Gender-related stress and health for Swedish working women

The validation of the women workplace culture questionnaire: gender-related stress and health for Swedish working women

Bodil Bergman

A Women Workplace Culture Questionnaire (WWC) was developed (Bergman & Hallberg, 2002) based on a grounded theory model (Bergman & Hallberg, 1997). Three factors with coefficient [alpha] > .70 were identified that concerned areas of perceived burdens on me (the individual woman), perceived burdens on women in general, and sexual harassment. A fourth factor, organizational support, also showed modest reliability. Factors I and IV were significantly correlated with self-reported ill health. Factors I and II were significantly related to psychological stress and job satisfaction. When self-reported somatic health was used as a dependent variable, 35% of the variability was explained by the factor workplace culture, which combined Factor I and Factor II (Bergman & Wright, 2000). It was suggested that masculine values were perceived as the norm, which created uncertainty and loss of control for the women and produced reactions of alarm and illness problems.

Although the women who participated in the survey were experienced in a male-dominated workplace, the data came from a limited sample in one industry, a rather high age group, and a rather high status group. There is thus a need for empirical evidence of its utility for other samples and settings. In this work of the verification of the data we have chosen women working in qualified positions or rank-and-file positions at a university, a university hospital, an outpatient clinic, and another male-dominated industry.

The hypotheses were: Do women working in male-dominated organizations such as university faculties, university hospitals, and industry give more indications of gender-related problems than women in more traditional female-dominated work organizations such as administrative staff, hospital laboratories, and out-patient clinics?

From the view of public health, gender-related harassment (in a broader sense than sexual harassment) is a problem faced by women at their workplaces and has to be viewed in the context of institutionalized male power, that is, the “chilly climate” (Cohorn, Chernyshenko, & Fitzgerald, 2002; Janz & Pyke, 2000; Pyke, 1996; Steele, James, & Barnett, 2002; Wenneras & Vold, 1997).

The desire to live in mutually beneficial relationships is in conflict with established patterns in male-dominated workplaces (Taylor, Kemeny, Reed, Bower, & Grunewald, 2000). Gender harassment constitutes a common workplace stressor that warrants serious attention (Piotrkowski, 1998). Because gender harassment is about the abuse of power and status rather than being restricted to unwelcome sexual invitations, it can lead to adverse psychological consequences as well as to impaired work performance (Robbins, Bender, & Finnis, 1997; Seymor, 1999). Important variables for health and a good work environment (Frankenhaeuser, 1996; Karasek & Theorell, 1990), such as control over work, influence, meaningfulness, support, and professionalism, seem to be negatively influenced by gender-related personal offences. Therefore, we suggest that a gender-related workplace measure might be relevant to our understanding of the aforementioned physical and mental health outcome dependent variables.

The main purpose of this study is

1. To conduct a psychometric validation of the WWQ (Bergman & Hallberg, 2002).

2. To present normative data for Swedish groups of White working women.

3. To investigate the relationship between gender-role perceptions/experiences and self-reported medical symptoms, psychological stress, and work satisfaction.

METHOD

Participants and Procedure

The empirical test of the WWC was carried out at three different work places: (1) a university, (2) a university hospital, and (3) a male-dominated industry (15% women). Our intention was to reach women with different experiences of working together with men. Nine different types of work were represented. In total, 588 women were asked to participate, and only 11 of them declined to take part in the study.

Two faculties of the university participated: the faculty of arts and the faculty of medicine. The faculty of arts included all women employed, such as the professional staff, the graduate students, and the administrative staff (N = 236), and the faculty of medicine included all the female professors (N = 91). From the university hospital women from three different professional groups were recruited with the intention to reach women in both female-dominated and male-dominated settings. These groups were convenience samples and included (a) the supervisors (N = 23) of the university hospital laboratory, which had only female employees, (b) a group of female doctors (N = 53) working in male-dominated specialties such as emergency medicine, surgery, and neurology, and (c) the staff (N = 118) of an outpatient female-dominated division (with few male colleagues) working with hearing disabled patients. Finally, a female group in a manufacturing industry was represented in the study; one white collar group (N = 48) working with computer technology as system analysts and one blue-collar group (N = 19) employed in the production of components. These groups were also convenience samples.

The questionnaires were mailed to the women employed at the university and to the doctors at the hospital. The laboratory supervisors and outpatient staff were asked to participate in the study at staff meetings, and the questionnaires were handed out there. The women employed in the industry were reached through an employee at the personnel office who was responsible for issues of gender equality at the work place. A cover letter was included, which described the purpose of the study as well as the acknowledgment of the study by the equality committees. The respondents who consented to participate were asked to mail the completed questionnaires to the research leader (BB).

The Questionnaires

The Women Workplace Culture Questionnaire (WWC)

Twenty-six of the items consist of pairs of questions in which respondents are asked (1) to compare the general status of women at work with that of men and (2) to describe their specific experiences in this regard. Here is an example of such a pair: (1) “Do you think that women have fewer opportunities than men for professional development at a work place?” and (2) “Estimate your own opportunities for professional development.”

