Leave Your Job at Work

Abstract

Even though work–family conflict has been linked to negative outcomes for both correctional facilities and staff, little research has been conducted on how workplace factors may be related to it. This study tested nine hypotheses based on the job demand–resource model. Ordinary least squares (OLS) regression analysis of survey data from 160 staff revealed that while most job demand variables (i.e., the role stressors of role conflict, role overload, and perceived dangerousness) predicted work–family conflict, the same did not hold for job resource variables (i.e., the workplace factors of job autonomy, supervision, job variety, instrumental communication, and integration).

Keywords

correctional staff private prison work–family conflict workplace

Although many aspects of the correctional workplace have been examined, there is a pressing need for additional research on some other neglected areas to understand how they affect correctional staff. Work–family conflict is one of these issues. Work and home are the two primary domains for most adults (Byron, 2005). Work–family conflict occurs when one domain spills over into the other, causing conflict for an individual (Michel, Mitchelson, Kotrba, LeBreton, & Baltes, 2009). Work–family conflict is known to be a problem for correctional staff (Lambert & Hogan, 2006). Corrections work is unique because correctional institutions are responsible for holding offenders against their will—unwilling clients who may be maladapted, troublemakers, and hostile (Brough & Williams, 2007). In addition, correctional institutions can be sullen environments where, in addition to the bureaucratic dynamics found in other complex organizations (e.g., staff politics), there exists an oscillation between stimulus overload and deprivation/boredom, as well as the ongoing potential for danger and violence. Thus, working in a correctional institution can lead to problems that can spill over and cause conflict at home.

The small body of correctional literature on work–family conflict has mainly focused on the consequences of work–family conflict, such as its effects on job stress, job burnout, job satisfaction, and organizational commitment. These studies support the conclusion that work–family conflict can have detrimental outcomes for correctional staff and their employing organizations; however, little research has been conducted on how workplace factors are associated as antecedents with work–family conflict among correctional staff (Lambert & Hogan, 2006; Obidoa, Reeves, Warren, Reisine, & Cherniack, 2011). Work–family conflict is bidirectional, which means that problems in each domain can cause conflict in the other (Michel et al., 2009). Work-on-family conflict occurs when problems and issues from work spill over and cause a person conflict at home. Family-on-work conflict occurs when problems at home spill over and cause conflict at work. While not discounting the importance of family-on-work conflict, the focus of this study is on work-on-family conflict because workplace factors should be more likely related to work-on-family conflict, and they are also more controllable by correctional administrators. This study explored the relationship among role stressors (role conflict, role ambiguity, role overload, and perceived dangerousness of the job) and workplace resources (job autonomy, supervision, job variety, instrumental communication, and integration) with work-on-family conflict.

Literature Review

Work–Family Conflict

In an ideal world, home and work domains coexist peacefully; however, the world is not ideal for most working adults, and conflict from spillover can occur (Greenhaus & Beutell, 1985). In the literature, this is referred to as work–family conflict. Two common work-on-family conflict issues are time-based conflicts and strain-based conflicts (Lambert & Hogan, 2006). Time-based conflicts occur when work’s time demands or scheduling interferes with or causes conflict with the home life of a person. Time-based conflicts are common in organizations, including corrections, which require staffing around the clock every day of the year, including holidays. For example, a staff member may not be able to miss work to attend a play or school event of a child or a staff member may be forced to work mandatory overtime. Strain-based conflict occurs when the demands and tensions from work negatively affect the quality of a person’s home life. Strain-based conflict is more likely to occur in correctional institutions than in many other types of organizations because staff must deal with unwilling clients who can be sometimes manipulative and dangerous. While staff may be told that what happens in the correctional institution should remain there when the person goes home, this is not realistic. For example, a correctional staff member who has dealt with an uncooperative and verbally abusive inmate may take his or her frustration and anger home and take it out on family and friends.

Past Correctional Staff Studies

Research suggests that work–family conflict may vary by different demographic characteristics, with female staff reporting higher levels of this type of conflict (Triplett, Mullings, & Scarborough, 1999). In addition, correctional officers have been observed to experience higher levels of work–family conflict than noncustody staff (Lambert & Hogan, 2006). Studies have also linked work–family conflict to salient correctional staff outcomes. Work–family conflict has been shown to be related to higher job stress (Lambert, Hogan, & Griffin, 2007; Triplett et al., 1999), lower job satisfaction (Lambert, Hogan, Camp, & Ventura, 2006), lower organizational commitment (Lambert et al., 2006), increased job burnout (Lambert, Hogan, & Altheimer, 2010), lower support for the rehabilitation of inmates/offenders, as well as greater support for punishment (Lambert & Hogan, 2009). Work–family conflict has also been linked with lower perceptions of distributive justice (fairness in outcomes) and procedural justice (fairness in the decision-making process), as well as lower life satisfaction (Lambert et al., 2007; Lambert, Hogan, Paoline, & Baker, 2005). Finally, work-on-family conflict and family-on-work were positively related to higher levels of depression (Obidoa et al., 2011). There is evidence that work–family conflict has negative outcomes in the field of institutional corrections.

