Legal and Extralegal Factors Associated with Success on Misdemeanor Probation ()
1. Introduction
The high incarceration rate in the United States is the subject of national and international scrutiny, however, it is probationers that make up the largest share of the nation’s correctional population (Maruschak & Minton, 2020). In 2018, there were 3.5 million Americans on probation, compared to the 2.1 million people incarcerated in prisons or jails that same year. Furthermore, while probation and parole are often combined under the banner of “community supervision”, there are four times as many people on probation than on parole (Maruschak & Minton, 2020). Recent data indicate that about 1 in 72 American adults is on probation (Kaeble & Alper, 2020). While these numbers are staggering, they represent a notable decline from a decade ago. However, the demographic profile of probationers has remained virtually unchanged. According to the Bureau of Justice Statistics, 78% of probationers were male in 2000, as opposed to 75% in 2016. Additionally, in 2000, 54% of probationers were white, 31% were African American, and 13% were Latino. By 2016, those numbers had only shifted to 55%, 28%, and 14%, respectively (Kaeble, 2018). Men, who make up half of Americans, are vastly overrepresented in these figures. The proportion of Black probationers also deserves attention, considering only 13.4% of the population identifies as African American (United States Census Bureau, 2020).
Considering the scale of the probation system, the lack of a large body of scholarship focused on probation outcomes is surprising. Questions about the nature and effectiveness of community probation have been raised both nationally and internationally (Villeneuve et al., 2021; Yang, 2020). Since the 1960s, probation has been marketed as a cost-effective, rehabilitation-oriented sentencing alternative. Yet, probation may not truly be an ideal alternative if people are frequently unsuccessful in complying with the terms of their supervision—resulting in financial costs, prolonged contact with the criminal justice system, and stints of incarceration (Phelps, 2013). Nationwide, only 54% of people who exited their probation in 2018 were marked as successfully completing their supervision. Of the remaining individuals with alternative outcomes, over half returned to prison or jail (Kaeble & Alper, 2020). The policies adopted by each jurisdiction significantly impact whether probation serves as a “net-widener” that draws more people into imprisonment or a genuine form of diversion that reduces mass incarceration (Phelps, 2018: p. 53). As such, it is critical to determine the factors that contribute to individual successes and failures so that supervision procedures can be designed to promote success. There is some existing literature to guide our understanding of probation outcomes, particularly related to race, gender, socioeconomic indicators, and case variables. However, a full understanding of factors related to success on probation is wanting. There is a particular dearth of research specific to misdemeanor probation, despite its potential negative impact through disruption of employment, bans on employment and assistance opportunities, and discrimination by employers and others (Phelps, 2020). This study sought to address this gap and examine both demographic (sex, race, education, employment, age) and probation (time on probation, offense category, first offender status, fines assigned and unpaid fines) factors associated with successful misdemeanor probation completion in one county in a southern state.
2. Literature Review
2.1. Factors Associated with Differential Probation Outcomes
Generally, the most common lines of inquiry in the existing literature revolve around the effects of race and gender on probation outcomes. In 2000, Olson and Lurigio studied predicting factors for rearrest, revocation, and technical violations among a sample of over 2400 felony and misdemeanor probationers in the Midwest. The authors were mainly concerned with measuring negative case outcomes, which they defined as 1) new arrests, 2) technical violations, and 3) probation revocation (Olson & Lurigio, 2000). Multivariate logistic regression showed that age, income, prior convictions, a history of drug abuse, and geographic setting—urban versus rural—were significant predictors of all three negative outcomes. Prior convictions were a particularly strong predictor; one previous conviction doubled the chances that the individual would have their probation revoked or be rearrested. While not related to probation revocation, race was a predicting factor for new arrests and technical violations (Olson & Lurigio, 2000).