Nine pairs of items concern central questions about working life and its influence on women and men. These include questions about competence and professional development, equal judgment of men and women, organizational support and trust, management attitude toward women at work, freedom to be oneself, employment security, attention to what women say, demands on performance and accomplishment, and assertiveness required in order to obtain fair compensation, promotions, and development opportunities. Five pairs of items deal with support for managing professional work and domestic responsibilities, as well as support from their partners, the influential role of parents, and sexual harassment at work. The remaining two items concern the availability of somebody to turn to and to speak with openly, as well as thoughts about leaving the job because of gender-related problems.

All items have fixed response alternatives. Depending on the formulation of the specific questions, however, one feature of the questionnaire is the use of different adjectives or behavioral descriptors (Bergman & Hallberg, 2002). In Table II the numbers of the scale steps are shown as well as the point rating scales. The items 22, 23, and 29 are reversed. Low scores on all items indicate perceived general negative attitudes toward women and negative attitudes specifically experienced by the women themselves.

The Work Environment and Health Questionnaire

The standardized questionnaire Work Environment and Health (Wallin & Wright, 1986; Wright, Bengtsson, & Frankenberg, 1994) was used. Questions on background data such as age, education, duration of employment, number of children and their ages were included. The scale is a 121-item questionnaire with five main areas (factor groups) covered: Physical Work Environment, Medical Symptoms, Psychological Stress, Work Satisfaction, and Quality of Life. The individual questions are answered on a 4-point scale: (1) often, (2) sometimes, (3) seldom, and (4) never. Low scores indicate symptoms or problems. In this study, the three factor groups Medical Symptoms, Psychological Stress, and Work Satisfaction were used to measure self-reported ill health, distress, and work satisfaction.

Within the factor group Medical Symptoms of 23 items, seven subfactors are included: (1) gastrointestinal symptoms, for example, stomach-ache, flatulence, diarrhea; (2) musculoskeletal symptoms, for example, pain in joints and muscles, neck, or back; (3) psychological problems, for example tiredness, restlessness, irritability; (4) sleeping problems; (5) cardiovascular symptoms that is, chest pain/tachycardia; (6) other medical symptoms, for example, headache, eye-strain, skin problems, and itching; and (7) worry about health hazards.

Within the factor group Psychological Stress of 26 items, eight subfactors are included of which we have applied four: (1) work obstacles; (2) incompatible demands; (3) work regulated by others; (4) workload.

The factor group Work Satisfaction of 17 items includes seven subfactors labelled (1) pleasure in work; (2) opportunities for development; (3) relationship with boss; (4) relationship with colleagues; (5) relationship with work group; (6) influence/recognition; and (7) tiredness/listlessness.

RESULTS

Sample Characteristics

A description of the women under study as well as the previously studied group is presented in Table I. The response rate between the participating subgroups varied from 77 to 97%, the lowest from the industry subgroup. The participants in the study were 446 White women, aged 20-60 years, with a median age range of 40-49 years. Seventy-two percent were married or in long-term relationships, and 53% had children less than 17 years of age. The education level was high, and the median length of employment was long.

The group consists of a proportionally large number of women with university degrees such as assistant and associate professors, physicians, and systems analysts. The respondents work as professors (29%), graduate students (8.5%), and administrative staff (12.6%) in areas of academia (the faculties of art and medicine), physicians with in-patients (9.4%), laboratory supervisors (4.4%) in a university hospital, rehabilitation staff with outpatients (25.9%), and finally as computer staff or blue-collar workers in a male-dominated industry (10.1%). The composition of the group thus enabled us to study White female employees in different occupations and with different experiences of working with men in different settings and at different workplaces. Compared to the previously studied group of women employed in a male-dominated industry (Bergman & Hallberg, 2002) this group has a higher level of education and younger children. The laboratory supervisors and the rehabilitation staff with outpatients work in female-dominated organizations.

With regard to self-reported somatic health the mean scores were within 1 SD of the mean scores of 673 adult females who had completed The Work Environment and Health Questionnaire at the aforementioned male-dominated industry.

Descriptive Statistics

Characteristics of the distributions of the raw scores were obtained by calculating means, standard deviations, and skewness of the distribution of raw scores for each item (see Table II). The largest standard deviations (1.02 and 0.99) were found in relation to items 14 and 11. These items deal with experiences of having to be extra prepared for talking to men professionally at meetings and greater employment security for men. On item 14 there were no significant deviations between the subgroups. However, on item 11 there were significant deviations between the professional staff at the university and the outpatient rehabilitation staff, laboratory supervisors, and the white collars, F(8, 424) = 3.81; p < .001. The three latter groups feel a better employment security than the professors, many of whom have time-limited employments. Responses to item 30, thoughts about leaving the job because of gender-related problems, were highly negatively skewed. The remaining items had 3-17 missing cases except for item 21, which had 74 (16.6%) missing cases. The number of missing cases on this item (support from partner for "double role") is related to the age of the respondents and their children; the question was no longer relevant to their situation. In these respects there were no significant discrepancies between the subgroups.

