicate that male and femaleemployees may well have different perceptions of their roles.Consequently, men and women can experience different lev-els of role conflict and role ambiguity from the work domain.For example, cultural and societal expectations of parentalresponsibility are strongest for mothers who may, therefore,feel more obligated to deal with family matters even at theexpense of the job—possibly resulting in the perception ofgreater WFC—whereas men may tend to put the job first(Gilbert 1992; Marsiglio 1993).

There are at least two possible theoretical explanations fordifferent outcomes of role stress for male and female sales-people. First, there is the concept of a physio-biological dif-ference leading to different desired attributes in the workenvironment. Tbis position suggests females seek work rolesthat do not interfere with family roles, especially mother-hood (Brown 1970; Browne 1998; Mackey and Coney 2000).The second perspective predicts women are socialized intocommunal behaviors and men into more agentic or instru-mental values (Eagly 1987). These different theoretical per-spectives suggest men and women will have differentexpectations of job conditions and to report different work-related outcomes based on role demands from both workand nonwork settings.

A recent theoretical perspective suggests a convergence ofthese two theoretical perspectives as they relate to career. Thisrecent perspective, the dual-impact model (Abele 2000), sug-gests that psychological role adaptations can be, but are notnecessarily, based both on biological differences as well as

adaptations to those domains requiring social role adjustmentssuch as careers. As an example, the biological necessity of child-bearing by the female in conjunction with societal expecta-tion of female child-rearing creates expectations of and bythe female salesperson that the female will focus on familyroles. These expectations can result in ambiguity in botb familyand career roles. However, for men, a societal expectation as"breadwinner" does not have biological necessity, but doesimpact both job and family roles.

The issue of possible differences in the work-related out-comes and attitudes of male and female salespeople has be-come of greater importance in recent years due to theincreasing numbers of women filling professional sales posi-tions. Whereas previous studies of differences in perceptionsof the workplace between male and female salespeople wereexpected to provide clear differences, results have generallyshown few differences in mean ratings between the percep-tions and attitudes of the two sexes (Busch and Bush 1978;Schul and Wren 1992; Sigauw and Honeycutt 1995). Thedifferences tbat bave been discovered have often been incon-clusive or contradictory, and possibly indicate that these dif-ferent levels of reported perceptions may be, at least somewhat,sample-specific. However, a recent study of service employ-ees found only one small difference in means between menand women, but reported substantial differences in bow theconstructs related to one another (Babin and Boles 1998).That finding has yet to be replicated in sales research, butmay offer an indication that the outcomes of various stressorsdiffer for male and female salespeople.

Despite the lack of conclusive evidence in differential cog-nitions about career by gender, otber research suggests tbatthere are systematic differences in the job-related values ofmen and women. Specifically, Mason proposes that "womenare satisfied witb jobs in which they can interact with othersin a supportive and cooperative way . . . whereas men's agenticorientation is manifested in self-assertion, self-expansion, andthe urge to master" (1995).

Based on the perspective that men and women have con-trasting expectations of jobs, it is proposed that role conflicts,ambiguous roles, as well as WFC will result in different evalu-ations of each facet of satisfaction. Therefore, it is hypoth-esized tbat:

H4: For men and women there is a structural difference inthe effect of the three role stressors on satisfaction with:

H4a: type of work;

H4b: coworkers;

H4c: pay;

H4d: supervisor;

H4e: promotions;

spring 2003 103

H4f: policy;

H4g; customers.

It appears from the perspective of the dual impact model,an important predictor of satisfaction is role stress related toWFC. The two most salient societal views that also supportthe agentic versus communal dichotomy of men and womenare that men are the primary source of family income andthat women are the primary source of familial cohesiveness,including child rearing. From this perspective, the followinghypothesis are tested.

H5; For males:

H5a: Work-family conflict will be negatively related tosatisfaction with pay.

H5b: Work-family conflict will be negatively related tosatisfaction with promotion.

H5c: Work-family conflict will be negatively related tosatisfaction with policy.

Previous research into tbe societal as well as personal ex-pectations of women suggests a particular susceptibility tostress related to conflicts between work and family. Tbe limi-tations of time to satisfy both work and family obligationswill lead to the occurrence of coworkers "covering" or substi-tuting for a time-constrained salesperson. This resolution ofa WFC will result in obligations to coworkers. The dual im-pact model mentioned previously suggests the following hy-pothesis.

H5: For females:

H5d: Work-family conflict will be negatively related tosatisfaction with work.

H5e: Work—family conflict will be negatively related tosatisfaction with coworkers.

H5f: Work-family conflict will be negatively related tosatisfaction with policy.

Eagly (1987) suggests that men's agentic or instrumentalwork roles are "status asserting." This perspective implies thatmen will experience significant role stress with those facets ofjob satisfaction closely associated with agentic role expecta-tions. For men, facets not salient to the agentic role will notbe significantly related to role stressors. Therefore, for men,perceptions of role ambiguity as they relate to interpersonalinteractions will negatively relate only to satisfaction withcoworkers and supervisors.

H6: For males:

H6a: Role ambiguity will be negatively related to satisfac-tion with coworkers.

