Mobile Edition
Print
Get the Mag
Weekly UpdatesHypothesis 2: When differences in satisfaction with facets of one's job (interesting job, pleasant environment, job security/fringe benefits, friendly co-workers, advancement opportunities, and a supportive work environment) are controlled for, gender differences in pay satisfaction and pay expectations are reduced.
The Impact of Turnover Intentions and Salary
The importance of current pay as a predictor of pay satisfaction and pay expectations also deserves attention. While no theoretically compelling argument exists, several research studies (Berkowitz et al., 1987; Dyer and Theriault, 1976; Ronan and Organt, 1973; Schwab and Wallace, 1974; Sweeney et al., 1990) have found pay satisfaction to be significantly related to current income. As noted by Sweeney et al. (1990), the failure of early studies (e.g., Major and Konar, 1984) to include current pay as a predictor of pay satisfaction and pay expectations may have produced some misleading relationships among the remaining predictor variables. Lawler and Porter (1963) investigated the relationship between current salary and perceptions of what pay "should be." Among a sample of vice presidents as well as a sample of lower-level managers, perceptions of what pay should be increased in a concave fashion with increasing pay. Accordingly, we will investigate the impact of current pay on gender differences in pay expe ctations. Finally, the impact of anticipated turnover on gender differences in pay expectations is investigated. Hom and Griffeth (1995) completed a meta-analytic study of the causes and correlates of turnover. Their conclusion, based on 15 studies of the relationship between gender and turnover, is that men are more likely to quit than women.
Borjas (1984) reports the results of an investigation of the effects of turnover on earnings. The study is based on the National Longitudinal Surveys of Young Men (age 14 to 24 in 1966) and mature men (age 45 to 59 in 1966). Data were collected for four two-year periods between 1966 and 1975. There were 8,153 young men and 7,408 mature men studied. Among young white men the average inflation adjusted percent change in pay per two-year period for stayers, voluntary quits and involuntary separations was +10%, + 13%, and +9%, respectively. Among mature white men the figures for stayers, voluntary quits, and involuntary separations was +4%, +3%, and -0.5%, respectively. A similar pattern was observed among black males.
Because previous research, as noted by Hom and Griffeth (1995), suggests that men have a higher incidence of voluntary turnover and because voluntary turnover appears to be associated with larger pay increases in compensation than received by those who do not leave, we investigate whether intention to leave accounts for part of the gender differences in pay expectations.
Hypothesis 3: When differences in current salary and intention to quit one's current job are controlled for, gender differences in pay expectations are reduced.
Method
Sample
This investigation analyzes data amassed by the National Opinion Research Center (NORC) for the Graduate Management Admissions Council (GMAC) during 1985. The data set is comprised of a random sample of first-year graduate students from colleges and universities offering programs leading to an MBA or MBA-equivalent degree. Of the one hundred schools contacted, 91 participated in the study. Schools accredited by the International Association for Management Education as well as nonaccredited schools were sampled. Students received the surveys from representatives of the participating schools and returned them directly to NORC in a pre-addressed, stamped envelope.
A total of 2,054 responses were received from a random sample of 2,794 full-time and part-time students, resulting in a 73.5% response rate. Given our interest in studying pay satisfaction, we limited our analysis to respondents who were currently employed full-time and had more than 12 months of work experience since graduating from college. This reduced the number of respondents for our analysis to 721. Mean substitutions for missing data were used in cases where only data for one variable was missing. Cases were eliminated if data were missing for more than one variable. The final sample was comprised of 716 respondents.
Typical Respondent
Respondents ranged in age from 21 to 53, and had been in the workforce as little as 12 months to as long as 29 years. A total of 359 individuals were employed as business professionals and managers, 172 were engineers, 22 worked as teachers, social service, or as health care professionals, 33 were in sales and 130 were categorized as working in either blue-collar or clerical positions. The modal respondent was a 28-year old male (62%) who works in an organization employing between 2,000 and 5,000 workers, has worked 5 years since college, has changed jobs once since graduation and has worked for 3.5 years with his or her current employer.
