The glass ceiling: Does perception have any impact in keeping it in place or removing it?
v Table of Contents
Acknowledgments iv List of Tables viii List of Figures xi CHAPTER 1. INTRODUCTION 1 Background of the Problem 2 Statement of the Problem 7 Theoretical Framework 7 Social Theory 13 Purpose of the Study 15 Rationale 16 Research Questions 16 Significance of the Study 17 Definitions of Terms 17 Assumptions and Limitations of the Study 18 Nature of the Study 18 Research Bias 19 Organization of the Remainder of the Study 20 CHAPTER 2. LITERATURE REVIEW 22 Obstacles to Success 23 Occupational Segregation 31 Organizational Culture 44
vi Colleges and Universities 48 Work Commitment 50 Leadership Styles 53 Summary 56 CHAPTER 3. METHODOLOGY 57 Introduction 57 Design of the Study 57 Data Collection 58 Analytical Procedures 58 Data Analysis 59 Validity and Reliability 62 CHAPTER 4. DATA COLLECTION AND ANALYSIS 65 Descriptive Statistics 65 Analyses of the 1997 and 2002 Data 68 Statistical Analysis 71 Logistic Regression 135 Quantile Regression 141 CHAPTER 5. RESULTS, CONCLUSIONS, AND RECOMMENDATIONS 150 Currently in School 152 Willing to Work Harder for Company Success 153 Effort Put Into Job Beyond What Is Required 153 Willing to Find a New Job 153
vii Recommendations for Further Research 155 REFERENCES 156 APPENDIX A. MODEL SIGNIFICANCE 165 APPENDIX B. RELIABILITY TEST OUTPUT 169 APPENDIX C. CORRELATION, 1-TAILED 174
viii List of Tables Table 1. Female/Male Hourly Wage Ratio, 1978–1998 33 Table 2. Mean Salary of All Federal Labor Relations Employees by Gender, 1991–2000 36 Table 3. Gender Distribution by Institutional Class 38 Table 4. Institutional Gender Distribution 39 Table 5. Gender Earnings Difference Probability 42 Table 6. Average Wages 44 Table 7. Firm Size Effect on Gender Wage 47 Table 8. Validity Correlation 63 Table 9. Reliability Statistics 64 Table 10. Summary Item Statistics 64 Table 11. Mean Scores of Organizational/Structural Variables by Race/Gender in 1997 67 Table 12. Mean Scores for Organizational/Structural Variables by Race/Gender in 2002 70 Table 13. Model 1 Regression Results 73 Table 14. Model 2 Regression Results 77 Table 15. Model 3 Regression Results 81 Table 16. Model 4 Regression Results 85 Table 17. Model 5 Regression Results 90 Table 18. Model 6 Regression Results 95 Table 19. Model 7 Regression Results 101 Table 20. Model 1 Regression Results 104 Table 21. Model 2 Regression Results 108
ix Table 22. Model 3 Regression Results 111 Table 23. Model 4 Regression Results 115 Table 24. Model 5 Regression Results 120 Table 25. Model 6 Regression Results 124 Table 26. Model 7 Regression Results 129 Table 27. Currently in School 132 Table 28. Willing to Work Harder 133 Table 29. Effort Beyond What Is Required 134 Table 30. Willing to Work Harder for Company Success 135 Table 31. Logistic Odds for Model 8 136 Table 32. Logistic Odds for Model 9 138 Table 33. Logistic Odds for Model 8 139 Table 34. Logistic Odds Using Model 9 140 Table 35. Quantile to Find New Job 144 Table 36. Quantile Currently in School 144 Table 37. Effect of Race on Structural/Organizational Variables 145 Table 38. Effect of Gender on Structural/Organizational Variables 146 Table 39. Correlation Matrix for the Quantile Regression 147 Table 40. Correlation Matrix for Logistic Regression 148 Table A1. Willing to Work Harder 165 Table A2. Effort Beyond What Is Required 166 Table A3. Currently in School 167
x Table A4. Work Harder for Company 168 Table A5. Reliability—Notes 169 Table A6. Case Processing Summary 170 Table A7. Reliability Statistics 170 Table A8. Interitem Correlation Matrix 171 Table A9. Interitem Covariance Matrix 172 Table A10. Summary Item Statistics 172 Table A11. Scale Statistics 173 Table A12. ANOVA 173 Table A13. Correlation, 1-Tailed Significance for 1997 174 Table A14. Correlation, 1-Tailed Significance for 2002 176