H6b: Role ambiguity will be negatively related to satisfac-tion with supervisors.

In addition, for men, tbe impediment of "status asserting"or instrumental goals of achievement will result in dissatis-faction with job facets related to status.

H6: For males:

H6c: Role conflict will be negatively related to satisfactionwith promotion.

H6d: Role conflict will be negatively related to satisfactionwith policy.

The desire of women to achieve communal goals suggeststhat interpersonal interactions are the "means" to relieve ambi-guity. For women, perceptions of role ambiguity as they re-late to interpersonal interactions will positively relate tosatisfaction with coworkers and supervisors.

H7: For females:

H7a: Role ambiguity will be positively related to satisfac-tion with coworkers.

H7b: Role ambiguity will be positively related to satisfac-tion with supervisor.

However, conflict will result in a negative relationship withthe interpersonal facets of satisfaction. The betrayal of theexpectation of support from the communal group will leadfemales to exhibit a stronger or higher level of a negative rela-tionship to each facet of job satisfaction.

H7: For females:

H7c: Role conflict will be negatively related to satisfactionwith coworkers.

H7d: Role conflict will be negatively related to satisfactionwith supervisor.

H7e: Role conflict will be negatively related to satisfactionwith policy.

METHODS AND MEASURES

Surveys were distributed to all 149 sales employees of a re-gional promotions firm. Thus, the study represents a censusof one firm's sales force. Of those, 129 completed responseswere received, for a response rate of over 80 percent. Thehigh response rate was due in part to the firm endorsing thestudy and helping distribute the surveys directly to the sales-people. The completed survey instruments were returned di-rectly to the research team in postage-paid envelopes. Resultsfrom the study were only reported to the sponsoring firm inaggregate so salespeople knew that their individual responses

104 foumal of Personal Selling & Sales Management

retnain confidential. Salespeople for this ftrm could beviewed as fulfilling the role of institutional seller (Moncrief1986).

Cotnpensation v as based on both salary and a bonus fortneeting the quota established by the firtn. Respondents aver-aged slightly over 32 years of age and had approximately fiveyears of sales experience. Approximately 60 percent of thesalespeople had a four-year college degree. Sixty percent indi-cated they were married or had children living at home. Sta-tistical tests of other demographic variables as indicators ofeither the dependent or independent variables are not signifi-cant. Additional tests of these demographic variables, such asmarital status, presence of dependents, education, or job ten-ure, do not suggest any confounding effects. All measures usedin the current study have seen previous use in organizationalresearch. Role conflict and role ambiguity were assessed us-ing the Rizzo, House, and Lirtzman (1970) instrument. Thismeasure has been used in many different work settings tomeasure these constructs. Role conflict was measured witheight items and role ambiguity with six items. Reliability forthese measures was 0.81 for both scales.

The various facets of job satisfaction were measured usinga reduced version of the INDSALES scales (Comer, Machleit,and Lagace 1989; Lagace, Goolsby, and Gassenheimer 1993).The scales assess satisfaction with work, coworkers, pay, pro-motion, supervisor, policy, and customers. Reliability for thesescales is 0.85, 0.75, 0.75, 0.80, 0.83, 0.74, and 0.68, respec-tively. Salesperson responses were given on 1—7 point scaleswhere smaller numbers indicated less job satisfaction. Thesereduced scales were used in an attempt to keep the surveyinstrument to a reasonable length since these scales have atotal of 28 items as opposed to the 95 items of the fullINDSALES measure (Churchill, Ford, and Walker 1974).

WFC was assessed using an adaptation of the Burke, Weir,and DuWors's (1979) scale assessing the influence of workactivities on home responsibilities. It has been used in otherresearch studies across a wide range of work settings (e.g.,Bedeian, Burke, and MofFet 1988; Burke 1989). Items fromthis scale are worded so that they are valid for both marriedand unmarried employees. Previous research suggests that un-married individuals also experience conflict between theirwork duties and nonwork responsibilities (Babin and Boles1998). Responses were based on a five-point scale where greaternumbers reflect increasing levels of conflict between the twodomains. Reliability for this measure was 0.80. Gender wasself-reported on a single item as one of many demographicitems included in the survey.

The relationship of the independent variables—role con-flict, role ambiguity, and WFC—^with the various facets ofjob satisfaction was assessed using linear regression. Responsesfor each role stress construct are mean-centered to avoid multi-coUinearity, whereas gender is dummy coded with one for

males and zero for females. The results of these fitted regres-sions are presented in Table 1. Then, to determine if gendersuggests a specific structural difference, each gender has thesame regression fitted for each facet of job satisfaction. Re-sults for each gender as well as the i^-statistics generated usingthe Chow test (1960) are calculated and also included in Table3. The i^-statistic is calculated by examining the ratio of theunexplained variance or error (sum of squares of the residu-als, SSRcombined) from regression line undifferentiated bygender to the unexplained variance from fitted regressionsthat are differentiated by gender (SSRfemale and SSRmale).The formula for calculating the i^-statistic is

SSRcombine d - (SSRfemale + SSRmale)r = /

K

SSRfemale -I- SSRmale

M + N-2P

The results from the Chow test examines if specific struc-tural diff^erence exists in a regression model (Bleaney 1990)between men and women using all the constructs of role stress.Significant i^-statistics indicate more variance is explained bydiff^erentiating the groups than by analyzing them as a homo-geneous group.