Steinmetz and Henderson (2016) addressed similar questions in a longitudinal study examining outcomes in over 115,000 probation cases over ten years. Probation outcomes were divided into expiration (standard completion), early discharge, revocation, and adjudication (Steinmetz & Henderson, 2016). In an initial multinomial model, property offenders were more likely to face revocation than person offenders, and younger offenders were more likely to experience failure overall. When interaction effects between race and gender were incorporated, the authors found that being a Black male or a Hispanic male was a significant predictor of probation failure (Steinmetz & Henderson, 2016). After incorporating offense type, Hispanic males convicted of property offenses were, surprisingly, more likely to be discharged early from their probation. On the other hand, Black male property offenders were less likely to be granted early discharge (Steinmetz & Henderson, 2016). The authors suggest that these findings might be directly related to the “socially constructed perception of African American men as dangerous” (Steinmetz & Henderson, 2016: p. 14).
Some academic debate remains regarding the specific influence of sex on probation outcomes. Few researchers have centered their work around female probationers, but one 2015 study investigated how parenting and connection to intimate partners influenced probation compliance among 257 women (Stalans & Lurigio, 2015). The authors examined marital status, the criminal and substance abuse history of intimate partners, if children were in the home or state custody, and housing stability. Instead of defining probation failure exclusively in terms of subsequent criminal behavior, Stalans and Lurigio (2015: p. 159) focused on “behavioral measures of non-compliance”, including missed office visits and treatment sessions. Women were more likely to miss treatment sessions if they had nonconforming intimate partners, children in foster care, or unstable housing situations. Still, women with children—regardless of whether they were in foster care—were 62% less likely to miss office appointments than those without children. The number of missed office appointments was, in turn, significantly related to an increased chance of new arrests for drug possession, property crimes, and misdemeanors (Stalans & Lurigio, 2015).
While much of the literature has emphasized race and gender, a multitude of factors may influence probation outcomes. Morgan (1994) found that probationers who were married and worked full-time were more likely to be successful—suggesting the importance of stable social bonds on probation outcomes. Gray, Fields and Maxwell (2001) also provide an expansive overview of variables associated with success and failure on supervision. Their sample included 1500 Michigan probationers, 64.2% of whom had one or more probation violations on their records (Gray et al., 2001). These violations were categorized as either “most serious”, “medium serious”, or “least serious”. From a new arrest to missing a curfew, these authors defined any violation as probation failure. Still, Gray et al. (2001: p. 554) concluded that most violations were for minor infractions that “may not necessarily pose a risk to the public at large”. Analyses showed that minorities, probationers with low educational attainment, and those with a history of drug use were more likely to have technical violations. The authors emphasize that substance abuse history plays an important role in probation failure since 22% of all recorded violations were for positive urinalysis screenings. Individuals with a higher number of technical violations were subsequently more likely to commit a new crime while under supervision (Gray et al., 2001).
A small body of research examines outcomes less in terms of probationer characteristics and more in terms of supervision practices. Jalbert, Rhodes, Flygare and Kane (2010) studied how reducing caseloads in conjunction with implementing evidence-based intensive supervision practices affected recidivism. In this case, “evidence-based practices” referred to utilizing risk assessments to determine the level of supervision required for each probationer (Jalbert et al., 2010). The study included 8000 probationers, about 20% of whom were labeled as high needs and assigned to Intensive Supervision Probation (ISP). ISP officers carried smaller caseloads due to the nature of their clients—about 30 cases as opposed to 50 for a typical probation officer (Jalbert et al., 2010). As such, the authors compared the outcomes for the clients of ISP officers to the outcomes of clients who remained on the regular supervision caseload. Using a regression discontinuity design, the study found individuals on ISP had significantly lower criminal recidivism rates compared to probationers on a regular caseload (Jalbert et al., 2010).
Some scholars have examined the impact of sentence requirements on probation outcomes. Ruhland, Homles and Petkus (2020) considered the influence of fines and fees on revocations, notably in a jurisdiction where probation officers spent “a significant amount of time trying to collect monetary sanctions” and depended on these collections “for a portion of officers’ salaries…” (Ruhland et al., 2020: p. 4). The authors engaged in a secondary data analysis of administrative records for 1600 probation cases, with the sample being overwhelmingly white (84%) and male (65%). They found that individuals who were assessed higher fee amounts were significantly more likely to have their probation revoked, either due to technical violations or new criminal offenses. People with a high percentage of delinquent fines and fees were also more likely to face revocation. However, the authors acknowledge that the data were limited in that they could not discern whether probationers were revoked explicitly due to non-payment or if a variety of ongoing violations led to revocation (Ruhland et al., 2020).
