Breadcrumb

  1. Home
  2. Special Topics Annual Report: Women in STEM

Special Topics Annual Report: Women in STEM

Supplement to the US Equal Employment Opportunity's Annual Report on the Federal Workforce Fiscal Year 2019 

Office of Federal Operations

Download this document as a PDF file

 

Executive Summary

As a supplement to the U.S. Equal Employment Opportunity’s (EEOC) Annual Report on the Federal Workforce Fiscal Year 2019, EEOC’s Office of Federal Operations (OFO) presents this report of its research on the participation and experiences of women who work in Science, Technology, Engineering, and Mathematics (STEM) within the federal government. Although there has been a great deal of focus on women in STEM in the private sector, little has been reported on the diversity and experiences of women working in STEM in the federal sector.

As such, we offer this report as an examination of this population. To produce this report, the EEOC focused on the occupational job series identified by the United States Office of Personnel Management (OPM) as STEM occupations. The EEOC gathered and analyzed information from numerous federal sources, including: (1) OPM’s Enterprise Human Resources Integration (EHRI) data; (2) EEOC Form 462 (EEO complaint) data; and (3) select OPM Federal Employee Viewpoint Survey (FEVS) responses. The EEOC combined the data to form several datasets, which we then analyzed to learn about these employees, focusing on the following metrics: Demographic composition by race, age, national origin, sex and disability; representation in leadership positions; employee viewpoints; thoughts about leaving their current agency; and the general nature of the complaints filed with the EEOC. In formulating our findings, we generally compared the women in STEM to their male colleagues. These findings appear below.

Findings

  • Overall, women accounted for 29.3 percent of STEM federal workers. Science occupations had the most (49,546), while math occupations in the federal sector had the fewest number of women (6,469). There were significantly fewer women in Technology and Engineering than expected.
  • The overall average age of women in STEM occupations in the federal sector was 45.5 years, compared to an average age of 47.4 years for men.
  • Most of the women working in STEM in the federal sector were White (66.02 percent). By comparison, 14.58 percent were African American or Black, 9.76 percent Asian, 6.42 percent Hispanic or Latina, 0.97 percent American Indian/Alaska Native, and 0.28 percent Hawaiian or Pacific Islander. In addition, 1.98 percent indicated they were more than one race.
  • Approximately 10.7 percent of women working in STEM were individuals with disabilities and 1.2 percent of those women had a targeted disability.[1]
  • The largest percentage of women working in entry-level STEM occupations were working in Science occupations. Similarly, the largest percentage of experienced women, which includes Grades 8 through 12, worked in Science occupations.
  • For senior professionals (Grades 13 through 15), the largest percentage of women in STEM were in Science occupations; in contrast, the male senior professionals were evenly split between Science and Engineering.
  • There was a total of 16,454 women in leadership roles compared to 47,167 men; only 25.9 percent of all STEM leaders were women.
  • In FY 2019, 34,483 women were counseled and 14,096 female federal employees filed formal complaints, which is important to understanding the experience of women in STEM work environments.
  • Despite the belief that sexual harassment is the foremost type of discrimination raised by women, generalized harassment was actually a larger issue in FY 2019—with 1,986 complaints filed, compared to 358 complaints for sexual harassment.
  • There was a strong relationship between women’s intentions to leave their current agencies and complaint activity; the more sex-related complaint activity, the more likely women were to state an intention to leave.
  • Women’s belief that their supervisors were committed to a diverse workforce was significantly correlated with fewer numbers of individuals receiving counseling and to fewer formal complainants.
  • There was no apparent impact of the level of representation of women in STEM occupations at a given agency and the number of individuals counseled or number of formal complainants filed.
  • After taking pay into account, women were about 40 percent less likely to work in Engineering, 33 percent more likely to work in Math, and nearly 92 percent more likely to work in Science than in Technology jobs.

This report presents an important examination of women working in STEM occupations within the federal government. This report is intended to assist federal agencies’ continuing efforts to make the federal government a model EEO employer.

Introduction

The U.S. Equal Employment Opportunity’s (EEOC) is responsible for providing guidance to federal agencies on all aspects of the federal government's equal opportunity program and assisting agencies in their efforts to become model EEO employers. In keeping with this responsibility, the EEOC researches and issues reports on various topics of interest to the federal sector community. Such reports may focus on a single agency or topic or may encompass a governmentwide trend or issue. As a supplement to the EEOC’s Annual Report on the Federal Workforce Fiscal Year 2019, EEOC’s Office of Federal Operations (OFO) presents this report of its research on the participation and experiences of women who work in STEM within the federal government. Although there has been a great deal of focus on women in STEM in the private sector, little has been reported on the diversity and experiences of women working in Science, Technology, Engineering, and Mathematics (STEM) in the federal sector. As such, we offer this report as an examination of this population.