The use of moderated regression to examine the specificinfluence of each role stress construct for each gender allowsfor interpretation of differential effects of gender with eachrole stressor by interpretation of the significance of the inter-action. The reported f-test for the coefficient is interpretableas indicative of each construct's relationship to each facet ofjob satisfaction. The results of the following fitted moderatedregressions are presented in Table 2.

B^ (WFC) +

5g (Gender*WFC) + e,

where/5. is each of the facets of job satisfaction.Correlations between the constructs and reliability mea-

sures for the constructs included in the study are presented inTable 3. Correlations between the constructs for men andwomen are presented separately in Table 4.

RESULTS

As the results in Table 1 indicate, each of the role stressors(role conflict, role ambiguity, and WFC) is differentially re-lated to the various facets of job satisfaction. The results inTables 1 and 2 indicate that the gender of the salespersondoes moderate the relationship of role conflict, role ambigu-ity, and WFC on the various facets of job satisfaction. In

105

I"o

-Iu

Wl

O'wi

4)

I

so

CMpO

rM

o

l

a.o

pd

oci

Opd

o

* * *hv ro ON ON ( N roro ro ^^ ^~ — ro

* OOd NOr^ON NONOQ^lOOOO O c N ^ . |N»roXjr o r o o o r o r o r o O O o

r O T T — f N l r o — O O o O r O L O ' ^ L O

— LO ro

— — O

^ 00 ^(N (N fNj

oddf fs O

— — rod d d

p p p fvj

d d d d d d

ro 00 D00 >o 1^IN ro fSd d d

CO "^ O NON fS 00ro ^ rod d d d d

< N — ; r N |USUSLO

ooNO

0 0 ^ — m r o f SO ^ O O L O h o O O O N

co

Ipd

106 Journal of Personal Selling & Sales Management

Table 2Results of Moderated Regressions

Mean R-squareRole

AmbiguityRole

ConflictWork-Family

Conflict

Satisfaction with Work (interaction effects)H2 a Males

FemalesSatisfaction with Coworkers (interaction effects)H2b Males

FemalesSatisfaction with Pay (interaction effects)H2c Males

FemalesSatisfaction with Supervisor (interaction effects)H2d Males

FemalesSatisfaction with Promotion (Interaction effects)H2e Males

FemalesSatisfaction with Policy (interaction effects)H2f Males

FemalesSatisfaction with Customer (interaction effects)H2g Males

Females

* Significant at the 0.05 level.

0.327*5.505.830.0685.175.270.274.334.680.0034.944.91

-0.711*4.734.02

-^.0774.904.820.368*4.745.12

0.3080.2890.2450.2110.1220.356O.IOI0.0780.0640.2050.1750.2470.3530.3680.2760.4180.4240.3890.0920.0410.031

0.319-0.466*-0.147

0.484*-0.362*

0.1220.003

-0.0220.0120.952*

-0.357*0.595*0.452

-0.541*-0.088

0.424*-0.376*

0.0480.062

-0.145-0.083

0.213-0.313*-0.099-0.201-0.102-0.303-0.186-0.084-0.271-0.810*

0.0 II-0.798*-0.412-0.388*-0.800*-0.052-0.327*-0.379*

0.065-0.036

0.029

-0.919*0.077

-0.841*-1.008*

0.004-1.004*

0.529-0.435*

0.094-0.007-0.552*-0.476

0.615-0.452*

0.163-0.192-0.649*-0.841*-0.078-0.113-0.192

fact, the signs of the coefficients are different for male andfemale salespeople for some facets, with females experiencinga significant positive relationship between role ambiguity andsatisfaction with supervisor. Each of the sub-hypotheses willbe examined in detail below.

Sub-Hypotheses Results

The results of the tests of H3a and H2a indicate that roleambiguity and role conflict are significantly and negativelyrelated to satisfaction with work for the total sample. How-ever, WFC is not a significant predictor of satisfaction withwork. When examining males and females separately, role am-biguity and role conflict are negatively related to satisfactionwith work for males. Conversely, WFC is negatively relatedto satisfaction with work for females, H5d, but not for males.A Chow test supports H4a, which shows that there is a sig-nificant difference between males and females in role con-flict, role ambiguity, WFC, and satisfaction with work.

Analysis of H3b and H2b indicates that role conflict androle ambiguity are significantly and negatively related to sat-isfaction with coworkers for the total sample. WFC is not asignificant predictor of satisfaction with coworkers. Whenexamining males and females separately, role ambiguity is nega-tively related to satisfaction with coworkers for males, sup-

porting H6a. Evidence for differential effects based on gen-der is also indicated by the significant interactions for WFCand role ambiguity between men and women, as shown inTable 2. As suggested, WFC is negatively related to satisfac-tion with coworkers for females, H5e. Role conflict is not asignificant predictor of satisfaction with coworkers for eithermales or females. A Chow test supports H5b, which indi-cates a significant difference between males and females inrole conflict, role ambiguity, WFC, and satisfaction with co-workers.