In light of how detrimental work–family conflict is, there has been surprisingly little research exploring the possible antecedents of correctional staff work–family conflict. Only two published studies could be located, which examined the possible antecedents of work-on-family conflict among correctional staff. Perceived dangerousness of the job and role conflict had positive associations with work-on-family conflict, while organizational fairness (a combined measure of distributive and procedural justice) had a negative association (Lambert & Hogan, 2006). In the second study, supervisory support had a negative effect on work-on-family conflict and perceived psychological demands of the job had a positive effect (Obidoa et al., 2011). Although these two studies support the postulation that workplace factors may be linked with work-on-family conflict, two studies are far from conclusive evidence. The present study was undertaken to address the need for more research by surveying staff at a private prison. This study is a replication and expansion of the study by Lambert and Hogan (2006), which studied staff at a state prison. A single study is not sufficient to make persuasive recommendations to correctional administrators. The possible antecedents of work-on-family conflict may be contextual and vary by different types of correctional facilities.

Theoretical Foundation

The job demands–resources model provides a theoretical foundation for understanding how different aspects of the correctional workplace may be linked to work-on-family conflict. This model divides the workplace into two basic areas, demands and resources (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001). Job demands place strain on people and, if nothing is done over time, the chances increase that these demands will result in negative outcomes such as burnout (Karasek & Theorell, 1990). Unbalanced job demands can lead to a poor fit, which results in a wide array of detrimental outcomes for workers (Demerouti et al., 2001). Four common strains (demands) for correctional staff are role conflict, role ambiguity, role overload, and perceived dangerousness of the job, and are often referred to as role stressors (Cullen, Link, Wolfe, & Frank, 1985; Triplett et al., 1999). Resources, on the other hand, are valued by employees and can provide them with positive outcomes over time (Lee & Ashford, 1996). Job autonomy, supervision, job variety, instrumental communication, and integration are salient workplace resources for many staff (Lambert & Hogan, 2006). They can provide more control for staff to avoid strain at work or, if strain arises, to deal with it more effectively.

When role behaviors, duties, and directions are inconsistent with one another, role conflict can occur (Rizzo, House, & Lirtzman, 1970). An example of role conflict is when a staff member is ordered to perform conflicting duties by different superiors, such as counseling versus disciplining a particular prisoner. Role conflict can lead to frustration, and this frustration can follow staff home. Role ambiguity results when there is a lack of clarity on how to perform job duties or what is expected from a person at work (i.e., the job duties are not clearly defined; Rizzo et al., 1970). An example of role ambiguity is when a staff member is told to keep inmates in his or her care under control and quiet without receiving guidance on how to accomplish this. Like role conflict, role ambiguity can be frustrating to staff who wish to do their job correctly and can spill over and cause conflict at home. Role overload occurs when a person is required to handle too many tasks and duties and/or is not provided the necessary resources to accomplish assigned job responsibilities (Triplett et al., 1999). For instance, a staff member may be asked to complete several reports in an unrealistic time frame or supervise numerous inmates, while answering the telephone and radio and conducting cell searches. Having feelings of being overwhelmed at work can wear a person down, which can spill over to one’s home life. Dangerousness of the job is a perception on the part of an employee that the job is threatening to the person’s safety (Cullen et al., 1985). There is the ever-present possibility of being assaulted by an inmate. Feelings of dangerousness can wear on a person and cause psychological strain, which can follow the employee home, again potentially causing conflict.

Workplace resources that may alleviate role stressors are job autonomy, supervision, job variety, instrumental communication, and integration. Job autonomy is the perceived degree of control a person feels he or she has over his or her job duties, such as how the job is done or the order of tasks (Saavedra & Kwun, 2000). As a workplace resource, it can provide a person with a sense of being in control and also of being a valued, trusted employee, which can help deal more effectively with job strains (Agho, Mueller, & Price, 1993). A lack of job autonomy can cause frustration for a person that may spill over to home life. Supervisors are the first line of management, and they provide guidance, direction, control, structure, and support for their subordinates (Brough & Williams, 2007). Quality supervisors are a resource for staff to be effective in their jobs, and being effective can make staff feel better about themselves and their jobs (Lambert et al., 2009). In addition, supervisors can be a resource for staff to deal with potential workplace strains and conflicts. Conversely, a lack of quality supervision can in itself be a stressor for staff, leading to problems at both work and at home. Job variety is the degree of variation in a job; some jobs have a high degree of variety, while other jobs are highly repetitive (Mueller, Boyer, Price, & Iverson, 1994). Job variety offers staff the chance to experience new things, to try new skills, and to grow in their jobs, which can be enriching. Positive feelings may spill over to home, leading to a more positive experience outside of work. Conversely, tedious, repetitive jobs provide little opportunity for mental stimulation and growth, which can psychologically wear a person down over time, causing frustration both at work and home.

Instrumental communication is the “degree to which information about the job is formally transmitted by an organization to its members” (Agho et al., 1993, p. 1009). Clear and relevant communication is a valuable resource for staff to be successful in their jobs. It provides them information so they can be effective not only at their jobs but within the organization, leading to less conflict. Also, it may be a resource that offers staff possible solutions if work–family conflict arises and to deal with conflict in a more effective manner. Lack of communication can hamper employees, keep them in the dark about important organizational changes and issues, and ultimately lead to problems at work that can spill over to home life. Integration refers to creating group cohesion among staff, work groups, departments, and divisions (Mueller et al., 1994). Integration is at one end of the cooperation continuum while competition and dissimilation is at the other end (Lambert & Hogan, 2006). Integration can create a more productive and pleasant working environment, which may lead to positive psychological feelings that may follow staff home. When work groups, departments, sections, and divisions of an organization are pitted against one another, unintended negative outcomes may result, such as an “us versus them” type of workplace. In the end, a lack of integration can lead to frustration for staff, which can spill over causing conflict at home.