Part I of this report provides the rationale and a review of the relevant recent literature. Part II sets forth the EEOC’s research objectives and methodology. Part III presents Women in STEM federal employee demographic data by age, sex/gender, race, national origin, disability, and leadership diversity. Part IV provides comprehensive data and analysis on complaint activity and its relationship to the experiences of women working in the federal government. Part V presents data on pay and the relationship between pay, gender, and STEM occupational group membership. Part VI sets forth recommendations and is followed by a Conclusion.

I. Research Context

For many years, the STEM workforce in the United States has been predominately male, and overwhelmingly White and Asian. Though their numbers have been growing in recent decades, women make up only 26 percent of employed scientists, with minority women representing only 11 percent of that total (National Academy of Sciences, National Academy of Engineering; Institute of Medicine, 2011). There are especially broad gender gaps in some of the fastest growing and highest paid industries, such as computer science and engineering (AAAW, 2020). Given the need for creative responses to modern day trends such as terrorism, pandemics, climate change, and an increasingly diverse U.S. population, it is imperative that both the federal government and the private sector diversify their STEM workforces to respond effectively to problems both at home and abroad.

The pipeline into STEM careers begins to leak as early as high school and college. Despite a reversal in the gender gap in terms of undergraduate enrollment, women still enroll in STEM majors at lower rates than their male counterparts (Legewie & Diprete, 2014; Riegle-Crumb, King, Grodsky, & Muller, 2012). Women are also less likely to pursue postgraduate degrees in STEM fields, and less likely than their male counterparts to work in STEM after graduating – male college graduates are more than twice as likely to work in STEM fields after receiving their degree as female college graduates (Lim, Haddad, Butler, & Giglio, 2013; Strayhorn, DeVita, & Blakewood, 2012). Researchers point to students’ concerns regarding work-life balance, gender bias and sexual harassment within their programs, and a limited pool of female role models and mentors at the academic level as culprits (Tan-Wilson & Stamp, 2015; Leaper & Starr, 2018; Strayhorn, DeVita, & Blakewood, 2012).

There is also a significant racial disparity in the STEM education pipeline, though part of this is due to gaps in educational attainment that are not necessarily STEM-specific (Lim, Haddad, Butler, & Giglio, 2013). Students from minority ethnic groups are less likely to have access to high school math and science courses that allow them to build the skills necessary for STEM degrees. As with women, both racial stereotypes and a lack of diversity in mentorship are cited as obstacles for students of color studying STEM (Grandy, 1998; Strayhorn, DeVita, & Blakewood, 2012). Though nearly two-thirds of students who initially major in STEM have not dropped that major three years later, the percentage of Black and Hispanic students who graduate with a STEM degree is far lower than the number who showed interest in STEM fields upon entering university (Anderson & Kim, 2006).

Even graduating with a STEM degree does not necessarily translate into a STEM career. A longitudinal study comparing the trajectories of women in STEM-related occupations to other professional occupations found that women in STEM were significantly more likely to leave the STEM field than other professional women, even as neither group exited the labor force (Glass, Sassler, Levitte, & Michelmore, 2013). The reasons women leave STEM professions are myriad. As work-life balance becomes a top priority for many people entering the workforce, both men and women believe STEM to be less family-friendly fields than other potential career options; women, more so than men, worry that careers in science will prevent them from having a family (Mason, Goulden, & Frasch, 2009, Tan-Wilson & Stamp, 2017). In a study of physics and chemistry doctoral students, women were more likely than men to make choices in order to meet familial obligations that would compromise their careers (Wyss & Tai, 2010).

Work-life balance is not the only reason for the gap, however. A survey done by the Pew Research Center in 2017 found that half (50 percent) of women in STEM jobs say that they have been discriminated against at work because of their gender, compared to 41 percent of women in non-STEM fields, and 19 percent of men in STEM fields (Funk & Parker, 2020). Women in STEM experience sexist comments, a lack of trust in their experience or skills, and the feeling of being “outsiders” in a male domain. A significant number of women in STEM careers report being subjected to sexual harassment within their workplace; sexist comments have been found to have a negative effect on women’s motivation to remain in the field (Leaper & Starr, 2018; Aycock et al., 2019; Burke, 2017). In addition to gender discrimination, women of color also report instances of racial discrimination and racist remarks in their workplaces (Burke, 2017; Funk & Parker, 2020).

Finally, while the gender wage gap is a problem across industries, it is particularly pronounced in STEM occupations, both in the public and private sectors. Early in their careers, female STEM graduates earn less than their male counterparts, a gap that only widens as women experience additional earning penalties upon having children – penalties that male workers do not (Xu, 2016; Lim, 2016). In a 2017 Pew Research Center survey on men and women in STEM, 29 percent of women found themselves earning less than a man for doing the same job. Women, especially women of color, are also more likely to be overlooked for promotions and other opportunities for advancement (Funk & Parker, 2020). Such discrimination also varies based on the type of STEM field. In the public sector,  men were more likely to be paid more than women – even at the same pay grade, and with the same individual characteristics, when employed in the science fields that are perceived as “masculine” (such as physical sciences and engineering),  (Smith-Doerr, Alegria, Fealing, Fitzpatrick, & Tomaskovic-Devey, 2019).