The results of H l a indicate that WFC is significantly andnegatively related to satisfaction with pay for the total sample.However, role ambiguity and role conflict are not significantpredictors of satisfaction with pay for the full sample for ei-ther gender. When examining males and females separately,WFC is negatively related to satisfaction with pay for males,H5a, but not for females. Results of the Chow test do notsupport H4c, indicating that there are no significant differ-ences between males and females in the relationships of roleconflict, role ambiguity, and WFC with satisfaction with pay.

The results of H i d indicate that WFC is significantly andnegatively related to satisfaction with the supervisor for thetotal sample, whereas role conflict, H2c, and role ambiguity,H3c, are not significant predictors of satisfaction with thesupervisor. When examining males and females separately.

107

2

•ooa.!S

ee

I =K oU

1.1I

cou X

If IIS

C 1.O Ou X •>ui i W

•S *• a

.2 ^

.2 §tJ X «

<S i t

J « ^44n

V)

" « >-

in

o00o

m m— T m00 m m

m o

O ^ O *— ro — 00s ^ n A

0 0 0

o O ^ O ^ rtCO ro (N <N (N ro

o d d o o o

u^ CO UI IT) — mo o sO rs 10 ro voCO fS - ^ p ^ ^

d d d d

CO0

o o o

( N CO ^O CO

*******C N r S K C N ^ N O ^ O s O

m r o — r ^ r o h s c o r v m

v o r ^ m r ^ t N N O m *L O v O C O O S L O f N — m O I ^

0 0 0 0 0 0 0 0 0 0

B ^ § i,? J S 5

U

0 0 0 0O

s4S

* * 1 « -_ c c .2P e E

•p M w « O P O u..§ .j^ S S i < u i

•5 •5 ^

a: a: >

"S

l/pd

108

III

I"

•i S

5

J?

52

co

J2

m o o t ( N

*00ro

* * * *i o O O O O r o

LO ro — lo ro • ^ —

— <N p p (NO O O d d

* &^2L&.(N vO -. LO >O•^ rn ro 10 roO O O O d

* * *— vO fSh« O 00ro *O NO

00 n o o— •«•_ — ( S

0 0 0 0ro —; p ro (N

d d d o c>

* «O Mro rro T

•< lO

r o*rsCOro

rM —

0 0 0 0 0 0

t»rMroLO— —00

r-(N — —OO^iorvroCTvr o p p p — —0 0 0 0

rooOfSCTr-.

p0 0 0

*ro

rod 11

<NtNOO

*LO

<N LO

d d

*00LOo

*00

o

fSr>4CNo1

*rM

oLOo1

c

odV

spring 2003 109

role ambiguity, H6b, and WFC are negatively related to satis-faction with the supervisor for males. For females, role ambi-guity, H7b, is positively related to satisfaction with thesupervisor. Evidently, men see role ambiguity as a problemrelated to their supervisor whereas females look to their su-pervisor as a potential source of resolution of ambiguity. Roleconflict is also a significant negative predictor of satisfactionwith the supervisor for females, H7d. The Chow test of H4dis supported, which suggests that there are differences betweenmales and females in the relationships of role conflict, role am-biguity, and WFC with satisfaction with the supervisor. Theresults of the moderated regression indicate that the significantdifferences occur with role ambiguity and role conflict.

The results of H l b indicate that WFC is significantly andnegatively related to satisfaction with promotion for the totalsample. Role conflict, H6c, as well as role ambiguity are sig-nificant predictors of satisfaction with promotion for males.Support for H5b indicates that males may view promotion asa form of compensation that mitigates WFC. The negativerelation between role conflict and satisfaction with promo-tion by females may indicate issues of time constraints or is-sues with lack of promotion opportunities. The outcome ofthe test of H4e again suggests a structural difference betweenmen and women. However, results in Table 2 do not indicatea significant interaction for any of the role stressors and gen-der, rather the significant result is with the intercept. Thissuggests overall females are less satisfied with opportunitiesfor promotion than males.

The result of H l c indicates that WFC is significantly andnegatively related to satisfaction with policy for the totalsample, as well as for males separately, H5b. WFC is the onlysignificant predictor of satisfaction with policy for females.H4f is not supported, which suggests that there are no differ-ences between males and females in role conflict, role ambi-guity, WFC, and satisfaction with policy.

The lack of support for H3d and H2d indicates that roleconflict, role ambiguity, and WFC are not significantly re-lated to satisfaction with the customer for the total sample, orfor males or females separately. H5g is not supported, whichsuggests that there are no differences between males and fe-males in role conflict, role ambiguity, WFC, and satisfactionwith the customer. These results may indicate that these re-spondents as boundary-spanners by nature have a customerorientation. The lack of support suggests that, for both menand women, a significant relationship between role stress andsatisfaction with customers does not exist.

DISCUSSION

This study indicates that the relationship between work-re-lated role stress, WFC, and job satisfaction is more complexthan previously reported. In earlier research, role conflict, role

ambiguity, and WFC were found to be negatively related tojob satisfaction. The current study indicates that each of thevarious dimensions of job satisfaction may have different (evennonsignificant) relationships with role conflict, role ambigu-ity, or WFC. Moreover, these relationships may be, and fre-quently are, moderated by the sex of the salesperson.

