Women in Construction and the Workforce Investment Act

Abstract

This article examines whether women’s representation among senior-level construction jobs increased over time during a long-term, large-scale construction project in Boston, Massachusetts, and whether enactment of the Workforce Investment Act (WIA) has fostered support and connection to construction career opportunities for women in this metropolitan area. Results show that the proportion of hours worked by women in senior-level positions on the project was relatively high for a few years, followed by a general decline across all job types, and the WIA did not significantly enhance opportunities for women in construction.

Keywords

women construction training Workforce Investment Act nontraditional employment

Introduction

In coming years, the construction industry is predicted to be a significant driver of employment opportunities (Bureau of Labor Statistics 2015). Careers in the construction industry offer the unique benefit of paid on-the-job training and opportunities for advancement. Workers in construction and extraction occupations earn median annual wages of $41,380, exceeding the national median of $35,540 across all occupations (Bureau of Labor Statistics 2015). Despite the promise of this growing sector of the economy, a very small proportion of construction jobs are filled by women. Barriers to entry and challenges to retention within construction occupations have led to women representing just 2.5 percent of construction workers nationwide (Bureau of Labor Statistics 2014b), which falls well below the federal goal of at least 6.9 percent (Moir, Thompson, and Kelleher 2011). This article examines women’s retention and advancement in construction careers and the ability of a public employment program to connect women to jobs in this industry.

Prior research has found that women consistently account for around 3 percent of individuals in entry-level construction apprenticeship programs (Moir, Thompson, and Kelleher 2011). Despite continually constituting 3 percent of entrants into these training programs, overall representation of women in the field remains at the already noted 2.5 percent (Bureau of Labor Statistics 2014b), which suggests that women are not developing an increasing presence in upper level construction positions. This research uses a newly released database containing over sixty-four million employee hours worked at the apprentice, journey worker, and foreperson levels on the twenty-year Central Artery/Tunnel construction project (commonly referred to as the Big Dig) in Boston, Massachusetts—with more than 1.9 million hours’ worth of women’s work recorded—to explore the topic in detail.

Furthermore, this research examines whether enactment of the Workforce Investment Act (WIA) and the associated creation of American Job Centers (AJCs)¹ have fostered support and connection to construction career opportunities for women in the Boston area. The creation of AJCs, which focus on linking individuals to training and employment, has the potential to connect women with opportunities in the construction industry in Boston. Given that the Centers were implemented in September 2000 (halfway through the Big Dig²), evidence of women being connected to construction careers through AJCs may have been observed through increased female employment on the Big Dig during the second half of the project. However, analysis of the Big Dig data found that the development of the AJC system is not strongly associated with an increase in women’s participation in the project. Women’s participation in all construction job types (apprentice, journey worker, and foreperson) generally declined over the course of the Big Dig.

The remainder of this paper is organized as follows: first, I detail benefits and barriers to women’s participation in the construction industry. Next, I provide a discussion of programs and policies that support, or have the potential to support, women in construction occupations. I then present research questions and methods, provide an analysis of the results, and discuss findings.

Benefits and Barriers to Women’s Participation in the Construction Industry

The construction industry entails (1) preparing building sites, (2) constructing buildings and engineering projects (e.g., highways and dams), and (3) maintaining, altering, and repairing these types of structures (Bureau of Labor Statistics 2014c). Occupations within this industry include carpenters, brick masons, equipment operators, laborers, electricians, plumbers, roofers, and sheet metal workers (Bureau of Labor Statistics 2014a). In addition to relatively high wages, construction positions also often offer health care and pension benefits, which are important facets of financial security (Moir, Thompson, and Kelleher 2011).

Jobs with relatively high wages and benefits in the construction industry hold particular promise for women. Women are overrepresented in the low-wage workforce, working in jobs that are less likely to offer benefits and more likely to be associated with high rates of employee turnover (Albelda 2009). Occupations that fall into the low-wage category include personal care and health care support positions (Albelda 2009). Among full-time workers in the female-dominated personal care and services field, a woman earns just seventy-five cents for every dollar earned by a man (Bureau of Labor Statistics 2013d). The gender wage gap is much less pronounced in the construction industry: among full-time workers in construction and extraction occupations, women earn ninety-eight cents for every dollar earned by a man (Bureau of Labor Statistics 2013d).

An additional benefit related to construction careers is that employees receive paid training while at work through apprenticeship programs (Torpey 2013). On-the-job training allows individuals to concurrently earn an income and learn marketable skills, which may be especially significant for working parents. Combining employment and training constitutes a more cost-effective and convenient way to build labor market skills than incurring additional child care and transportation costs to obtain training in a classroom outside working hours. Furthermore, construction workers report having more flexibility and freedom to change their schedules when needed than workers in other industries (Sweet and Pitt-Catsouphes 2010). This type of employment flexibility provides an important benefit for single parents, 77 percent of whom are female (U.S. Census Bureau 2010). Single mothers have reported stress when struggling to care for their children in the face of inflexible work arrangements in low-wage jobs (Kahn 2005).

Apart from pay equity, convenient on-the-job training, and flexibility, the overall employment potential for construction work shows promise. The Bureau of Labor Statistics has forecasted a 10.1 percent national growth rate in construction occupations through 2024, which is significantly higher than the 6.5 percent growth rate across all occupations (Bureau of Labor Statistics 2015).

A demographic analysis of American Community Survey data by the Institute for Women’s Policy Research found that among current female construction workers, two-thirds are white and nearly 20 percent are Hispanic. Less than 10 percent are African American, and 6 percent identify as Asian, American Indian, or more than one race (Hegewisch and O’Farrell 2014). Respondents to a recent exploratory survey of women in construction were largely the primary breadwinner in their households. A third of respondents were single, and 55 percent of respondents had children. The age distribution of surveyed workers was nearly evenly split between women under forty-five years of age, and women aged forty-five or older (Hegewisch and O’Farrell 2014).

While women working in construction experience multiple benefits related to this career choice, there are also significant barriers facing women who seek employment in this field. Women have reported difficulty getting and retaining construction jobs at all levels. Barriers faced by women include entering the industry with fewer skills than their male counterparts, not having access to informal networks commonly used to find work, being let go once effort has been made toward meeting diversity quotas, and encountering hostile work environments and sexual harassment on the job.

The first step toward a construction career often involves completion of an apprenticeship program to develop trade-based skills. In 2011, over 400,000 individuals were active in a formal apprenticeship program (Helmer and Altstadt 2013). Formal construction apprenticeship programs are typically sponsored by unions and contractor associations and require two to four years of training and instruction in technical skills (Bureau of Labor Statistics 2013a). Apprenticeships involve a curriculum that combines on-the-job training with classroom learning. The classroom portion of the training is often paid for by the organization sponsoring the apprenticeship (Conway, Gerber, and Helmer 2010). The wages received by apprentices working on job sites through an apprenticeship program increase over time (Mulligan-Hansel, Owens-Wilson, and Beach 2013).

