Workplace Control : Women and Minority Workers in America

In this paper, we examine the effects of gender and race on American workers’ workplace control. Scholarship on gender, work, and occupation states that gender and race are important predictors of the extent of control workers exercise in workplaces. Literature also posits that job satisfaction and work-family conflict also contribute substantially to workers’ workplace control. However, there exists hardly any empirical study that explores the impacts of gender, race, job satisfaction and work-family conflict altogether on their workplace control. That is what we accomplished in this study. Obtaining data from the 2008 National Study of Changing Workforce (NSCW), we ask: 1) Do women and men workers in America differ in their perceptions of workplace control? 2) Do non-white and white workers in America differ in their perceptions of workplace control? And 3) Do gender and race of the workers influence their workplace control when job satisfaction and work-family conflict are considered? Analyses are based on quantitative methods. Results show that women perceive to have less control over their workplace as compared to men. Moreover, job satisfaction is a more significant predictor of their workplace control than work-family spillover.


Introduction
Gender scholars have paid much attention to workplace control of the workers since the beginning of women's movement in the early 1970s (Binder et al., 2010).The general conception is that with their increasing participation in the labor market, women and minority workers gain more control over their workplace (Binder et al., 2010;Cohn, 2000).However, current empirical studies show that even at the onset of 21st century, women and minority members do not have as much control on their workplaces as compared to men and workers from the dominant group respectively (Cohn, 2000;Grönlund, 2007;Harkness et al., 2005).Thus, gender and race of the workers continue to be the important predictors of workers' workplace control.
Moreover, scholarship on gender, work and occupation also identifies two other key factors that influence workplace control of the workers: job satisfaction and work-family conflict (Bethge et al., 2009;Noor, 2002).Interestingly, these two factors are often considered as the human capital characteristics of the workers (Binder et al., 2010;Cohn, 2000).Nevertheless, scholars have found that workers with greater job satisfaction often perceive to have more workplace control than those with less job satisfaction (Blau & Devaro, 2007;Näswallet al., 2005).Research on workers' workplace control also suggests that workers with greater work-family conflict are more likely to experience less control in their workplaces (Grönlund, 2007;Harkness et al., 2005).
However, to the best of our knowledge, there hardly exists any study that examines the impacts of gender, race, job satis-faction, and work-family conflict together on workplace control of the workers.In this study, we address that concern and ask: 1) Do women and men workers in America differ in their perceptions of workplace control?2) Do non-white and white workers in America differ in their perceptions of workplace control?And 3) Do gender and race of the workers influence their workplace control when job satisfaction and work-family conflict are considered?
This study intends to extend the literature on gender that focuses on the work-related well-being of the workers by exploring workplace control of American workers.In this study, we use the 2008 National Study of Changing Workforce (NSCW) survey to conduct rigorous quantitative analyses.Therefore, it is also an attempt to contribute to the empirical literature on gender, work, and occupation that focuses on worker's workplace control.This paper is organized into 4 specific sections.
In the first section we outline a brief review of the literature on workplace control in terms of the scholarship of gender, work, and occupation.In the second section, data and methods of the study are presented.The third section includes the findings of the study.Finally, in the conclusion section, we interpret the findings with regards to the literature on gender, work, and occupation.

Workplace Control
Workplace control is primarily defined as the decision making abilities of workers within workplaces (Feldt et al., 2004).
Additionally it also means a worker's power to express her opinion in different aspects of her work (Drago et al., 2009), as well as her ability to make choices regarding important workrelated elements such as subordinates, starting time, break time, and office hours (Fischbacher et al., 2005).Scholarship on gender, work and occupation considers workplace control as key to the understanding of women and minority workers' well-being in workplaces (Feldt et al., 2004).This is because with an increased participation of women and minority workers in the labor market, there is a substantial consciousness-raising among the workers about their decision making abilities, and access to resources about different aspects of workplaces (Grönlund, 2007).Grönlund (2007) posits that women workers with greater decision making power within workplaces are happier than women with less decision making power.
Workplace control of workers has been studied extensively by the scholars of gender, work, and occupation.Empirical literature presents 4 key predictors of workers' workplace control: gender, race, job-satisfaction, and work-family conflict (Binder et al., 2010;Blau & Devaro, 2007;Drago et al., 2009;Grönlund, 2007).Whereas gender and race are considered as demographic factors, job satisfaction and work-family conflict are often viewed as human capital variables (Bethge et al., 2009;Binder et al., 2010;Fischbacher et al., 2005;Noor, 2002).In the following subsections we will present brief description of the literature based on each of these key predictors of workplace control.