Summarizing the findings of the study, role ambiguity orrole conflict are negatively related to males' satisfaction withwork, satisfaction with coworkers, satisfaction with supervi-sors, satisfaction with promotion, and satisfaction with policy.However, role ambiguity or role conflict are only negativelyrelated to females' satisfaction with supervisors, satisfactionwith coworkers, satisfaction with promotion, and satisfactionwith policy. However, for females, a positive relationship ex-ists between role ambiguity and satisfaction with supervisor.Conversely, WFC for women is negatively related to satisfac-tion with work, satisfaction with coworkers, and satisfactionwith policy. For men, decreased WFC is related to increasedlevels of satisfaction with pay, satisfaction with supervisor,satisfaction with promotion, and satisfaction with policy.These findings indicate that there is no one "best" policy thatwill positively influence all types of job satisfaction.

At first glance, it appears that management should takeefforts to reduce role stressors so that they could increaseemployee job satisfaction. Results of this study indicate thistype of management action could lead to erroneous decisions.Role stressors differentially affect each facet of job satisfac-tion. The gender of the salesperson plays an important role inboth the level of satisfaction on the different facets of jobsatisfaction and their relationships with role conflict, roleambiguity, and WFC. Activities undertaken to relieve a stres-sor and thus improve the employees' satisfaction with a par-ticular facet of job satisfaction may create counterproductiveeffects in another dimension of job satisfaction. Further, someresearch suggests that there may be a positive relationshipbetween moderate levels of role conflict and performance forsome marketing employees (Babin and Boles 1998). Thus,completely eliminating salesperson role conflict may result inan unintended decrease in performance.

Although employers cannot make significantly differentpolicies for men and women, employers need to recognizethat the outcomes of policy changes will have different effectson males and females. In order to improve satisfaction withwork, coworkers, supervisor satisfaction, promotion satisfac-tion, and policy satisfaction, employers need to determinethose aspects of the work environment and home environ-ment that generate role conflict, role ambiguity, and WFC.They then need to selectively develop policies and proceduresaimed at reducing levels of role conflict, role ambiguity, andWFC for specific groups of employees based on those factorsthat cause these negative perceptions.

Clearly, any attempt to reduce work-related role stress or

110 Journal of Personal Selling & Sales Management

WFC will have to balance differential affects on each gender.Consider, for instance, a specific attempt to lower males roleambiguity as it relates to the supervisor, which is related posi-tively to female salespeoples' satisfaction with the supervisor.Currently sales managers may limit their variability of re-sponses to specific stressors in an attempt to provide a senseof clarity. Males may welcome such clarity in role interac-tions, whereas females may experience an inverse response asthey welcome managers' ability to customize responses to in-dividual situations.

As another example, perhaps the company intervention inhandling stress is a weekend retreat managed by a sensitivitytrainer. For some females, this approach may actually increasethe level of WFC because it would take them away from theirfamily activities and responsibilities for the weekend. Like-wise, a supervisor sensitive to the results of this study wouldnot use an appeal tailored to promotion as a means to miti-gate the immediate conflict. Instead, he or she might allowadditional future vacation time (even unpaid) to help decreaseperceptions of WFC among this group of employees.

A way to reduce role stress is by providing supervisors withtraining on how to give clear and accurate guidance to em-ployees. Using the information from this study, clear objec-tives for the whole sales force can be integrated with agreedupon objectives for each salesperson, along with an actionplan to accomplish those objectives. Clearly defined job ex-pectations would assist this process. Moreover, supervisorsshould be provided training so that they have contacts in thecompany to assist the salespeople when they are engaged inconflicts that result from their boundary-spanning activities.

Whereas making changes discussed above should help maleemployees to be more satisfied with most aspects of their jobs,the impact of these changes on female job satisfaction appearto be more modest. Reducing WFC will produce better re-sults in making women satisfied with the dimensions of thejob than would efforts at easing role conflict. Women are of-ten stressed from trying to be a "good employee," a "goodwife," and a "good mother." Part of this stress is associatedwith cultural expectations, and part of this stress is a result ofwomen carrying more responsibility than their spouses formaintaining the family life (Crosby and Jaskar 1993). Oneway to address this concern is to offer training to help sales-people learn to better manage the interface between work andpersonal life. This training could address some of the root causesof WFC and how to avoid those problems when possible.

Employers could accommodate the demands on workingpeople in a variety of ways. One way to help reduce WFC isby providing on-site day care as well as care for sick children.Another approach might involve offering a job-sharing op-tion where two part-time salespeople could share accountmanagement duties—thereby providing full-time coverage byusing two part-time salespeople instead of one full-time sales-

person. Of course, sharing account management tasks wouldrequire excellent communication skills, team skills, and in-frastructure to ensure that accounts being managed jointlywould receive satisfactory service. It would also involve pro-viding training to the salespeople in such a program to ensurethat they understood their responsibilities toward one anotherand to customers.

