Gender and job access:
Homophily in progressive hiring practices at small and mediumIT businesses
Sara B. Haviland, Jennifer Craft Morgan, and Victor W. Marshall
Paper for presentation at the International Labor Process Conference
Rutgers University, March 15-17 2010
WORK IN PROGRESS
ABSTRACT
While women currently comprise half of the US workforce, in specific industries they remain underrepresented. One such industry is information technology (IT), a model of the new economy which offers jobs characterized by many positive features: autonomy, potential for income growth, skill development, and work flexibility. Using data from the Workforce Aging in the New Economy Study (WANE), an international study of small and medium IT firms utilizing a case study methodology, we analyze survey and interview data to determine the role of hiring practices in the recruitment of women to these desirable IT jobs. We begin with a quantitative assessment of firm homophily with regards to gender and age, placing each firm on a matrix of gender and age homophily. We then use the matrix to select the most and least homogenous for in-depth case analysis of hiring practices. We first explore organizational hiring practices using survey data. We then analyze interview data to determine how firm members describe their hiring processes and the team environment, and identify factors that appear to encourage or discourage the hiring of women and those of different ages in these firms, including interview processes, attitudes towards females, attitudes towards education, and firm size.
The authors wish to acknowledge the Social Sciences and Humanities Research Council of Canada for their support of the Workforce Aging in the New Economy Study. Further information about the study can be found at
INTRODUCTION
Information technology (IT) is a vanguard of the modern, knowledge economy. Typified by intellectual labor rather than physical labor or the transformation of natural resources, the knowledge economy is based on a continuous cycle of rapid advances and almost equally rapid obsolescence in technology and science(Powell and Snellman 2004). When it comes to IT, innovation is not limited to technology and services either; IT businesses are also the vanguard of new employee-friendly work practices. From the heyday of the IT boom in the 1990s, when IT introduced workplaces with a work hard, play hard ethic, to today, when IT companies have topped the Fortune Best Companies to Work for List for the past five years[1](Fortune Magazine), the IT industry has served as the model for other companies wishing to become employee-friendly, a testing ground for human resource practices that range from traditional (e.g. sharing profits with employees) to radically innovative (e.g. Google’s bus service for employees - see(Helft 2007)). For the most part, it would seem that when it comes to employment practices, IT is an industry that has its heart in the right place.
However, IT is not immune to issues surrounding workplace inequalities. Overwhelmingly dominated by young white men, IT is an industry with plenty of room for improvement when it comes to workplace segregation. Birds of a feather (McPherson, Smith-Lovin, and Cook 2001) often work together. The demographic composition of a workplace, and workplace segregation in particular, has long captured the attention of sociologists of work and inequality(Bielby and Baron 1986a; Bielby and Baron 1986b; Kmec 2008; Reskin, McBrier, and Kmec 1999; Reskin 1993; Rosenfeld 1983; Rosenfeld and Kalleberg 1991; Stewman 1988). The presence of a dominant in-group majority and a smaller, powerless out-group impacts not only position and opportunity for the out-groupbut also the day-to-day experiences of these workers, and can have a major impact on ambitions(Collins 1997; Kanter 1977).
A variety of factors affect the composition of organizations, including labor pools, statistical discrimination, group power, labor costs and group status, market incentives, and employment practices (Reskin, McBrier, and Kmec 1999). These break down into supply-side and demand-side factors. While larger organizations may have the ability to affect supply-side factors in tangible ways, a great deal of the US labor market is actually represented by smaller employers (Granovetter 1984; Wiatrowski 1994) who have less control over external market forces. Most small employers that would be interested in trying to achieve a diverse workplace only have threesimple factors in play: organizational entries (hires), exits (quits, fires, or layoffs), and general appeal to out-groups (work environment, human resource practices). This paper is concerned with the first factor. The hiring process serves as the portal to work organizations, and is the starting point for organizational composition when it comes to demography. Hiring is a critical source of difference in organizational access for out-groups (Kmec 2008), and likely more subjective than other parts of the employment relationship as it is less likely for rejected applicants to take legal action for unfair treatment than employees (Petersen, Saporta, and Seidel 2000).
We examine two particular sources of inequality: gender and age.[2]As noted by Ridgeway and Smith-Lovin (Ridgeway and Smith-Lovin 1999), gender is unlike other forms of social inequality where the in-group and out-groups are socially segregated; to the contrary, men and women have highly integrated social groupings as members of families, households, and throughout various role relations. Gender discrimination is not based in unfamiliarity; rather, men and women know each other quite well and even hold members of the other sex in high regard. We submit that age relations also occupy a similar strange location in social inequality, as intergenerational relationships are the cornerstone of family life, and group membership is constantly changing. The old have experience with life in the younger age groups, and with a bit of good fortune, the young will eventually become the old group. No other form of social inequality has such a dynamic, where membership in the in-group and out-group is on a continuum through which all members are moving.