Comparisons Between the Group Under Study and the Previously Studied Group

The item responses of the group currently studied and the previously studied group (Bergman & Hallberg, 2002) were compared by mean of t test for equality of means. Significant differences between several of the items questioning perceived attitudes to women in general were observed (items 1, 5, 7, 11, 15, 17, and 25), which indicate lesser burdens on women in the presently studied group. To compare differences in item response between the respondent groups the statistical method ANOVA was applied. On items 1, 7, 15, and 17 there are no significant differences in means between the subgroups. On items 5, 11, and 25 there are, however, significant differences in the means between the subgroups. Thus the professional staff groups and the hospital doctors more frequently perceive organizational support and trust in men, F(8, 426)= 4.20; p < .001, the aforementioned employment security to men F(8, 424) = 3.81; p < .001, and sexual connotations, F(8, 427) = 5.17; p < .001, than the outpatient rehabilitation staff and the laboratory supervisors.

There are four significant differences between the items questioning experienced attitudes toward themselves as women, three positive (items 8, 16, and 26) and one negative (item 12). The discrepancies indicate more experienced negative attitudes among the previously studied industry group, with the exception of employment security. On these items the differences in means are significant between the subgroups in the presently studied group. The university employees and the hospital doctors experience less positive attitudes from superiors at work than does the outpatient rehabilitation group, F(8, 426) = 5.00; p < .001, the professional staff groups and graduate students worry more about professional security than the other more traditionally employed subgroups, F(8, 426) = 10.95; p < .001, and the professional staff groups and university hospital doctors feel a greater need to be more accomplished at work, F(8, 422) = 3.97; p < .001. On item 26 the graduates more often reported sexual connotations than did the outpatient rehabilitation group, F(8, 428) = 2.37; p < .02. The difference on item 12 depends on time-limited employment due to limited funding for professors who do research as well as for graduate students (representing 38%).

Exploratory Factor Analyses

Questionnaire item data (n = 446) were intercorrelated and subjected to principal component analyses and a varimax rotation. Pairwise deletions of missing cases were applied in the calculations. Three items were excluded: item 24 due to the categories as response alternatives, item 29 due to general neutrality, and item 30 due to strongly negative skewness. In the factor analysis five factors were extracted and rotated. This five-factor solution was supported by a graphic scree-test and Kaiser’s criterion for number of factors retained (Kline, 1995; Nunnally & Bernstein, 1994). The five unrotated factors had eigenvalues above 1. The five factors retained accounted for 55.8% of the total variance. Table III presents the factors.

Factor I, perceived burdens on women, consisted of 11 items with factor loading >.40 that explained 19% of the total variance. Four of these items (8, 14, 16, 18) also had a high factor loading in Factor II. Seven items that constituted the factor clearly addressed the women’s perceptions of general attitudes toward women at their work place, and four items addressed their own experiences of gender-related attitudes (8, 14, 16, 18). The factor contained items dealing with perceived general attitudes that women must meet at work, for example: “women have to be more accomplished in their work than men in order to be promoted,” “women have fewer opportunities for professional development than men,” “men receive more organizational support and trust than women,” “women receive more unfair judgments of their work performance than men,” “men fail to pay attention to what women say at meetings,” “men have greater employment security than women.” Shared with Factor II are the items about negative attitudes from superiors, preparation, experiences of accomplishment, and assertiveness. Factor I appears mostly to describe a perceived general, more or less conscious, :negative attitude toward women.

Factor II, personally experienced burdens, contains nine items, including the four aforementioned common items with factor I, with factor loading >.40. The explained variance was 14%. The factor contained items dealing with women’s own personal experiences of equality at work. Three of these items dealt with restricted inequality: “unfairly judged,” “I would liked to have received more organizational support and trust,” and “fewer developmental opportunities than wished for.” Two items concerned restricted liberty: “to be myself at work,” “worry about professional security.” This factor appears to describe a need for mental strength to cope with inequality in order to be as accepted as men are.

Factor III, sexual harassment, explained 10% of the total variance and included two item pairs that addressed sexual harassment. One pair was “unwelcome body contact or unwelcome suggestions at work,” and the second pair included “unwelcome connotations, glances, gestures, or comments at work,” and experiences of this type of connotations and comments. The content of the factor suggested both perceived and experienced sexual harassment of women at work.

Factor IV, inadequate organizational support, explained 8% of the total variance and consisted of three items: “women receive enough organizational support to manage their professional work and their domestic responsibilities (double role),” women’s experiences of specific “support to manage the double role,” and the item of “sufficient support from your partner.” This factor reflects the practical difficulties that working women are confronted with because, in addition to their jobs, they generally have the main responsibility for their families, regardless of number of children and professional position (Lundberg, Mardberg, & Frankenhaeuser, 1994).

In Factor V, influence by parents and siblings, one positive and one negative loading explained 5% of the total variance. The items “influence of family relationships on women’s ability to face difficulties at work” and the specific experienced “support from parents and siblings,” which deals with the respondent’s history or childhood had high loadings. This suggests that the women consider their relationships to their parents to be important for their ability to handle future events in life.

Corrected Item-Total Correlation Analyses

Corrected item-total correlations (Nunnally & Bernstein, 1994), that is, the correlation of an item with the sum of the remaining items, were calculated for each factor and are presented in Table III. The item-total correlations for the Factors I-IV exceeded, r = .25; p < .001. The items of Factor V were not statistically significant. The mean item-total correlation for Factor I, perceived burdens on women, was r = .63; p < .001; for Factor II, personally experienced burdens, was r = .60; p < .001; for Factor III, sexual harassment, r = .64; p < .001; and for Factor IV, inadequate organizational support, r = .53; p < .001.