Family friendly policies aimed at alleviating the negativeinfluences of WFC could help increase women's satisfactionwith work, coworkers, and policy. Whereas these benefitsshould also be extended to men, the reduction in their WFCis likely to be related to different results—making men hap-pier with their supervisors, or happier with their prospects ofbeing promoted. Overall, helping all employees learn how tomanage their time and responsibilities better, both at workand at home, should help to reduce WFC and work-relatedrole stress.

Satisfaction with pay has another useful managerial out-come based on gender. It appears that management may suc-cessfully use compensating pay as a tactic for alleviating WFCfor males, but this particular tactic may have little impact onfemales. One reason for this may be that, in many families,men are expected to conform to the "breadwinner" stereo-type. By making more money, men may feel less conflictedabout the possibility of spending less time with their wife andchildren. However, pay satisfaction is not related to femalesalespeople's perceptions of conflict that results from work-ing outside of the home. In some instances, such as for em-ployees who are feeling high levels of WFC, it may be moreeffective from an organizational standpoint to hire additionalsalespeople instead of adding accounts to each salesperson.Conversely, for some salespeople, the opportunity to makemore money by working longer hours on additional accountsmay reduce the negative effects of WFC because of the orien-tation of those salespeople toward making more money—allowing them to do more for their families financially. If asalesperson does exceed performance expectations, acknowl-edging their efforts formally in an award ceremony, attendedby the spouse and family, may also help eliminate some WFCby demonstrating to the family the importance of theemployee's work to overall firm success.

FUTURE RESEARCH AND LIMITATIONS

Results from this research suggest a number of avenues forfuture research. First, future sales force research should exam-ine the effects of family—work conflict (FWC) as well as fur-ther examine the work and nonwork effects of WFC. This isimportant because FWC has been demonstrated to have dif-ferent relationships with a number of constructs when com-pared to WFC. An additional step would involve determiningif male and female salespeople react differently to FWC.

spring 2003 111

Another avenue of study could include specific analyses(possibly via qualitative analysis of diary information) con-cerning daily events that trigger perceptions of WFC and FWCamong professional salespeople and other boundary-spanningemployees. Since boundary-spanners appear particularly sus-ceptible to WFC and its effects, analyzing diaries from sales-people might provide valuable, managerially actionableinformation for sales managers. Finally, examining the mod-erating effects of gender in models of salesperson behaviormight uncover other differences regarding how men andwomen respond to various aspects of the sales environmentthat could help improve sales force management practices.

This study is limited by its sample. A larger study mayprovide a more "balanced" test of the interrelationships be-tween the constructs. However, the study does represent acensus of all salespeople of one firm who were willing to par-ticipate. Finally, the results of this study are not necessarilyrelevant for other types of sales positions. Future research needsto be conducted in other sales environments to determine ifthe results reported here are generalizable to a broad range ofsales settings.

CONCLUSION

Findings from this study indicate that employers need totake steps to reduce role stress and WFC in order to in-crease employee's satisfaction in the different dimensions oftheir jobs. Reductions in role stress and WFC will have dif-ferent effects on men and women's satisfaction with the vari-ous dimensions of job satisfaction. Moreover, changes thatwill improve one dimension of job satisfaction may not berelated to increases in job satisfaction in a different dimen-sion. Consequently, management should take the demo-graphics of their workforce into consideration, along withthe potential dimensions of job dissatisfaction before mak-ing policy changes aimed at reducing role stress and WFC.This is very important, given the role job satisfaction playsin generating organizational commitment and reducing quit-ting intentions.

REFERENCES

Abele, Andrea E. (2000), "A Dual-Impact Model of Gender andCareer-Related Process," in The Developmental Social Psy-chology of Gender, Thomas Eckes and Hanns M. Trautner,eds., Mahwah, NJ: Lawrence Erlbaum, 361—388.

Babin, BarryJ., and James S. Boles (1998), "Employee Behaviorin a Service Environment: A Model and Test of PotentialDifiFerences Between Men and ^omen," Journal of Market-ing, G2 {A^n\), 77-n.

Bacharach, Samuel B., Peter Bamberger, and Sharon Conley(1991), "Work-Home Conflict Among Nurses and Engi-neers: Mediating the Impact of Role Stress on Burnout and

Satisfaction at Work," foumal of Organizational Behavior,12 (1), 39-54.

Bedeian, Arthur G., Beverly G. Burke, and Richard G. MofTet(1988), "Outcomes of Work-Family Conflict Among Mar-ried Male and Female Professionals," foumal of Manage-ment, 14(3), 475-491.

Behrman, Douglas H., and William D. Perreault, Jr. (1984), "ARole Stress Model of the Performance and Satisfaction ofIndustrial SAes^trsons," foumal of Marketing, 48 (Fall), 9 -21.

Bleaney, Michael (1990), "Some Comparisons ofthe RelativePower of Simple Tests for the Structural Change in Regres-sion MoAds," foumal of Forecasting, 9 (5), 437-444.

Boles, James S., and BarryJ. Babin (1996), "On the Front-Lines:Stress, Conflict, and the Customer Service Provider,"/o«r-nal of Business Research, 37 (September), 41—50.