Although the job demands–resources theory has been used to examine work–family conflict among police officers (Hall, Dollard, Tuckey, Winefield, & Thompson, 2010), this model has not been applied conceptually to such conflict among correctional staff. Hence, the present study extended the literature by examining hypotheses relevant to job demands and job resources among correctional staff.

Hypotheses

Hypothesis 1: Role conflict was predicted to have a positive association with work-on-family conflict among correctional staff.

Hypothesis 2: Role ambiguity was predicted to have a positive association with work-on-family conflict among correctional staff.

Hypothesis 3: Role overload was predicted to have a positive association with work-on-family conflict among correctional staff.

Hypothesis 4: Perceived dangerousness of the job was predicted to have a positive association with work-on-family conflict among correctional staff.

Hypothesis 5: Job autonomy was predicted to have a negative association with work-on-family conflict among correctional staff.

Hypothesis 6: Supervision was predicted to have a negative association with work-on-family conflict among correctional staff.

Hypothesis 7: Job variety was predicted to have a negative association with work-on-family conflict among correctional staff.

Hypothesis 8: Instrumental communication was predicted to have a negative association with work-on-family conflict among correctional staff.

Hypothesis 9: Integration was predicted to have a negative association with work-on-family conflict among correctional staff.

Method

Participants

All the available staff at a private maximum closed-security prison that housed 450 juvenile inmates adjudicated as adult offenders were provided a survey packet.¹ The packet contained a letter, the survey, a bifurcated raffle ticket, and a return envelope. The cover letter indicated the purpose of the study, that participation was voluntary, and that responses would be anonymous. Completed surveys were returned using a stamped pre-addressed envelope that was mailed to the research team. To encourage staff to participate, a random drawing raffle was held in which several cash awards ranging from US$50 to US$100 were provided. To participate in the raffle, all the staff member had to do was return half of the raffle ticket, regardless of whether or not he or she had completed the survey. To ensure that there was no possibility of linking a particular staff member to a returned survey, the returned raffle tickets were removed from the envelopes and separated from the surveys before the surveys were examined. A total of US$500 was given away at the raffle. Because of various reasons (e.g., sickness, vacation, travel, etc.), only 200 of the 220 total staff complement were provided the survey packet. A total of 160 surveys were completed (i.e., 80% response rate).

The responding staff represented all work areas of the prison except upper administration and included correctional officers, counselors, work supervisors, and so forth. In terms of work position, about 62% of the participants indicated that they were correctional officers and the other 38% worked in other positions, such as unit management, education, medical department, or other areas. Fifty-nine percent of the participants were men. The age of participants ranged from 19 to 68 years, with a median age of 33 years. Tenure at the facility ranged from 1 to 53 months, and a median tenure of 17 months. The reason for the low average tenure was that the private prison had been opened for less than 5 years at the time of survey. Six percent indicated they had a high school diploma or general educational development degree (GED), 47% had some college but no degree, 24% had an associate’s degree, 16% had a bachelor’s degree, and 7% had a graduate or professional degree. Approximately 79% of the participants indicated that they were White, 11% were Black, 2% were Hispanic, 3% were Native American, and 4% were another race. Demographically, the participants appeared to be representative of the entire correctional staff population. Among those who responded, 79% were White, 59% were men, and 62% held a custody position; the median age was 33 years and the median tenure was 17 months. At the time of the survey, institutional records indicated that approximately 81% of the all employees at the private prison were White, 61% were male, 66% held a custody position; the median age range was about 33 years, and the average tenure was about 19 months.

Variables

Dependent variable

Work-on-family conflict was the dependent variable. It was measured using an eight-item index that measured possible conflicts resulting from work that may cause problems at home (i.e., My job keeps me away from my family too much; my time off from work does not really match other family members’ schedules and/or my social needs; work makes me too tired or irritable to fully enjoy my family social life; I frequently argue with my spouse/family members about my job; when I get home from work, I am often too frazzled to participate with family or friends; the uncertainty of my work schedule interferes with my family and/or social life; I find that my job has negatively affected my home life; I often have to miss important family or social activities/events because of my job). The work-on-family conflict items were based on several studies (Bohen & Viveros-Long, 1981; Carlson, Kacmar, & Williams, 2000; Triplett et al., 1999). The items were answered using a 5-point Likert-type scale ranging from strongly disagree (coded 1) to strongly agree (coded 5). The responses to these items were summed together to form an index measuring work-on-family conflict, which had a Cronbach’s alpha value of .86.

Independent variables

The independent variables in this study were the four workplace role stressors of role conflict, role ambiguity, role overload, and perceived dangerousness of the job, and the five workplace resources of job autonomy, supervision, job variety, instrumental communication, and integration.

Role conflict was measured using five items from Ivancevich and Matteson (1980; that is, I regularly receive conflicting requests at work from two or more people; when a problem comes up here, people seldom agree on how it should be handled; sometimes I am criticized by one supervisor for doing something ordered by another supervisor; I sometimes have to bend a rule or policy to get an assignment done; I often receive an assignment with conflicting tasks or guidance to get it done). The five role conflict items were used to form an additive index, which had a Cronbach’s alpha of .73.