Gender and Workplace Control
There exist a number of studies that establish significant connection between gender and workplace control.For example, Binder et al. (2010) examine data for faculty at a public research university in the United States between 1995 and 2004.They explore whether or not gender wage gap can be explained by the faculty's decision making power.Their findings suggest that wage differentials result more from faculty's decision making abilities regarding courses taught, enrollment, grant dollars, and number and impact of publications and less from them being a woman or a man.Using a large sample of datum from the Multi-City Study of Urban Inequality Employer Survey, Blau and Devaro (2007) examines gender differences in workers' decision making abilities and promotional opportunities.Their results indicate that women have lower decision making power and less promotional opportunities than men even with comparable education, training, and job experience.
Gender has also been identified as an important component of workplace control within economics professions (McDowell et al., 2001).Evidence suggests that over the period from the 1960s through the early 1980s, female economists had lower levels of workplace control and career advancement than their male counterparts with similar educational and experiencebased qualifications.These gender differences persist irrespective of workers' racial/ethnic backgrounds and self-selection between profit and not-for-profit jobs (McDowell et al., 2001).Again, studies have presented interesting relations among gender, work-family conflict and workplace control (Noor, 2002).In a study of 310 employed women with families in Malaysia, gender as well work-family spillover are found to be the 2 vital factors for women's decision making abilities within their work settings (Noor, 2002).
Literature on gender, work, and occupation pays much atten-tion to transition of women workers from unpaid to paid work and suggests that women enter the paid labor market primarily for 2 reasons: 1) out of sheer economic necessity, and 2) to increase their decision making abilities both within and outside workplaces (Zipp & Plutzer, 2000).Understanding the decision making power of married women in workplaces and dual-career families, provides a perspective for addressing important issues like workers' well-being and happiness in the current economy (Zipp & Plutzer, 2000).Furthermore, gender has an explanatory power, and it is primarily due to the employment experiences like, autonomy, control, responsibility, self-direction, that shape women's work-life experiences within the labor market (Zipp & Plutzer, 2000).

Race and Workplace Control
Although race is defined as the sociological differences among people based on skin color, and ethnicity primarily refers to people's ancestry, in the literature of gender, work, and occupation these categories are often used interchangeably (Cohn, 2000).Literature suggests that people's race/ethnicity has a profound influence on their workplace control (Cohn, 2000).Using discourse analysis to examine work-life experiences of female Canadian clerical employees, Harkness et al. (2005) found that minority people are subjected to lower workplace decision making power than the dominant group.Another study found that the power of decision making within workplaces is an outcome of different factors that include race and job satisfaction (Näswall et al., 2005).Information from 400 nurses at a Swedish health care institute (91% of respondents being women) showed that both decision making ability and job security of nurses result from their race and job satisfaction (Näswall et al., 2005).
Race has also been identified as a strong determinant of jobrelated control among the social workers.Robinson (2009) suggests that English social workers from minority groups often experience lower self-esteem as compared to the workers from the dominant group.Thus, the former group often report lower work control than the latter (Robinson, 2009).Race affects work control of the workers-irrespective of their educational qualification.For example, Skinner (2002) studies gender and racial/ethnic differences in work experiences for high schooleducated residents of New York City.His study measures work-control in terms of workers' power to make choices about several aspects in their workplaces.Results show that workplace control varies across racial/ethnic backgrounds of the workers.Furthermore, African American and Hispanic American women perceived to have the least workplace control as compared to the other groups (Skinner, 2002).