From their initial experiences in the workforce, women in STEM face a long list of limiting factors that can prevent them from pursuing a STEM career, continuing in the field, or advancing to high-level STEM positions. As the United States attempts to diversify its STEM workforce from its historically White and male roots, it is imperative to research, understand, and mitigate these biases.

II. Research Objectives

The goal of this report is to provide useful information to federal EEO Directors, Chief Human Capital Officers, and Chief Information Officers, on women working in STEM within the federal government, and to facilitate mission readiness and the delivery of services that are dependent upon advancement in STEM applications. Accordingly, this research report addresses several important questions:

  1. What is the demographic makeup of the women and men within the federal STEM workforce?
  2. What are specific demographics of the women and men within each STEM occupational category (i.e., in Science, in Technology, in Engineering, and in Mathematics)?
  3. What is the expected participation of women within each STEM category?
  4. Is there a relationship between EEO complaint activity and gender composition within agencies that have predominately male STEM workforces reporting more gender/sex-related complaints?
  5. For women working in STEM within the federal government, is there a relationship between employee perceptions, as measured by the FEVS, and sex/gender-related complaints?
  6. For men and women, what is the average pay by STEM occupational category?
  7. Do gender and pay predict an employee’s STEM category?

Data and Methodology

To better understand the population of women working in STEM, we analyzed FY 2019 data from the following sources:  

  • Office of Personnel Management (OPM Enterprise) Human Resources Integration (EHRI) data (December 2018 Status Data);
  • EEOC Form 462 EEO complaint data; and
  • Select OPM Federal Employee Viewpoint Survey (FEVS) responses.

An effort was made to gather data on all federal agencies from the sources listed above. We did not have data on Women in STEM contractors working in federal agencies. Some agency data was unavailable (e.g., FEVS data was not available for all federal agencies).   

We combined the data obtained to form several different datasets, which were analyzed to answer specific research questions. For example, the EEOC complaint data and OPM EHRI data were combined to examine potential relationships between complaints and the relative representation of women in STEM occupations. Likewise, FEVS data was combined with complaint data to analyze the potential relationship between specific EEO-related viewpoint questions and complaint data.

Not all data could be combined to research all questions of interest, as not all the data sources contained the same identifiers required to combine the data. For example, the FEVS data was reported at the agency level and not the individual employee level; thus, it is not possible to combine that data on a one-to-one basis with EHRI data, which is individual level data. Furthermore, EEOC complaint data does not include occupation; thus, it is not possible to report the number of EEO complaints that are specific to Federal Women in STEM employees. Also, as noted above, this report examines only full-time, permanent federal employees in OPM classifications identified as STEM occupations; the list of STEM occupations was obtained from FedScope and may be reviewed there.[2]

III. Federal Women in STEM Workforce Demographics

OPM EHRI data for STEM occupations was analyzed in order to discover important information concerning the demographics of the overall federal STEM workforce.

Demographic Overview

Before delving into specific demographic characteristics of women in STEM occupations, it is necessary to understand the basic participation of women in STEM occupations, relative to men. Table 1 below provides the overall STEM demographics for women in STEM occupational groups.

Table 1: STEM Job Category by Gender

STEM Job Category

Male/Man

Female/Woman

Total

Science

70,568

49,546

120,114

Percent

21.04%

14.77%

35.81%

Row Percent

58.75%

41.25%

 

Column Percent

29.75%

50.43%

 

Technology

63,599

23,378

86,977

Percent

18.96%

6.97%

25.93%

Row Percent

73.12%

26.88%

 

Column Percent

26.82%

23.79%

 

Engineering

89,447

18,855

108,302

Percent

26.67%

5.62%

32.29%

Row Percent

82.59%

17.41%

 

Column Percent

37.72%

19.19%

 

Math

13,550

6,469

20,019

Percent

4.04%

1.93%

5.97%

Row Percent

67.69%

32.31%

 

Column Percent

5.71%

6.58%

 

Total

237,164

98,248

335,412

70.71%

29.29%

100.00%

Math has the fewest number of women (n=6,469), while Science has the most (n=49,546).

Demographic Characteristics of Women in Federal STEM Occupations

When examining the overall participation of women in STEM occupations, it is important to understand the specific demographic composition of the women STEM occupational group.

Age

The overall average age of women in STEM occupations is 45.5 years compared to the average age of 47.4 years for men. Figure 1 below demonstrates that women in STEM occupations are, on average, younger than their male counterparts. Women in engineering are the youngest (x=42.68 years), while women in Technology (x=49.74 years) are the oldest.

Figure 1:
STEM Job Category Women Men
Science  44.89 48.52
Technology  49.74 48.41
Engineering 42.68 46.13
Math  43.54 45.14
All 45.53 47.4

Figure 1: Average Age by STEM Job Category and Gender

Table 2 below provides the average age of women in the different STEM occupational groups.