Role ambiguity was measured by four items from Cullen et al. (1985; that is, I clearly know what my work responsibilities are [reverse coded]; the rules that we’re supposed to follow seem to be very clear [reverse coded]; I am unclear to whom I report and/or who reports to me; I do not always understand what is expected of me at work). These four items were used to form an additive index of role ambiguity, which had a Cronbach’s alpha of .70.

An index for role overload was created using three items from Ivancevich and Matteson (1980; that is, I am responsible for almost an unmanageable number of assignments and/or inmates; the amount of work required in my job is unreasonable; the amount of work I am required to do seems to be increasing all the time). The responses for the items were summed together to form an index, which had a Cronbach’s alpha value of .77.

Perceived dangerousness of the job was measured using five items from Cullen, Lutze, Link, and Wolfe (1989; that is, Most of the time when I’m at work I don’t feel that I have much to worry about [reverse coded]; in my job, a person stands a good chance of getting hurt; I work at a dangerous job; my job is a lot more dangerous than most other jobs; a lot of people I work with have been physically injured on the job). The items had a Cronbach’s alpha of .81.

Job autonomy was measured with an additive index created using three items from Curry, Wakefield, Price, and Mueller (1986; that is, My job does not allow me much opportunity to make my own decisions [reverse coded]; I have input into matters that affect me at my job; I have a great deal of freedom as to how I do my job). The three items had a Cronbach’s alpha of .69.

Six items based on Teas (1981) were used to measure perceptions of supervisors (i.e., My supervisor looks out for my personal welfare; my supervisor does little to make it pleasant to work here [reverse coded]; my supervisor is friendly and approachable; my supervisor gives me advance notice of changes; my supervisor is very strict about following rules; my supervisor maintains a definite standard of performance for all employees under his or her command). The supervision items were summed to form an index, which had a Cronbach’s alpha value of .81.

Based on five items from Curry et al. (1986) and Mueller et al. (1994), an additive index was created to measure job variety (i.e., I rarely get to do different things on my job [reverse coded]; my job requires that I must constantly learn new things; my job is mainly concerned with routine matters [reverse coded]; my job has a lot of variety in it; my job requires that I be very creative). The five items had a Cronbach’s alpha of .70.

Instrumental communication was measured by the five items adapted from Curry et al. (1986; that is, How well are you informed by prison management about the following aspects of your job: what is to be done; what is most important about the job; how the equipment is used; rules and regulations; what you need to know to do the job correctly). An index was formed by summing the responses to these five items, which had a Cronbach’s alpha of .89.

Integration was measured using four items from Miller and Droge (1986; that is, At this prison, interdepartmental committees are frequently set up to allow for joint decision making; at this prison, task forces are frequently set up to help interdepartmental collaboration on specific projects; at this prison, there is generally bargaining among different department heads rather than competition; at this prison, there is a great deal of departmental interaction on most decisions). The summed integration index had a Cronbach’s alpha value of .85.

Except for instrumental communication, the items were measured using a 5-point Likert-type scale ranging from strongly disagree (coded 1) to strongly agree (coded 5). The items for instrumental communication were answered using a scale of not informed at all (coded 1), informed very little (coded 2), informed somewhat (coded 3), informed (coded 4), and very well informed (coded 5). Except for job autonomy, all the indexes had a Cronbach’s alpha value at or above .70, which indicates good internal reliability. Job autonomy had a Cronbach’s alpha of .69, which is close to the conventional standard of .70.

Control variables

The six demographic characteristics of position, gender, age, tenure, educational level, and race were included more as control variables than predictor variables. Position was measured where those who worked in a custody position (i.e., correctional officer) were coded as 1 and those who worked in noncustody positions were coded as 0. Gender was coded as female = 0 and male = 1. Age was measured in continuous years. Tenure was measured as the number of months the participant had worked at the private prison. Educational level was measured as a dichotomous variable representing whether the respondent had earned a college degree (coded 1) or not (coded 0). Finally race was measured as a dichotomous variable representing if the participant was Nonwhite (coded 0) or White (coded 1).

Results

The descriptive statistics for the variables are presented in Table 1. Statistical tests indicated that the variables were normally distributed with no significant problems of kurtosis or skewness.

Table 1.

Descriptive Statistics.

Variable	Description	Minimum	Maximum	Median	M	SD
Position	38% non-CO (coded 0)	0	1	1	0.62	0.49
	62% CO (coded 1)
Gender	41% female (coded 0)	0	1	1	0.59	0.49
	59% male (coded 1)
Age	Measured in continuous years	19	68	33	35.77	10.82
Tenure	Measured in continuous months employed at the private prison	0	53	17	20.64	13.84
Education	53% had no college degree (coded 0)	0	1	0	0.47	0.50
	47% had a college degree (coded 1)
Race	21% Non-White (coded 0)	0	1	1	0.79	0.40
	79% White (coded 1)
Role conflict	Five-item additive index, α = .73	8	25	15	15.29	3.96
Role Ambig	Four-item additive index, α = .70	4	19	9	9.22	2.86
Role Overload	Three-item additive index, α = .77	3	15	8	8.44	2.67
Danger	Five-item additive index, α = .81	6	25	16	16.01	4.32
Job Auto	Three-item additive index, α = .69	3	15	9	9.12	2.67
Supervision	Six-item additive index, α = .81	6	29	21	19.68	5.38
Job variety	Five-item additive index, α = .70	3	15	9	9.12	2.67
Inst Comm	Five-item additive index, α = .89	7	25	19	17.95	4.07
Integration	Four-item additive index, α = .85	4	20	11	10.60	3.42
Work–family	Eight-item additive index, α = .86	10	39	21	22.78	6.71

Note. Min = minimum value; Max = maximum value; CO = correctional officer; Education = educational level; Role Ambig = role ambiguity; Danger = perceived dangerousness of the job; Job Auto = job autonomy; Inst Comm = instrumental communication; Work–family = work-on-family conflict; α = Cronbach’s alpha. The number of cases was 160.