Due to gender role socialization, women may embark upon the path toward a construction career with fewer relevant skills and less experience with tools than their male counterparts (Moir, Thompson, and Kelleher 2011). Women have also reported having trouble finding jobs as apprentices and having less access to the informal networks used by men to gain entry into the industry (Byrd 1999). Trade unions offer an important route to construction jobs for women. Union apprenticeship programs have had greater success at recruiting women, along with higher completion rates, than nonunion programs (Argyres and Moir 2008). A majority (80%) of female apprentices participate in union-sponsored programs (Mulligan-Hansel, Owens-Wilson, and Beach 2013).

Workers who complete an apprenticeship program graduate to journey-level status within the industry. Journey workers are paid a higher wage rate and are entitled to additional benefits (Mulligan-Hansel, Owens-Wilson, and Beach 2013). The most upper level position on construction jobs is the foreperson. Foreperson managers of construction jobs receive higher wages than journey-level workers and have fewer physical work requirements (Eisenberg 1998).

Women able to enter the construction industry often face subsequent challenges. Those who have worked in construction have described being hired to ensure a contractor is meeting a certain diversity quota, only to be let go once a requisite number of hours have been worked to meet compliance standards (Eisenberg 1998; Moir and Azaroff 2008). This type of behavior, called “checkerboarding” (Moir and Azaroff 2008, 205), harms women’s advancement within the industry. For those women able to maintain a position in the industry, hostile work environments (due to the perception that female construction workers are outsiders in a men’s occupation) and sexual harassment are common (Moir, Thompson, and Kelleher 2011). In the long term, continued harassment can negatively affect female workers’ ability to persevere in the field (Byrd 1999). Women who have persisted in the industry and have achieved foreperson status describe struggling with uncooperative male team members, sabotage, heightened scrutiny, and discrimination (Eisenberg 1998).

Despite these challenges, women in construction often describe a feeling of fulfillment associated with mastering a skilled trade and a sense of accomplishment that comes with finishing a major project. A comment from one woman in construction illustrates the point: “My kids know which bridges in town are mine. I have a lot of pride in that” (Eisenberg 1998, 90). This is not surprising; women, like men, derive great satisfaction from independently exercising learned skills to produce a building, highway, or other structure (Moir, Thompson, and Kelleher 2011).

Programs and Policies Applicable to Women in Construction

Various nonprofit programs and federal antidiscrimination policies have been implemented to specifically support women seeking construction careers. A number of nonprofit organizations offer pre-apprenticeship training programs to help women gain the skills needed to compete for apprenticeship opportunities. Pre-apprenticeship programs generally span one to three months and may require either full-time or part-time/flexible participation in hands-on and classroom training activities (Conway, Gerber, and Helmer 2010).

Federal policies prohibit gender-based employment discrimination and require affirmative action to guarantee equal employment opportunity (Office of Federal Contract Compliance Programs 2009). The earliest antidiscrimination law was the Civil Rights Act of 1964, which made employment discrimination illegal. Executive Order (EO) 11246 was issued one year later, prohibiting discrimination and providing affirmative action for minorities in federally funded construction projects. In 1967, EO 11246 was amended to extend the same protections to women (Eisenberg 1998). Subsequent amendments set an affirmative action goal of using female construction workers for 3.1 percent of hours on federal or federally assisted contracts and later revised this goal to 6.9 percent (Eisenberg 1998).

According to the Office of Federal Contract Compliance Programs (2009, 8), “Affirmative action goals under EO 11246 are targets for recruitment and outreach and should be reasonably attainable by means of applying good faith efforts. The standard of compliance is good faith.” This good faith effort has had little impact on women’s participation in the construction industry, however. While women’s participation in construction has increased over the past thirty years, the percentage of women working in construction remains low. In 1983, women comprised just 1.8 percent of construction workers (Eisenberg 1998). By 2012, women made up 2.5 percent of workers in construction and extraction occupations (Bureau of Labor Statistics 2014b). Although the percentage change between 1983 and 2012 reflects a nearly 40 percent increase over approximately three decades, women’s prevalence within the industry grew by less than one percentage point and is still well below the 6.9 percent federal affirmative action goal.

While some policies specifically aim to promote women’s involvement in construction, other policies, like WIA, work more generally to connect jobseekers with training and employment opportunities. A major goal of WIA is to streamline the process of matching local business needs for trained employees with area workers looking for training and employment (Department of Labor 2010). Local AJCs were developed to connect individuals with promising employment and training opportunities in their communities (Department of Labor 2010).

WIA was initially passed in August 1998, with the broad purpose of providing

workforce investment activities, through statewide and local workforce investment systems, that increase the employment, retention, and earnings of participants, and increase occupational skill attainment by participants, and, as a result, improve the quality of the workforce, reduce welfare dependency, and enhance the productivity and competitiveness of the Nation.

This piece of legislation mandated development of AJCs where individuals could access information about job training and education services and search for employment (WIA of 1998). The final rule concerning the implementation of AJCs was made effective in September 2000, paving the way for the Centers to open and operate on a national scale (Department of Labor 2010).

Currently, there are nearly three thousand AJCs (both comprehensive and affiliate centers) operating nationwide (Jacobson 2009). The Centers typically provide training and job placement services to over fifteen million individuals each year, around 10 percent of the labor force (Bureau of Labor Statistics 2013c; Jacobson 2009). Services are offered to adult jobseekers under two different programs: the adult program and the dislocated worker program. The adult program is meant to provide training and employment assistance to adults who have low initial skills levels, while the dislocated worker program targets services for workers who have experienced a recent layoff (Employment and Training Administration 2016). Among individuals receiving services through the adult and dislocated worker programs, roughly half are women. Program users are commonly high school graduates without postsecondary credentials. More than one-quarter of participants in the adult program receive some form of public assistance, nearly half are considered low income, and 12 percent are single parents (Department of Labor 2015a, 2015b).

