Examining the Link Between Parole Officers’ Bases of Power and Their Exercise of Power

Abstract

Parole officers are responsible for supervising offenders conditionally released from prisons into communities. This structural arrangement creates a power relationship, with officers’ views providing the foundation for various bases of power and possibly influencing their exercise of discretionary power. This study is an examination of parole officers’ perceptions of their bases of power and whether those perceptions influenced officers’ use of their power to revoke offenders’ parole. Findings revealed that officers identified legitimate and reward power as the primary means by which they gain compliance; however, only legitimate power and expert power were linked to officers’ use of power.

Keywords

parole power revocation offender

Offenders released from prisons on parole are released conditionally; the offenders’ liberty is contingent on their compliance with certain rules during a period of supervision in the community. Parole officers are responsible for the supervision of offenders released from prisons (Glaser, 1969; Petersilia, 2003), This structural arrangement creates a power relationship between parole officers and paroled offenders in which parole officers employ various mechanisms to influence and control the behavior of the offenders under their supervision. Parole officers may counsel offenders, refer them to treatment services, search their residences for drugs and other illegal contraband, or sanction them for violations of their release conditions (Petersilia, 2003). However, the ultimate control mechanism parole officers have at their disposal is their discretion to pursue revocation hearings that often result in the return of offenders to prison (Lynch, 1998).¹ Although parole officers’ discretionary power to revoke offenders’ parole is bounded in part by legal and bureaucratic constraints (McCleary, 1978; Simon, 1993), researchers have found that parole officers still have considerable influence over whether offenders are revoked (Lynch, 1998; Simon, 1993; Steiner, Makarios, Travis, & Meade, 2011). Very few studies, however, have examined whether differences between officers influence their revocation decisions.² Even though the discretion to revoke offenders is constant across officers, there may be variation between officers that coincides with officers’ exercise of power.

How officers view their bases of power may influence the use of their discretionary power. The relationship between parole officers and offenders under their supervision provides the basis for various sources of power, and officers may view this relationship differently, which could lead them to mobilize alternative bases of power. Since individuals’ bases of power are resources that are mobilized to attain compliance (Hepburn, 1985; Smith, Applegate, Sitren, & Springer, 2009; Stojkovic, 1986), it seems reasonable to expect that parole officers’ perceptions regarding their individual bases of power may affect the rate at which they exercise their discretionary power over offenders. To date, however, this potential relationship has not been examined. Using data collected from a sample of parole officers in Ohio, we investigate the relative effects of officers’ bases of power on the use of their discretionary power to revoke offenders’ parole.

Power Relations, Bases of Power, and Parole Officers’ Exercise of Power

The study of power relations has a rich tradition that has contributed to various conceptual definitions of power (see Tedeschi, Bonoma, Schlenker, & Lindskold, 1970, for a review of this literature). Power has been conceived as one dimensional, purely coercive. Others view power as the result of exchange relationships (Stojkovic, Kalinich, & Klofas, 2012). Weber (1968), for instance, has defined power as the probability that an individual within a social relationship will be able to impose his or her will, even in the face of resistance, and regardless of the basis on which the probability rests. Criminologists have typically defined power as one person’s ability to influence the behavior of another (e.g., Hepburn, 1985; Stichman, 2002; Stojkovic, 1984). The basis of power refers to the relationship between the power holder and the power recipient that is the source of the power (French & Raven, 1959). Hepburn (1985) observed that it is the base of power rather than the definition of power that is relevant to understanding how justice system actors exercise their power. Power can be analyzed from the point of view of either the power holder or the power recipient (French & Raven, 1959), and researchers have examined both perspectives (e.g., Hepburn, 1985; Smith et al., 2009; Stojkovic, 1986). Power can also be examined at the individual or organizational level (Stojkovic et al., 2012). In this study, we focus on the perspective of individual parole officers, the power holders in an exchange relationship.

French and Raven (1959) articulated five distinct power bases that characterize interactions between power holders and power recipients—reward power, coercive power, legitimate power, referent power, and expert power. Reward power refers to the ability to give positive incentives or withhold negative consequences. Coercive power is the authority to impose punishment for noncompliance. Legitimate power emanates from an internalized respect for the power holder’s structural position. Referent power stems from the respect for and identification with the individual who is the power holder. Finally, expert power is rooted in the unique knowledge, skills, or expertise held by the power exerciser.³

Prior research on power relations in the justice system has typically focused on identifying the bases of power that power recipients or power holders deem most important for attaining compliance (e.g., Hepburn, 1985; Stichman, 2002; Stojkovic, 1984, 1986). As far as we are aware, only two studies have linked criminal justice actors’ bases of power to behavioral outcomes: Tifft’s (1975) ethnographic study of police supervisors’ exercise of power across different units in a city police department and Smith et al.’s (2009) study of the relationship between offenders’ perceptions of county probation officers’ bases of power and their compliance with probation. Both studies revealed that the power holders’ bases of power had relatively weak effects on the outcomes examined. We build on these studies by examining the relationship between parole officers’ bases of power and their use of the discretionary power to revoke offenders’ parole. We also extend the limited research on criminal justice actors’ bases of power to the context of parole. As discussed above, parole officers’ discretionary capacity to revoke offenders is the ultimate control mechanism that officers have within their power (Lynch, 1998), and it may be that officers’ views regarding their bases of power affect the rate at which they exercise this power.

Parole officers who identify reward power as a base of their power may be less likely to revoke offenders. Parole officers typically have the discretion to issue rewards (e.g., travel permits, early release) to the offenders under their supervision and the capacity to withhold sanctions. Yet, the capacity to use reward power does not ensure that officers will mobilize this base of their power. Reward power has typically been ranked low in studies of justice system actors’ bases of power (e.g., Hepburn, 1985; Stichman, 2002), and Smith et al. (2009) found that only half of the offenders in their sample agreed that probation officers could reward good behavior. In addition, reward power was not associated with offenders’ compliance (Smith et al., 2009). Still, parole officers who do identify reward power as a reason why offenders comply may be more likely to use rewards to attain compliance; punishment would only be used as a last resort.