The Pearson correlations for the variables are presented in Table 2. Among the control demographic variables, position (i.e., being a correctional officer or not) and age had statistically significant correlations with the work-on-family conflict variable. In general, correctional officers reported higher levels of work-on-family conflict than staff who worked in other positions. Older staff in general reported lower work-on-family conflict. Gender, tenure, educational level, and race each had nonsignificant correlations with the dependent variable. Role conflict, role ambiguity, role overload, and perceived dangerousness of the job each had a significant positive correlation with work-on-family conflict. Higher levels of reported job demands were associated with increased work-on-family conflict. Job autonomy, supervision, job variety, instrumental communication, and integration all had significant negative correlations with the dependent variable. Higher levels among the job resource variables were associated with decreased reported work-on-family conflict. The size of the correlations ranged from small to moderate, with all the workplace variables having equal or larger correlations than the demographic variables.

Table 2.

Pearson Correlations.

Variables	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.	11.	12.	13.	14.	15.
1. Position	1.00
2. Gender	.28**	1.00
3. Age	−.53**	−.17*	1.00
4. Tenure	−.20**	.00	.07	1.00
5. Educ	−.29**	−.13	.16*	.12	1.00
6. Race	−.02	.08	−.05	.06	.08	1.00
7. Role Conflict	.35**	.08	−.26**	−.01	−.08	.04	1.00
8. Role Ambig	.14**	−.06	−.11	.02	.02	−.03	.63**	1.00
9. Role Overload	.09	.07	−.04	.06	.03	.01	.48**	.51**	1.00
10. Danger	.61**	.20	−.39**	.09	−.16*	−.14	.44**	.34**	.33**	1.00
11. Job Auto	−.36**	−.03	.22**	−.05	.08	.06	−.59**	−.55**	−.48**	−.44**	1.00
12. Supervision	−.27**	.08	.24**	−.06	−.01	.09	−.64**	−.59**	−.40**	−.31**	.59**	1.00
13. Job Variety	−.28**	.15*	.15	.09	.12	.10	−.39**	−.38**	−.15*	−.28**	.49**	.45**	1.00
14. Inst Comm	−.14	.05	.12	−.11	−.02	.07	−.65**	−.69**	−.46**	−.29**	.60**	.62**	.40**	1.00
15. Integration	−.25**	−.03	.18*	−.02	.04	.01	−.48**	−.36**	−.26**	−.35**	.54**	.37**	.36**	.46**	1.00
16. Work–family	.28**	.04	−.31**	.07	−.08	−.07	.52**	.45**	.46**	.47**	−.45**	−.41**	−.32**	−.40**	−.32**

Note. See Table 1 for a description of the variables and how they were measured. Educ = educational level; Role Ambig = role ambiguity; Danger = perceived dangerousness of the job; Job Auto = job autonomy; Inst Comm = instrumental communication; Work–family = work-on-family conflict. The number of cases was 160.

p ≤ .05. **p ≤ .01.

An ordinary least squares (OLS) regression equation was estimated with work-on-family conflict as the dependent variable. The independent/control variables were the demographic characteristics (i.e., position, gender, age, tenure, educational level, and race) and the workplace indexes (i.e., role conflict, role ambiguity, role overload, perceived dangerousness of the job, job autonomy, supervision, job variety, instrumental communication, and integration) were the independent/predictor variables of interest. The results are reported in Table 3.

Table 3.

OLS Regression Results With Work-on-Family Conflict as the Dependent Variable.

Variable	B	β	Tolerance	VIF
Position (0 = noncustody staff; 1 =correctional officer)	−1.11	−.08	.40	2.52
Gender (0 = female; 1= male)	−0.34	−.02	.82	1.21
Age (in years)	−0.12	−.19*	.69	1.45
Tenure (months at facility)	0.03	.05	.82	1.22
Educational level (0 = no college degree; 1= college degree)	−0.33	−.02	.88	1.14
Race (0 = Non-White; 1 = White)	−0.90	−.06	.90	1.12
Role conflict	0.38	.22*	.37	2.70
Role ambiguity	0.23	.10	.40	2.50
Role overload	0.61	.24**	.60	1.67
Perceived dangerousness of the job	0.30	.20*	.44	2.25
Job autonomy	−0.09	−.04	.40	2.48
Supervision	0.01	.01	.43	2.32
Job variety	−0.19	−.10	.63	1.59
Instrumental communication	0.12	.08	.37	2.70
Integration	−0.03	−.02	.64	1.56
R ²	.43**	F = 7.19**
Adjusted R²	.37**

Note. See Table 1 for a description of the variables and how they are coded. OLS = ordinary least squares; B = unstandardized coefficient; β = standardized coefficient; Tolerance = tolerance value; VIF = variance inflation factor score. The number of cases was 160.

p ≤ .05. **p ≤ .01.