Thus, probation research overall (not specific to misdemeanor probation, where, as stated before, there is a paucity of research) indicates that age, income, prior convictions, substance abuse, race, low education, and fees are associated with probation failure. Being married, employed full time, served by a specialized probation program, and being a woman with children were associated with greater probation success.
2.2. The Implications of Misdemeanor Probation Failure
One of the primary ways the current study differs from previous literature is its sole focus on misdemeanor probation outcomes. The most recent statistics suggest that approximately 36% of probationers are on misdemeanor probation, representing about 1.27 million Americans (Kaeble & Alper, 2020). Kohler-Hausmass (2013: p. 353) argues that “misdemeanor justice…is one of the dominant components of contemporary criminal justice, and its operations represent an underappreciated modality of social control”. There is no clear explanation as to why the research focused on misdemeanor probation is scarce. One possible hypothesis relates to the fact that misdemeanor probation is more than twice as likely to be handled at the county-level when compared to felony probation, which is typically controlled by state-level agencies (National Center for State Courts, 2011). County-level data may be challenging to obtain and analyze, as was our experience in the current study, or may simply be unavailable to researchers.
The research that does exist suggests that misdemeanor probation is an important part of the broader conversation about mass incarceration and mass supervision that should not be overlooked. Olson and Lurigio (2000) used a mixed sample including both misdemeanants and felons on probation and found that 23.7% of their misdemeanor participants were rearrested while under supervision, and 33.4% had technical violations. In addition, 9.4% of misdemeanor participants had their probation revoked—thus likely sentenced to serve time in jail (Olson & Lurigio, 2000). Even short-term incarceration has been linked to a host of collateral consequences, including lost wages, eviction, and strained familial relationships (Pogrebin, Dodge, & Katsampes, 2001). One study focused on incarcerated misdemeanants found that 23% of participants were evicted from their rental housing while incarcerated (Weisheit & Klofas, 1989). Furthermore, a misdemeanor conviction has a lasting impact on socio-economic outcomes—mainly employment status. In an experimental correspondence study, Leasure (2019) examined the effects of misdemeanor convictions on hiring outcomes. Across the 582 sample resumes submitted to potential employers, 37% had no criminal record listed, 32% had a felony conviction listed, and 29% had a misdemeanor conviction listed. Resumes were also divided so that approximately half of the “applicants” had a racially distinct African American sounding-name, and the other half had a racially distinct white-sounding name. For both white and Black applicants, individuals with misdemeanor convictions were about half as likely to receive a callback than someone without a criminal record—but the impact of a misdemeanor conviction was even more severe for Black applicants (Leasure, 2019).
3. Methods
3.1. Data Set
The data for this study were provided by a county probation agency located in a mid-sized city in Georgia. The dataset was drawn from the full computerized record of every misdemeanor probation case opened in the jurisdiction between 2009 and 2018, pulled directly from the agency’s online case management system. These data are entered into the system by the probation officer throughout the time or probation. Data were entered through both dropdown menus and open fields. A thorough examination was necessary to remove obvious signs of entry error. All cases that 1) were actively open, 2) where the only listed charges were felonies, 3) where people were on probation as a bond condition and thus had not been convicted, and 4) where the dates provided for the probation term were clearly entered incorrectly (e.g. start date of 1/1/1900), were removed. The analyses for this study examine individuals placed on probation from January 1, 2016, to December 31, 2018. The decision to focus on these three years was driven by the potential confounding effects of policy changes made to the Georgia misdemeanor probation system in 2015. Georgia House Bill 310, which went into effect on July 1, 2015, restructured a variety of practices related to misdemeanor probation to reduce the overall probation population (Wiltz, 2017). The bill created more regulation surrounding private probation companies and codified judges’ power to waive probation-related fees and give misdemeanants suspended probation sentences (GA HB 310, 2015). In the 12 months following the enactment of HB 310, the county agency highlighted in this study saw a 43% decline in misdemeanor probation cases. Considering the substantial impact HB 310 had on the partner agency, it was decided that the analysis should only include the information available after the policy change took effect. The final sample included 6600 misdemeanor probation cases.