Parole is, in part, designed to be coercive, and so parole officers who recognize coercive power as a base of their power might be more likely to revoke offenders under their supervision. Evidence from studies of power recipients and power holders in the justice system, however, has been mixed regarding the efficacy of coercive power (Hepburn, 1985; Stichman, 2002; Stojkovic, 1984, 1986). Smith et al. (2009) found that offenders on probation rated coercive power as a strong reason why they complied; however, coercive power was unrelated to offenders’ odds of compliance. Thus, the effects of coercive power remain unclear.

Parole officers who identify legitimate power as a source of their power may be less likely to revoke offenders’ parole. Legitimate power emanates from the structural position of parole officers (French & Raven, 1959; Hepburn, 1985), and the officers’ position provides them with the formal authority to direct offenders’ behavior (Hepburn, 1985; Smith et al., 2009). However, the strength of legitimate power depends on the power recipients’ perception of the power holder as legitimate. Use of power that is outside the range of power that is viewed as legitimate may weaken power recipients’ perceptions of power holders as legitimate (French & Raven, 1959). Officers who identify legitimate power as a source of their power are likely to be aware of the tenuous nature of this relationship, and so they may be cautious regarding the frequent exercise of their power to punish noncompliance. Hepburn (1985) found that prison guards identified legitimate power as a reason inmates comply. Stichman (2002) and Smith et al. (2009) observed similar findings among inmates and offenders on probation, respectively. However, Smith et al. (2009) did not observe a relationship between legitimate power and compliance.

For justice system actors, referent power may be difficult to conceive. Individuals come to their attention in difficult situations and generally not because they desire to do so. Thus, attaining respect and admiration from power recipients (e.g., defendants, inmates) may seem unlikely. Indeed, Hepburn (1985) and Stojkovic (1986) found that referent power was not identified frequently among correctional staff. Findings from studies of the perceptions of power recipients (e.g., inmates), however, have revealed referent power is a base of power than can be effective in attaining compliance (Smith et al., 2009; Stichman, 2002). Smith et al. (2009) also observed a bivariate relationship between referent power and offender compliance, although this relationship did not persist in their multivariate analyses. It could be that those offenders who view their supervising officers with admiration are more likely to comply with their directives out of a desire to please them. If officers perceive this possibility and identify referent power as a source of their power, then they may be less likely to punish offenders because they wish to maintain this relationship (French & Raven, 1959).

Officers who rely on expert power may revoke offenders less frequently. Expert power is based in part on the knowledge or skill of the power holder (French & Raven, 1959; Hepburn, 1985). Officers who use expert power may believe that they have the skills and expertise to manage offenders in the community. It could also be that officers who identify expert power may view revocation as failure on their part, as well as the offenders, and so they may only revoke offenders after all other options are exhausted. Hepburn (1985) found that many correctional staff identified expert power as a source of their power; however, Stojkovic (1986) did not. Smith et al. (2009) found that expert power was linked to compliance among offenders on probation.

Other Potential Effects on Officers’ Exercise of Power

A reliable examination of the effects of officers’ bases of power on the exercise of their power to revoke offenders requires consideration of other variables that may confound the potential bases of power–exercise of power relationship. For reasons discussed below, factors that may be relevant could include officers’ background characteristics and their caseload characteristics.⁴

Officers’ Background Characteristics

Researchers of justice system actors’ attitudes and behaviors have observed that the social background of actors may affect their professional and socialization experiences; these processes can in turn affect actors’ attitudes and behaviors (e.g., Kautt & Spohn, 2007; Myers, 1988). For instance, scholars have argued that age could be positively related to justice system actors’ sanction decisions because actors who are older may be more conservative or hold more traditional views (Kautt & Spohn, 2007; Myers, 1988). Researchers have also posited that justice system actors who are female or racial/ethnic minorities may be less likely to sanction offenders because they are often more liberal and/or sympathetic to the individuals they encounter during the course of their jobs (Gruhl, Spohn, & Welch, 1981; Johnson, 2006). Gruhl et al. (1981), for example, suggested that females are socialized to be sympathetic and more understanding than men. Similarly, minorities often view themselves as more liberal than Whites, and liberal views are often supportive of individuals who are poor or minorities (such as many offenders on parole) (Welch, Combs, & Gruhl, 1988). Findings across studies in which researchers have examined the effects of justice system actors’ demographic characteristics on sanction or social control decisions, however, have been mixed (see, e.g., Dembo, 1972; Gruhl et al., 1981; Johnson, 2006; Kautt & Spohn, 2007; Mastrofski, Snipes, Parks, & Maxwell, 2000; Myers, 1988; Steffensmeier & Britt, 2001; Steffensmeier & Herbert, 1999; Welch et al., 1988).

In addition to their demographic background characteristics, officers’ employment characteristics (education, length of service, and rank) may also affect their attitudes and behaviors. For instance, researchers have suggested that justice system actors who have a higher level of education may be less enforcement oriented (Anderson & Spanier, 1980; Reese, Curtis, & Whitworth, 1988; Terrill & Mastrofski, 2002). Similarly, parole officers who are less experienced may be more willing to revoke offenders’ parole because newly hired individuals often embrace the enforcement-oriented functions of their job (Petersilia, 2003; Steffensmeier & Hebert, 1999; Terrill & Mastrofski, 2002). Supervisors may also be less likely to revoke offenders’ parole because of their responsibilities to support all of the agencies’ objectives (Engel & Worden, 2003; Farrell, Young, & Taxman, 2011; McCleary, 1978; Rudes, 2012). Findings regarding the effects of these characteristics on sanction or social control decisions, however, have been inconsistent across studies (see, e.g., Anderson & Spanier, 1980; Dembo, 1972; Farrell et al., 2011; Johnson, 2006; Mastrofski et al., 2000; Mastrofski & Ritti, 1996; McCleary, 1978; Reese et al., 1988; Steffensmeier & Hebert, 1999; Terrill & Mastrofski, 2002; Welch et al., 1988).