xi List of Figures Figure 1. Gender Inequality Only 28 Figure 2. High Gender Discrimination Outcomes 29 Figure 3. Gender Career Discrimination 30 Figure 4. Gender Wage Convergence 32
1 CHAPTER 1. INTRODUCTION Managers whose performance is visible to powerful people in the organization tend to have more power than those in jobs where their performance is less obvious . . . to get ahead, managers have to make sure that people are familiar with their performance. (Hill, 1995, p. 7) Women have been segregated in the lower ranks of the labor market over the years. This occupational segregation has become so obvious that children now have perceptions about the types of jobs with which people of different genders identify (Trepanier-Street, Romatowski, & McNair, 1990). Trepanier-Street et al. found that 93% of school-age children believed nursing is a female occupation, while only 63% believed mechanics is a female occupation. The effect of gender discrimination over the years has resulted in numerous actions, both by female gender groups and the government. Comparable worth has been considered the cornerstone of inequality against women. In the debate about comparable worth, some have argued that meritocracy has been used in formulating rewards. Brenner (1987) found that, despite all their activism, feminist campaigns have failed to produce a much-needed bridge for societal equality. Her findings also showed that men were reluctant to allow equitable pay for women for the same work. She stated, Gender divisions within the work force have played an important part in perpetuating women’s low pay. Male-dominated trade unions have been willing to take up the issue only as women workers become organized. The kind of strategies used for raising the pay of women workers can undermine or aggravate men’s resistance, increase or decrease the possibilities of overcoming employers’ opposition, and create a more unified—or more divided—workforce. (p. 460)
2 State intervention was required to establish equitable opportunities for women and men. Brenner argued that the goal of twentieth century policy was a just distribution of individuals within a hierarchy of rewards and power. Thomas-Hunt and Phillips (2004) found that, despite women’s expertise, women are still perceived to be less than men with similar expertise. Women tend to be less influential than their male counterparts. According to Thomas-Hunt and Phillips, “Consistent with previous research, we found that, in general, women were perceived by others as less expert than men, were less influential, and felt less confident about their impact on the group.” (p. 1594). This study will focus on the impact of perception on the discrimination, or glass ceiling, women experience.
Background of the Problem Women have suffered from the effects of discriminatory employment practices throughout American history. Women have traditionally earned less and enjoyed fewer management positions than have their male counterparts. This is changing, however, and females are increasingly busting through the glass ceiling. The disparities between men and women in the workplace seem to be eroding, and many women today are succeeding as executives and leaders in professions in which their presence was previously restricted to clerical roles. What makes this phenomenon important is the manner in which these successful women have managed to maintain some type of perspective of their lives, and have been able to provide a lifestyle that embraces traditional family values, while recognizing the opportunities now available for many women in the United States.