Using data from the Workforce Aging in the New Economy study (WANE), we investigate hiring practices at firms that are particularly heterogeneous and those that are particularly homogenous with respect to gender and age. The WANE study utilized a mixed-methods case study approach, and therefore allows analysis of quantitative survey data and in-depth interviews in the context of work organizations. The WANE cases are all small and medium enterprises, reflecting the organizational environment of a great deal of workers. The cases are distributed among four western countries (the US, Canada, Australia, and the UK). We analyze the way that firms describe their hiring processes and the team environment, paying particular attention to who is involved in the hiring processes, the language surrounding descriptions of ideal candidates, and characteristics of the team with respect to gender and age.
What follows is a preliminary analysis, the first step in analyzing the role of hiring in the composition of small enterprises. We begin with an assessment of the literature on work segregation, hiring, and the specific situation of the IT industry. Next we describe in greater detail the WANE study and our approach to these data. We create homogeneity assessments for each firm and assign them placement on a homogeneity matrix. This matrix is then used to select cases for in-depth case study analysis, using interview data and web survey data to delve further into specific hiring practices. Finally, we offer some lessons for firms looking to diversify their workforce, and describe our future research objectives on these points.
BACKGROUND
The IT industry is a knowledge-driven, change-oriented field that has recently been a lead industry in modeling positive employment relationships. IT jobs have many of the hallmarks of ‘good jobs’ (Kalleberg, Reskin, and Hudson 2000) and are typified by relatively good pay, high autonomy, and good opportunities for advancement. This can make them very attractive to applicants and employees on the one hand, but can also give the incumbents great incentive to exclude outsiders. (Reskin, McBrier, and Kmec 1999). As an industry, IT experienced a rapid phase of growth in the boom years of the 1990s, which can be associated with increasing gender diversity (Reskin 1993); however, after the bust and slow recovery, we might expect gains in diversity to decline. Examining the hiring practices of a mid-sized technology firm and its applicant pool, Petersen et. al. found that gender was of little consequence in the hiring process for men and women, and all differences between the sexes could be explained by differences in age and education. (Petersen, Saporta, and Seidel 2000).
Firm size presents both challenges and opportunities when it comes to diverse workforce compositions. Employer stereotypes and preferences can become custom in a company, creating organizational inertia that privileges certain groups (Baron, Mittman, and Newman 1991; Reskin 1993). Young, small entrepreneurial firms are often positioned better than large, bureaucratic establishments to rapidly integrate their workforces, as they do not face the same level of organizational inertia (Baron, Mittman, and Newman 1991,Reskin 1993). However, while younger organizations may not face liabilities like that of aging and organizational inertia, younger organizations do face liabilities related to newness and size that challenge organizational legitimacy and survival (Aldrich and Auster 1986). Organizations often mimic other successful organizations to enhance survival odds in a process known as institutional isomorphism; this process may improve survival odds but it does not always improve efficiency (DiMaggio and Powell 1983) or enhance the achievement of organizational objectives. These survival pressures can inhibit innovation in human resource practices or the structure of work life. Additionally, as small firms, these organizations are often exempt from regulation meant to curb discrimination (Reskin and Hartmann 1986, Reskin 1993). This lack of regulation may allow a higher level of discrimination in small business, indicating an implicitpublic acceptance of homogeneity in small firms.Taken small firm by small firm, this seems innocuous. However, in the aggregate, these small firms employ a large percentage of the workforce. (Granovetter 1984). For example, in the US, employers with fewer than twenty employees employ nearly a fifth of the working population, and firms with fewer than 100 employees represent more than one-third of the working population (U.S. Small Business Administration, Office of Advocacy). One half of Americans working in private businesses are employed by companies with fewer than 100 employees (Wiatrowski 1994).
Homophily is the principle that similar individuals are more likely to be socially proximate than dissimilar individuals. This means that organizations depending on social networks to recruit members are likely to have a homogeneous membership (McPherson, Smith-Lovin, and Cook 2001). As noted by (Cohen, Broschak, and Haveman 1998), organizations are more likely to hire and promote women when a higher proportion of women already work there. The placementof womenwithin a firm also impacts hiring outcomes, as female-owned firms are more likely than male-owned firms to hire women (Carrington and Troske 1995). Other hiring issues that have the potential to influence diversity include the level of formality in hiring practices and the level of discretion on the part of the decision-makers (Kmec 2008). Proportions of in- and out-groups have particularly strong effects on the working life of out-groups, with more balanced groups leading to the least discrimination and the least pressure from visibility and polarization (Blau 1977; Kanter 1977). Sex segregation affects a large proportion of businesses (Bielby and Baron 1986a). Reskin et. al. (1999) noted outcomes of segregation for both workers (cross-group contact, stress, satisfaction, turnover) and organizations (performance, human resource practices, levels of job segregation, and benefits).