Based on the Variance Inflation Factor (VIF) scores and the Tolerance values (see Table 3), which did not meet or exceed the threshold limits of VIF scores greater than 5 or 6 and Tolerance values lesser than .20, multicollinearity was not a problem (Tabachnick & Fidell, 1996). The R² value for the regression equation was .43 (adjusted R² was .37), which means that the independent variables explained approximately 37% to 43% of the observed variance of the work-on-family conflict variable.

Among the independent variables, role ambiguity, job autonomy, supervision, job variety, instrumental communication, and integration all had nonsignificant relationships with the dependent variable in the regression analysis (i.e., Hypotheses 2, 5, 6, 7, 8, and 9 were not supported). Of the control variables, only age was significant and had a negative relationship with work-on-family conflict (i.e., older staff were less likely to report work-family conflict). Role conflict, role overload, and perceived dangerousness of the job each had a significant positive association with work-on-family conflict (i.e., Hypotheses 1, 3, and 4 were supported). Higher levels of these variables were linked with higher levels of work-on-family conflict. By examining the standardized regression coefficients (i.e., the values in the β column in Table 3), the size of the impact can be estimated. The significant variables were similar in their effects, with role overload having the largest standardized regression coefficient and age having the smallest.

Discussion and Conclusion

Despite evidence that work–family conflict is associated with negative outcomes for correctional employees and their organizations (Lambert, Hogan & Altheimer, 2010; Triplett et al., 1999), there is a dearth of research examining the antecedents of work-on-family conflict among these staff. The present study was undertaken to address this void, and findings revealed that work-on-family conflict was significantly greater among staff who were younger and who reported higher levels of role conflict, role overload, and perceived job dangerousness of the job.

In contrast with previous findings (Lambert & Hogan, 2006), the data did not show that correctional officers experienced significantly greater work–family conflict than noncustody staff, and neither education nor tenure was predictive of such conflict. The only significant demographic variable was age. A possible explanation is that younger staff are more likely to have younger children at home, and this could present more opportunities for conflict between the work and family domains. In addition, in many cases younger staff are newer to their careers in corrections and likely to have accumulated less experience at effectively balancing the work–family domains and preempting or coping with possible conflicts. Alternatively, younger staff may have more active social lives or they may have less desirable posts and assignments at the prison, such as working the graveyard shift.

Consistent with the only previously published article of antecedents of work–family conflict among correctional staff (Lambert & Hogan, 2006), which was a study of public prison employees, the present research using a sample of private prison staff found that both perceived job dangerousness and role conflict were positively associated with conflict spillover from work to home. In the current study, another significant predictor was role overload. In fact, all job demands variables studied, except role ambiguity, were significantly related to work-on-family conflict. It makes sense that staff who experience conflicting job expectations, those who feel overwhelmed by work demands, and those who perceive their jobs as dangerous are most likely to report conflict spilling over from work to home. Being unclear about one’s job duties and expectations may be a comparatively mild role stressor, one less likely to promote work–family conflict once the effects of the others are taken into account.

The present study also contributed to the literature by being the first to apply the job demands–resources theory to conceptualize work–family conflict among prison staff. The findings revealed that three of four job demands variables were directly predictive of work-on-family conflict; none of the five job resources variables were significant predictors, however. This suggests role stressors may do much to promote work-family conflict directly. At the same time, the workplace may present few resources to mitigate work-on-family conflict directly.

Job resources should not be dismissed as unimportant. This study was not designed to investigate the possible indirect relationship of these resources to work–family conflict through the medium of other variables, and resources have been shown to directly predict other relevant variables such as job satisfaction (Lambert, Hogan, & Barton, 2002). In fact, the effect of job resources on work–family conflict could be mediated through job satisfaction. Staff who perceive their jobs as presenting greater resources might like their jobs better and thus be less likely to experience negative spillover to their families. Another possibility is that job resource effects are mediated by job demands. For instance, a staff member may experience some job autonomy and have a favorable perception of his or her supervisor, but the effects of these variables may be overshadowed by more proximal conflicts, inducing effects of role conflict, role overload, and perceptions of job dangerousness. Indeed the magnitude of work-on-family conflict might have been even greater in the absence of indirect curtailing effects from job resources. This interpretation is consistent with the finding that the significant negative bivariate relationship of each job resources variable to work-on-family did not hold up in the multivariate analysis when combined with job demands variables.

The insignificance of job resources variables in this study should not be interpreted as being inconsistent with the job demands–resources model. Bakker and Demerouti (2007) argued that job demands variables are likely to promote job strain under conditions where job resources are deficient. The logic is that, if present in sufficient degree, job resources can buffer strain-inducing job demands. Although Bakker and Demerouti (2007) posited an interaction between demands and resources, they also make clear the core assertion of the model that whereas “job demands predict job strain . . . job resources predict motivation . . .” (p. 321, emphasis added). Given that motivation is not synonymous with the mere absence of job strain, it can be argued that the present study did not measure the construct job resources are posited to predict directly.

As with all studies, the current research had shortcomings. The results are based on the survey responses of staff at a single prison. Staff at other correctional institutions need to be surveyed to determine whether the results of the current study can be replicated. A handful of studies are far from conclusive. Without further research, how different aspects of the workplace are associated with work-on-family conflict among correctional staff will remain unclear. Moreover, the staff in this study worked for a private Midwestern prison that incarcerated juvenile offenders sentenced as adults. It could be the relationships between workplace factors and work-on-family conflict may be situational and contextual such that relationships vary by different types of facilities (e.g., jails, prisons, public, private, adult, juvenile, etc.), or by different regions of the country.