Race and National Origin

Table 2: Race and National Origin for Women in STEM

Race/National Origin

Women

Men

Total

% Women

African American/Black

14,066

19,256

33,322

42.2%

American Indian/Alaska Native

932

1,856

2,788

33.4%

Asian

9,423

19,926

29,349

32.1%

Hawaiian/Pacific Islander

268

697

965

27.8%

White

63,710

169,030

232,740

27.4%

More than one race

1,910

3,956

5,866

32.6%

Hispanic/Latinx

6,198

14,139

20,337

30.5%

Total

96,507

228,860

325,367

29.7%

Women in STEM have diverse racial and ethnic backgrounds, however, most of the women working in STEM are White (66.02 percent). Comparatively, there were: 14.58 percent African American/Black, 0.97 percent American Indian/Alaska Native, 9.76 percent Asian, and 0.28 percent Hawaiian/Pacific Islander women. There were 1.98 percent who indicated they were more than one race, while there were 6.42 percent STEM employees who identified as Hispanic/Latina.

Disability

Approximately 10.7 percent of women working in STEM are persons with disabilities (PWD), in comparison to 12.8 percent of men with disabilities. Furthermore, only 1.2 percent of women working in STEM have a targeted disability (PWTD)[3], compared to 1.5 percent of men. These percentages do not meet the federal goals of 12 and 2 percent for hiring PWD and PWTD, respectively.[4]

Figure 2: Decorative graph of the data presented in Table 3

Figure 2: Disability Status by STEM Job Category and Gender

Figure 2 depicts disability status (PWD or Non-Disabled/ND) by STEM occupational group. The largest number of women STEM employees is in Science occupations, which includes the largest number of women with disabilities.

Within Science, Technology, Engineering, and Math, women with disabilities represent 9.7, 14.2, 8.9, and 10.6 percent, respectively.

Table 3 below presents the complete participation of persons with disabilities in STEM occupations.

Table 3: Women with Disabilities Working in STEM Occupations

Gender by Disability Status

Science

Technology

Engineering

Math

All

Total Persons with Disabilities (PWD)

12,767

15,501

9,664

2,379

40,311

Men with Disabilities

7,992

12,219

7,995

1,695

29,901

Women with Disabilities

4,775

3,282

1,669

684

10,410

Total Persons with Targeted Disabilities (PWTD)

1,594

1,896

949

314

4,753

Men with Targeted Disabilities

1,051

1,500

798

219

3,568

Women with Targeted Disabilities

543

396

151

95

1,185

Total Non-Disabled (ND)

105,962

70,242

97,334

17,434

290,972

Men with No Disability

61,663

50,439

80,290

11,686

204,078

Women with No Disability

44,299

19,803

17,044

5,748

86,894

All STEM Employees

118,729

85,743

106,998

19,813

331,283

More women with disabilities are needed within Science, Engineering, and Math occupations, since only the participation of women with disabilities in Technology met the 12 percent goal threshold for the federal workforce. None of the occupations met the 2 percent goal.

Experience and Leadership

Beyond the traditional demographic descriptors, it is important to understand both the experience level of women working in STEM and their representation in leadership roles. There are very few women in entry-level (Grades 1 through 7) jobs; the largest number of entry-level women are working in Science occupations. This is also true for experienced women, which includes Grades 8 through 12.

Figure 3
Women's STEM Experience
Science 829 21,166 20,939
Technology 261 9,402 10,727
Engineering 515 4,793 6,758
Math 105 1,245 2,397
Figure 3: Continued
Men's STEM Experience
  Entry Level Experienced Senior Professional
Science 1115 28322 29836
Technology 710 27643 26369
Engineering 1865 19522 29839
Math 305 1771 4315

Figure 3: STEM Experience Level by Gender

For senior professionals (Grades 13 through 15), as shown in Figure 3 above, the largest number of women in STEM are in Science occupations; in contrast, the male senior professionals in STEM are most likely to be in Science and Engineering, closely followed by Technology. Figure 4 below shows that most leaders in STEM are men.

Figure 4
Senior Manager/Executive
  Women Men
Math (N=1940) 30.5% 69.5%
Engineering (N=15276) 16.0% 84.0%
Technology (N=11330) 24.8% 75.2%
Science (N=15750) 32.5% 67.5%
Figure 4 Continued 
Frontline/Program Manager
  Women Men
Math (N=1027) 38.3% 61.7%
Engineering (N=3210) 15.2% 84.8%
Technology (N=2406) 26.9% 73.1%
Science (N=8751) 33.2% 66.8%

Figure 4: Women in Leadership (Senior Leaders)

Figure 5
Team Leader or Lead Worker
  Women Men
Math (N=179) 36.9% 63.1%
Engineering (N=1663) 16.2% 83.8%
Technology (N=1245) 33.5% 66.5%
Science (N=844) 37.1% 62.9%
Figure 5: Continued
Individual Contributor
  Women Men
Math (N=179) 32.1% 67.9%
Engineering (N=1663) 17.8% 82.2%
Technology (N=1245) 27.1% 72.9%
Science (N=844) 43.5% 56.5%

Figure 5: Women in Initial Leadership Roles

Figures 4 and 5 above indicate that there was a total of 16,454 women in leadership roles compared to 47,167 men; only 25.9 percent of all STEM leaders are women.