Caseload Characteristics

The type of offender on parole officers’ caseloads can influence the rate at which officers exercise their power. For instance, some officers may supervise more high-risk offenders, who would inevitably provide more opportunities for officers to revoke their parole (Committee on Community Supervision and Desistance from Crime, 2008; Petersilia, 2003). Simon (1993) has discussed how modern parole supervision has shifted to a managerial approach, where parole officers recognize offenders only by their level of risk, rather than the individual differences between offenders. In this managerial model, parole officers simply attempt to manage and control high-risk offenders until they are ultimately returned to prison. However, Simon (1993) and Lynch (1998) have also observed that some officers resist these managerial strategies and attempt to supervise offenders according to their individual differences and needs. For these officers, it may be the offenders who they perceive as more likely to reoffend that are subjected to greater surveillance and control, rather than those offenders who are labeled high risk by an assessment tool. Regardless of the approach, officers who supervise greater levels of offenders classified as high risk or offenders they perceive as high risk may be more likely to revoke offenders because these types of offenders are predicted to (or perceived to) engage in more violation behaviors that motivate a response (Petersilia, 2003; Simon, 1993).

The size of officers’ caseloads may also be relevant. Larger caseload will often contribute to a greater number of violations that go undetected (Committee on Community Supervision and Desistance from Crime, 2008; Petersilia & Turner, 1993; Piehl & LoBuglio, 2005). So, officers who supervise larger caseloads may revoke fewer of their offenders or revoke offenders less frequently simply because they are responsible for managing more offenders (Petersilia, 2003; Petersilia & Turner, 1993; Simon, 1993).

Officers who manage sex offender caseloads may also be more likely to revoke offenders’ parole. Specialized caseloads such as sex offender caseloads have become institutionalized in most large parole agencies because they are viewed as a direct response to offender types that are considered problematic for more general supervision approaches (Farrell et al., 2011; Latessa & Allen, 2003; see also Crank & Langworthy, 1992 for their application of this idea to police organizations). The specialized caseload legitimizes the unique treatment of the relevant offender type, and officers who manage these caseloads often embrace this position and view their assignment as a legitimate, essential function of the parole organization. Since sex offenders are generally perceived as a high risk to public safety (see, e.g., Huebner & Bynum, 2006), it seems logical to expect that officers who supervise sex offenders may revoke offenders’ parole more frequently.

The location in which the offenders on officers’ caseloads reside may also be relevant (Farrell et al., 2011; Olson, Weisheit, & Ellsworth, 2001; Petersilia, 2003). For example, officers who supervise offenders in more urban areas may be prone to revoke more offenders because the offenders they supervise are often higher risk (Ellsworth & Weisheit, 1997). Offenders located in urban areas are often subject to stricter supervision than offenders in rural areas (Ellsworth & Weisheit, 1997; Olson et al., 2001). Subjecting offenders to stricter supervision could influence officers’ detection of violations and ultimately those officers’ level of response (see, more generally, Committee on Community Supervision and Desistance from Crime, 2008; Petersilia & Turner, 1993). For instance, Olson et al. (2001) observed that, compared to offenders on probation in rural areas, offenders on probation in urban areas were subject to more urinalyses, were arrested more frequently, and were revoked more often. Researchers of other justice system actors’ behavior have uncovered differences between the styles and sanctioning patterns of the actors who work in rural areas versus those actors who work in urban areas (see, e.g., Crank, 1990; Johnson, 2006; Liederbach & Travis, 2008; Ulmer & Johnson, 2004).

Parole Context in Ohio

The state of Ohio has been a determinate sentencing state since 1996. Although the implementation of sentencing guidelines abolished discretionary parole release, the guidelines still provide for post release control (PRC) supervision for those offenders who would have previously received parole and discretionary PRC placement for nonviolent offenders. The Ohio Department of Rehabilitation and Correction (ODRC) is responsible for supervising all adult felony offenders in the state of Ohio. The Ohio Adult Parole Authority (APA), which is contained within the ODRC, is responsible for the release and supervision of adult felony inmates returning to communities from prison. The APA is organized into seven regional offices, each of which contains multiple supervision units. In addition to a parole services supervisor, each supervision unit contains approximately 5 to 10 APA parole officers.

APA parole officers are responsible for aiding offenders in their transition to the community (e.g., making treatment and employment referrals) as well as monitoring and enforcing the conditions of their release (e.g., collecting urinalysis). If offenders do not comply with their conditions of their release, officers may pursue violation hearings that can result in offenders return to prison. Ultimately, decisions to revoke or return offenders to prison are made by APA hearing officers during violation hearings; however, as discussed above, the majority of all hearings generally result in revocation. It is presumed that officers’ decisions to pursue revocation hearings are the key decision in Ohio’s revocation process in Ohio (Martin & Van Dine, 2008; Steiner et al., 2011). It is important to note, that APA officers rarely pursue violation hearings exclusively for new felony offenses. The reason is that Ohio criminal court judges have the option of imposing a sentencing enhancement of 12 months (or the remainder of the offender’s period of PRC) when an offender on PRC is convicted of new felony. ODRC policy requires APA officers to notify the relevant prosecutors’ office when an offender who is on PRC has been charged with a new felony offense. Based on ODRC administrative records for the time frame of the study, less than 10% of all revocations were pursued exclusively for new felony offenses.