As a publicly funded resource with a goal of connecting individuals to promising job opportunities, AJCs have the potential to connect low-income women to training and employment opportunities in construction, creating a pathway to financial security. However, surveys with female construction workers suggest that few are directed into construction work through the AJC system (Hegewisch and O’Farrell 2014). A disconnect between women and construction careers may stem from the way success is measured within the AJC system. WIA has been criticized for maintaining a work-first orientation, where jobseekers are placed directly into jobs wherever possible, at the expense of developing skills and training that could lead to long-term financial stability. One measure of a successful AJC client outcome is a job placement—regardless of whether that job presents a promising pathway out of poverty (Gatta 2014). This has negative implications for workers, and particularly for low-income single mothers, who may be placed directly into low-skill jobs without the education and training needed to exit poverty (Gatta 2010). The Workforce Investment and Opportunity Act (WIOA), passed in 2014 to revise and replace WIA, works to address this concern by placing a heavier emphasis on the pursuit of skills and training for workers (Department of Labor n.d.). However, WIOA continues to measure AJC performance by the number of users who enter employment and by employment retention rates and average earnings (Employment and Training Administration 2016).

Shaw and Rab (2003) find that performance measures which emphasize success in obtaining a job and maintaining employment and earnings over time give AJCs an incentive to “cream” their clients by choosing to focus the most time and resources (i.e., training and educational opportunities) on those who have the best chance of flourishing under these metrics. Creaming for successful outcomes could prevent female AJC clients from being matched to employment and training opportunities in construction. In other words, because women face barriers to finding this type of work and staying employed, they may be less likely to be trained and matched to construction jobs. A lack of connection to construction careers through AJCs would create an additional obstacle to women’s participation in a field in which they already face barriers to participation.

In the Boston area, the construction industry flourished over the course of the multi-billion dollar Big Dig project, which created thousands of job opportunities in the city. The project involved removing and replacing an antiquated elevated highway with a high-capacity underground highway and building two new bridges over the Charles River (Massachusetts Department of Transportation 2014b). Initial planning began in 1982, and construction work spanned 1989 through 2010. The Big Dig project excavated sixteen million cubic yards of dirt and laid 3.8 million cubic yards of concrete in the process of building 161 lane miles of highway (Massachusetts Department of Transportation 2014b). The project improved traffic flow, established green open space and a beautified skyline, and ultimately created the most expensive highway ever built in the United States (Flint 2015). At its height, over five thousand construction workers were employed on the project (Massachusetts Department of Transportation 2014b).

Enactment of AJCs midway through the Big Dig, and widespread use of the Centers by over fifteen million individuals seeking employment and training each year (Jacobson 2009), had the potential to connect women to construction careers during the latter years of the project. If AJCs are able to facilitate connections for women to employment and training opportunities in construction, we might expect women’s presence in this field to have increased after AJCs were established. Such a change may be visible in Big Dig employment data, which tracked working hours for women and men on contracts related to the project for ten years before, and ten years after, the final rule implementing AJCs was enacted. However, prior research has suggested that women are not generally matched with employment in nontraditional fields³ through AJCs. For example, one study of large-scale WIA training program data showed that women are typically directed into traditionally “women’s jobs” (which do not include construction work), and men are directed to “men’s jobs” (Hegewisch and Luyri 2009). Furthermore, other factors external to the AJC process of connecting workers with employment—such as the availability of training opportunities, the role of unions in connecting individuals with construction work, and women’s inclination to enter a field that has traditionally been perceived as masculine—could have influenced women’s participation in the industry over time.

Research Questions and Hypotheses

This research focuses on two main questions as follows:

Research Question 1: To what extent were women working at various levels of employment (particularly senior levels) throughout the Big Dig?

Research Question 2: Is there an association between the enactment of AJCs and women’s participation in the Big Dig?

These questions are examined using the Big Dig payroll data set assembled by the Massachusetts Department of Transportation.

Four hypotheses are associated with the analyses as follows:

Hypothesis 1: The prevalence of women working on the Big Dig will not increase subsequent to enactment of AJCs under WIA.

Hypothesis 2: Women’s prevalence in all job types will decrease as contract length increased.

Hypothesis 3: Women will be more prevalent at upper levels of employment when employed by woman-owned businesses (WBEs).

Hypothesis 4: As the total number of monthly project hours worked in a certain position (e.g., foreperson, journey worker) increases, prevalence of women within that position will increase.

Due to reported practices like checkerboarding, the variety of external factors that may affect women’s participation in construction, and the AJC’s focus on job placements and retention as successful outcomes, it is hypothesized that the creation of AJCs and related efforts to support individuals seeking employment and training will not be associated with an increase in women’s prevalence on the Big Dig (Hypothesis 1). Prevalence is measured as the likelihood of women working 1 percent or more of hours on construction vendor-submitted monthly payroll reports, as described in the “Method” section below.

Prior research detailing the experiences of women who have been laid off contracts after certain diversity quotas have been met (Eisenberg 1998; Moir and Azaroff 2008) also informs Hypothesis 2. Hours worked by women and men on the Big Dig were recorded under 147 contracts of varying lengths. As contract length increased, women’s prevalence at all levels of employment was expected to decrease, which could indicate women being let go once an effort to meet diversity quotas has been expressed.

Through a variable capturing whether employee hours worked were performed at a WBE, the data set provides a unique opportunity to explore women’s prevalence in upper level construction positions when employed by a WBE. A 2012 Catalyst report of employment and leadership in the business world finds that upper level women are more likely than their male counterparts to invest in the career development of female workers, for example, by mentoring and opening doors to opportunities (Dinolfo, Silva, and Carter 2012). This phenomenon may occur in the construction industry as well, with female business owners more likely to support the promotion of female employees, as described in Hypothesis 3.

Historically, times of economic expansion have enabled women to gain traction in previously male-dominated occupations (Amott and Matthaei 1996). By extension, when a greater demand for labor exists within a male-dominated industry, more women may be able to access work in this industry. This theory informs Hypothesis 4.

Method

The Big Dig data set contains payroll data for the entire Big Dig project, covering October 1989 through April 2010. These data were compiled from construction vendors’ Monthly Employment Utilization (MEU) reports. Construction vendors completed MEU reports for each aspect of project-based work. Vendors often submitted multiple reports each month, reflecting different projects and/or construction trade types. For example, in one month, a construction company submitted two separate MEU reports for project-based laborer and operating engineer work.

Variables in the data set include monthly hours worked by employees based on gender, race (with hours worked by black, Hispanic, Asian, and Native American workers reported separately), and employment level (foreperson, journey worker, and apprentice⁴). Also included is information on whether work hours were performed by a disadvantaged business enterprise (woman-owned and/or minority-owned business) and the type of construction trade associated with the MEU work. There are 102 separate trades represented in the data. As prior research on women in construction has largely focused on qualitative studies (see, for example, Byrd 1999; Goldenhar et al. 2003; Schroedel 1985), availability of a large quantitative data set to examine participation and advancement trends allows for an innovative approach to this topic.