Expected Representation of Women in STEM Occupational Groups

In addition to understanding where women are currently working, it is important to understand where there should be greater female representation. Thus, we analyzed the expected participation of women in specific categories based upon their overall numbers in STEM occupations; the results are presented in Table 4 below.

Table 4: Expected Representation for Women in STEM Job Categories

Science

Engineering

Women

49,546

Women

Expected

35,183

Women

18,855

Women

Expected 31,724

Men

70,568

Men

89,447

Technology

Mathematics

Women

23,378

Women Expected 25,477

Women

6,469

Women

Expected 5,863.9

Men

63,599

Men

13,550

The difference between the actual number of women and their expected participation rates was significant (χ2 (3, N=335412) = 16007.3164, p<.0001). There were significantly fewer women in Technology and Engineering than expected.

IV. Complaint Activity and Employee Viewpoints

Women may face discrimination and harassment in many forms. This report focuses on complaints related to sex, such as sexual harassment and pregnancy discrimination.

Sex-Based Complaints by Federal Employees

EEOC annually collects a variety of data from federal agencies on sex-related female complaint activity. [5] Key data concerning the number of counselings, formal complaints, bases, and issues for women-centered discrimination and harassment is presented in the following tables.

In FY 2019, there were 34,483 female federal employees counseled for EEO related matters. As table 5 shows, less than half (or 14,096) of those women decided to file formal complaints.

Table 5: Sex-Related Complaint Activity for Women in FY 2019

Female Complaint Activity

Total

Average Per Agency

Counselings Initiated

36,550

393.0

Individuals Counseled

34,483

370.8

Complaints Filed

14,637

157.4

Complainants

14,096

151.6

The vast majority of sex-related bases in EEO complaints stemmed from perceived violations of Title VII of the 1964 Civil Rights Act, as amended (table 6). By comparison, the other two clearly female-related bases—the Equal Pay Act (EPA) and the Pregnancy Discrimination Act (PDA)—had a relatively small number of complaints filed by women.

Table 6: Sex-Related Bases for FY 2019 Female Complaint Activity

Sex-Related Bases

Total

Average Per Agency

Title VII (Sex/Gender)

3,534

38.0

Pregnancy Discrimination Act

131

1.4

Equal Pay Act

34

0.4

The number of formal complaints filed for sex-related issues is presented in Table 7 below. Complaints based on promotion were included due to the previously presented analysis that shows underrepresentation of women in leadership roles within STEM occupations.

Table 7: Sex-Related Issues in FY 2019 Female Complaint Activity

Sex-Related Issues

Total

Average Per Agency

Total Harassment

2,344

25.2

  • Generalized Harassment

1,986

21.4

  • Sexual Harassment

358

3.8

  • Pregnancy Harassment

60

0.6

Promotion

465

5.0

While sexual harassment is often perceived as the most common issue women face in the workplace, the data shows that generalized harassment was actually a larger issue in FY 2019. In total, female federal employees filed 1,986 complaints due to generalized harassment, compared to 358 complaints for sexual harassment. In addition, promotion was an issue in 465 complaints.

Women’s Viewpoints on Federal Employment

In FY 2019, 227,506 women completed the Federal Employee Viewpoint Survey (FEVS). The FEVS data deemed most relevant to the goals of this report relate to employment practices and the employee viewpoints related to work environment, including promotional opportunities and job satisfaction. Although data for STEM occupations alone was not available, the FY2019 FEVS data—which includes all occupations for female federal employees—is instructive.   

For example, about a third of these women (72,783) indicated that they intended to leave their current employer, either for another federal agency, the private sector, or for another unspecified reason (such as retirement).

The following FEVS questions may provide insights into the reasons why women choose to leave their current employer:

17.  I can disclose a suspected violation of any law, rule or regulation w/o fear of reprisal.

20.  The people I work with cooperate to get the job done.

22.  Promotions in my work unit are based on merit.

38.  Prohibited Personnel Practices are not tolerated.

45.  My supervisor is committed to a workforce representative of all segments of society.

49.  My supervisor treats me with respect.

55.  Supervisors leaders work well with employees of different backgrounds.

69.  Considering everything, how satisfied are you with your job?

70.  Considering everything, how satisfied are you with your pay?

71.  Considering everything, how satisfied are you with your organization?

Figure 6 below shows these data using ratings from 0 to 5. On average, female federal employees felt that merit-based promotions were an issue, giving it the lowest score (3.0) out of all the EEO-related FEVS questions. Satisfaction with pay and with the organization also got relatively low scores of 3.6 each. In contrast, women gave the highest score (4.2) to the question about supervisors treating them with respect.