, Mark W.Johnston, and Joseph E Hair, Jr. (1997), "RoleStress, Work-Family Conflict, and Emotional Exhaustion:Inter-Relationships and Effects on Some Work-RelatedConsequences," foumal of Personal Selling and Sales Man-agement, 17, 1 (Winter), 17-28.

Brown, Judith K. (1970), "A Note on the Division of Labor,"American Anthropologist, 72 (5), 1073-1078.

Brown, Steven P, and Robert A. Peterson (1993), "Antecedentsand Consequences of Salesperson Job Satisfaction: Meta-AnaJysis and Assessment of Causal ^iitcts," foumal of Mar-keting Research, 30 (February), Gi—77.

, and (1994), "The Effect of Effort on Sales Per-formance and Job Satisfaction," foumal of Marketing, 58(April), 70-81.

Browne, Beverly A. (1998), "Gender Stereotypes in Advertisingon Children's Television in the 1990s: A Cross-NationalAndlysK," fournal of Advertising, 27 (Spring) 83-96.

Burke, Ronald J. (1989), "Some Antecedents and Consequencesof Work-Family Conflict," in Work and Family Theory, Re-search and Applications, Elizabeth Goldsmith, ed., NewburyPark, CA: Sage, 287-302.

,Tamara Weir, and Richard E. DuWors, Jr. (1979), "TypeA Behavior of Administrators and Wives' Reports of Mari-tal Satisfaction 3ind'We\[-Bemg," Journal of Applied Psychol-ogy, 64 (January), 57-65.

Busch, Paul, and Ronald E Bush (1978), "Women Contrastedto Men in the Industrial Sales Force: Job Satisfaction, Val-ues, Role Clarity, Performance, and Propensity to Leave,"Journal of Marketing Research, 15 (August), 438-448.

Churchill, Gilbert A., Jr., Neil M. Eord, and OrviUe C. Walker(1974), "Measuring the Job Satisfaction of Industrial Sales-people," Journal of Marketing Research, 11 (Augtist), 254—260.

, , , Mark W. Johnston, and John F. Tan-ner (2000), Sales Force Management, 6th ed., Boston: IrwinMcGraw Hill.

Comer, James M., Karen A. Machleit, and Rosemary R. Lagace(1989), "Psychometric Assessment ofa Reduced Version ofINDSALES," Journal of Business Research, 18 (4), 291-309.

Conger, Rand D., and Martha Rueter (1999), "Couple Resil-ience to Economic Viessurc," Journal of Personality and So-cial Psychology, 76 (1), 54-72.

112 foumal of Personal Selling & Sales Management

Cox, Thomas (1979), "Psychophysiological Response to Occu-pational Stress," in Response to Stress: Occupational Aspects—A Collection of Papers Presented at the Ergonomics Society'sConference, Colin McKay and Thomas Cox, eds., Cam-bridge, MA: IPC Science and Technology Press, 1-30.

Crosby, Faye J., and Karen L. Jaskar (1993), "Women and Menat Home and at Work: Realities and Illusions," in GenderIssues in Contemporary Society, Stuart Oskamp and MarkCostanzao, eds., Newbury Park, CA: Sage, 143-172.

Eagly, Alice H. (1987), "Sex Differences in Social Behavior: A So-cial-Role Interpretation, "Hillsdale, NJ: Lawrence Eribaum.

Fisher, Cynrhia D., and Richard Gitelson (1983), "A Meta-Analy-sis of the Correlates of Role Conflict and Ambiguity, "/o«r-nal of Applied Psychology, 68 (May), 320-333.

Frone, Michael R., Marcia Russell, and M. Lynne Cooper (1992),"Antecedents and Outcomes of Work-Family Conflict: Test-ing a Model of the Work-Family Interhce," foumal of Ap-plied Psychology, 77 (February), 65-78.

Futrell, Charles M., and A. Parasuraman (1984), "The Relation-ship of Satisfaction and Performance to Salesforce Turn-over,"/o«r«^/o/A/<jr>&fr%, 48, (Fall), 33-40.

Gilbert, Richard K. (1992), "Revisiting the Psychology of Men,"foumal of Humanistic Psychology, 32 (2), 41-68.

Good, Linda K., Grovalynn F. Sisler, and James W Genrry (1988),"Antecedents of Turnover Intentions Among Retail Man-agement Personnel," foumal of Retailing, 64 (Fall), 295-314.

Goolsby, Jerry R. (1992), "A Theory of Role Stress in BoundarySpanning Positions of Marketing Organizations," foumalof the Academy of Marketing Science, 30 (2), 155-164.

Greenhaus, Jeffrey H., and Nicholas J. Beutell (1985), "Sourcesof Conflict Between Work and Family Roles," Academy ofManagement Review, 10 (January), 200—215.

Hammonds, Keith (1996), "Balancing Work and Family," Busi-ness Week, 6, 3493 (September 16), 74-80.

Higgins, Christopher A., and Linda E. Duxbury (1992), "Work-Family Conflict: A Comparison of Dual-Career and Tradi-tional-Career Men," foumal of Organizational Behavior, 13,4 Quly), 389-411.