The data set contains information on hours performed under 147 different contracts (120 of which span multiple years) throughout the course of the Big Dig. For each of the 120 multi-year contracts, chi-square tests are performed to assess whether the percentage of total women’s hours worked in upper level positions (journey worker and foreperson) on the contract was significantly greater in the last year of the contract relative to the first. If women were increasingly represented in higher level positions over time while employed on multi-year contracts, this analysis will show a greater percentage of total women’s hours being worked in upper level positions during the final year of each contract.

Next, a logistic regression model is estimated to understand trends in the likelihood of women’s participation in upper level positions on the Big Dig both before and after the enactment of AJCs. The model focuses on measuring how the likelihood of women working at least 1 percent of monthly MEU hours in upper level construction positions (foreperson and journey worker) changes over time, particularly in the months subsequent to the final rule implementing AJCs.

The model is expressed as follows⁵:

P = \Pr (Y = 1),

\log (\frac{p}{(1 - p)}) = β_{0} + β_{1} months + β_{2} post_WIA + β_{3} contract_months + β_{4} WBE + β_{5} MEU_hours + e .

The dependent variable, Y, is an indicator variable equal to one if women worked 1 percent or more of total hours worked in upper level positions (foreperson or journey worker) on a given MEU form in a given month and zero otherwise. The model is also run using apprentice data (with the dependent variable indicating whether women worked 1 percent or more of total apprentice hours recorded on MEU reports) for comparison. One independent variable (months) represents the number of months elapsed since the first month in which MEU reports were submitted for the Big Dig project. The first month during which payroll data were submitted (October 1989) is assigned 0, the next month (November 1989) 1, and so on.

The “post_WIA” variable is used to examine differences in the likelihood of women working 1 percent or more of hours in different job types in the time period spanning September 2000, when the final rule concerning the implementation of AJCs under WIA was made effective (Department of Labor 2010), through April 2010. This variable is set to equal 1 if the MEU month was September 2000 or later, and 0 if the report month was prior to September 2000.

Another independent variable, “contract_months,” reflects total length (in months) of the contract for which the MEU was submitted. An independent dummy variable, WBE, indicates whether MEU hours were performed for a WBE. A final independent variable, “MEU_hours,” reflects total number of hours recorded in the relevant position (e.g., foreperson when examining the foreperson dependent variable, journey worker when examining the journey worker dependent variable) on each MEU form in a given month.

When examining foreperson data, cases are only included if some foreperson hours were worked on a given MEU in a given month. For example, if an MEU did not include any hours worked in a foreperson position by either a male or a female foreperson in a given month, this case is excluded from the model. Analysis of journey worker and apprentice data uses a similar method. When analyzing each job type, robustness tests are performed using a dependent dummy variable indicating whether women are working at least 5 percent or at least 25 percent of hours on an MEU in a given month. Descriptive statistics for all variables, reflecting data collected on 62,858 MEU forms and used in logistic regression modeling, are displayed in Table 1.

Table 1.

Descriptive Statistics, Logistic Regression Variables.

Quantitative variables^a
Variable	n	M	SD	Minimum	Maximum
Months	62,858	123.7	43.1	0	246
Total months in contract	62,858	77.9	32.8	1	146
Total foreperson hours worked on MEU	62,858	167.6	568.8	0	11,196
Total journey worker hours worked on MEU	62,855	812.5	2,010.6	0	33,950
Total apprentice hours worked on MEU	62,858	40.9	173.3	0	8,104
Categorical variables^b
Variable	n	Frequency (%)
Woman-owned business	7,110	11.30
Non-woman-owned business	55,748	88.70
Total	62,858	100.00
Variable	Foreperson		Journey worker		Apprentice
Variable	n	Frequency (%)	n	Frequency (%)	n	Frequency (%)
Women work ≥ 1% of total MEU hours in position	530	2.30	8,308	14.70	1,713	17.10
Women work < 1% of total MEU hours in position	22,368	97.70	48,126	85.30	8,322	82.90
Total	22,898	100.00	56,434	100.00	10,035	100.00
Women work ≥ 5% of total MEU hours in position	418	1.80	5,922	10.50	1,657	16.50
Women work < 5% of total MEU hours in position	22,480	98.20	50,512	89.50	8,378	83.50
Total	22,898	100.00	56,434	100.00	10,035	100.00
Women work ≥ 25% of total MEU hours in position	136	0.60	1,391	2.50	1,249	12.40
Women work < 25% of total MEU hours in position	22,762	99.40	55,043	97.50	8,786	87.60
Total	22,898	100.00	56,434	100.00	10,035	100.00

Total MEU cases = 62,858. MEU = monthly employment utilization.

An error was found in which three MEU forms listed journey worker hours as negative numbers. These instances were removed from the analysis, resulting in a total of 62,855 MEU forms being included in the analysis of journey worker hours.

MEU forms may contain hours worked in either one, two, or all three of the different job types; therefore, the total n in each Total row will not add up to 62,858.

Findings

Preliminary analysis of the Big Dig data finds that women performed 3 percent of all construction hours on the project—well under half of the 6.9 percent federal goal. Figures 1 and 2 display the percentage of hours worked by women in foreperson and journey worker positions (Figure 1) as well as apprentice positions (Figure 2) each year over the course of the project. Although hours varied, women tended to work the greatest portion of hours in apprentice positions and the smallest portion of hours in foreperson positions. As displayed in Table 2, female apprentices worked an average of 8.2 percent of annual apprentice hours, with female journey workers and forepersons working smaller portions of annual hours in journey worker and foreperson positions (2.9% and 0.9%, respectively). Women’s participation peaked in the early 1990s. Female apprentices worked 29.1 percent of annual apprentice hours in 1993, and female forepersons and journey workers both completed a high of 5.2 percent of total hours worked in each position type in 1991. A trend line (shown in Figures 1 and 2) indicating the percentage of hours worked in all three job types shows a general downward tendency over the course of the project, with a small upward movement in 2007 before dropping off in the final months.

Figure 1.

Percent of total annual foreperson and journey worker hours worked by women on the Big Dig, 1989-2010.

Figure 2.

Percent of total annual apprentice hours worked by women on the Big Dig, 1989-2010.

Table 2.

Percent of Total Annual Foreperson, Journey Worker, and Apprentice Hours Worked by Women on the Big Dig, 1989-2010.