Method

The study described here focused on determining the relative effects of parole officers’ bases of power on the use of their discretionary power to revoke offenders’ parole. The data were collected as part of a larger project designed to examine officers’ perceptions of a violation response policy implemented by the APA (see Martin & Van Dine, 2008).⁵

Sample and Data

The target population for this study included all of the APA officers who were responsible for the supervision of offenders released from prisons under supervision in Ohio (N = 452), a subsample of the sampling frame for the larger project; all the parole services supervisors and lower-level officers employed by the APA. Each officer was sent a survey through the mail, and follow-up mailings were sent in general accordance with mixed mode procedures described by Dillman, Smyth, and Christian (2008). Each officer in the sampling frame was sent a survey and cover letter. The initial mailing was followed with an electronic mail reminder. Approximately 3 to 4 weeks after the initial mailing, a replacement survey and reminder letter was sent to each of the nonrespondents. Another electronic mail reminder was sent to nonresponding officers about 2 weeks after the replacement survey. Finally, a fourth and final follow-up letter and replacement survey were mailed to those officers who were identified by the ODRC as either a parole services supervisor or a caseload carrying officer at the time the sampling frame was generated. As such, this group of officers received four follow-up reminders, as opposed to the three reminders sent to the entire sampling frame. These procedures resulted in 372 returned surveys of the 452 officers who were in the target population for this study, a response rate of 82%. The sample was further reduced to 366 officers because six surveys were determined to be unusable due to substantial missing data (usable response rate = 81%). Despite the nonresponses and missing data, the final sample was generalizable to the target population; univariate hypothesis tests revealed that the sample statistics for the variables age, race, gender, rank, and years employed by the ODRC were not significantly different from the corresponding population parameters generated from official data provided by the ODRC.

Measures

All of the measures included in the final models are described in Table 1. The outcome measure, revocation rate, was measured with a question that asked officers how many revocation hearings they had pursued in the past month. Rates were computed by standardizing responses by the number of offenders on each officer’s caseload. Supervisors answered the question regarding revocation hearings for their entire unit, and a rate was computed by dividing their response by the number of offenders their unit supervised. For reasons discussed above, revocations pursued by parole officers in Ohio were primarily for technical violations and not new felony offenses.

Table 1.

Sample Means (Standard Deviations)

	$\bar{x}$	s
Outcome
Revocation rate	.02	(.03)
Background characteristics
Age	39.20	(7.79)
Female	.39	(.49)
African American	.16	(.37)
Education > bachelor’s degree	.25	(.43)
Years in current position	6.58	(4.58)
Parole services supervisor	.13	(.33)
Caseload characteristics
Caseload size	77.16	(24.27)
Proportion caseload high risk	.19	(.18)
Proportion caseload will commit new felony	.27	(.18)
Supervises sex offenders	.24	(.43)
Most of caseload lives in urban area	.57	(.50)
Most of caseload lives in suburban area	.10	(.30)
Bases of Power
Reward	.54	(.50)
Coercive	.26	(.44)
Legitimate	.73	(.44)
Referent	.38	(.49)
Expert	.37	(.48)

Note: Ratio scales include revocation rate, age, years in current position, caseload size, proportion caseload high risk, proportion caseload will commit new felony. All other measures are dummy coded. Officer-level N = 366, Unit-level N = 63.

The predictor variables included measures of officers’ bases of power, background characteristics, and caseload characteristics. The primary independent variables of interest were officers’ bases of power, which were each measured by inquiring about officers’ level of agreement with the following survey questions:

Offenders typically do what I ask them to because I can give them special help or benefits (reward power).

Offenders typically do what I ask them to because they fear sanctions (coercive power).

Offenders typically do what I ask them to because they believe I have the authority to tell them what to do (legitimate power).

Offenders typically do what I ask them to because they want my respect (referent power).

Offenders typically do what I ask them to because they think I know what is best for them (expert power).

The questions are similar to those that have been used in prior studies of justice system actors’ bases of power (e.g., Hepburn, 1985; Smith et al., 2009; Stichman, 2002), although each question was modified for the parole context. For the multivariate analyses described below, the ordinal scales were collapsed into dichotomous indicators of parole officers’ bases of power (agree/disagree) because equal distances between the categories of the ordinal scales could not be assumed.

Regarding the control variables, measures of officers’ background characteristics included age, sex (female), race (African American), education > bachelor’s degree, number of years in current position, and whether officers were a parole services supervisor. Characteristics of officers’ caseloads included proportion caseload high risk, proportion caseload will commit new felony, caseload size, supervises sex offenders, most of caseload lives in urban area, most of caseload lives in suburban area.⁶

Statistical Analysis

Prior studies of justice system actors have uncovered important variation in the attitudes and behaviors of actors across different organizations, departments, counties, and so forth (Clear & Latessa, 1993; Crank, 1990; Engel & Worden, 2003; Farrell et al., 2011; Johnson, 2006; Liederbach & Travis, 2008; Ulmer & Johnson, 2004), and so bi-level models were estimated (with officers at Level-1 and supervision units at Level-2) that controlled for cross-unit differences in the outcome examined. Specifically, the bi-level modeling strategy used here adjusted for the correlated error among officers nested within the same supervision units (i.e., officers in the same unit are not truly “independent” of each other). The technique also removed (through group mean-centering) between-unit variation in officer characteristics that might have corresponded with differences in the outcomes across units (Raudenbush & Bryk, 2002). Although bi-level models were estimated to address the issues discussed above, it is important to note that the models that are ultimately displayed here only include measures at the first level of analysis (officers). The decision to only model officer-level effects was based on the focus on the bases of power–use of power relationships and because we did not collect any measures of unit-level processes.