Measures
Respondents provided information regarding their college major, current job title, current salary, spouse's salary, industry of employment, months of employment since graduation from college, months of employment with their current employer, hours worked per week, the dollar amount of the annual budget supervised, and number of employees supervised. Respondents were also asked if they intended to work for their current employer after receiving their MBA. College majors were grouped in the following categories: business, engineering, physical science and health-related majors, and social science, humanities and other majors. Current job titles were categorized as follows: business, engineering, social and health service/teaching, and blue-collar/clerical jobs. Industry of employment was categorized as manufacturing and other.
Pay satisfaction, operationalized as pay level, was measured by a single item, "We would like to know how true you feel each item is of your current or most recent employment: The pay is good." Responses ranged from "very true" to "not at all true." This item was initially developed in 1969 for a research project conducted by the Survey Research Center of the Institute for Social Research (ISR) at the University of Michigan entitled "Quality of Employment Survey." Sweeney et al. (1990) used this item when measuring pay level in a study of pay satisfaction. As an indicator of construct validity, Sweeney et al. (1990) found comparable results when using the above single-item measure and a three-item measure developed by Andrews and Withey (1976). Salary was measured by asking "What are your current earnings before taxes (including salary, bonuses, and commissions)." Spouse's income was assessed by asking a similar question, "What are your spouse's earnings before taxes (including salary, bonuses, and commissio ns.)" Expected salary was measured with the following question, "There is a 50% chance my earnings will be above _____ for my first job after graduation." Single-item measures assessed the importance attached to pay ("How important is it that your pay is good?"), the importance of one's career ("How important is your career and work?"), and the soundness of one's family financial situation ("So far as you and your family are concerned, would you say that you are pretty well satisfied with your present financial situation?"). Managerial responsibilities were measured with two questions: "How many persons do you directly supervise?" and "What is the total annual budget over which you have primary managerial responsibility?".
Thirty-four items were used to measure six facets of job satisfaction. These items were initially developed for use in the 1972-1973 Quality of Employment Survey (Quinn et al., 1975), and were coded on a scale from one (not at all important) to four (very important). A factor analysis of the Quality of Employment Survey data by Kalleberg (1977) revealed the existence of six factors: interesting work, pleasant environment, job security and fringe benefits, friendly co-workers, advancement, and supportive environment. Analysis of the current data confirmed the findings of Kalleberg (1977). Consistent with Kalleberg (1977), scales were formed using the mean of the unweighted scores on component items. Alpha coefficients for the scales were greater than .70, except for the pleasant environment variable (.66) and the two-item-job security/fringe benefit variable (.40).
Results
Table 1 reports the correlation coefficients for the variables included in the analysis. Mean scores for males and females for noncategorical variables and significance level of t-tests of gender differences for these measures are listed in Table 2. Gender is not related to pay satisfaction, but is related to expected salary (p[less than].001) and to expected percent change in salary (p[less than].01). As anticipated, based on BLS salary data, men report significantly higher salaries than women. Further, men report higher pay expectations than women. Using a p[less than].10 level of significance, women attached greater importance to their career and reported greater satisfaction with their family financial situation.
Gender differences in college major, job title, industry of employment, and intention to leave one's current employer after one's MBA studies were also assessed. In each case gender was significant. Women were underrepresented among those with a major in engineering and overrepresented among those with a major in the social sciences, humanities and other category (chi-square = 16.85, df = 3, p [less than] .001). Women were underrepresented in engineering jobs and overrepresented in service/teaching jobs (chi-square 29.05, df 4, p [less than] .001). With respect to industry of employment, women were underrepresented in manufacturing (chi-square = 37.77, df = 1, p [less than] .001). Regarding intention to change employers, women were overrepresented among those intending to leave and underrepresented among those intending to stay (chi-square = 23.40, df = 2, p [less than] .001).
FEED