Year	Foreperson (%)	Journey worker (%)	Apprentice (%)	All positions (%)
1989	^a	0.0	0.0	0.0
1990	0.0	1.9	0.0	1.3
1991	5.2	5.2	1.0	4.5
1992	0.3	4.4	25.5	4.9
1993	1.7	4.1	29.1	4.7
1994	3.0	4.0	22.1	4.9
1995	0.3	4.0	13.7	4.3
1996	1.0	4.0	16.4	4.0
1997	0.8	3.5	18.2	3.6
1998	0.6	3.4	12.6	3.3
1999	0.6	3.4	8.8	3.2
2000	1.0	2.8	8.5	2.8
2001	0.9	2.4	7.5	2.3
2002	1.3	2.3	6.6	2.3
2003	0.6	2.3	6.3	2.1
2004	0.1	2.5	3.0	1.9
2005	0.2	2.0	0.6	1.5
2006	1.1	1.8	0.3	1.6
2007	0.0	2.8	0.1	1.8
2008	0.0	4.5	0.0	3.4
2009	0.0	3.3	0.0	2.4
2010	0.0	0.0	0.0	0.0
Average	0.9	2.9	8.2	2.8
Minimum	0.0	0.0	0.0	0.0
Maximum	5.2	5.2	29.1	4.9

Zero hours worked by forepersons (male or female).

Table 3 displays summary contract-level data, showing that among upper level positions, women were more likely to work as journey workers than as forepersons. While more than 86 percent of contracts employed female journey workers, a smaller 42 percent of contracts in which foreperson hours were recorded employed female forepersons. Overall, journey workers worked more hours than either of the other construction job types on the project. Male and female workers in this position type recorded over fifty-one million hours’ worth of effort on all contract MEU forms, compared with around 10.5 million hours worked by forepersons and 2.6 million hours worked by apprentices. Women worked the greatest number of hours as journey workers (1.6 million or 3.1% of total journey worker effort). However, the portion of work performed by women was largest at the apprentice level, where over 254,000 hours representing 9.9 percent of all apprentice work was completed by female employees.

Table 3.

Summary Contract Data.

	Foreperson	Journey worker	Apprentice	All positions
Number of contracts in which some hours in position were worked	115	147	113	147
Number of contracts in which some hours in position were worked by women	48	127	74	129
Percent of contracts in which some hours in position were worked by women	41.7%	86.4%	65.5%	87.8%
Average percent of total hours in position worked by women on a contract	1.8%	3.7%	12.2%	3.6%
Median percent of total hours in position worked by women on a contract	0.0%	3.1%	5.0%	3.0%
Minimum percent of total hours in position worked by women on a contract	0.0%	0.0%	0.0%	0.0%
Maximum percent of total hours worked by women on a contract	100.0%	24.7%	100.0%	18.2%
Total hours worked in position on all contracts	10,536,633	51,072,247	2,569,369	64,178,248
Total hours worked by women in position on all contracts	85,533	1,562,794	254,386	1,902,713
Percent of hours worked by women in position on all contracts	0.8%	3.1%	9.9%	3.0%

Each multi-year contract was examined to determine whether women were working a greater percentage of hours in upper level positions in the last year of the contract versus the first. For most contracts, this was not the case. Women were working a smaller percentage of hours in upper level positions in the last year of a contract relative to the first in seventy-eight out of 120 cases. Chi square testing reveals that none of the changes in percentage of hours worked by women on multi-year contracts are statistically significant. Results of these tests are displayed in Table 4.

Table 4.

Differences in the Percent of Foreperson and Journey Worker Hours Worked by Women in the First and Last Year of Each Multi-Year Contract.

Contract number	Percent of total hours worked by women		χ²	Contract number	Percent of total hours worked by women		χ²	Contract number	Percent of total hours worked by women		χ²
Contract number	First year	Last year	χ²	Contract number	First year	Last year	χ²	Contract number	First year	Last year	χ²
1	5.9	3.6	0.01	41	8.6	6.0	0.01	81	7.1	0.0	0.07
2	1.8	2.8	0.00	42	0.0	2.6	0.03	82	22.4	0.0	0.25
3	3.2	0.0	0.03	43	1.5	4.7	0.02	83	0.0	0.0	^a
4	0.0	2.4	0.02	44	3.4	3.5	0.00	84	2.8	0.0	0.03
5	1.1	8.4	0.06	45	3.0	3.4	0.00	85	2.0	3.8	0.01
6	0.0	0.0	0.00	46	3.3	0.0	0.03	86	0.0	0.0	^a
7	2.0	4.7	0.01	47	3.7	2.5	0.00	87	0.0	0.0	^a
8	2.9	0.9	0.01	48	0.0	0.0	^a	88	2.2	0.0	0.00
9	0.0	5.0	0.05	49	0.0	0.0	^a	89	0.0	7.7	0.08
10	3.5	0.0	0.04	50	1.9	4.2	0.01	90	0.0	0.0	^a
11	2.9	1.6	0.00	51	1.0	9.6	0.07	91	0.0	0.0	^a
12	0.0	4.0	0.04	52	11.7	0.9	0.10	92	0.0	2.7	0.03
13	7.5	2.2	0.03	53	2.2	0.0	0.02	93	0.0	0.0	^a
14	6.0	0.0	0.06	54	3.2	7.3	0.02	94	0.0	5.1	0.05
15	8.3	0.0	0.09	55	0.0	0.1	0.00	95	0.0	4.2	0.04
16	7.0	7.9	0.00	56	0.0	0.0	^a	96	0.0	3.2	0.03
17	2.1	0.0	0.02	57	2.1	0.0	0.02	97	5.5	0.0	0.06
18	6.8	4.9	0.00	58	0.0	0.0	^a	98	28.0	10.9	0.09
19	4.8	0.0	0.05	59	5.1	2.8	0.01	99	13.1	10.7	0.00
20	2.9	0.0	0.03	60	0.0	0.0	^a	100	3.9	2.4	0.00
21	3.8	0.0	0.04	61	0.0	0.0	^a	101	0.0	2.0	0.02
22	0.0	0.5	0.01	62	9.1	0.0	0.10	102	0.0	10.4	0.11
23	1.9	0.0	0.02	63	2.5	26.1	0.23	103	10.1	0.0	0.11
24	0.0	0.0	^a	64	4.2	0.0	0.04	104	0.0	27.5	0.32
25	1.7	0.0	0.02	65	0.2	0.0	0.00	105	0.0	0.0	^a
26	5.2	0.0	0.05	66	3.7	0.8	0.02	106	4.0	0.0	0.04
27	1.2	0.0	0.01	67	1.0	0.0	0.01	107	0.0	0.0	0.00
28	11.0	0.0	0.12	68	8.7	0.6	0.07	108	0.0	0.0	^a
29	1.9	3.4	0.00	69	2.8	0.0	0.03	109	4.8	4.3	0.00
30	2.0	0.0	0.02	70	0.0	0.0	^a	110	0.0	0.0	^a
31	0.0	8.3	0.09	71	0.0	6.3	0.01	111	0.0	1.8	0.02
32	2.8	0.0	0.03	72	8.6	0.0	0.09	112	0.0	0.0	^a
33	2.0	0.0	0.02	73	3.2	0.0	0.03	113	1.7	0.0	0.02
34	0.4	4.6	0.04	74	1.0	12.6	0.11	114	5.5	6.0	0.00
35	3.6	0.0	0.04	75	4.4	1.2	0.02	115	12.0	14.5	0.00
36	2.1	0.0	0.02	76	0.0	13.1	0.14	116	1.2	0.0	0.01
37	0.0	0.0	^a	77	0.0	0.0	^a	117	2.2	0.7	0.01
38	3.2	5.4	0.01	78	1.7	3.3	0.01	118	0.0	0.0	^a
39	5.6	2.2	0.02	79	0.0	0.0	^a	119	0.0	3.2	0.03
40	0.0	1.8	0.02	80	8.3	10.4	0.00	120	5.8	5.4	0.00

All χ² tests used one degree of freedom.