Figure 6: : Average Ratings for EEO-Related FEVS Questions

FEVS Indicator Average Rating
17. I can disclose w/o fear of reprisal. 3.7
20. People cooperate to get the job done. 3.9
22. Promotions are based on merit. 3.0
38. Prohibited Personnel Practices are not tolerated. 3.8
45. Supervisor is committed representativeness. 3.9
49. Supervisor treats me with respect. 4.2
55. Leaders work well w/different backgrounds. 3.8
69. How satisfied are you with your job? 3.7
70. How satisfied are you with your pay? 3.6
71. How satisfied are you with your organization? 3.6

Figure 6: Average Ratings for EEO-Related FEVS Questions

Identifying issues regarding the retention of women in federal employment may help improve employee engagement and foster a better work environment. Such improvements could reduce the amount of complaint activity.

Relationship between Employee Viewpoints and Complaint Activity

While understanding both the EEO complaint activity and employee viewpoints of women working in the federal sector is valuable, it is equally important to know if there is a correlation (or relationship) between the views of women regarding their experiences in federal employment and EEO complaint activity. Specifically, understanding the potential relationship between female employee perceptions (as measured by the FEVS) and sex/gender-related complaints could indicate if women believe that the federal sector is unwelcoming to them. We examine the same set of questions presented in Figure 6 above in order to determine any relationship between FEVS ratings and complaint activity.

For the intention to leave their current agency, there was a significant positive correlation[6] with all complaint activity analyzed. We found significant correlations for the number of formal complainants (.88), general harassment complaints (.88), individuals counseled (.85), sex-based complaints filed by women (.85), sexual harassment complaints (.81), promotion complaints filed by women (.79), EPA complaints filed by women (.71), and pregnancy complaints (.53). Larger correlations suggest a stronger relationship between the category and the intention to leave.

Women’s view that the people they work with cooperate to get the job done was significantly and negatively related to the number of formal complainants (-.37), general harassment complaints (-.37), individuals counseled (-.36), sex-based complaints filed by women (-.36), promotion complaints filed by women (-.34), and sexual harassment complaints (-.32). The closer the correlation is to -1, the larger the negative effect. There was no correlation to EPA or PDA complaints.

Women’s belief that their supervisors are committed to a diverse workforce was significantly and negatively related to the number of formal complainants (-.33), sex-based complaints filed by women (-.33), general harassment complaints (-.33), and individuals counseled (-.32). There was no correlation with EPA, PDA, sexual harassment, or promotion complaints.

Women’s feelings that their supervisor treats them with respect were significantly and negatively related to the number of general harassment complaints (-.35), formal complainants (-.34), sex-based complaints filed by women (-.34), individuals counseled (-.33), and promotion complaints filed by women (-.31). There was no correlation with EPA, PDA, or sexual harassment complaints. It is important to note that the potential lack of correlation of EPA and PDA complaints with viewpoint data could be attributed to the very small number of complaints with those bases.

Promotions based upon merit, prohibited personnel practices, supervisors working well with diverse employees, being generally satisfied with their job, being satisfied with their pay, or being generally satisfied with their agency had no correlation with any kind of complaint activity. It is worth noting that feeling respected by a direct supervisor was more important to turnover intention than was believing that their supervisor works well with people of different backgrounds or being satisfied with the nature of their work or their agency. This seems to offer support for the old adage that an employee joins an organization, but they leave their boss.

Workforce Gender Composition and Complaint Activity

An analysis was conducted to determine if agencies with predominately male workforces within STEM have more complaint activity (i.e., both individuals counseled and formal complainants) than agencies which have more equal numbers of men and women working in their STEM occupations. A multiple analysis of variance (MANOVA), an analysis to look at relationships between multiple variables at once, was modeled using an indicator of the percentage of women working in STEM within federal agencies as a predictor of individuals counseled and the number of formal complainants. The analysis included 93 agencies with both complaint activity and EHRI data available. Agencies that were in the bottom 50 percent in terms of percentage of women working in STEM occupations were regarded as being predominately male agencies with respect to STEM employment.

The results of the MANOVA suggests that there is no impact on the level of representation of women in STEM occupations at a given agency and the number of individuals counseled [F value (1, N=93) = 1.10, p=.30]. The analysis didn’t find a relationship for the number of formal complainants either [F value (1, N=93) = 1.18, p=.30].

V. Women’s Average Pay in STEM Occupations

Examining employee pay offers some interesting information relevant to the value federal agencies place on their ability to recruit and retain female STEM workers. The average difference in pay between men (M= $88,914.40, SD= $26,769.80) and women (M= $84,608.60, SD= $25,244.20) is about $4,305.90 per year, which is a significant difference (t(334,980) = 43.07, p < .01).

Table 8: Average Base Salary by STEM Job Category and Gender

Women

Men

STEM Job Category

Average Base Pay

STEM Job Category

Average Base Pay

Science

$81,623.40

Science

$87,929.12

Technology

$86,564.78

Technology

$84,555.97

Engineering

$88,364.34

Engineering

$92,220.63

Math

$89,423.85

Math

$92,660.56

All

$84,608.56

All

$88,914.45

N=335,412

Table 8 above demonstrates that women are paid, on average, less than men in Science, Engineering, and Math, but are paid more than men in Technology jobs because women in Technology are fewer in number, but occupy higher grades, in contrast to the fact that there are much more men than women at all grade levels.