The study was approved by the university’s Institutional Review Board. Data were thoroughly anonymized to meet Institutional Review Board requirements, thus there was no way to determine if the same person had multiple open cases within these three years. As such, this sample may not necessarily represent 6660 individual probationers.
3.2. Sample
Demographic characteristics are summarized in Table 1. Description of the sample was somewhat limited by the categories provided in the agency’s case
![]()
Table 1. Demographic characteristics.
management system. For example, 72.1% of the sample is listed as male, and 27.9% as female, with no non-binary options. In terms of race and ethnicity, 47.6% of the sampled probationers were white, 40.9% were Black, 1.6% were Asian. Native American and bi-racial individuals each made up less than 1% of the sample. An additional 9% were listed as Hispanic, though ethnicity was not considered a distinct category from race in the case management system. Approximately 75% of the probationers had a high school equivalent education or above, and about 45% were known to be employed. The relatively high levels of education, yet low employment figures, are likely due to the jurisdiction’s large college student population. In this analysis, being a student was categorized as its own employment status. The average age of the sample was 29.58 years.
As described in Table 2, characteristics of probation were examined in this study. The variable of interest was the probation outcome upon case closure (1 = closed successful, 0 = closed unsuccessful). Nearly 71% of cases were closed successfully. Cases listed as “closed successful” by the supervising officer were those
![]()
Table 2. Probation characteristics.
where the probationer completed their required conditions, recognizing that conditions can be added or removed throughout the life of a probation case. The category of “closed unsuccessful” included outcomes such as probation revocation, absconding, and re-incarceration. There were also instances where cases were labeled as “closed unsuccessful”, without any indication of the specific cause of failure. One notable distinction between the current study and past research is that individuals with probation violations were not automatically included in the “unsuccessful” category. The partner agency would frequently close cases as successful even if probationers had a few minor violations throughout their supervision.
On average, probationers in the sample spent about a year on probation (375.08 days). Offense category varied, with violent offenses (28.6%) and ordinance violations (25.6%) most common. Offenses also included traffic offenses, substance violations, property crimes, and other. Since these were misdemeanors convictions, “violent offenses” typically refer to simple battery charges, and “property offenses” are predominately low-level theft cases. On average, probationers were assessed $538.15 in fines at sentencing. At case closure, a majority of probationers (53.7%) had unpaid fines.
First Offender status was held by almost 10% of the sample, however, “First Offender status”, in this context refers to a specific sentencing designation under the Georgia First Offender Act. This statute allows people with certain first-time offenses to have their convictions expunged and their charges sealed upon completing their sentence (Georgia Justice Project (GJP), 2020). First Offender status is not guaranteed simply because the defendant does not have a previous criminal history—a defense attorney must ask that the defendant be sentenced as a First Offender, and the Judge must agree. In addition, some common misdemeanors, such as DUIs, are ineligible to receive the First Offender designation (GJP, 2020).
3.3. Analyses
Initially, we completed bivariate analyses to examine the relationship between each factor and the dependent variable Case Outcome, using chi square or two-tailed independent t test. All factors were significant at the p < 0.01 level. We then used stepwise multivariate logistic regression analysis to determine the effects of race, gender, education, employment, fines, first-offender status, age, days on probation, and offense category, on the likelihood of success. All analyses were conducted using the software IBM SPSS Statistics 27.
4. Results
The logistic regression model was statistically significant (χ2(20) = 648.84, p < 0.001) explaining 27% of variance (Nagelkerke R2 = 0.267). Women were about 1.5 times more likely to be successful than men (p < 0.001). Older probationers and those with the highest levels of education were also significantly more likely to be successful (2% per year of age and 6%, respectively). In terms of employment status, non-student employment was not significantly correlated to success. Still, students were found to be 1.8 times more successful when compared to the unemployed. Black probationers were about half as likely to succeed as white probationers (p < 0.001). Hispanic probationers, on the other hand, were about two times more likely to be successful than white probationers (p < 0.001). No other racial categories had a significant relationship to successful case closure.