The independent variables in this study accounted for 37% to 43% of the variance in the work-on-family conflict measure. This means that more than half of the variance in work-on-family conflict was the result of other independent variables. These other variables need to be identified and tested in future research. For example, role underload is a role stressor that should be included in future studies. Role underload is the perception of being underutilized at work (Triplett et al., 1999). Role underload could lead to feelings of boredom and frustration, of not being able to be fully engaged at work, which could spill over to home, leading to work-on-family conflict. Another area that could be explored is how different forms of social support (e.g., supervisory, peer, administrative, family, friends, etc.) are related to both work-on-family conflict and family-on-work conflict. Bakker and Demerouti (2007) conceptualized social support as important in buffering job demands.

As this study was cross-sectional, it is not possible to show a causal relationship between the role stressors and work-on-family conflict. Longitudinal research is needed to determine the causal direction of the associations. Future research may wish to use more complex measures of the latent workplace variables. In this study, role overload, job autonomy, and job variety were measured using three item indexes. More in-depth measures of these (and other) workplace variables should be used in future research. In addition, the Cronbach’s alpha value of .69 for the job autonomy index was low. Furthermore, the effects of workplace variables on work-on-family conflict could vary across different groups of correctional staff, such as by gender, age, tenure, position, and so forth. With the sample size of 160 cases, it was not possible to explore how the effects of the selected workplace variables on work–family conflict may vary across subgroups of staff. Having a larger number of participants would allow for a greater exploration of the nuances of the relationships between various workplace factors and work–family conflict.

Future researchers may also wish to differentiate time- from strain-based conflicts, as these could be linked with separate workplace factors. For example, role overload could promote time-based conflict, whereas role conflict or perceptions of dangerousness could be more strongly associated with strain-based conflict. A more complex conceptual expansion is implied from the bidirectionality of work–family conflict. This expansion involves juxtaposing the distinction between job demands and resources with an analogous distinction between family demands and resources. Such a model would allow interactions to be examined between the respective sets of demands and resources, crisscrossing work and family life domains, in the study of both work-on-family and family-on-work conflicts.

There is clearly a need for more research on identifying and understanding the factors that may lead to or increase the level of work-on-family conflict among correctional staff. This is critical information if scholars and administrators are to develop strategies to deal effectively with work–family conflict. For instance, if future research replicates the present findings, an implication for reducing work-on-family conflict would be that administrators should focus on reducing stressors (e.g., role conflict, role overload, and perceived dangerousness of the job) among their staff. Concomitantly, administrators should seek ways to improve job resources in ways that allow employees to draw directly on these to reduce the spillover of job demands into family life. Although the ideal arrangement might be to achieve demand–resource commensurability, such that deleterious effects of demands are mitigated by resources, very limited fiscal resources and a traditionally top-down decision structure are engrained realities of many correctional workplaces. The parameters such realities place (or are perceived as placing) on job resource development suggest the importance of administrators attending also to the development of employees’ personal resources (Xanthopoulou, Bakker, Demerouti, & Schaufeli, 2007). For example, during basic and/or in-service training, attention could be given to instilling awareness among staff of the particular job demands substantiated by research as most likely to promote conflict at home, and staff could then be trained to develop personal resources (e.g., organizational self-esteem, stress-coping skills) to prepare themselves to preempt and mitigate the negative effects of job on home.

In closing, correctional staff are a valuable resource for correctional facilities. They are responsible for numerous tasks and duties necessary for the operation of a safe, secure, and humane correctional facility. The workplace can also affect staff as well. One potential negative impact is work–family conflict. Problems at work can spill over and cause problems at home for correctional staff. Likewise, problems at home can spill over and result in conflict at work for staff. There needs to be far greater research on work–family conflict, particularly on how workplace variables may be linked with work–family conflict. The authors hope that this study will spark future research and interest by correctional administrators and scholars in this area.

Footnotes

Acknowledgements

The authors thank Janet Lambert for editing and proofreading the article. The authors also thank the editor and anonymous reviewers for their comments and suggestions.

Authors’ Note

All the authors contributed equally to the study and are listed in random order.

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) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Eric G. Lambert, PhD, is professor and chair in the Department of Legal Studies, University of Mississippi. His research interests include organizational issues, job, and organizational effects on the attitudes, intentions, and behaviors of criminal justice employees, and international perceptions, attitudes, and views on criminal justice issues.

Kevin I. Minor, PhD, is a professor in the School of Justice Studies, Eastern Kentucky University. His areas of professional interest are institutional and community corrections, theories of crime and punishment, juvenile delinquency/justice, and evaluation and applied research.

James B. Wells, PhD, is a professor in the School of Justice Studies, Eastern Kentucky University. A specialist in evaluation and survey research, he is the principal investigator and project director in a training evaluation study being conducted for the National Institute of Corrections.

Nancy L. Hogan, PhD, is a faculty member in the School of Criminal Justice and Graduate Program Coordinator, Ferris State University. She conducts research on job satisfaction and organizational issues of correctional staff, use of force, and cognitive behavioral treatment for inmates.

References

Agho

Mueller

Price

(1993). Determinants of employee job satisfaction: An empirical test of a causal model. Human Relations, 46, 1007-1027.

Bakker

A. B.

Demerouti

(2007). The job demands-resources model: State of the art. Journal of Managerial Psychology, 22, 309-328.