3 Two important legislative acts passed in the 1960s altered many aspects of American society. The first of these was the Equal Pay Act of 1963 (EPA), which prohibited employers from paying men and women different rates for the same work strictly on the basis of gender. A year later, the Civil Rights Act of 1964 was passed into law, prohibiting discrimination based on race, color, religion, national origin, or gender. One part of that act in particular, Title VII, addressed economic and employment issues. Title VII was implemented in order to ensure equal employment opportunities in American businesses for people of all races, religions, national origins, and genders. Since the enactment of these laws, much research has been conducted about the nature and degree of the changes caused by Title VII—whether those changes were influenced directly by the Civil Rights Act or whether the changes occurred in spite of legislative progressions and societal changes (Truman & Baroudi, 1994). Researchers (Blau & Kahn, 1997; Ragins, Townsend, & Mattis, 1998; Renner, Rives, & Bowlin, 2002) have been particularly interested in whether legislation has had any effect on women’s ability to get high-power jobs and earn salaries equal to those of their male counterparts. Most new managers’ eagerness to implement their ideas about how to run an effective organization often falls short of empowerment, thereby neglecting their obligations and duties (Hill, 1994, 1995; Janssen, 2001). Women managers must be able to establish trust and credibility with their subordinates before they can reap the benefits of organizational equity. For women and minorities in upper management to be able to exercise organizational influence, they need to both develop shared power and build networks. According to Hill (1994), “Accomplishing this job requires. . . equal amounts of expertise, analytical ability,
4 and interpersonal and group dynamics skills” (p. 3). Another problem that affects the career of both women and men is visibility to influential managers. Managers who are assigned very visible tasks often are exposed to upper management, resulting in pay increases and promotions. Hill (1995) stated, Managers whose performance is visible to powerful people in the organization tend to have more power than those in jobs where their performance is less obvious. It is not enough just to perform effectively and contribute to organizational success. If others are unaware that the success of an important initiative should be attributed to a particular manager, that manager will not receive credit’. . . unless influential people are aware of a manager’s positive performance, he or she will not reap such rewards as a desirable assignments, salary increases, and promotions. (p. 7) Janssen (2001) presented equity theory, as discussed by other researchers (Adams, 1963, 1965; Mowday, 1991). According to equity theory, perceptions of fairness form a job- related motivational base that influences employees’ behaviors and responses to organizational behavior. Hence, according to equity theory, employees compare rewards in terms of a ratio between what they receive from their affiliated organization and the effort they put into the organization. Job efforts on the part of the employees include comprehensive investments (e.g., intelligence, experience, training, skill, and seniority) and specific work actions needed to accomplish the task. On the other hand, inducements an employee expects to receive from the organization (e.g., money or other values) are considered job rewards (Janssen, p. 1041). An employee might feel discriminated against if rewards are unequal to the efforts she or he made. Workers often reduce their efforts to counter the perceived inequality. Porter (2003) found that women have the same problem of obscurity as do men. In her words, “For more African American women to obtain senior management status, more
5 mentors are needed. . . mentors play an important role in getting women to upper- management positions” (p. 125). For women to enjoy visibility in the workplace, upper management needs to propel junior employees into job assignments that have high visibility. Upper-management employees act as mentors and nurture the junior ones. Mentoring plays a significant role in advancing junior managers to higher management positions (Athey, Avery, &Zemsky, 2000; Blau & Kahn, 2000; Cotter, Hermsen, Ovadia, & Vanneman, 2001; Ragins &Cotton, 1991; Wrigley, 2002). Thus, equity theory plays a vital role in understanding one of the barriers confounding women and minorities in businesses. Most women and minority managers do not enjoy the privilege of equity theory’s law of reciprocity. The primary cause of this is the fact most male mentors feel more comfortable mentoring men than women (Athey et al., 2000). Because the majority of the upper management is male, fewer female mentors are available to mentor entry-level women. According to Athey et al., “The direct effect of type-based mentoring is that entry-level employees of the majority type acquire more human capital, and thus firms who base promotions solely on ability promote more majority employees” (p. 766). Daily and Dalton (1999) reported that much work remains to be done in order for women to have the same opportunities as men do. Bilimoria observed, “Even when women do all the right things and have all the right stuff, they continue to be blocked from the innermost circles of power” (as quoted in Daily & Dalton, p. 5). In fact, studies (Daily & Dalton, 1999; Troy, 2001) have reported that, in 2000, female CEOs took home far less in pay than did their male counterparts and even less than did many lower ranked female
6 executives. Troy reported that, of the 30 highest paid female executives, only three were CEOs, and those women were ranked only as the 14 th ,24 th ,and 26 th .The highest of these women was still out earned “by multiples of three and four by lower ranked executives at other companies” (p. 2). Furthermore, this disparity in executive positions is not confined to one industry or marketplace, but rather reflects the pervasive nature of the problem itself. For instance, Herron and Hubbard (1998) reported on a study by the National Study of Arts Managers (NSAM), conducted to find out if gender discrimination exists in the arts industry. According to NSAM, the greatest gender discrepancies were found in upper management. Although differences in the level of education, experience, and age of men and women may account for some disparity in salary levels, NSAM’s survey showed that, even when women were on par with men in terms of education, experience, and age, a salary discrepancy still existed. Men predominantly held the upper management positions and, at the same time, has significantly higher salaries than did their female counterparts. Thus, NSAM suggested that a glass ceiling might exist in the field of arts management. Other studies (Olan, Carroll, & Giannantonio, 1993; Ragins et al., 1998; U.S. Department of Labor, 1991) have suggested the same premise: women are still not gaining access to the top jobs or the top salaries in business, even though women comprise more then 50% of the total workforce in America. Lai, Yoon, and Carlson (1999) reported that, in the United States, women earned between $0.71 and $0.74 for every dollar earned by men, regardless of occupation. Women have advanced in education and training, but are still being held back by a number of factors.