Test statistic cannot be calculated due to inability to divide by zero.

p < .1. **p < .05. ***p < .01.

The results of the logistic regression model are displayed in Table 5. The estimated odds ratio associated with the “post_WIA” variable shows whether the likelihood of women working at least 1 percent of total hours increased or decreased after AJCs were established. An odds ratio greater than 1 means the likelihood increased, and a ratio less than 1 means that it decreased. For female apprentices, the likelihood of women working 1 percent or more of hours recorded on MEU forms subsequent to the enactment of AJCs in September 2000 decreased by 6.6 % however, this change was not statistically significant. The odds of 6.6 percent were calculated by taking the odds ratio associated with the post_WIA variable (0.934) and subtracting 1, then multiplying by 100. Both journey workers and forepersons experienced statistically significant changes in the odds of working at least 1 percent of total MEU hours after AJCs were established. At the journey worker level, odds decreased by about 33.5 percent. For forepersons, in contrast, the likelihood of working 1 percent of more of MEU hours increased significantly after AJCs were established in September 2000, by nearly 84 percent. The findings for apprentices and journey workers support Hypothesis 1, with female apprentices maintaining similar prevalence subsequent to the enactment of AJCs and journey workers experiencing declining prevalence. Foreperson findings do not support this hypothesis.

Table 5.

Logistic Regression Model Results.

Variable	Odds of working 1% or more of MEU hours model	Odds of working 5% or more of MEU hours model	Odds of working 25% or more of MEU hours model
Variable	OR (SE)	OR (SE)	OR (SE)
Forepersons
Months	0.992*** (0.002)	0.995** (0.002)	0.994 (0.004)
Post WIA	1.838*** (0.289)	1.539** (0.265)	2.993*** (0.961)
Contract months	1.001 (0.002)	1.003* (0.002)	0.996 (0.003)
Woman-owned business	1.441** (0.249)	1.539** (0.279)	2.852*** (0.651)
MEU hours	1.001*** (0.000)	1.000*** (0.000)	1.000 (0.000)
Journey workers
Months	0.997*** (0.001)	0.997*** (0.001)	0.997*** (0.001)
Post WIA	0.665*** (0.031)	0.681*** (0.033)	0.949 (0.092)
Contract months	0.998*** (0.000)	0.998*** (0.000)	0.995*** (0.001)
Woman-owned business	0.878*** (0.040)	0.830*** (0.040)	1.029 (0.084)
MEU hours	1.001*** (0.000)	1.000*** (0.000)	0.999*** (0.000)
Apprentices
Months	0.987*** (0.001)	0.987*** (0.001)	0.986*** (0.001)
Post WIA	0.934 (0.087)	0.946 (0.089)	1.079 (0.113)
Contract months	1.002 (0.001)	1.001 (0.001)	1.002* (0.001)
Woman-owned business	0.276*** (0.041)	0.232*** (0.037)	0.315*** (0.054)
MEU hours	1.002*** (0.000)	1.001*** (0.000)	1.000*** (0.000)

ORs are converted to relative differences by subtracting 1. MEU = monthly employment utilization; OR = odds ratio; WIA = Workforce Investment Act.

p < .1. **p < .05. ***p < .01.

For each additional month in contract length, the likelihood of female journey workers working 1 percent or more of hours on MEU forms decreases slightly, by 0.2 percent. Odds of working 1 percent or more of MEU hours do not change significantly for forepersons or apprentices based on contract length. The journey worker findings support Hypothesis 2, while the results found among apprentices and forepersons do not.

Journey workers were less likely to work at least 1 percent of hours on MEU forms when employed by WBEs. Conversely, forepersons were much more likely to represent 1 percent or more of hours worked on MEUs when employed by WBEs. The odds of women working 1 percent or more of foreperson hours on a given MEU form are 44.1 percent greater for women working for WBEs relative to those working for non-WBEs. Findings related to the most upper level foreperson position support Hypothesis 3, while journey worker findings do not support this hypothesis.

In all job types, an increase in the total number of hours reported on an MEU typically increases the likelihood of women working at least 1 percent of those hours. Among forepersons, the odds of women working 1 percent or more of foreperson hours on an MEU project increase by about 0.1 percent for each additional foreperson hour reflected on that MEU. Similar and slightly higher odds are found for journey workers and apprentices, respectively. These findings support Hypothesis 4 across all position types.

Overall, there is a decline in women’s prevalence in each job type over the course of the Big Dig. As measured by the likelihood of women working 1 percent or more of hours on a given MEU, the prevalence of women working on projects decreases at every level of employment (foreperson, journey worker, and apprentice) each month for the duration of the project. Within the context of this overall declining trend, women’s prevalence decreases in the journey worker position subsequent to the enactment of AJCs, and increases in the most upper level foreperson position. Among apprentices, there is no significant change. A variety of factors (including, for example, women’s propensity to enter a male-dominated field and the availability of training opportunities in construction) could have influenced women’s participation on the project in different job types in the months subsequent to the enactment of AJCs. However, if AJCs had the capacity to advertise and connect women to construction careers, a marked increase in participation (particularly in entry-level apprentice positions and potentially extending to upper level positions over time) would have been consistent with the enactment of the Centers in September 2000.

Although the findings associated with this research are significant, they are subject to some limitations. First, the data set does not include variables measuring the host of other factors that may have affected women’s participation in the project over time. For instance, regression modeling is unable to control for factors like the presence or absence of specific project-related or vendor-related participation goals, which could have been associated with checkerboarding. Furthermore, the data did not include individual employee identifiers, which could have been used to track the persistence of individual women as employees and to examine their movements up the career ladder over time.