The continuous outcome measure was examined with hierarchical linear regression using HLM 6.08 (Raudenbush, Bryk, Cheong, Congdon, & du Toit, 2004). The analyses proceeded in two stages. First, an unconditional model (with no predictors) revealed the variance estimates in revocation rate at Level-1 (among officers within supervision units) and Level-2 (between supervision units). Next, the Level-1 predictors were added to the model. Due to the limited numbers of officers within units, however, we could not examine the random effects of the officer-level measures across units (i.e., these effects were “fixed” or averaged across all units). Although there is no rule of thumb per se regarding how many cases per aggregate are required for reliable random effects models to be estimated, the supervision units examined here only contained, on average, approximately six officers per unit, which is considerably below the number generally needed for reliable estimation of random effects (Raudenbush & Bryk, 2002). Even though the data restricted our ability to take advantage of a unique aspect of multilevel modeling, we could still allow the Level-1 model intercepts to vary randomly across supervision units. Allowing the intercepts to vary across supervision units partitioned the variance in the outcome into variance to be explained within and between units and rendered the officer-level analyses independent of cross-unit differences in the outcome examined. All of the officer-level measures were also centered on the means for each unit to reduce the odds of finding spurious Level-1 effects due to unmeasured unit-level effects that might also be related to compositional differences in officer populations across APA supervision units. Prior to estimation of the final models, the measures were examined for multicollinearity. Multicollinearity was not determined to be a problem here.

Findings

The officers’ responses to the survey items tapping their perceptions regarding the bases of their power are contained in Table 2. Over 70% of the officers agreed that legitimate power was a source of their power, and over 50% of the officers agreed that reward power was a source of their power. Nearly 40% of the officers identified referent and expert power, whereas just over 25% agreed that coercive power was a source of their power.

Table 2.

Description of Parole Officers’ Perceptions of Their Bases of Power

Base of Power
Offenders do what I ask them to because:	SD	D	A	SA
Reward power	.08	.38	.50	.04
I can give them special help or benefits.
Coercive power	.31	.43	.24	.03
They fear sanctions.
Legitimate power	.04	.23	.64	.09
They believe I have the authority to tell them what to do.
Referent power	.15	.47	.35	.03
They want my respect.
Expert power	.10	.53	.34	.04
They think I know what is best for them.
N	366

Note: SD = strongly disagree, D = disagree, A = agree, SA = strongly agree.

The results of the analysis of officers’ use of their power to revoke offenders’ parole are contained in Table 3. Before discussing the specific findings, it is worth noting that 12% of the variation in revocation rates was between supervision units. This suggests that there were unmeasured unit-level influences that affected officers’ use of their power to revoke parole, although these effects were parceled out of the officer-level analysis reported in Table 3.

Table 3.

Officer-Level Effects on Officers’ Exercise of Discretionary Power

	Revocation rate
Intercept	.02
Background characteristics
Age	.0004*
	(.0002)
Female	−.003
	(.003)
African American	.005
	(.004)
Education > bachelor’s degree	−.003
	(.003)
Years in current position	−.001
	(.0004)
Parole services supervisor	−.02*
	(.005)
Caseload characteristics
Caseload size	−.0003*
	(.0001)
Proportion caseload high risk	−.001
	(.01)
Proportion caseload will commit new felony	.02*
	(.01)
Supervises sex offenders	.005
	(.005)
Most of caseload lives in urban area	.01*
	(.004)
Most of caseload lives in suburban area	.02*
	(.01)
Bases of power
Reward	.0001
	(.002)
Coercive	−.003
	(.003)
Legitimate	−.01*
	(.003)
Referent	.005
	(.004)
Expert	−.01*
	(.002)
N	366
Proportion variation within supervision units	.88
Proportion variation within supervision units explained	.12
Proportion variation between supervision units	.12

Note: Unstandardized coefficients reported (with standard errors in parentheses); Level-1 effects “fixed” across supervision units.

p < .05.

Table 3 shows that officers who identified expert power or legitimate power as the reason why offenders comply were less likely to revoke offenders’ parole. Neither coercive power, reward power, nor referent power was related to officers’ use of their discretionary power. Older officers and officers who supervised more offenders who they perceived would commit a new felony were more likely to revoke offenders. Similarly, officers who supervised more offenders who lived in an urban area or more offenders who lived in a suburban area also revoked offenders more frequently. Supervisors, and officers who supervised larger caseloads, were less likely to revoke offenders under their supervision. Altogether, the significant predictors in the model accounted for 12% of the variation in hearing rates.

Discussion and Conclusions

The conditional release of prison inmates on parole establishes a power relationship between the offenders under supervision and the parole officers charged with supervising them in the community. Parole officers use a variety of methods to influence the behavior of the offenders under their supervision, including their discretionary power to revoke parole (Glaser, 1969; Petersilia, 2003; Steiner, Makarios, Travis, & Meade, 2012). This study involved an examination of whether differences between officers influenced the rate at which officers exercise this power. Specifically, we investigated how Ohio parole officers perceived their bases of power and whether officers’ perceptions of their bases of power affected their use of their discretionary power to revoke offenders’ parole.

The findings from this study suggest that most officers view legitimate power and reward power as the reasons why offenders comply with their directions, whereas the fewest number of officers identified coercive power as a source of offender compliance. Hepburn (1985) found that correctional officers also identified legitimate power as a primary source of their power, whereas coercive power was also recognized as one of the weakest sources of officers’ power. In contrast to the findings from this study, however, Hepburn (1985) observed that reward power was infrequently identified among correctional officers. The differences between studies can be attributed to the different populations examined. Parole officers are in a position to legitimately reward offenders under their supervision (e.g., extended curfew). Correctional officers, in contrast, are typically not in a position to legitimately reward the inmates they are charged with supervising (see, e.g., Hepburn, 1985; Marquart, 1986).

Interestingly, the results from this study of power holders (parole officers) were only partially consistent with findings from studies of power recipients. In their study of offenders on probation, Smith et al. (2009) observed that offenders perceived that probation officers had legitimate power; however, they generally did not agree that officers had reward power. Moreover, the offenders seemed to agree that probation officers had expert and coercive power. Still, only expert power was linked to offenders’ odds of compliance with probation (Smith et al., 2009). There are clearly differences between probation and parole; however, the divergence in the findings from this study of power holders and the only study of power recipients within the context of community corrections suggests further research is needed to clarify the differences between the perceptions of power holders and power recipients. It may also be worthwhile to examine the perceptions of power holders and power recipients within the same jurisdiction to control for potential differences that can be attributed to variation between jurisdictions.