Relationship among Gender, Pay, and STEM Category Participation

Based on the results presented earlier in this report, a multinomial logistic regression was performed to determine if an employee’s gender and pay could accurately predict their STEM category. There were 334,982 observations used because 480 observations were deleted due to missing values for the response or explanatory variables; the category counts used in the analyses for each category is as follows: Engineering (n= 108,251), Math (n= 20,000), Science (n=119,810), and Technology (n= 86,921). Technology was used as the reference group and Gender was dummy coded as 0= Man and 1= Woman. The results are presented in Table 9 below.

The analysis of the parameters shows that gender and average base pay do predict STEM occupational group membership. However, the effect of average base pay is potentially an artifact of the large sample size, as the estimates are extremely small, despite being highly significant. This is reflected in the fact that the odds ratios for Math, Science, and Engineering are all 1.00 with an invariant confidence interval (i.e., both the upper and lower bounds are 1.00).

Table 9: Pay and Gender Estimates for Predicting STEM Occupational Group

Analysis of Maximum Likelihood Estimates

Parameter

STEM Occupational Group

DF

Estimate (β)

Standard Error

Wald Chi-Squared

Probability

Intercept

Engineering

1

-0.4445

0.0164

730.2912

<.0001

Intercept

Math

1

-2.4078

0.0282

7314.0214

<.0001

Intercept

Science

1

0.00783

0.0163

0.2309

0.6309

Gender (Female)

Engineering

1

-0.5266

0.0111

2238.7866

<.0001

Gender (Female)

Math

1

0.2955

0.017

302.0345

<.0001

Gender (Female)

Science

1

0.6512

0.00966

4544.953

<.0001

Average Base Pay

Engineering

1

8.83E-06

1.77E-07

2505.0457

<.0001

Average Base Pay

Math

1

9.65E-06

2.90E-07

1105.0596

<.0001

Average Base Pay

Science

1

1.09E-06

1.78E-07

37.4325

<.0001

Engineering (n= 108,251), Math (n= 20,000), Science (n=119,810), and Technology (n= 86,921).

With respect to gender and working in Engineering relative to Technology, the female parameter estimate (β= -.5266) is significantly different from zero, as is the case with Math relative to Technology (β= .2955) and Science relative to Technology (β= .6512). This indicates that women are less likely to be expected to be in Engineering compared to Technology, but more likely to be expected in Math and Science, when salary is taken into consideration. To be more precise, the odds ratio for Engineering is 0.591, while it is 1.334 for Math and 1.918 for Science. This indicates that women are about 40 percent less likely to work in Engineering, while being 33 percent more likely to work in Math and nearly 92 percent more likely to work in Science than those working in Technology jobs. These results mirror the results depicted in Table 4, where women’s actual participation in STEM occupations versus their expected numbers are presented.

VI. Recommendations

As a result of the above described research and analysis, we propose the following recommendations.

Recommendations

  • Agencies should recruit more women in Technology and Engineering job series.
  • Agencies should develop recruitment plans that encourage more women to join STEM occupations in the Federal Sector.
  • Agencies should develop targeted recruitment plans for women from historically underrepresented racial and ethnic backgrounds.
  • Agencies should ensure that all STEM occupational groups meet the 12 percent target for hiring women with disabilities.
  • Agencies should create leadership development programs to help prepare women for leadership roles in STEM occupations.
  • Agencies should conduct organizational assessments in STEM occupational groups to determine how to reduce the amount of sex-related complaint activity and improve women’s viewpoints of their work environments.
  • Agencies should collect and analyze exit interview data for women in STEM to determine if there were issues related to discrimination or harassment that led to their departure.
  • Agencies should ensure that leaders in STEM demonstrate a commitment to a diverse workforce.

Conclusion

This report is designed to present an important examination of women working in STEM occupations in the federal government—who they are, their viewpoints, and workplace experiences concerning discrimination and harassment. We anticipate that the information and analyses herein will be helpful to federal agencies as they embark upon their own barrier analyses of Women in STEM occupations. In addition to barrier analyses, agencies should conduct other relevant analyses, such as salary studies and complaint audits. This report is intended to assist federal agencies’ continuing efforts to make the federal government a model EEO employer and able to compete with the private sector for critical talent.