Reliability

An estimate of internal consistency, a coefficient alpha, was computed for the factors (see Table III). Factor I, perceived burdens on women, had an estimated alpha of .87, Factor II, personally experienced burdens, was [alpha] = .84, for Factor III, sexual harassment, [alpha] = .80 and for Factor IV, inadequate organizational support, [alpha] = .71. Because Factor V, influence by parents and siblings contained a negative loading the coefficient alpha was not estimated. Considering the failure of Factor V to produce a significant item-total correlation and coefficient alpha, Factor V will not be discussed further.

Normative Data

The estimated reliabilities were acceptable for four factors ([alpha] > .70). The distribution of the scores was positively skewed, that is, the average score was above the midpoint. The standard deviations were large enough, however, to allow individual differences to be captured.

Calculation of the mean scores for items that belong to each factor forms the values of the factors (Kline, 1995). Means and standard deviations of the factors are presented in Table IV for the subgroups, the total group, and the previously studied group (Bergman & Hallberg, 2002). The average score of Factor I, perceived burdens on women, was 2.08 (SD = 0.55) and of Factor II, personally experienced burdens, was 2.54 (SD = 0.55), which indicates that the responses of the typical respondent ranged from “yes, sometimes” to “no, seldom.” The average score of Factor III, sexual harassment, was higher, 3.20 (SD = 0.44), and ranged from “no, seldom” to “no, never.” Finally, the average score of Factor IV, inadequate organizational support, was 2.37 (SD = 0.54), which meant that the typical answer varied between “need a better support” and “it is OK.” Analyses of differences between the total group under study and the previously studied group resulted in one significant value for the first factor, t = 3.77; df = 546, p = .001. This indicates that the typical answer for the previously studied group ranged from “yes, often” to “yes, sometimes,” that is, the group perceived more burdens on women than the currently studied group did.

To compare differences in response set between the respondent groups the statistical method ANOVA for comparison of equality of means for each factor was applied. The values used were transformed to z-scores. Two contrasting subgroups were found: (1) the academic professors and university hospital doctors; and (2) the rehabilitation outpatient staff. There were significant differences on Factor I, perceived burdens on women, F(8, 431) = 3.59; p < .001, between the university professional staff groups and the rehabilitation outpatient staff such that the outpatient staff perceived lesser burdens on women. On Factor II, personally experienced burdens, F(8, 431) = 6.93; p < .001, the university professional staff at the faculty of arts scored lower than the administrative university subgroup, the outpatient rehabilitation group, and the white-collar group, which indicates that the professors at the faculty of arts more often experienced gender equality problems. In comparison with the outpatient rehabilitation staff the professors at the faculty of medicine and university hospital doctors also perceived greater gender equality problems. On Factor III, sexual harassment, F(8, 431) = 3.45; p < .001, the outpatient group differed positively and significantly from the administrative staff and graduate student groups, which indicates lesser problems in this respect for the rehabilitation group working in a more female-dominated organization. The low score for the blue-collar group was probably due to the small sample size. Concerning Factor IV, inadequate organizational support, F(8, 428) = 4.24; p < .001, there was a significant difference, between on the one side, the university professional staff group at the faculty of arts and the university hospital doctors and, on the other side, the outpatient rehabilitation staff and the white-collar groups. This indicates lesser organizational support to the professors and doctors and a greater support to the rehabilitation group and the white-collar group.

The hypotheses that women working in male-dominated organizations such as university faculties and university hospitals would give more indications of gender-related problems than women in more traditional female-dominated organizations such as outpatient clinics and administrative staff group was confirmed. The gender-related problems seemed greater at the university than in the male-dominated industry. The reason may be that the white-collar group was employed in a less traditional organization on a department and group level.

Simple Correlations

A simple correlation matrix (see Table V) was constructed for the gender-related standardized factors and the factors of self-reported medical health, psychological stress, and work satisfaction. The correlations for Factor I ranged from 0.20 to 0.38 and for Factor II from 0.34 to 0.57, whereas correlations for Factor III and Factor IV were somewhat lower, 0.10-0.24 and 0.13-0.26, respectively.