Bohen

Viveros-Long

(1981). Balancing jobs and family life: Do flexible work schedules help? Philadelphia, PA: Temple University Press.

Brough

Williams

(2007). Managing occupational stress in a high-risk industry: Measuring the job demands of correctional officers. Criminal Justice and Behavior, 34, 555-567.

Byron

(2005). A meta-analytic review of work-family conflict and its antecedents. Journal of Vocational Behavior, 67, 169-198.

Carlson

Kacmar

Williams

(2000). Construction and initial validation of a multidimensional measure of work-family conflict. Journal of Vocational Behavior, 56, 249-276.

Cullen

Link

Wolfe

Frank

(1985). The social dimensions of correctional officer stress. Justice Quarterly, 2, 505-533.

Cullen

Lutze

Link

Wolfe

(1989). The correctional orientation of prison guards: Do officers support rehabilitation? Federal Probation, 53, 33-42.

Curry

Wakefield

Price

Mueller

(1986). On the causal ordering of job satisfaction and organizational commitment. Academy of Management Journal, 29, 847-858.

10.

Demerouti

Bakker

Nachreiner

Schaufeli

(2001). The jobs demands-resources model of burnout. Journal of Applied Psychology, 86, 499-512.

11.

Greenhaus

Beutell

(1985). Sources of conflict between work and family roles. Academy of Management Review, 10, 76-88.

12.

Hall

Dollard

Tuckey

Winefield

Thompson

(2010). Job demands, work-family conflict, and emotional exhaustion in police officers: A longitudinal test of competing theories. Journal of Occupational and Organizational Psychology, 83, 237-250.

13.

Ivancevich

Matteson

(1980). Stress and work: A managerial perspective. Glenview, IL: Scott Foresman.

14.

Karasek

Theorell

(1990). Healthy work: Stress, productivity, and the reconstruction of working life. New York, NY: Basic Books.

15.

Lambert

Altheimer

Hogan

(2010). An exploratory examination of a gendered model of the effects of role stressors. Women & Criminal Justice, 20, 193-217.

16.

Lambert

Hogan

(2006). Correlates of correctional staff work-on-family conflict. Professional Issues in Criminal Justice, 1(2), 17-34.

17.

Lambert

Hogan

(2009). Exploring the predictors of treatment views of correctional staff: A test of an integrated work model. Journal of Offender Rehabilitation, 48, 504-528.

18.

Lambert

Hogan

Altheimer

(2010). The association between work-family conflict and job burnout among correctional staff: A preliminary study. American Journal of Criminal Justice, 35, 37-55.

19.

Lambert

Hogan

Barton

(2002). Satisfied correctional staff: A review of the literature on the antecedents and consequences of correctional staff job satisfaction. Criminal Justice and Behavior, 29, 115-143.

20.

Lambert

Hogan

Camp

Ventura

(2006). The impact of work-family conflict on correctional staff: A preliminary study among private prison staff. Criminology & Criminal Justice: An International Journal, 6, 371-387.

21.

Lambert

Hogan

Griffin

(2007). The impact of distributive and procedural justice on correctional staff job stress, job satisfaction, and organizational commitment. Journal of Criminal Justice, 35, 644-656.

22.

Lambert

Hogan

Moore

Tucker

Jenkins

Stevenson

Jiang

(2009). The impact of the work environment on prison staff: The issue of consideration, structure, job variety, and training. American Journal of Criminal Justice, 34, 166-180.

23.

Lambert

Hogan

Paoline

Baker

(2005). The good life: The impact of job satisfaction and occupational stressors on correctional life satisfaction. An exploratory study. Journal of Crime & Justice, 28, 1-26.

24.

Lee

Ashford

(1996). A meta-analytic examination of the correlates of the three dimensions of job burnout. Journal of Applied Psychology, 81, 123-133.

25.

Michel

Mitchelson

Kotrba

LeBreton

Baltes

(2009). A comparative test of work-family conflict models and critical examination of work-family linkages. Journal of Vocational Behavior, 74, 199-218.

26.

Miller

Droge

(1986). Psychological and traditional determinants of structure. Administrative Science Quarterly, 17, 333-357.

27.

Mueller

Boyer

Price

Iverson

(1994). Employee attachment and noncoercive conditions of work. Work and Occupations, 21, 179-212.

28.

Obidoa

Reeves

Warren

Reisine

Cherniack

(2011). Depression and work family conflict among corrections officers. Journal of Occupational and Environmental Medicine, 53, 1294-1301.

29.

Rizzo

House

Lirtzman

(1970). Role conflict and ambiguity in complex organizations. Administrative Science Quarterly, 15, 150-163.

30.

Saavedra

Kwun

(2000). Affective states in job characteristics theory. Journal of Organizational Behavior, 21, 131-146.

31.

Tabachnick

Fidell

(1996). Using multivariate statistics (3rd ed.). New York, NY: HarperCollins.

32.

Teas

(1981). A test of a model of department store salespeople’s job satisfaction. Journal of Retailing, 57, 3-35.

33.

Triplett

Mullings

Scarborough

(1999). Examining the effect of work-home conflict on work-related stress among correctional officers. Journal of Criminal Justice, 27, 371-384.

34.

Xanthopoulou

Bakker

Demerouti

Schaufeli

W. B.

(2007). The role of personal resources in the job demands-resources model. International Journal of Stress Management, 14, 121-141.