Discussion and Conclusion

Preliminary analysis of the Big Dig data reveals that women’s prevalence in all positions in construction decreased over the course of the project. On most multi-year contracts, women worked a smaller percentage of hours in upper level positions during the last year than in the first. Consistent with trends observed in participation, logistic regression modeling indicates a decreased likelihood of women working 1 percent or more of MEU hours at all levels of employment as the project progressed. Odds of female forepersons working 1 percent or more of MEU hours increased after September 2000, while odds declined for female journey workers and remained the same for female apprentices. An increase in women’s prevalence on MEUs in the latter half of the Big Dig (post-AJC enactment), particularly at the apprentice level, would have supported an association between the development of AJCs and women’s ability to enter the construction industry. However, odds for apprentices did not change, and odds for journey workers declined. The finding that odds increased for female forepersons is surprising, since the prevalence of women working on the Big Dig was not expected to increase subsequent to AJC enactment.

The decrease in the likelihood of women working 1 percent or more of hours on MEUs in each position type over the course of the Big Dig project could have numerous causes. Checkerboarding may play a part, with contractors letting female workers go after fulfilling project-related diversity quotas. In 65 percent of multi-year contracts, women worked a larger portion of hours during the first year of the contract than in the last. Furthermore, women’s overall participation in the project peaked during the first five years and dropped off during the last five years of the project.⁶ These trends could indicate an attempt by contracted construction employers to meet federal and/or project-based female participation goals that drop off once a requisite good faith effort has been demonstrated.

Other factors could also have contributed to the declining likelihood of women working 1 percent or more of hours on MEUs over time. Socially stressful work environments, including sexual harassment and verbal abuse as cataloged by Byrd (1999), may have harmed women’s ability to persevere, particularly in upper level positions. Furthermore, a severe economic downturn occurred toward the end of the project, with a recession spanning December 2007 through June 2009 and a subsequent slow recovery that hit the construction industry particularly hard (Tasci and Lindner 2010). Economic contraction could have led to women being overlooked in favor of male workers as the overall availability of construction work declined. This phenomenon would have been consistent with women gaining ground in male-dominated industries during times of economic expansion and losing ground during periods of contraction and may explain the sharp decline in women’s participation during the final years of the project. Other explanations are also possible.

Interestingly, working for a WBE has mixed effects on the odds of increasing women’s participation in upper level positions over time. Among journey workers, employment at a WBE does not have a significant impact on the likelihood of women working 1 percent or more of hours on a project. Yet for forepersons, working for a WBE is associated with an increased likelihood of working 1 percent, 5 percent, and 25 percent or more of hours reported on MEU forms. This finding lends some support to the theory that women may be more likely to mentor and support female workers in the upper most levels of employment.

Consistent with Hypothesis 4, increases in the number of hours recorded in a particular position on an MEU form increase the likelihood of women working 1 percent or more of hours on the project. This finding supports the idea that, as opportunity within the industry increases (as measured by increasing available hours), women’s participation increases. As the demand for construction work is predicted to experience above-average growth in the near future (Bureau of Labor Statistics 2015), the coming years may provide enhanced prospects for women to enter and gain footholds in the industry.

Further research is called for to explore factors that promote and discourage women’s retention and advancement on construction projects, and the impact of working for a WBE on women’s experiences. In addition, research may work to develop a more in-depth understanding of the processes by which AJCs connect workers to employment and the extent to which connections exist between unions, construction firms, and public employment and training programs.

Footnotes

Acknowledgements

The author thanks Susan Moir, director of the University of Massachusetts Boston Labor Resource Center, and Scott Davis, research associate at IMPAQ International, for their advice and feedback on this paper.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This paper was funded with Federal funds from the U.S. Department of Labor, Employment and Training Administration under Contract Number DOLQ121A21885, Task Order DOLU121A21908. The contents of this publication do not necessarily reflect the views or policies of the Department, nor does mention of trade names, commercial products, or organizations imply endorsement of same by the U.S. Government.

Notes

Author Biography

Carolyn Arcand is a lecturer in political science and public administration at the University of New Hampshire. Her research interests include women, labor policy, and economic policy. She earned a PhD in public policy from the University of Massachusetts Boston in 2014.

References

Albelda

Randy

. 2009. “Why It’s Harder (and Different) for Single Mothers: Gender, Motherhood, Labor Markets and Public Work Supports.” Center for Social Policy, University of Massachusetts Boston.

Amott

Teresa

Matthaei

Julie

. 1996. Race, Gender, and Work. Boston: South End Press.

Argyres

Anneta

Moir

Susan

. 2008. “Building Trades Apprentice Training in Massachusetts: An Analysis of Union and Non-Union Programs, 1997-2007.” The Future of Work Paper Series, Paper No. 3a, Labor Resource Center, University of Massachusetts Boston.

Bureau of Labor Statistics. 2013a. “Construction Laborers and Helpers.” In Occupational Outlook Handbook, 2012-13. http://www.bls.gov/ooh/construction-and-extraction/construction-laborers-and-helpers.htm.

Bureau of Labor Statistics. 2013b. “Labor Force Characteristics by Race and Ethnicity, 2012.” Report 1044. http://www.bls.gov/opub/reports/race-and-ethnicity/archive/race_ethnicity_2012.pdf.

Bureau of Labor Statistics. 2013c. “Table 3. Employment Status of the Civilian Noninstitutional Population by Age, Sex, and Race.” Labor Force Statistics from the Current Population Survey. http://www.bls.gov/cps/tables.htm.

Bureau of Labor Statistics. 2013d. “Table 39. Median Weekly Earnings of Full-Time Wage and Salary Workers by Detailed Occupation and Sex.” Labor Force Statistics from the Current Population Survey. http://www.bls.gov/cps/tables.htm.

Bureau of Labor Statistics. 2014a. “Construction and Extraction Occupations.” In Occupational Outlook Handbook. http://www.bls.gov/ooh/construction-and-extraction/home.htm.

Bureau of Labor Statistics. 2014b. “Current Population Survey. Employed Persons by Intermediate Occupation, Sex, Race, and Hispanic or Latino Ethnicity, Annual Averages 2000-2013.” Data File. Received electronically from Karen Kosanovich, Economist, Office of Employment and Unemployment Statistics, Bureau of Labor Statistics, January 30.

10.

Bureau of Labor Statistics. 2014c. “Industries at a Glance.” Construction: NAICS 23. http://www.bls.gov/iag/tgs/iag23.htm.

11.

Bureau of Labor Statistics. 2015. “Table 4. Employment by Major Occupational Group, 2014 and Projected 2024.” Economic News Release. http://www.bls.gov/news.release/ecopro.t04.htm.

12.

Byrd

Barbara

. 1999. “Women in Carpentry Apprenticeship: A Case Study.” Labor Studies Journal 24 (3): 3-22.