When examining characteristics of probation, the status of unpaid fines was the strongest predictor of success by far—individuals without unpaid fines were 7.2 times more likely to be successful than those with unpaid fines. Despite this strong correlation, the dollar amount of fines assessed at sentencing was not significant. Time on probation was also not significantly associated with success. Out of the offense categories, only property crimes were significantly associated with successful case closure. Probationers convicted of property crimes were about half as likely to be successful than those convicted of violent offenses (p < 0.001). Finally, probationers without First Offender status were half as likely to be successful as those with First Offender status. In summary, gender, age, education level, race, student status, having unpaid fines, type of crime, and first offender status were significantly associated with misdemeanor probation outcome. The logistic regression model is shown in Table 3.
![]()
Table 3. Multivariate association between case outcome and probationer characteristics.
1Nagelkerke R2 value for this logistic regression was 0.267.
5. Discussion
Our findings that women and older, more highly educated individuals are more likely to be successful on probation are consistent with decades of criminological research (see Gray, Fields, & Maxwell, 2001). Some of the findings related to financial assessment and offense categories also align with recent probation-specific studies. Our finding that the dollar amount of fines assessed was not associated with success is similar to Ruhland et al.’s (2020) finding that fines were not a predictor of probation revocation. Our finding that individuals with property crimes are less likely to have a successful case outcome is similar to the work on differential probation outcomes of Steinmetz and Henderson (2016), who also found that individuals convicted of property crimes had a greater chance of probation failure. The fact that this study of a small jurisdiction produced results comparable to nationwide research adds to the evidence that certain patterns in the probation system may be ubiquitous, no matter the location.
Still, the analyses produced several unique findings that warrant further exploration. The most surprising outcome was that Hispanic individuals in the sample were twice as likely to be successful on probation than non-Hispanic white individuals. In prior research, being a person of color has consistently been tied to poor probation outcomes (see Steinmetz & Henderson, 2016; Gray et al., 2001). However, many previous studies have also utilized a white/nonwhite binary that fails to examine differences that may arise within various nonwhite groups. After consulting with the agency partner that provided the data, one possible hypothesis is that Hispanic probationers were more successful due to the offense types that were common among the group. Data revealed that 64.7% of the Hispanic individuals in the sample were on probation for traffic offenses—predominately driving without a valid driver’s license. According to the partner agency, typically individuals with these types of crimes have limited supervision requirements outside of paying necessary fines and reporting to an officer. Although we did not have the required data to explore supervision conditions in this study, previous research supports the idea that having fewer supervision requirements is correlated to success (Doleac, 2018).
The conclusion that paying off fines was the strongest indicator of success also deserves significant attention. Probationers who paid off all their fines by the time their case was closed were 7.2 times more likely to be successful than those who had outstanding, unpaid fines. Since paying fines is one of the most common conditions of a probation sentence, it is logical that those who do not pay their fines are highly susceptible to failure. The magnitude of the effect, however, is striking. The finding regarding unpaid fines should be considered in the context of a secondary result: that the dollar amount of fines assessed at sentencing was not correlated to success. Since paying off fines is critical to success, one might assume that those required to pay less would be more successful—yet this was not the case. If someone is genuinely unable to pay, they are at significant risk of failure, even if the amount assessed is relatively “small”. This may suggest that one of the most crucial roles of judges and probation officers in increasing probation success is accurately assessing clients’ ability to pay and only requiring financial payments when any kind of payment is realistically within a probationer’s means.
5.1. Implications for Policy and Practice
The findings of this study have implications for both practice and policy. Identifying factors related to probation success can guide policies and decision-making, and potentially heighten awareness among probation officers of probationers who may be at greater risk of failure, allowing jurisdictions to develop community-specific evidence-based policies and supervision practices. Knowing, for example, that unpaid fines are associated with failure, an agency could develop an income threshold below which fees are automatically waived, rather than waiting until a probationer is already behind on payments to explore the idea of a fee exception. Knowing that time spent on probation is not correlated with success, an agency could rethink the notion that giving someone a shorter sentence will increase the likelihood they will complete their supervision without incident. Practices focused on increasing success rates are especially important in misdemeanor probation, considering that the consequences of probation failure could lead to prolonged criminal justice involvement and lifelong repercussions for minor offenses (Phelps, 2018).