DISCUSSION

This study of 446 White women in different work positions verified the dominant factors called perceived burdens on women and personally experienced burdens. Compared to the first factor analysis (Bergman & Hallberg, 2002), the two factors came out in reversed order, and the coefficient alpha was stronger and accepted for all four factors. The WWC seems to be a psychometrically sound measure of women’s perceptions/experiences of gender-related problems at work, as defined in terms of item-total correlations and internal consistency reliability. On the basis of the factor analyses, four items (17, 22, 23, and 24) were excluded, as they neither represented a meaningful separate factor nor had satisfactory factor loadings on consistent factors. Two items (29 and 30) were excluded due to a very skewed frequency distribution. Thus 24 items remained. When the mean ratings for items belonging to each factor were calculated four factors were formed: (1) perceived burdens on women; (2) personally experienced burdens; (3) sexual harassment; and (4) inadequate organizational support. Significant differences between the respondent subgroups factors indicated the usefulness of the inventory. The academic staff groups, the graduates, and the university hospital doctors working in male-dominated organizations suffered more from gender-related attitudes than did the university administrative staff who work in traditional women’s positions and the outpatient rehabilitation staff who work in organizations dominated by women. The white-collar group, who worked in a male-dominated organization (15% women) suffered comparatively less from gender-related attitudes, and it is suggested that this depends on their type of work as computer analysts in a more modern organization (Bergman, Wright, & Carlsson, 1996). On the other hand, the response rate (77%) was lower for this group, which may mean that employees with more negative experiences refused to take part in this study. Comparisons between the groups in the present study and the previously studied group showed significant differences between the scores for the factor perceived burdens on women, which indicated that the previously studied group experienced greater gender equality problems.

Our data definitely suggest a need to change aspects of the structure and culture of academic and occupational arts and sciences in order to better accommodate the socialized preferences of women and avoid systemic discrimination and microinequalities, which discriminate against women. This is in line with suggestions by Acker (1996), Pyke (1996), and Seymor (1999).

Further evidence of the construct validity of the instrument was provided by the facts that significant correlations were found between the WWC factors and health, psychosocial stress, and work satisfaction. This suggested that the gender-related factors perceived burdens on women, personally experienced burdens, sexual harassment, and inadequate organizational support are important measures of women’s health and well-being. The perceived superficial, social inequalities appear to be more related to health outcome than sexual harassment per se. On the other hand, the women reported low frequencies of sexual abuse, which may be due to the high median age of the women most of whom live in long-term relationships and the fact that some of them work in places dominated by women.

Orth-Gomer (1998) has found that good social and emotional support is beneficial to women’s health. Depressive symptoms seem to play an important role as possible prime risk factors for coronary heart disease in women (Orth-Gomer, 2001). It is important to improve the awareness of women’s discrimination as well as the understanding of the nature and sources of constraints that they encounter in their disciplinary fields. Professions that are traditionally dominated by women appear to be valued less, which can be seen in their comparatively lower salaries as well as in the reduced funding for the social sector where the percentage of women is high. Variables such as control, influence, collaboration, and professional skills, which are all characteristics of a good environment, and affect health, are important. Lack of control over work is related to heightened levels of catecolamine secretion (Frankenhaeuser, 1996). It is therefore crucial that new ways be found to alter the negative views of women because these attitudes may critically affect their health and well-being (Frankenhaeuser et al., 1989). I consider this to be an important step in the process of adequate actions against gender-related problems at work. My hope is that the WWC can be useful in these actions to reduce gender-related occupational stressors. The questionnaire is easy to use because it is moderate in length and the items are easy to understand. The WWC, however, is still under development; the consideration of race, education, and job status are important future directions and a limitation of the study. There is thus a need for further empirical evidence of the utility of WWQ for other samples and settings, such as younger women, women with lower levels of education, with young children, holding other job positions.

Table I. Description of the Presently Studied Sample and the

Previously Studied Sample

Presently Previously

studied sample studied sample

(n = 446) (n = 104)

Age

20-29 years of age 9.3% 8%

30-39 years of age 21.5% 17%

40-49 years of age 32.7% 36%

50-59 years of age 30.6% 32%

60 years or older 5.8% 6%

Married or in long-term 72.0% 77%

relationships

Children younger than 19.1% 6%

7 years of age

Children older than 33.3% 19%

7 years of age

Education

Secondary school 10.8% 23%

High school 16.7% 43%

University 70% 33%

Median years of employment 10 years or more 10 years or more

Table II. Descriptive Statistics for the Items of the WWQ (n = 446)

The group

WWQ under study

Item

no Abbreviated items Scale steps Mean SD

29 Somebody to speak openly about 2 (a) 1.83 0.37

experience of a particular

difficulty?

30 Thoughts about leaving your 2 1.94 0.24

job because of gender-

related problems?

6 Would have liked more 3 (b) 2.24 0.71

organizational support and

trust

7 Working life is characterized 3 2.10 0.58

by a negative attitude

toward women

8 Superiors have a negative 3 2.58 0.55

attitude toward you as a

woman

10 Experience of being “myself” 3 2.63 0.57

at work

12 Worry about professional 3 2.36 0.69

security

13 Men fail to pay attention to 3 2.06 0.65

what women say

19 Inadequate organizational 3 2.11 0.72

support for double role

20 Inadequate support for 3 2.47 0.62

managing the “double role”