Job Satisfaction and Workplace Control
When workers perceive that they are happy with what they do in their job settings, they tend to gain higher control over their work (Bethge et al., 2009).Work-related satisfaction results from a number of factors such as work place policies, benefits and rewards, relationship with supervisors, and physical facilities within workplaces (Bethge et al., 2009;Binder et al., 2010).Empirical studies on workplace control show that job satisfaction reduces work stress of the workers.Thus, they perceive to have more control on their work than workers with lower level of job-satisfaction (Bethge et al., 2009).
Gender also interacts with job-satisfaction to impact workplace control of the workers."Women's job satisfaction is one of the key elements of their workplace control."To examine this proposition, Bethge et al. (2009) used the data from the baseline survey of German SPE where 1463 working women and men (aged 30 -59 years) participated.Job satisfaction was defined in terms of workers' perceptions about workplace benefits, relationship with supervisors, and flexibility in work hours.Women employees with higher job satisfaction reported greater workplace control as compared to women employees with low job satisfaction.In their study of the faculty in a public research institute in the United States, Binder et al. (2010) found that women faculty who view themselves to be contented with their teaching and research seemed to have more control on their work than women faculty who are not happy with their jobs.
Job satisfaction and gender have been recognized as vital predictors of workplace control within economics professions as well (McDowell et al., 2001).McDowell et al. (2001) suggest that over a period of 2 decades (1960s to 1980s), female economists with lower job satisfaction had less control on their work as compared to women with greater job satisfaction.Again, Blau and Devaro (2007) used information from the Multi-City Study of Urban Inequality Employer to examine gender differences in workers' job-satisfaction and decision making abilities within their workplaces.Findings suggest that women employees with higher level of happiness in workplaces are more likely to make important work-related decisions than women with lower level of happiness.
In addition to race/ethnicity, Näswall et al. (2005) found that workplace control is also impacted by job satisfaction.Specifically, workplace decision making ability is least among the people of the minority group who also perceive a lower extent of job-satisfaction as compared to other groups of nurses.

Work-Family Conflict and Workplace Control
Just like gender, race, and job-satisfaction, work-family conflict has also been identified as one of the key determinants of workplace control of the workers (Bethge et al., 2009;Grönlund, 2007).Moreover, studies suggest that work-family conflict is more important factor for women than men in terms of workplace control.Grönlund (2007) makes a connection between work-family spillover and women's work-life experiences.She uses a survey data of 800 Swedish employees and studies the impacts of gender and work-family tension on workers' decision making power within workplaces.Her findings suggest that important factor for women's decision making is not only the quantity of work but also its quality.This aggravates work-family tension of women which in turn negatively impacts their decision making power within job settings.
There are studies that address the interaction of race/ethnicity and work-family conflict on workers' workplace control.For example, Harkness et al. (2005) conducted discourse analyses to explore work-life dynamics of female clerical employees of Canada.The authors suggest that minority employees with higher work-family spillover participate in workplace decision making processes much less than the other groups of workers.
Using discourse analysis to examine work-life experiences of female Canadian clerical employees Harkness et al. (2005) found that minority people are subjected to lower workplace decision making processes than the dominant group.However, to the best of our knowledge there exists hardly any study that examines workplace control in terms of all the 4 key aspects (gender, race, job-satisfaction, and work-family conflict) together.This is what we have accomplished in this paper.Therefore, based on the literature review, we hypothesize that: H1: Gender has a significant impact on workers' workplace control; H2: Race/ethnicity has a significant impact on workers' workplace control; H3: Job-satisfaction of the workers have significant impacts on workers' workplace control; and H4: Work-family conflicts of the workers have significant impacts on workers' workplace control.

Data
Data for this study are derived from The 2008 National Study of Changing Workforce (NSCW), which was conducted by the Family and Work Institute.The NSCW is a nationally representative sample of workers across all workplaces in the US.A total of 3502 interviews were completed with a nationwide cross-section of employed adults.Interviews were conducted by using the computer-assisted telephone interviewing (CATI) system.Calls were made to a stratified (by region) un-clustered random probability sample generated by random-digit-dial methods.
Sample eligibility was limited to the workers who (1) worked at a paid job or operated an income-producing business, (2) were 18 years or older, (3) were in the civilian labor force, (4) resided in the contiguous 48 states, and (5) lived in a non-institutional residence (household with a telephone).In households with more than one eligible person, one was randomly selected to be interviewed.Although interviewing began in 2007, 88% of interviews were completed in 2008.Thus, this survey is referred to as the 2008 NSCW.
Of the total 42,000 telephone numbers called, 24,115 were found to be non-residential or non-working numbers and 6970 were determined to be ineligible residences (1389 because no one spoke English or Spanish well enough to be interviewed).Of the remaining telephone numbers, 3547 were determined to represent eligible households, and interviews were completed between November 2007 and April 2008 for 3502 of these-a completion rate of 99%.However, eligibility or ineligibility could not be determined in the remaining 7368 cases.This study focuses on workplace identities of salaried workers accounting for gender and race.The total number of salaried male workers in the sample is 1424 and that of female workers is 1345.Also, there are 2233 white and 505 non-white salaried workers.