The analyses of officers’ use of power revealed that officers who identified legitimate and/or expert power as a source of their power were less likely to revoke offenders’ parole. Legitimate power is based on the structural position of parole officers, which provides them with the formal authority to regulate offenders’ behavior (French & Raven, 1959; Hepburn, 1985; Smith et al., 2009). It may seem that failure to comply with the directives of officers who believe they have the inherent right to direct offender behavior would result in revocation. However, officers who recognize legitimate power are presumably aware that the strength of legitimate power depends on the power recipients’ perception of the power holder as legitimate. Since any use of power that is outside the range of power viewed as legitimate may weaken the legitimate power of the power holder (French & Raven, 1959), officers who identified legitimate power may have been less willing to exercise their power to punish noncompliance as frequently as officers who identified other sources of their power.

Regarding expert power, officers who identified this source of power may have revoked offenders less often because they believe that they have the skills and expertise to manage offenders’ noncompliance in the community. For these officers, the use of revocation may be viewed as failure on their part, as well as the offenders, and so these officers could have been more willing to exhaust all other options prior to revoking offenders’ parole.

Officers’ perceptions concerning coercive, reward, or referent power had no effect on their willingness to exercise their power to revoke parole. Regarding coercive and reward power, although some officers identified with these sources of power, their use of these types of power may have been restricted by bureaucratic constraints. McCleary (1978) observed that officers’ intentions as well as the use of their discretionary power are often controlled by bureaucratic processes. Tifft (1975) also found that police supervisors’ exercise of power was constrained by structural and environmental conditions. Perhaps more so than other forms of power, the mobilization of coercive power or reward power requires officers to engage in behaviors that might be subject to oversight. As such, officers’ actual initiation of these types of power may be the most likely to be constrained. In particular, the officers under study here had recently undergone a policy change that was designed to structure officers’ responses to offender noncompliance (see Martin & Van Dine, 2008; Steiner et al., 2011; Steiner, Travis, & Makarios, 2011). In the wake of the implementation of this policy, officers who relied of the use of coercion, rewards, or their discretion to withhold sanctions (reward power) may simply have felt restricted from using their power.

Referent power stems from respect and admiration for the power holder. Even though power recipients have identified this base of power as a reason they comply (see, e.g., Smith et al., 2009), and a sizeable minority of officers in this study recognized it as a source of their power, it may simply be that parole officers cannot mobilize this base of their power. Parole officers supervise large numbers of offenders, who turn over quickly (Petersilia, 2003; Simon, 1993). This situation leaves officers with little time to establish rapport with the offenders under their supervision. Therefore, although officers recognize its potential, it is possible that referent power is rarely mobilized in the context of modern day parole supervision.

The findings and preceding discussion regarding the bases of power–exercise of power relationships must all be tempered by an alternative explanation of the study results. Officers’ decisions to revoke parole may not be due to differences between officers, but instead due to differences in the behavior of the offenders’ under their supervision. Under this perspective, revocation rates would reflect offenders’ noncompliance, and officers who identified legitimate and expert power as the reason offenders comply were simply more effective in achieving offender compliance. Officers who identified coercive power, referent power, or reward power were not effective in achieving offender compliance, although they were not ineffective either. Although this explanation of the study findings is plausible, there are at least two reasons why we chose to frame the study and interpret the study findings within the existing research on justice system actors’ use of their discretionary power. First, for reasons discussed above, the majority of parole revocations in Ohio during the time of the study were for technical violations, as opposed to new offenses. Responses to technical violations have historically been left primarily to the discretion of probation/parole officers (see, e.g., Glaser, 1969; Petersilia, 2003; Simon, 1993). Thus, we followed from the existing research and viewed revocation decisions as discretionary, keeping in mind that officers have alternative mechanisms at their disposal for responding to noncompliance (e.g., community-based sanctions; Steiner et al., 2012). Second, we assumed that once the effects of the composition of officers’ caseloads were controlled (e.g., proportion high risk), offender behavior was relatively constant across officers’ caseloads. Thus, only the responses to offenders’ noncompliance might vary across officers. Although these justifications are potentially subject to some criticism, we think that they are more likely to reflect reality than the alternative which would require the assumption that officers respond to technical violations in a uniform manner.

In addition to the findings regarding officers’ bases of power, the analyses also revealed several relationships between the control variables and officers’ use of power. Consistent with our expectations, older officers were more likely to exercise their power to revoke offenders’ parole. In support of the idea that supervisors are responsible to all of the agencies’ objectives (Engel & Worden, 2003; McCleary, 1978), parole supervisors were less likely to revoke parole. The other background characteristics of officers, however, did not affect their use of power. It may simply be that officers’ revocation decisions are influenced more by agency policy, caseload characteristics, and to some extent their perceptions regarding their base of power (see, e.g., Clear & Latessa, 1993).

The size and characteristics of officers’ caseloads did affect their use of power. In contrast to Simon’s (1993) observations regarding actuarial parole supervision, the proportion of officers’ caseloads that were classified high risk had no effect on officer willingness to revoke offenders’ parole. Instead, officers were more likely to exercise their power to revoke offenders’ parole if they perceived that a higher proportion of their caseload would commit a new felony. Similarly, officers were more likely to revoke offenders’ parole if the majority of their caseload resided in a suburban or urban area. These findings are consistent with those derived from studies of other justice system actors’ sanctioning decisions (e.g., Crank, 1990; Johnson, 2006; Liederbach & Travis, 2008; Olson et al., 2001; Ulmer & Johnson, 2004). Finally, officers with larger caseloads were less likely to revoke parole, perhaps because they are simply responsible for managing more offenders, which may result in a greater number of violations that go undetected (Petersilia, 2003; Petersilia & Turner, 1993; Simon, 1993). Altogether, the influence of these structural factors surrounding officers caseloads appear to have a strong influence on officers’ supervision practices, perhaps even more so than their perceptions regarding their bases of power.