21 Experience of sufficient 3 2.56 0.61

support from partner

22 Influence from parents and 3 1.55 0.60

siblings

26 Experience of unwelcome sexual 3 2.76 0.52

connotations

28 Experience of unwelcome body 3 2.88 0.39

contact

1 Fewer developmental 4 (c) 1.74 0.64

opportunities for women

3 Unfair judgments of women’s 4 1.98 0.73

work

4 Unfairly judged 4 2.71 0.79

5 Organizational support and 4 1.83 0.74

trust in men

11 Men’s greater employment 4 3.07 0.99

security

14 Have to be extra prepared 4 2.52 1.02

15 Women have to be more 4 1.72 0.74

accomplished

16 Need to be more accomplished 4 2.12 0.87

17 Women are less assertive in 4 1.91 0.75

obtaining a fair salary

18 Had to be more assertive 4 2.30 0.90

23 Feelings of support from 4 3.09 0.98

parents and siblings

25 Unwelcome sexual connotations 4 3.52 0.74

27 Unwelcome body contact 4 3.69 0.56

2 Fewer developmental 5 (d) 2.67 0.95

opportunities than I would

wish for

9 Difficult for women to “be 5 2.58 0.96

themselves” at work

24 Relationships with the members Category (e)

of the family of origin

The

group

under The previous

study group

WWQ

t value of

Item mean-

no Abbreviated items Skewness Mean differences

29 Somebody to speak openly about -1.78 1.82 0.56

experience of a particular

difficulty?

30 Thoughts about leaving your -3.58 1.95 -0.57

job because of gender-

related problems?

6 Would have liked more -0.38 2.19 0.57

organizational support and

trust

7 Working life is characterized -0.005 1.9 3.18 **

by a negative attitude

toward women

8 Superiors have a negative -0.88 2.37 3.42 ***

attitude toward you as a

woman

10 Experience of being “myself” -1.17 2.56 1.07

at work

12 Worry about professional -0.63 2.52 -2.13 *

security

13 Men fail to pay attention to -0.003 1.96 1.38

what women say

19 Inadequate organizational -0.16 2.13 -0.31

support for double role

20 Inadequate support for -0.75 2.47 0.01

managing the “double role”

21 Experience of sufficient -1.08 2.61 -0.80

support from partner

22 Influence from parents and 0.59 2.57 -15.49 ***

siblings

26 Experience of unwelcome sexual -2.08 2.63 2.20 *

connotations

28 Experience of unwelcome body -3.27 2.85 0.72

contact

1 Fewer developmental 0.46 1.49 3.64 ***

opportunities for women

3 Unfair judgments of women’s 0.36 1.87 1.33

work

4 Unfairly judged -0.08 2.61 1.2

5 Organizational support and 0.50 1.6 3.36 ***

trust in men

11 Men’s greater employment -0.64 2.61 4.15 ***

security

14 Have to be extra prepared 0.01 2.46 0.53

15 Women have to be more 1.07 1.5 2.88 **

accomplished

16 Need to be more accomplished 0.53 1.87 2.64 **

17 Women are less assertive in 0.58 1.72 2.3 *

obtaining a fair salary

18 Had to be more assertive 0.17 2.23 0.63

23 Feelings of support from -0.87 3.17 -0.69

parents and siblings

25 Unwelcome sexual connotations -1.35 3.29 2.81 **

27 Unwelcome body contact -1.76 3.73 -0.64

2 Fewer developmental 0.37 2.67 -0.06

opportunities than I would

wish for

9 Difficult for women to “be 0.02 2.53 0.48

themselves” at work

24 Relationships with the members

of the family of origin

Note. The items are ranked according to number of scale steps.

(a) Yes; no.

(b) Definitely; to some degree; not at all.

(c) Often; sometimes; seldom; never.

(d) Fewer, than I would wish; sometimes fewer; about what I had

wished; somewhat more frequent; much more frequent.

(e) Both my parents and/or my siblings have had about the same

importance for me; my mother has been most important for me; my

father had been most important for me; my siblings had been most

important for me.

* p < .05 (two-tailed). ** p < .01 (two-tailed).

*** p < .001 (two-tailed).

Table III. Item-Total Correlations and Factor Structure

Factor

loadings

No. Abbreviations [r.sup.it] I II

15 Women have to be more accomplished .70 .76# .13

1 Fewer developmental opportunities .63 .73# .16

for women

5 Organizational support and trust .66 .73# .27

in men

3 Unfair judgments of women’s work .70 .72# .28

7 A negative attitude toward women .62 .67# .17

13 Men fail to pay attention to what .64 .66# .23

women say

16 Need to be more accomplished .68 .63# .42#

18 Had to be more assertive .60 .49# .49#

14 Have to be extra prepared .59 .46# .46#

11 Men’s greater employment security .51 .45# .26

4 Unfairly judged .67 .33 .75#

6 Would have liked more .68 .33 .72#

organizational support and trust

2 Fewer developmental opportunities .51 .21 .66#

than I would wish for

10 Never “be myself” at work .53 .18 .61#

12 Worry about professional security .39 -.01 .55#

8 Superiors have a negative attitude .58 .48# .52#

toward you as a woman

9 Difficult for women to “be .48 .36 .39

themselves” at work

28 Experience of unwelcome body .69 -.05 .03

contact

25 Unwelcome sexual connotations .63 .15 .09

27 Unwelcome body contact .61 .06 .08

26 Experience of unwelcome sexual .63 .08 .07

connotations

20 Inadequate support for managing .62 .13 .22

the “double role”