References

Anderson, E., & Kim, D. (2006). Increasing the success of minority students in Science and Technology. American Counsel on Higher Education. https://www.acenet.edu/Documents/Increasing-the-Success-of-Minority-Students-in-Science-and-Technology-2006.pdf

Aycock, L. M., Hazari, Z., Brewe, E., Clancy, K. B., Hodapp, T., & Goertzen, R. M. (2019). Sexual harassment reported by undergraduate female physicists. Physical Review Physics Education Research, 15(1), 1-13. doi:10.1103/physrevphyseducres.15.010121

Burke, K. (2017). Harassment in Science. American Scientist, 105(5), 262. https://www.americanscientist.org/article/harassment-in-science 

Funk, C., & Parker, K. (2020, May 30). Women and men in STEM often at odds over workplace equity. Pew Research Center. https://www.pewsocialtrends.org/2018/01/09/women-and-men-in-stem-often-at-odds-over-workplace-equity  

Glass, J., Sassler, S., Levitte, Y., & Michelmore, K. (2013). What's so special about STEM? A comparison of women's retention in STEM and professional occupations. Social Forces, 92(2), 723-756. www.jstor.org/stable/43287810

Grandy, J. (1998). Persistence in Science of high-ability minority students: Results of a longitudinal study. The Journal of Higher Education, 69(6), 589-620. doi:10.2307/2649210

Leaper, C., & Starr, C. R. (2018). Helping and hindering undergraduate women’s STEM motivation: Experiences with STEM encouragement, STEM-related gender bias, and sexual harassment. Psychology of Women Quarterly, 43(2), 165-183. doi:10.1177/0361684318806302

Legewie, J., & Diprete, T. A. (2014). The high school environment and the gender gap in Science and Engineering. Sociology of Education, 87(4), 259-280. doi:10.1177/0038040714547770

Lim, K. (2016). Major matters: Exploration of the gender wage gap among STEM graduates. UCLA. https://escholarship.org/uc/item/5d87v86c  

Lim, N., Haddad, A., Butler, D. M., & Giglio, K. (2013). Demographic trends and the DoD STEM workforce. In first steps toward improving DoD STEM workforce diversity: Response to the 2012 Department of Defense STEM Diversity Summit (pp. 17–30). RAND Corporation. http://www.jstor.org/stable/10.7249/j.ctt5vjw4b.11   

Mason, M., Goulden, M., & Frasch, K. (2009). Why graduate students reject the fast track. Academe, 95(1), 11-16. https://www.aaup.org/article/why-graduate-students-reject-fast-track  

National Academy of Sciences, National Academy of Engineering, & Institute of Medicine (2011). Expanding underrepresented minority participation: America's Science and Technology talent at the crossroads. Washington, DC: National Academies Press. https://doi.org/10.17226/12984  

Riegle-Crumb, C., King, B., Grodsky, E., & Muller, C. (2012). The more things change, the more they stay the same? Prior achievement fails to explain gender inequality in entry into STEM college majors over time. American Educational Research Journal, 49(6), 1048-1073. doi:10.3102/0002831211435229

Smith-Doerr, L., Alegria, S., Fealing, K. H., Fitzpatrick, D., & Tomaskovic-Devey, D. (2019). Gender pay gaps in U.S. Federal science agencies: An organizational approach. American Journal of Sociology, 125(2), 534-576. doi:10.1086/705514

Sorenson, S. (2013). How employee engagement drives growth [Green Paper]. Gallup. https://www.gallup.com/workplace/236927/employee-engagement-drives-growth.aspx

Strayhorn, T., DeVita, J., & Blakewood, A. (2012). Broadening participation among women and racial/ethnic minorities in Science, Technology, Engineering and Maths. In Basit T. & Tomlinson S. (Eds.), Social inclusion and higher education (pp. 65-82). Bristol University Press. doi:10.2307/j.ctt1t891n1.8

Tan-Wilson, A., & Stamp, N. (2015). College students’ views of work–life balance in STEM research careers: Addressing negative preconceptions. CBE: Life Sciences Education, 14(3). doi:10.1187/cbe.14-11-0210

American Association of University Women (2020, March 31). The STEM gap: Women and girls in Science, Technology, Engineering and Math. https://www.aauw.org/resources/research/the-stem-gap

Wyss, V. L., & Tai, R. H. (2010). Conflicts between graduate study in Science and family life. College Student Journal, 44(2), 475-491. https://www.researchgate.net/publication/303208976_Conflicts_between_graduate_study_in_science_and_family_life  

Xu, Y. (2015). Focusing on women in STEM: A longitudinal examination of gender-based earning gap of college graduates. The Journal of Higher Education, 86(4), 489-523. doi:10.1353/jhe.2015.0020

 

[1] A targeted disability is a type of disability deemed to potentially have a profoundly limiting impact on an individual’s opportunity to gain meaningful employment.

[2] The list of occupations was obtained from https://www.fedscope.opm.gov. FedScope is a public data warehouse for federal employment data. The data that EEOC receives from OPM is similar to that provided via FedScope.

[3] A targeted disability is a type of disability deemed to potentially have a profoundly limiting impact on an individual’s opportunity to gain meaningful employment. For specific examples of targeted disabilities, see OPM Standard Form 256 at: https://www.opm.gov/forms/pdf_fill/sf256.pdf

[4] EEOC’s Final Rule on Affirmative Action for People with Disabilities in Federal Employment established participation goals.  https://www.eeoc.gov/laws/guidance/questions-answers-eeocs-final-rule-affirmative-action-people-disabilities-Federal

[5] As describe in the Methodology section of this report, EEOC does not collect complaint data by occupation.

[6] All described significant correlations are Pearson’s r values with p-values of less than 0.05.