Hiring happens at the firm or establishment level. Reskin et. al. (1999) note “occupations and industries do not employ workers or constitute the settings in which people work. Establishments are both actors in employment decisions and the settings in which workers perform.” The employment organization sits at the intersection of micro and macro forces (Baron and Bielby 1980), a setting that weds societal-level trends to individual outcomes. Understanding hiring practices in a cross-section of organizations was part of a call issued by Reskin (2000), who outlined a research agenda:
These accounts must be based on interviews with informants at multiple levels in the organization. Researchers must learn what if any organizational constraints determine how the opening came about, who specifies the necessary qualifications, and who can authorize exceptions? Who defines the applicant pool, whether anyone has an inside track, and if so, who, why, and with what result? What is the demographic composition of the pool? Of the decision makers? At what levels are relevant decisions made, and what are their consequences? Does the decision create expectations about future job-assignment decisions? Given the way that a position is filled, how likely is it that someone of an atypical race or sex could be appointed? Are there other ways the job in question could be filled and that other positions are filled that would alter these probabilities? In short, we need to identify the actors, how they define the situation, whose definitions prevail, and why.
Our efforts to begin to address this agenda are further outlined below.
DATA AND METHODS
This article is centered in mixed-methods analysis, and has a two-step approach. First, we use quantitative data to generate an overview of the state of homogeneity in small and medium IT firms. These data are then used to pinpoint cases for the second step: an in-depth case analysis,centered in qualitative data, of hiring practices and the team environment in firms that are particularly homogenous or particularly heterogeneous on the dimensions of gender and age.
The analysis employs data collected using a case study approach. Case studies are particularly useful in the study of the IT industry, which is demographically young and known for its rapid evolution in both technology and the organization and implementation of work. Given the rapid evolution of the industry, case studies are particularly appropriate sources for in-depth, contextualized data(Marshall 1999). As noted by Ragin (1989)“[c]ase studies, by their nature, are sensitive to complexity and historical specificity” thus making them an ideal research approach in which to analyze this ever-shifting industry. This chapter triangulates web survey data and firm-based interview data to explore the relationships between firm homogeneity and hiring practices.
The Workforce Aging in the New Economy project is an international, cross-comparative study of information technology (IT) workers. Centered at the University of Western Ontario, the study included cases in six countries: the U.S., Australia, Canada, England, Germany, and the Netherlands. Due to language and other cultural barriers, methodology was most consistent across Canada, Australia, England, and the U.S., and therefore we have focused on these four countries in our analyses.Data collection for the study occurred from 2004 to 2006, after the IT boom of the 1990s and the IT bust of the early 2000s, but prior to the deep worldwide recession of the late 2000s. This was a time when the IT industry was in transition, stabilizing from the industry downturn but not yet subject to the heavy blows of the coming economic freefall.
Each country study was centered within a region that had a thriving IT industry, and the study teams were located at research universities within these regions. For example, in the U.S., the team was centered at the University of North Carolina Institute on Aging, located in the larger Triangle region of North Carolina, itself home to the Research Triangle Park (RTP). RTP is a major IT region, smaller than Silicon Valley but larger than many others. A satellite U.S. team was centered at Florida State University, in Tallahassee, Florida. The Tallahassee area hasa smaller but still active IT industry; the bulk of cases were snowball sampled from major regional professional organizations.
The international WANE study used two major data collection instruments uniformly across regions: semi-structured interviews with selected employees at all levels in WANE firms, and a web-based quantitative survey of all IT employees in WANE firms. Altogether 47 cases were completed internationally (11 in Australia, 18 in Canada, 7 in the U.K., 11 in the U.S.).
Web surveys for case studies
Invitations to participate in the survey were issued on paper and by email at the beginning of each case study, and reminder emails were sent to non-responders. All IT employees in the WANE firms were invited to participate. The survey took approximately one hour to complete. A total of 452 surveys were completed internationally, with an international response rate of 46%[3]
To determine firm homogeneity for the present analysis, we examine gender, and age. Analysis was limited to firms with five or more respondents for the age and gender questions in the web survey, thereby limiting analysis to firms with a large enough group of employees to be comprised of more employees than simply the founding teams. This resulted in a sample of 30 firms internationally (12 Canadian firms, 5 Australian firms, 5 English firms, and 8 U.S. firms) out of the original 47 WANE firms.