Jackson, Susan E., and Randal S. Schuler (1985), "A Meta-Analy-sis and Conceptual Critique of Research on Role Ambigu-ity and Role Conflict in Work Settings," OrganizationalBehavior and Human Decision Processes, 36 (August), 16—78.

Johnston, Mark W., Charles M. Futrell, A. Parasuraman, andWilliam C. Black (1990), "A Longitudinal Assessment ofthe Impact of Selected Organizational Influences onSalespeople's Organizational Commitment During EarlyEmployment," foumal of Marketing Research, 27 (August),333-344.

Katz, Daniel, and Robert L. Kahn (1966), The Social Psychologyof Organizations, New York: Wiley.

Kossek, Ellen Ernst, and Cynthia Ozeki (1998), "Work-FamilyConflict, Policies, and the Job-Life Satisfaction Relation-ship: A Review and Directions for Organizational Behav-ior—Human Resources Research," foumal of AppliedPsychology, 83 (2), 139-149.

Lagace, Rosemary, Jerry R. Goolsby, and Jule B. Gassenheimer

(1993), "Scaling and Measurement: A Quasi-ReplicativeAssessment of a Revised Version of INDSALES," foumal ofPersonal Selling and Sales Management, 13,1 (Winter), 6 5 -72.

Livingston, Mary M., and Kim Burley (1996), "The Importanceof Being Feminine: Gender, Sex Role, Occupational andMarital Role Commitment, and Their Relationship to An-ticipated Work-Family Conflict," foumal of Social Behaviorand Personality, 11 (5), 179-194.

Lynagh, Peter M., and Paul R. Murphy (1996), "Career-RelatedPerspectives Regarding Women in Logistics: A Compara-tive Analysis," Transportation foumal, 56 (1), 35—43.

Mackey, Wade C , and Nancy S. Coney (2000), "Cultural Evo-lution and Gender Roles: A Re-AfFirmation of J.K. Brown'sNote," International foumal of Comparative Sociology, 36( 3 ^ ) 285-298.

Marsiglio, William (1993), "Contemporary Scholarship on Fa-therhood," foumal of Family Issues, 14 (December),484-510.

Martins, Luis L. (2002), "Moderators of the Relationship Be-tween Work-Family Conflict and Career Satisfaction," Acad-emy of Management foumal, 45 (April), 399-409.

Mason, Sharon E. (1995), "Gender Differences in Job Satisfac-tion,"/owrw/z/o/'5oai2//'yyc/^o/ogy, 135 (April), 143-151.

McNeilly, Kevin, and Ronald E. Goldsmith (1991), "The Mod-erating Effects of Gender and Performance on Job Satisfac-tion and Intentions to Leave in the Sales Eorct," fournal ofBusiness Research, 22 (3), 219-232.

Moncrief, William C , III (1986), "Selling Activity and SalesPositional Taxonomies for Industrial Salesforces," Journalof Marketing Research, 23 (August), 261-270.

Netemeyer, Richard G., James S. Boles, and Robert McMurrian(1996), "Development and Validation of Work-FamilyConflict and Family-Work Conflict ScAts," foumal of Ap-plied Psychology, 81 (August), 400-410.

, Mark W. Johnston, and Scott Burton (1990), "Analysisof Role Conflict and Ambiguity in a Structural EquationsFramework," foumal of Applied Psychology, 75 (April),148-157.

Phillips-Miller, Diane L., Jo N. Campbell, and Charles R.Morrison (2000), "Work and Family: Satisfaction, Stress,and Spousal Support," foumal of Employment Counseling,37(1), 16-31.

Rizzo,JohnR., Robert J. House, and Sidney L. Lirtzman (1970),"Role Conflict and Ambiguity in Complex Organizations,"Administrative Science Quarterly, 15, 150-163.

Sager, Jeffrey K. (1994), "A Structural Model DepictingSalespeople's Job Stress, "foumal of the Academy of Market-ing Science, 22 (Winter), 74-85.

Schul, Patrick L, and Brent M. Wren (1992), "The EmergingRole of Women in Industrial Selling: A Decade of Change,"foumal of Marketing, 56 Quly), 38-54.

Siguaw, Judy A., and Earl D. Honeycutt (1995), "An Examina-tion of Gender Differences in Selling Behaviors and JobAttitudes," Industrial Marketing Management, 24 (January),45-52.

Singh, Jagdip (1998), "Striking a Balance in Boundary-Spanning

spring 2003 113

Positions: An Investigation of Some Unconventional Influ- Wilson, Wayne J. (1997), "From the Practitioner's Desk: A Com-ences of Role Stressors and Job Characteristics on Job Out- ment on Role Stress, Work-Family Conflict, and Emotionalcomes o( Salespeople," Journal of Marketing, 62 QxAy), 69-87. Exhaustion: Inter-Relationships and Effects on Some Work-

Wiersma, Uco J. (1990), "Gender Differences in Job Attribute Related Constructs," Journal of Personal Selling and SalesPreferences: Work-Home Role Conflict and Job Level as Management, 17, 3 (Summer), 51—52.Mediating Variables," Journal of Occupational and Organi- Zedeck, Sheldon (1992), Work, Families and Organizations, SanzationalPsychology, 63 (September), 231—244. Francisco: Jossey-Bass.