Finally, there are several limitations that should be kept in mind when interpreting the study findings. First, the study was limited to one state, and so the findings may not be generalizable to other jurisdictions. Second, the outcome variable was based on officers’ self-reported behaviors. Self-report measures have been found to contain systematic error resulting from underreporting by some groups and over reporting from others. Although self-report data have been found to be a valid indicator of behavior (Hindelang, Hirschi, & Weis, 1981), it may be worthwhile for future researchers to examine the influence of officers’ bases of power on official indicators of their behaviors. Third, the use of single measures of officers’ bases of power may be unreliable (Stichman, 2002), and so future studies may want to replicate the analyses performed here using indices comprised of multiple measures of officers’ bases of power. Finally, the data used here were initially collected for another purpose, and therefore may be subject to limitations imposed by design of the larger study.

The limitations of the study aside, the findings here point to the potential relevance of officers’ bases of power for influencing their exercise of power. Future studies may want to continue to examine these linkages, and consider the effects of officers’ bases of power on other outcomes. For instance, officers who identify reward power may be more likely to recommend early release. Officers who identify legitimate power may be more likely perform their duties in a procedurally fair manner. Officers’ recognition of expert power may be related to a greater use of referrals and so forth. Future studies should also examine the relevance of parole officers’ sources of power from the perspective of power recipients (offenders under parole supervision). It is only after a number of studies of power holders and power recipients have been carried out across the components of the justice system, that a better understanding of the power dynamics within the system can be realized.

Footnotes

Declaration of Conflicting Interests

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

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was indirectly supported by award number 2005–IJ–CX–0038 from the National Institute of Justice, Principal Investigators: Ohio Department of Rehabilitation and Correction.

Notes

Bios

Benjamin Steiner is an assistant professor in the School of Criminology and Criminal Justice at the University of Nebraska—Omaha. He holds a PhD from the University of Cincinnati.

Rhys Hester is an assistant professor in the Department of Government and Justice Studies at Appalachian State University. His research interests include sentencing, justice actor decision making, and criminal procedure. His recent publications have appeared in the New England Journal on Criminal and Civil Confinement and Policing: An International Journal of Police Strategies & Management.

Matthew D. Makarios is an assistant professor in the Department of Criminal Justice at the University of Wisconsin—Parkside. His recent research has been published in Criminal Justice and Behavior, Crime and Delinquency, and Federal Probation.

Lawrence F. Travis III is a professor of Criminal Justice at the University of Cincinnati. He formerly worked as the research director of the Oregon State Board of Parole and as a research analyst with the National Parole Institutes. He is the author of Introduction to Criminal Justice (Cincinnati: Anderson), and coauthor, with Robert Langworthy, of Policing in America (Englewood Cliffs: Prentice Hall).

References

Anderson

E. A.

Spanier

G. B.

(1980). Treatment of delinquent youth: The influence of the juvenile probation officer’s perceptions of self and work. Criminology, 17, 505-514.

Bacharach

Lawler

(1980). Power and politics in organizations. San Francisco, CA: Jossey-Bass.

Clear

T. R.

Latessa

(1993). Probation officer roles in intensive supervision: Surveillance versus treatment. Justice Quarterly, 10, 441-462.

Committee on Community Supervision and Desistance from Crime.(2008). Parole, Desistance from Crime, and Community Integration. Washington, DC: The National Academies Press.

Crank

J. P.

(1990). The influence of environmental and organizational factors on police style in urban and rural environments. Journal of Research in Crime and Delinquency, 27, 166-189.

Crank

J. P.

Langworthy

(1992). An institutional perspective of policing. Journal of Criminal Law and Criminology, 83, 338-363.

Dembo

(1972). Orientation and activities of the parole officer. Criminology, 10, 193-215.

Dillman

Smyth

Christian

L. M.

(2008). Internet, mail, and mixed mode surveys: The tailored design method. Hoboken, NJ: John Wiley.

Ellsworth

Weisheit

R. A.

(1997). The supervision and treatment of offenders on probation: Understanding rural and urban differences. Prison Journal, 77, 209-228.

10.

Engel

Worden

(2003). Police officers’ attitudes, behavior, and supervisory influences: An analysis of problem solving. Criminology, 41, 131-166.

11.

Etzioni

(1961). A comparative analysis of complex organizations: On power, involvement, and their correlates. New York, NY: Free Press.

12.

Farrell

J. L.

Young

D. W.

Taxman

F. S.

(2011). Effects of organizational factors on use of juvenile supervision practices. Criminal Justice and Behavior, 38, 565-583.

13.

French

Raven

(1959). The bases of social power. In Cartwright

(Ed.), Studies in social power (pp. 259-269). Ann Arbor, MI: University of Michigan.

14.

Glaser

(1969). The effectiveness of a prison and parole system. New York, NY: Bobbs-Merrill.

15.

Gruhl

Spohn

C. C.

Welch

(1981). Women as policymakers: The case of trial judges. American Journal of Political Science, 25, 308-322.

16.

Hepburn

(1985). The exercise of power in coercive organizations: A study of prison guards. Criminology, 23, 145-164.

17.

Hindelang

Hirschi

Weis

(1981). Measuring delinquency. Beverly Hills, CA: Sage.

18.

Huebner

B. M.

Bynum

T. S.

(2006). An analysis of parole decision making using a sample of sex offenders: A focal concerns perspective. Criminology, 44, 961-991.

19.

Johnson

(2006). The multilevel context of criminal sentencing: Integrating judge- and county-level influences. Criminology, 44, 259-298.

20.