21 Experience of sufficient support .43 .13 .07

from partner

19 Inadequate organizational support .54 .38 .16

for double role

22 Influence from parents and siblings -.05 .01 -.04

23 Support from parents and siblings -.16 -.01 .03

17 Women are less assertive in .02 .36 -.11

obtaining a fair salary

Explained variance 19% 14%

Reliability coefficient alpha .87 .80

Factor

loadings

No. Abbreviations III IV V

15 Women have to be more accomplished -.00 .21 .07

1 Fewer developmental opportunities .02 .09 .05

for women

5 Organizational support and trust .03 -.06 .06

in men

3 Unfair judgments of women’s work .06 .06 .14

7 A negative attitude toward women .12 .16 -.16

13 Men fail to pay attention to what .18 .12 .01

women say

16 Need to be more accomplished .01 .20 .05

18 Had to be more assertive .02 .14 -.01

14 Have to be extra prepared .17 .29 .08

11 Men’s greater employment security .04 .38 -.19

4 Unfairly judged .02 .04 -.06

6 Would have liked more .01 .09 -.06

organizational support and trust

2 Fewer developmental opportunities .00 .07 -.07

than I would wish for

10 Never “be myself” at work .23 .13 -.01

12 Worry about professional security .10 .39 .07

8 Superiors have a negative attitude .15 .10 -.18

toward you as a woman

9 Difficult for women to “be .33 .06 -.10

themselves” at work

28 Experience of unwelcome body .79# .09 .10

contact

25 Unwelcome sexual connotations .79# .11 -.15

27 Unwelcome body contact .75# -.07 -.03

26 Experience of unwelcome sexual .75# .03 .08

connotations

20 Inadequate support for managing .10 .80# -.03

the “double role”

21 Experience of sufficient support -.02 .67# .06

from partner

19 Inadequate organizational support .07 .64# -.00

for double role

22 Influence from parents and siblings .01 .16 .74#

23 Support from parents and siblings -.02 .16 -.71#

17 Women are less assertive in -.02 .23 .37

obtaining a fair salary

Explained variance 11% 8% 5%

Reliability coefficient alpha .80 .71

Note. The factor items are sorted by size. Extraction method:

Principal component analysis. Rotation method: Varimax with Kaiser

normalization. Rotation converged in five iterations,

[r.sup.it] = corrected item-total correlation, calculated for each

factor. r = .10, p < .001; r = .13, p < .01; Degrees of

freedom = 400. Factor loading >.40 are marked in bold.

Note: Factor loading >.40 are marked in #.

Table IV. Normative Data of the WWC Based on Swedish Working Women

(n = 446)

Factor I Factor II

Profession n M SD M SD

University faculty of medicine 64 1.93 0.51 2.43 0.53

University faculty of arts 68 1.93 0.60 2.25 0.48

University graduate students 37 2.01 0.61 2.47 0.56

University administrative staff 55 2.09 0.61 2.55 0.57

University hospital doctors 41 2.02 0.57 2.42 0.50

University laboratory supervisors 19 2.15 0.36 2.56 0.37

Outpatient staff 114 2.22 0.49 2.77 0.42

White-collar workers 30 2.25 0.43 2.71 0.47

Blue-collar workers 14 2.37 0.38 2.52 0.40

Total group 446 2.08 0.55 2.54 0.55

Previously studied group 104 1.86 (a) 0.45 2.49 0.56

Factor III Factor IV

Profession M SD M SD

University faculty of medicine 3.28 0.34 2.31 0.55

University faculty of arts 3.14 0.40 2.22 0.54

University graduate students 3.00 0.56 2.32 0.57

University administrative staff 3.13 0.63 2.28 0.52

University hospital doctors 3.20 0.46 2.15 0.56

University laboratory supervisors 3.24 0.36 2.59 0.39

Outpatient staff 3.36 0.29 2.50 0.49

White-collar workers 3.13 0.41 2.66 0.42

Blue-collar workers 2.98 0.50 2.43 0.62

Total group 3.20 0.44 2.37 0.54

Previously studied group 3.12 0.47 2.41 0.45

(a) The difference between means in relation to the total group is

significant at the .001 level (two-tailed).

Table V. Pearson’s Product-Moment Correlation Between the Factors

and Medical Symptoms, Psychological Stress, and Work Satisfaction

With Subfactors (n = 446)

Factor Factor Factor Factor

I II III IV

Medical symptoms .23 *** .39 *** .17 *** .30 ***

Gastrointestinal symptoms .15 ** .32 *** .22 *** .26 ***

Musculoskeletal symptoms .18 *** .29 *** .13 ** .20 ***

Psychological problems .25 *** .44 *** .19 ** .32 ***

Sleeping problems .15 *** .26 *** .12 * .20 ***

Cardiovascular symptoms .19 *** .24 *** .09 .21 ***

Psychological stress .20 *** .30 *** .10 * .13 **

Work obstacles .20 *** .29 *** .17 *** .09

Incompatible demands .18 *** .26 *** .10 * .13 **

Work regulated by others .22 *** .27 *** .08 .12 **

Workload .15 *** .21 *** .04 .12 **

Work satisfaction .38 *** .56 *** .24 *** .26 ***

Pleasure in work .12 ** .25 *** .15 ** .11 *

Opportunities for development .08 .15 ** .05 .03

Relationship with boss .33 *** .43 *** .15 ** .20 ***

Relationship with colleagues .26 *** .39 *** .15 *** .16 ***

Relationship within work .35 *** .51 *** .26 *** .16 ***

group

Influence/recognition .12 ** .35 *** .14 ** .10 *

Tiredness/listlessness .26 *** .40 *** .18 *** .36 ***

* p < .05 (two-tailed). ** < .01 (two-tailed).