Measurement Dependent Variable
Workplace control: This variable includes 5 items.Four of them are "I have the freedom to decide what I do on my job," "It is basically my own responsibility to decide how my job gets done," "I have a lot of say about what happens on my job," "I have the freedom to decide when I take breaks."The responses are strongly disagree (1), somewhat disagree (2), somewhat agree (3), and strongly agree (4).The fifth item is "[In your workplace] are you allowed to choose your own starting and quitting times within some range of hours?"Responses are no (1), and yes (2).Alpha for this variable is 0.70.

Independent Variables: Demographics
Gender is a dummy variable that is based on the question: "Please excuse me, but I have to ask whether you are a man or woman."Here "female" is coded as 1.
Race is also a dummy variable with "white" coded as 1.The variable is measured by the question: "What is your race?" Response categories are: white (1); black or African American (2); native American or Alaskan native (3); Asian, Pacific Islander, or Indian (4); other, including mixed (5).All non-white respondents are grouped together because there are too few from any one category to analyze separately.
Partnered family is also a dummy variable: "Are you presently married, remarried, living with someone as a couple, single and never married, divorced, widowed, or separated?"The first three categories are coded as 1.
Parent to any children is a dummy variable.It is measured by: "Are you the parent or guardian of any child of any age?Please include your own children, stepchildren, adopted children, grandchildren or others for whom you act as a parent."Yes is coded as 1.

Independent Variables: Human Capital
Education is determined by the question: "What is the highest level of schooling you have completed?"The responses are: less than high school (1), high school or GED (2), trade or technical school beyond high school (3), Some college (4), twoyear Associate's degree (5), four/five-year Bachelor's Degree (6), some college after BA or BS but without degree (7), professional degree in medicine, law, dentistry (8), Master's Degree or Doctorate (9).Education is used as a continuous variable.
The variable years worked in the current job is measured by the question: "How long have you worked for your current employer or been involved in your main line of job?"This is an interval-level variable.
Occupation is a dummy variable measured by the openended question: "What kind of work do you do or what is your occupation?"In the dataset there is a variable that has 2 categories of occupation: managerial or professional (1) and others (2).Here "managerial or professional" is coded as 1.
Work-family conflict is an index of 4 items: "In the past three months, how often have you NOT had enough time for your family or other important people in your life because of your job?" "In the past three months, how often have you NOT had the energy to do things with your family or other important people in your life because of your job?" "In the past three months how often has work kept you from doing as good a job at home as you could?" "In the past three months, how often have you NOT been in as good mood as you would like to be at home because of your job?"The responses are: never (1), rarely (2), sometimes (3), often (4), very often (5).The alpha is 0.59.
Satisfaction with income is determined by the question: "How satisfied are you with how much you earn in your main job?"The response categories are: not satisfied at all (1), not too satisfied (2), somewhat satisfied (3), very satisfied (4).
Perceived promotional opportunity is measured by the question: "How would you rate your own chance to advance in your organization?"The responses are: poor (1), fair (2), good (3), excellent (4).This variable is used as a continuous variable.
Job satisfaction is measured by the question: "All in all, how satisfied are you with your job?" Responses are not satisfied at all (1), not too satisfied (2), somewhat satisfied (3), and very satisfied (4).