Currently, the most widely used practice to predict probation success is use of a generalized risk assessment to determine the likelihood of recidivism, and subsequent assignment of high-risk probationers to differential supervision requirements. However, general models may not reflect the relevant risk factors present in a specific community, and matching assessments to context may be critical to ensure that these tools provide effective results (Vilijeon, Cochrane, & Johnson, 2018). Risk assessments could also be misleading if, for example, the same tool is used for both felony and misdemeanor probationers, as it should not be assumed that the risk factors are the same for both groups without empirical evidence.
Many jurisdictions could benefit from engaging in data-driven evaluations of their probation populations to develop their own risk assessment tools and models of supervision that acknowledge unique local circumstances. Thus, in addition to supporting the extant literature and providing some insight into additional factors associated with probation success, our study is an illustration of collaboration at the community level. Community-specific probation data analyses may provide insights into characteristics unique to the community and shape local practice. An example is seen in our finding that Hispanic probationers are twice as likely to succeed as white probationers. If this is, as posited by our partner agency, a result of offenses that are predominantly driving without a license, that may be a location specific phenomenon not seen in jurisdictions with a more established Latinx population.
Beyond the utility of this project for an individual probation agency, the findings related to unpaid fines have wide-reaching policy implications. Scholars and activists in recent years have become increasingly vocal about the devastating impact legal fines and fees can have on low-income defendants. Required to pay court costs well beyond their means, people can quickly fall into debt and face significant economic losses, prolonged contact with community supervision agencies, and even incarceration due to failure to pay (Martin et al., 2018). By revealing that paying off one’s fines is the most significant factor in misdemeanor probation success, this study contributes to the list of collateral consequences that can arise if an individual is assessed court fines that they cannot pay.
5.2. Conclusions and Recommendations
When considering the broader implications of this study, we recognize that the findings are limited and not generalizable beyond this one jurisdiction in Georgia. A significant portion of criminal justice research relies on data that may not be representative of the whole country but nevertheless contributes to an understanding of frequent patterns or issues that deserve future consideration. A limitation that is more specific to this analysis is the quality of the data. As stated in the methods section, there were inconsistencies in the database, which varied by the care and level of detail officers were willing to invest in data entry, and from officer to officer over time. There were also some data that, while technically available in the case files, were scanned handwritten note. These could not feasibly be extracted for a large sample. Consequently, we could not examine some variables, such as information on probation violations, that would have further enriched our analyses. Finally, as noted in Methods, our Institutional Review Board required total anonymity of the data and therefore there was no way to identify individuals who may have had multiple probations during the studied time period. Thus, our sample consisted of 6600 probation cases, but not necessarily 6600 individuals.
Despite the power of local criminal justice agencies and the vast number of people in this country convicted of minor criminal offenses, most literature on probation research focuses on state-wide felony data. This study addressed a significant gap in the literature by analyzing the outcomes of misdemeanor probationers and by providing a framework for using local-level data in an analysis of probation success. Our findings raise many questions worth future exploration. Further study is needed to explore the relationship among fees, socioeconomic status, and probation success. Mixed-methods studies could also be useful to determine if probation officer’s stated beliefs about factors related to success and failure are supported by quantitative analyses. Furthermore, exploring questions of racial disparity in probation must continue to be a priority for researchers. This study demonstrated the importance of considering identities beyond a simple white/nonwhite binary in order to uncover more nuanced findings about the role of race and ethnicity in probation success.
Our final call-to-action is to encourage more criminal justice researchers to focus on local-level analysis. Local control and variation are hallmarks of the American justice system, yet many agencies rely on research using data from vastly different jurisdictions to make decisions about their own practices. Creating researcher-agency partnerships is critical to provide the information needed for evidence-based reform at a local level and to contribute to a broader knowledge base about issues that exist regardless of geography. Challenges certainly exist for researchers in this space, particularly acknowledging that some criminal justice agencies might be wary of the intentions of researchers and thus protective of their data. Still, working to overcome these obstacles is worthwhile in the effort to produce meaningful information that can be used to support sustained reform in the criminal justice system.
Acknowledgements
The authors would like to thank Orion Mowbray for his assistance with data analysis, and the staff of the county Probation Office for their collegiality and collaboration. Since the completion of this study, author L.D.A. has passed away. This work would not have been possible without his tireless dedication to probation and rehabilitation.