7 Statement of the Problem Traditionally, women and minorities have earned less than their male and Caucasian counterparts have, even when they have the same education, work experience, and training. Recent research (American Federation of Labor and Congress of Industrial Organizations [AFL-CIO], 1999; Lai et al., 1999) has shown that women are still earning 11% to 23% less than are their male counterparts, even though the EPA strictly prohibits that practice. Some researchers have suggested that women earn less because their human capital is less. Singh (2002) suggested that one of the reasons women earn less than men do in comparable jobs is that they do not have the same skill levels as men or they may not understand how to network within the system to gain access to higher paying jobs.
Theoretical Framework This study draws from diverse theories about gender and disadvantage related to participation in the workplace. These theories span psychology, sociology, economics, and business in order to identify how gender plays a role either in the recruitment, retention, advancement, or willing of an individual to remain with any given business organization. In terms of the psychological perspective, an individual’s attitude, perception, and behavior can be segmented into two primary categories: those of the employed individual and those of the individual responsible for hiring, retaining, or promoting the employed. These perspectives tend to be salient to the many existing studies about the glass ceiling. Research rarely examines perceptions of the glass ceiling. This study is intended to take these perceptions of
8 the glass ceiling into account. This study focuses on gender issues and places little or no emphasis on race. Two broad theories of perception constitute the foundation of this study: theories about the individual and theories about the institution (organization). The Individual Individual attributes tend to shift the burden of advancement for women in business organization to gender deficiencies on the part of women. These deficiencies include but are not limited to human capital, productivity, and experience. Yammarino and Dubinsky (1988) tested these arguments and found they were not supported. Regardless of gender, individuals at the same or similar job position levels share the same aspirations, goals, and values. Yammarino and Dubinsky found that differences in employees’ work-related responses were attributable to factors that were more internal for women than for men (p. 367). Williams (2000) explained that the human capital model only accounts for less than 25% of the pay differential between women and men. She cited research by Davila (1993, as cited in Williams, 2000) as concurring. Research has not looked as closely at the impact of perceptions on the career path choices of men and women as it has at the outcomes of their career choices (Williams, 2000). Perceptions about women’s abilities and productivity often result in the devaluation of productivity. Reichman and Sterling (2004) found that women’s productivity was perceived and treated as a “soft” organizational practice. In some cases, high female productivity was considered just luck: “Kramer also described a situation where her hard work was assumed to
9 be just ‘luck’” (p. 68). Reichman and Sterling also found that women’s work was taken for granted, and in many instances, women were not given the same opportunities to enhance their positions that men were given. In their studies of legal careers, Reichman and Sterling reported that women who were promised partner positions found that recently hired male associates were paid more than the females’ partners-to-be. Similarly, when women self- promote their work accomplishments, they are considered to be “the bitch” (p. 69). Ragins and Sundstrom (1989) found that women cave into societal pressure to self- select into female-typed specialties, thus resulting in fewer women available for occupations deemed masculine specialties. Parsons (1980) reported that Black males have lower labor force participation than do their White counterparts. Parsons further theorized that the cause of this disparity was largely due to wage differences, when other human capital characteristics were similar. These differences in labor force participation behavior between Black and White males raise a serious concern: what are the causes of this attitudinal difference toward the labor force? Parsons asked, “Are the large differences in market behavior among Black and White males the result of relatively straightforward differences in economic incentive between the two groups, or the result of deeper, or at least less well understood, social forces?” (p. 911).