*** p < .001 (two-tailed).

REFERENCES

Acker, J. (1996, September). Gender and work: Does difference make a difference? Key-note address presented at the 25th International Congress on Occupational Health (ICOH), Stockholm, Sweden.

Bergman, B., & Hallberg, L. R.-M. (1997). Swedish women in a male-dominated industry: A qualitative study of gender-related problems. Journal of Gender, Culture, and Health, 1, 45-63.

Bergman, B., & Hallberg, L. R.-M. (2002). Women in a male-dominated industry: Factor analyses of a woman workplace culture questionnaire based on a grounded theory model. Sex Roles, 46, 305-316.

Bergman, B., & Wright, I. (2000). Self-reported health in relation to medical health and gender-specific problems in women. Journal of Occupational and Environmental Medicine, 42, 311-317.

Bergman, B., Wright, I., & Carlsson, S. G. (1996). Women’s work experiences and health in a male-dominated industry. Journal of Occupational and Environmental Medicine, 38, 663-672.

Cohorn, C. A., Chernyshenko, S., & Fitzgerald, L. F. (2002, April). The development and validation of a Gender-Related Work Discrimination Scale. In A. J. Ormerod (Chair), Negative experiences in the workplace: Discrimination, harassment, and aggression. Symposium presented at the 17th Annual Conference of the Society for Industrial and Organizational Psychology, Toronto, Canada.

Frankenhaeuser, M. (1996). Stress and gender. European Review, 4, 313-327.

Frankenhaeuser, M., Lundberg, U., Fredrikson, M., Melin, B., Tuomisto, M., Myrsten, A.-L., Bergman-Losman, B., Hedman, M., & Wallin, L. (1989). Stress on and off the job as related to sex and occupational status in white-collar workers. Journal of Occupational Behaviour, 10, 321-346.

Janz, T. A., & Pyke, S. W. (2000). A scale to assess student perceptions of academic climates. Canadian Journal of Higher Education, 30, 89-122.

Karasek, R., & Theorell, T (1990). Healthy work. Stress, productivity and the reconstruction of working life. New York: Basic Books.

Kline, P. (1995). The handbook of psychological testing. London: Routledge.

Lundberg, U., Mardberg, B., & Frankenhaeuser, M. (1994). The total workload of male and female white-collar workers as related to age, occupational level, and number of children. Scandinavian Journal of Psychology, 4, 315-327.

Nunnally, J., & Bernstein, I. H. (1994). Psychometric: theory (3rd ed.). New York: McGraw-Hill.

Orth-Gomer, K. (1998). Psychosocial risk factor profile in women with coronary heart disease. In M. Chesney & K. Orth-Gomer (Eds.), Women, stress, and heart disease (pp. 25-38). Mahwah, NJ: Erlbaum.

Orth-Gomer, K. (2001). Women and heart disease: New evidence for psychosocial, behavioral, and biological mediators of risk and prognosis. International Journal of Behavioral Medicine, 8, 251-269.

Piotrkowski, C. (1998). Gender harassment, job satisfaction, and distress among employed White and minority women. Journal of Occupational Health Psychology, 3, 33-43.

Pyke, S. W. (1996). Education and the “woman question.” Canadian Psychology, 38, 154-163.

Robbins, I., Bender, M. P., & Finnis, S. J. (1997). Sexual harassment in nursing. Journal of Advanced Nursing, 25, 163-169.

Seymor, E. (1999). The role of socialization in shaping the career-related choices of undergraduate women in science, mathematics and engineering majors. In C. Selby (Ed.), Women in science and engineering: Choices for success (pp. 118-126). New York: New York Academy of Sciences.

Steele, J., James, J. B., & Barnett, R. C. (2002). Learning in a man’s world: Examining the perceptions of undergraduates in male-dominated academic areas. Psychology of Women Quarterly, 26, 46-50.

Taylor, S. E., Kemeny, M. E., Reed, G. M., Bower J. E., & Grunewald, T. L. (2000). Psychological resources, positive illusions, and health. American Psychologist, 55, 99-109.

Wallin, L., & Wright, I. (1986). Psychosocial aspects of the work environment: A group approach. Journal of Occupational Medicine, 28, 384-393.

Wenneras, C., & Void, A. (1997). Nepotism and sexism in peer review. Nature, 387, 341-343.

Wright, I., Bengtsson, C., & Frankenberg, K. (1994). Aspects of psychological work environment and health among male and female white-collar and blue-collar workers in a big Swedish industry. Journal of Organizational Behaviour, 15, 177-183.

Bodil Bergman (1)

(1) To whom correspondence should be addressed at Department of Psychology, Box 500, Goteborg University, 405 30 Goteborg, Sweden; e-mail: bodil.bergman@psy.gu.se.

COPYRIGHT 2003 Plenum Publishing Corporation

COPYRIGHT 2003 Gale Group