Independent Variables: Workplace Context
Supportive workplace culture is a scale of 5 items: "There is an unwritten rule at my place of employment that you can't take care of family needs on company time.""At my place of employment, employees who put their family or personal needs ahead of their jobs are not looked on favorably.""If you have a problem managing your work and family responsibilities, the attitude at my place of employment is: "You made your bed, now lie in it!""At my place of employment, employees have to choose between advancing in their jobs or devoting attention to their family or personal lives."Response categories are: strongly agree (1), somewhat agree (2), somewhat disagree (3), and strongly disagree (4).The fifth item is, "At my company or organization where I work, I am treated with respect."Responses are strongly disagree (1), somewhat disagree (2), somewhat agree (3) and, strongly agree (4).The alpha is 0.72.
Supportive supervisor is a scale of 10 items: "My supervisor or manager keeps me informed of the things I need to know to do my job well;" "My supervisor or manager has expectations of my performance on the job that are realistic;" "My supervisor or manager recognizes when I do a good job;" "My supervisor or manager is supportive when I have a work problem;" "My supervisor or manager is fair and doesn't show favoritism in responding to employees' personal or family needs;" "My supervisor or manager accommodates me when I have family or personal business to take care of;" "My supervisor or manager is understanding when I talk about personal or family issues that affect my work;" "I feel comfortable bringing up personal or family issues with my supervisor or manager;" "My supervisor or manager really cares about the effects that work demands have on my personal and family life;" "I consider my supervisor or manager to be a friend both at work and off the job."The responses are: strongly disagree (1), somewhat disagree (2), somewhat agree (3), strongly agree (4).The alpha is 0.90.
Coworkers' support is a scale of 2 items.The questions are: "I have the support from coworkers that I need to do a good job;" and "I have support from coworkers that helps me to manage my work and personal and family life."The responses are: strongly disagree (1), somewhat disagree (2), somewhat agree (3), and strongly agree (4).The alpha is 0.68.

Methods of Analyses
Data analyses for this study are based on quantitative methods.The variability of all the variables was tested by running frequency distributions.Variables with more-or-less normal distributions with acceptable skewness and kurtosis were included.Next, factor analyses were conducted to construct scales for the variables that consist of more than one item.Items with factor loadings greater than 0.50, were included.
First, to examine the impacts of gender and race on the workplace control of workers, we conducted independent sample t-tests.Second, via OLS regression we tested the impacts of work-family conflict and job satisfaction (along with gender and race, and other control variables) on workplace control.We conducted the OLS regression in 3 sets following the examples of Binder et al. (2010), Cohn (2000), and Grönlund (2007).In the first set we introduced the demographics that included gender and race among other variables.In the second set we introduced human capital variables which included work-family conflict and job satisfaction of the workers among other variables.In the third set we introduced workplace context variables.Human capital and workplace context are often considered as important factors that shape workers' control over their workplaces (Binder et al., 2010;Cohn, 2000;Grönlund, 2007).Analyses were conducted by using SPSS 19.

Findings
The t-test results (in Tables 1(a) and (b)) suggest male has significantly more (0.19) workplace control than female.In terms of race, whites are more advantageous than non-whites (0.86; p < 0.000) over workplace control.In order to further explore which set of independent variables has more impact on workplace control, we run progressive regression models in this project (please refer to Table 2).
The base model examines only the relationship between workplace control and demographic characteristics (race, gender, marital status, and number of children).The results show that all of the four factors except number of children have significant impacts on workplace control.Firstly, the coefficient of −0.407 (p < 0.002) indicates that females are much more disadvantaged in workplace control than males.Secondly, being Whites means having more control in workplace (B = 0.793 at p < 0.000).Thirdly, Workers with families tend to have more control (B = 0.414 at p < 0.019) in workplace as well than peowho are single.In the second model, we added human capital variables (including job satisfaction and work-family conflict) in to control to see their impacts on workplace control.In this model, gender and race remained significant in predicting how much control one could have over workplace.Being female means significant much less control over workplace compared to being male (B = −0.669).However, the amount of control for being whites decreased to 0.309 (p < 0.039) after taking human capital vari-ables into consideration.These results suggest that human capital is stronger than race in predicting one's ability ofcontrolling over workplace.However, gender continues to be a strong predictor of workplace control.For every increase in educational level, there is a significant increase of 0.190 (p < 0.000) in workplace control.Having as managerial/professional job gives individuals much greater control (B = 0.925 at p < 0.000) over workplace than others.This result indicates that prestigious job allows workers more freedom and control in their workplaces.For every increase in work-family conflict, there is a significant decrease in workplace control (B = −0.051).Satisfaction and promotional opportunity are also positively associated with workplace control in a significant way.Last but not the least, the more satisfied one is with one's job, the higher control one could have over his workplace (B = 0.423 at p < 0.001).The last three variables suggest that positive experiences in workplaces give workers greater sense of control over their working environment.
In the third model, we added workplace context into our regression analysis to see if a supportive working environment would help with one's control over workplace.The results show that having a supportive workplace culture and a supportive supervisor have positive significant impacts on workers' control over workplace.Nevertheless, being female continued to be significantly disadvantaged with regards to workplace control (B = −0.712).This result indicates that gender is a strong predicator of one's workplace control and it is independent of all other factors.This supports our H1.Race became insignificant in model 3, which suggests that workplace context plays a more important role than race in individuals' control over workplace.This partly supports our H2.Our H3 is also supported because workers' job satisfaction remained a powerful predictor of their workplace control.Initially (in model 2), work-family conflict was a significant predictor of workplace control.However, its impact was reduced with the introduction of workplace context variables (in model 3).This suggests that worker's work-family conflict reduces when others support them in their workplaces.This result partly supports our H4.