Kautt

P. M.

Spohn

C. C.

(2007). Assessing blameworthiness and assigning punishment: Theoretical perspectives on judicial decision making. In Duffee

D. D.

Maguire

E. R.

(Eds.), Criminal justice theory: Explaining the nature and behavior of criminal justice (pp. 155-180). New York, NY: Routledge.

21.

Latessa

Allen

(2003). Corrections in the community. Cincinnati, OH: Anderson.

22.

Liederbach

Travis

L. F.

(2008). Wilson redux: Another look at varieties of police behavior. Police Quarterly, 11, 447-467.

23.

Lynch

(1998). Waste managers? The new penology, crime fighting, and parole agent identity. Law and Society Review, 32, 839-870.

24.

Martin

Van Dine

(2008). Examining the impact of Ohio’s progressive sanction grid. Washington, DC: National Institute of Justice.

25.

Marquart

J. W.

(1986). Prison guards and the use of physical coercion as a mechanism of prisoner control. Criminology, 24, 347-366.

26.

Mastrofski

S. D.

Ritti

R. R.

(1996). Police training and the effects of organization on drunk driving enforcement. Justice Quarterly, 13, 291-320.

27.

Mastrofski

S. D.

Snipes

Parks

Maxwell

(2000). The helping hand of the law: Police control of citizens on request. Criminology, 38, 307-342.

28.

McCleary

(1978). Dangerous men: The sociology of parole. Beverly Hills, CA: Sage.

29.

Molm

L. D.

(1988). The structure and use of power: A comparison of reward and punishment power. Social Psychology Quarterly, 51, 108-122.

30.

Molm

L. D.

(1989). Punishment power: A balancing process in power-dependence relations. American Journal of Sociology, 94, 1392-1418.

31.

Myers

M. A.

(1988). Social background and sentencing behavior of judges. Criminology, 26, 649-675.

32.

Olson

D. E.

Weisheit

R. A.

Ellsworth

(2001). Getting down to business: A comparison of rural and urban probationers, probation sentences, and probation outcomes. Journal of Contemporary Criminal Justice, 17, 4-18.

33.

Petersilia

(2003). When prisoners come home: Parole and prisoner reentry. New York, NY: Oxford University Press.

34.

Petersilia

Turner

(1993). Intensive probation and parole. In Tonry

(Ed.), Crime and justice: A review of the research (Vol. 17, pp. 281-335). Chicago, IL: University of Chicago Press.

35.

Piehl

A. M.

LoBuglio

S. F.

(2005). Does supervision matter? In Travis

Visher

(Eds.), Prisoner reentry and crime in America (pp. 105-138). New York, NY: Cambridge University Press.

36.

Raudenbush

Bryk

(2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage.

37.

Raudenbush

Bryk

Cheong

Y. F.

Congdon

du Toit

(2004). HLM 6: Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International.

38.

Raven

B. H.

(2008). The bases of power and the power/interaction model of interpersonal influence. Analyses of Social Issues and Public Policy, 8, 1-22.

39.

Reese

W. A.

Curtis

R. L.

Whitworth

J. R.

(1988). Dispositional discretion or disparity: The juvenile probation officer’s role in delinquency processing. Journal of Applied Behavioral Science, 24, 81-100.

40.

Rudes

D. S.

(2012). Framing organizational reform: Misalignment and disputes among parole and union middle managers. Law and Policy, 34(1), 1-31.

41.

Simon

(1993). Poor discipline: Parole and the social control of the underclass, 1890-1990. Chicago, IL: University of Chicago Press.

42.

Smith

H. P.

Applegate

B. K.

Sitren

A. H.

Springer

N. F.

(2009). The limits of individual control? Perceived officer power and probationer compliance. Journal of Criminal Justice, 37, 241-247.

43.

Steffensmeier

Britt

(2001). Judges’ race and judicial decision making: Do black judges sentence differently? Social Science Quarterly, 82, 749-764.

44.

Steffensmeier

Herbert

(1999). Women and men policymakers: Does the judge’s gender affect the sentencing of criminal defendants? Social Forces, 77, 1163-1196.

45.

Steiner

Makarios

Travis

Meade

(2011). Short-term effects of sanctioning reform on parole officers’ revocation decisions. Law and Society Review, 45, 371-400.

46.

Steiner

Makarios

Travis

Meade

(2012). Examining the effects of community-based sanctions on offender recidivism. Justice Quarterly, 29(2), 229-257.

47.

Steiner

Travis

Makarios

(2011). Understanding parole officers’ responses to sanctioning reform. Crime and Delinquency, 57, 222-246.

48.

Stichman

(2002). The sources and impact of inmate perceptions of correctional officers’ bases of power. Cincinnati, OH: University of Cincinnati.

49.

Stojkovic

(1984). Social bases of power and control mechanisms among prisoners in a prison organization. Justice Quarterly, 1, 511-528.

50.

Stojkovic

(1986). Social bases of power and control mechanisms among correctional administrators in a prison organization. Journal of Criminal Justice, 14, 157-166.

51.

Stojkovic

Kalinich

Klofas

(2012). Criminal justice organizations: Administration and management (5th ed.). Belmont, CA: Thompson Wadsworth.

52.

Tedeschi

J. T.

Bonoma

T. V.

Schlenker

B. R.

Lindskold

(1970). Power, influence, and behavioral compliance. Law & Society Review, 4, 521-544.

53.

Terrill

Mastrofski

(2002). Situational and officer-based determinants of police coercion. Justice Quarterly, 19, 215-248.

54.

Tifft

L. L.

(1975). Control systems, social bases of power and power exercise in police organizations. Journal of Police Science and Administration, 3, 66-76.

55.

Ulmer

Johnson

(2004). Sentencing in context: A multilevel analysis. Criminology, 42, 137-177.

56.

Weber

(1968). Economy and Society. New York, NY: Bedminster.

57.

Welch

Combs

Gruhl

(1988). Do Black judges make a difference? American Journal of Political Science, 32, 126-136.