The Organization The core of the organizational effect with respect to the glass ceiling is the establishment’s position that the worth of both genders adds value to an organization. As such, the organization’s perception of the female gender’s worth tends to impose an
10 unnecessary barrier on women in terms of hiring, salaries, advancement, and job segregation. The organizational effect can be described as the demand side of the glass ceiling. Organizational impact can take several forms, including taste discrimination, statistical discrimination, error discrimination, collusion, and harassment (Hull & Nelson, 2000). Hull and Nelson further stated that discrimination sometimes can be conscious and intentional, which they refer to as a “subtler fashion” (p. 233). Other research (Baron, 1991, as cited in Hull & Nelson, 2000; Bielby, 1991, as cited in Hull & Nelson, 2000; Marini & Fan, 1997, as cited in Hull & Nelson, 2000; Ridgeway, 1992, 1997, as cited in Hull & Nelson, 2000; Tomaskovic-Devey, 1993, as cited in Hull & Nelson, 2000) has shown that social interaction and reproduction of gender inequality emphasize the pervasive but often unrecognized role of gender in the employment context. In this arena, gender operates as a “cultural superschema” that influences perceptions of workers’ competence and reinforces existing inequalities. Structural approaches often assume the presence of micro-level sorting processes that channel men and women workers with comparable aspirations and qualifications into gender- typed jobs. Such organizational practices often appear neutral at face value, but affect men and women workers differently in terms of recruitment practices, entrance requirements, job- assignment policies, seniority and promotion systems, mentoring practices, and retention strategies. Roos and Reskin (1994, as cited in Hull & Nelson, 2000) stated that such institutionalized employer practices that are related to work schedules (e.g., time off from work for family emergencies) place a greater burden on some employees (mostly female) than on others (men). Cohen and Huffman (2003) found that organizations tend to under-
11 reward the skills typically associated with women and do not accept the contributions of women as profitable to the organization. Supervisory behaviors always play an important role in determining if a person is attached to his or her job. A person’s commitment and attachment, whether strong or not, is often the result of the situational characteristics of the environment in which the person works (Brown, 1996). An employee’s job satisfaction stems from person’s intrinsic satisfaction with the content of the work, regardless of the job-involved person’s supervisory perceptions of the job-involved person. A less involved person would be dissatisfied under similar circumstances. A person who is satisfied with job content exhibits strong affective ties to the organization. This forms a stronger bond between the person and the organization than does the person’s longevity in the organization. As a result of such bonds, this person is less likely to leave the organization. Role conflicts and ambiguity have less stressful effects on a person who is greatly involved with his or her job than they have on a person who is less involved. According to Brown, The job-involved person is not especially bothered by conflicting role requirements and expectations imposed by various role partners or by uncertainty regarding role requirements and expectation. Role conflict and ambiguity are not significantly related to the individual’s ability to identify strongly with the job. (p. 252) The psychological, social, or physical ill effects of identifying strongly with a job appear to have minimal effects on job-involved persons. Such people do not display any signs of high work-related stress, anxiety, somatic health complaints, or even work-family conflicts. The variables for this study have been shown by previous research (Brown, 1996; Cooper-Hakim & Viswesvaran, 2005; Eagly, Johannesen-Schmidt, & van Engen, 2003; Ragins, 1989) to be factors that contribute to perceived gender differences in the workplace.