Conclusion
With the intention to study the impacts of gender, race, jobsatisfaction, and work-family conflicts on American workers' workplace control, we asked three questions in this project: 1) Do women and men workers in America differ in their perceptions of workplace control?2) Do non-white and white workers in America differ in their perceptions of workplace control?And 3) Do gender and race of the workers influence their workplace control when job satisfaction and work-family conflict are considered?Our findings show that women workers perceive a lesser workplace control than their male colleagues, net of their work-family conflicts and job satisfaction.Again, non-white workers have less workplace control than whites.
But, the effect of race diminishes when human capital enters the equation.Work-family conflict significantly reduces the workplace control of the workers unless others in the workplace support them.In addition, job satisfaction remarkably increases the workplace control of the American workers.
Our study resonates with the literature that states that gender has a significant influence on workers' workplace control, and most of the time that influence is negative (McDowell et al., 2001;Blau & Devaro, 2007).With this data, we showed that women have less workplace control than men even when we control for workers' human capital and workplace contexts.Thus, it is possible that women are not completely empowered in the 21st century American workplaces.However, the situation is not that bad as far as racial relations are concerned.Our research suggests that racial differences decrease when we control for workers' human capital, and they disappear when we control for workers' workplace contexts.Thus, these findings do not confirm the empirical studies that suggest thatrace negatively impacts workers' workplace control (Skinner, 2002;Robinson, 2009).This study also shows that work-family conflict reduces workplace control of workers, more for women.But its effect vanishes when supportive workplace context enters the equation.Thus, we can say that having supportive workplace culture and supportive supervisor helps women workers to reduce their workload considerably, which in turn reduces their work-family conflict, and consequently, enhances their workplace control.Most importantly, our findings suggest that job satisfaction is a vital predictor of women workers' workplace control.This result also confirms the literature that posits that job satisfaction of the workers improves their workplace control (Bethge et al., 2009).
Workplace control is one of the key components of workers' well being (Robinson, 2009).Thus, we intend to inform the policy makers about the importance of women workers' job satisfaction.Our research shows that job satisfaction of the women workers considerably improves their workplace control.Thus by enhancing their job satisfaction, employers also contribute towards their work related well being.While gender and race continue to impact workers' workplace control, but we argue that a lot of gender and race related issues can be reduced if employers provide a supportive workplace culture to diminish the work-family conflict of the women workers.At the same time they should also try to take care of the job satisfaction of the women workers to increase their workplace control.We believe that taking positive measures towards women's work-family conflict and job satisfaction, gender related social change in the workplaces is inevitable.

Table 1 .
(a) Independent sample t-test comparing workplace control of women and men workers; (b) Independent sample t-test comparing workplace control of white and non-white workers.
Note: N is the total number of cases; Numbers in parentheses are standard deviations; **** Significance at p < 0.001.

Table 2 .
Un-standardized coefficients from the regression using workplace identity as dependent variable and demographics, self-perceived competence and reflected appraisals as independent variables.N = total number of cases; Numbers within parentheses are standard deviations; **** Significance at p < 0.001; ** Significance at p < 0.05.