13.

CareerOneStop. 2014. “Learn More About the American Job Center Finder.” http://www.careeronestop.org/localhelp/americanjobcenters/find-american-job-centers-help.aspx.

14.

Conway

Maureen

Gerber

Allison

Helmer

Matt

. 2010. “Construction Pre-apprenticeship Programs: Interviews with Field Leaders.” The Aspen Institute, Washington, DC. http://www.aspeninstitute.org/.

15.

Department of Labor. 2009. “Quick Facts on Nontraditional Occupations for Women.” http://www.dol.gov/wb/factsheets/nontra2008.htm.

16.

Department of Labor. 2010. “20 CFR Part 652; Part 660 et al. RIN 1205-AB20. Workforce Investment Act.” http://www.doleta.gov/usworkforce/finalrule.cfm.

17.

Department of Labor. 2015a. “PY 2014 WIA Summary Report—Adults.” https://www.doleta.gov/.

18.

Department of Labor. 2015b. “PY 2014 WIA Summary Report—Dislocated Workers.” https://www.doleta.gov/.

19.

Department of Labor. n.d. “WIOA Overview.” https://www.doleta.gov/wioa/Overview.cfm.

20.

Dinolfo

Sarah

Silva

Christine

Carter

Nancy

. 2012. “High Potentials in the Pipeline: Leaders Pay It Forward.” Catalyst, New York. http://www.catalyst.org/knowledge/high-potentials-pipeline-leaders-pay-it-forward.

21.

Eisenberg

Susan

. 1998. We’ll Call You if We Need You: Experiences of Women Working Construction. Ithaca, NY: Cornell University Press.

22.

Employment and Training Administration. 2016. “Workforce System Results.” U.S. Department of Labor, Washington, DC.

23.

Flint

Anthony

. 2015. “10 Years Later, Did the Big Dig Deliver?” The Boston Globe, December 29. https://www.bostonglobe.com/magazine/2015/12/29/years-later-did-big-dig-deliver/tSb8PIMS4QJUETsMpA7SpI/story.html.

24.

Gatta

Mary

. 2010. “It’s More than Just Job Placement! Rethinking Workforce Development Policy to Raise Single Mothers Out of Poverty.” In Interactions and Intersections of Gendered Bodies at Work, at Home, and at Play. Advances in Gender Research, vol. 14, edited by Segal

Marcia Texler

, 213-31. Bingley, UK: Emerald.

25.

Gatta

Mary

. 2014. All I Want Is a Job! Unemployed Women Navigating the Public Workforce System. Stanford: Stanford University Press.

26.

Goldenhar

Linda

Hecker

Steven

Moir

Susan

Rosecrance

John

. 2003. “The ‘Goldilocks Model’ of Overtime in Construction: Not Too Much, Not Too Little, but Just Right.” Journal of Safety Research 34 (2): 215-26.

27.

Hegewisch

Ariane

Luyri

Helen

. 2009. “The Workforce Investment Act and Women’s Progress: Does WIA Funded Training Reinforce Sex Segregation in the Labor Market and the Gender Wage Gap?” Institute for Women’s Policy Research, Washington, DC.

28.

Hegewisch

Ariane

O’Farrell.

Brigid

2014. “Women in Construction and the Economic Recovery: Results from 2013 IWPR Tradeswomen Survey.” Institute for Women’s Policy Research, Washington, DC.

29.

Helmer

Matt

Altstadt

Dave

. 2013. “Apprenticeship: Completion and Cancellation in the Building Trades.” The Aspen Institute, Washington, DC.

30.

Jacobson

Louis

. 2009. “Strengthening One-Stop Career Centers: Helping More Unemployed Workers Find Jobs and Build Skills.” Discussion Paper 2009-01. The Brookings Institution, Washington, DC.

31.

Kahn

Peggy

. 2005. “The Work-Family Time Binds of Low-Income Mothers: Nurse Aides Struggle to Care.” Journal of Women, Politics & Policy 27 (3): 97-111.

32.

Massachusetts Department of Transportation. 2014a. “The Big Dig.” https://www.massdot.state.ma.us/highway/TheBigDig.aspx.

33.

Massachusetts Department of Transportation. 2014b. “The Big Dig: Facts and Figures.” http://www.massdot.state.ma.us/highway/TheBigDig/FactsFigures.aspx.

34.

Moir

Susan

Azaroff

Lenore S.

2008. “The Boston-Area HASWIC Research Circle: An Innovative Participatory Method for Coloring in the Picture of a Special Work Environment.” In At the Point of Production: The Social Analysis of Occupational and Environmental Health. Work, Health and Environment, edited by Levenstein

Charles

. Amityville, NY: Baywood Publishing.

35.

Moir

Susan

Thompson

Meryl

Kelleher

Christa

. 2011. “Unfinished Business: Building Equality for Women in the Construction Trades.” Labor Resource Center, University of Massachusetts Boston.

36.

Mulligan-Hansel

Kathleen

Owens-Wilson

Sebrina

Beach

Ben

. 2013. The Construction Careers Handbook. Washington, DC: Partnership for Working Families. http://www.forworkingfamilies.org/.

37.

Office of Federal Contract Compliance Programs. 2009. “Technical Assistance Guide for Federal Construction Contractors.” U.S. Department of Labor, Washington, DC.

38.

Schroedel

Jean Reith

. 1985. Alone in a Crowd: Women in the Trades Tell Their Stories. Philadelphia: Temple University Press.

39.

Shaw

Kathleen

Rab

Sara

. 2003. “Rhetoric versus Reality in Policy and Practice: The Workforce Investment Act and Access to Community College Education and Training.” The Annals of the American Academy of Political and Social Science 586 (1): 172-93.

40.

Sweet

Stephen

Pitt-Catsouphes

Marcie

. 2010. “Talent Pressures and the Aging Workforce: Responsive Action Steps for the Construction Sector.” Industry Sector Report 5.1. Sloan Center on Aging & Work at Boston College, Chestnut Hill, MA.

41.

Tasci

Murat

Lindner

John

. 2010. “The Great Recession and Its Impact on Different Industries.” Economic Trends. Federal Reserve Bank of Cleveland, OH. http://www.clevelandfed.org/research/trends.

42.

Torpey

Elka

. 2013. “Apprenticeship: Earn While You Learn.” Occupational Outlook Quarterly 57 (2): 2-13.

43.

U.S. Census Bureau. 2010. “Table DP02. Selected Social Characteristics in the United States.” 2006-2010 American Community Survey 5-Year Estimates. http://factfinder2.census.gov.

44.

Workforce Investment Act of 1998. 1998. Pub. L. No. 105-220, § 106, 112 Stat. 936.