12 These variables are grouped into two categories: dependent variables and independent variables (ascribed variables). The dependent variables are human capital, organization (structural /discrimination), willing to work harder, effort beyond what is required, and genuine effort to find a new job. Recent research (Cooper-Hakim & Viswesvaran, 2005; Davis-Blake et al., 2003; Reichman & Sterling, 2004; Williams, 2000) has shown support for these variables as factors that influence or affect outcomes in organizations. Research (van Hooft et al., 2004) has indicated that genuine effort to find a new job is not associated with high levels of job search self-efficacy; rather, job search behavior is more strongly associated with personal experience. An individual’s job search behavior is stimulated not only by positive instrumental attitudes such as regarding job seeking as wise and useful, but also by affective attitudes such as regarding job seeking as interesting and enjoyable. . . we found a significant relation between job search behavior and job attainment. (p. 385) Ek et al. (2005) found that education is a strong social predictor for successful entrance into the labor market (p. 474). Bishop (1988) reported that staying in school to acquire more skills is a factor in retaining a job or finding a new one; “Consequently, there are strong incentives to stay in school” (p. 414). Thus, acquiring additional skills—either through formal education or corporate training—is deemed an extra or genuine effort to work harder and fine a new job if a person’s current job does not meet the individual’s goals and expectations. Parsons (1973) reported that the decision to quit one’s job and look for another one is quite complex. Most people search for new jobs while still employed. Parsons argued that a cost benefit analysis of quitting one’s job emphasizes reasons other than a monetary difference that might account for why someone would quit a job. According to Parsons,
13 “Relative wages are somewhat less so, while fluctuations in nominal wage levels generally have no particular relationship to quit rate fluctuations. Perhaps not surprisingly, ‘real’ factors seem to dominate the movement of quit rates” (p. 400). Parsons was interested in determining whether the glass ceiling (or the perception of a glass ceiling) is the “real” factor that influences quitting behavior. Yammarino and Dubinsky (1988) reported that job- or position-related differences (i.e., men are in a more advantageous position than are women) tend to result in quitting behavior differences: “Employees’ work-related behaviors and attitudes can be understood in terms of job- or position-related differences” (p. 368). Their empirical research revealed greater support for a situation-centered perspective than for a person-centered explanation (p. 378).
Social Theory Individuals often evaluate themselves in a way that allows them to acquire a social identity. Social identity is, in fact, an identification of one’s self in relation to a group. Ruthann and Crocker (1992) discussed the individual’s need to identify with a group. An individual’s group identification enables that person to acquire or acknowledge certain behaviors that are identified with the social group to which the individual belongs. Self-concept has two distinct aspects. One is personal identity, which includes specific attributes of the individual such as competence, talent, and sociability. The other aspect is one’s social identity, defined as part of an individual’s self-concept which derives from his knowledge of his membership in a social group (or groups) together with the value and emotional significance attached to that membership. (p. 302) Thus, when a group of individuals shares certain characteristics in common, the individuals can be said to be members of the same social group. These characteristics could be race,
14 gender, religion, or even occupation. The common characteristics they share could affect their views or experiences. These experiences can be work related or other social experiences. Social theorists (Cheek, 1989, as cited in Luhtanen & Crocker, 1992; Cheek, Underwood, & Cutler, 1985, as cited in Luhtanen & Crocker, 1992) grouped social identity into three domains: personal, social, and collective. This study expands social theory to embrace interpersonal domains, or how the self relates to others (Luhtanen & Crocker, 1992). Further, it includes other aspects of self-concept that result from race, ethnic background, religion, and feelings of belonging to a particular community (e.g., workplace). This aspect of social identity was supported by Tajfel and Turner (1979, as cited in Luhtanen & Crocker, 1992). These aspects of social identity and self-concept influence and individual’s perceptions and behaviors. The behaviors a person exhibits can be ascribed or acquired. This study will focus on ascribed behavior rather than acquired behavior. Research has shown that, while acquired behavior is idiosyncratic, ascribed behavior is peculiar to everyone in a social group. The group can be members of the same gender, race, or even occupation (Luhtanen & Crocker, 1992). The ascribed characteristics of the groups in this study will aid in determining how members of each group, gender, or race perceive the impact of the glass ceiling and its effects on them. The effects are not limited to finding another job or working harder than is necessary for advancement. Social identity has a considerable impact on the members of a social class.