Abstract
This research examines access to justice (AtoJ) through the lens of parole decisions, which is a pivotal stage in the criminal justice system, particularly in jurisdictions where post-release access to rehabilitation programs and other services are only available to parolees. The study analyzes 2,361 parole cases in Israel and explores the factors contributing to disparities in parole outcomes, with particular attention to way that biases interact with individual-level and procedural factors. The results reveal a significantly lower odds of parole for the minority group, but only after accounting for a range of other factors. Furthermore, the odds of early release among the minority inmates varied more according to individual-level characteristics like prior incarcerations and offense type, while parole decisions for majority group members were influenced more by procedural-level factors like the timing of the discussion and procedural delays. This suggests that judges’ decisions about minority inmates is more influenced by ethnic bias, while decisions about majority member inmates is more influenced by familiarity bias. By examining these dynamics, the study provides valuable insights into how personal and procedural factors intersect with biases in shaping parole decisions. The findings emphasize the importance of recognizing biases for promoting fairness in the parole process and ensuring equal access to justice.
Keywords
Introduction
Parole is an integral component of access to justice (AtoJ), understood as a fundamental human right guaranteeing the ability of individuals to seek and obtain fair remedies through the legal system on an equal basis (Gerards & Glas, 2017; Gutterman, 2022). The process of parole decision making and its implementation can raise several concerns related to AtoJ. Systemic biases such as racial disparities in sentencing and parole decisions can exacerbate inequalities in access to parole opportunities. Research indicates that individuals from marginalized communities, including racial minorities and socioeconomically disadvantaged groups, are disproportionately represented in the prison population and may encounter additional obstacles in securing parole (Huebner & Bynum, 2008). Procedural factors such as legal representation, procedural timing and delays can significantly impact AtoJ in parole decisions (Bass et al., 2005; Danziger et al., 2011; Lucy, 2020; MacDonald, 1992). Individuals may face barriers in accessing adequate legal representation or navigating the parole process, thereby affecting their chances of obtaining parole.
The importance of parole for AtoJ is even more pronounced in legal systems such as Finland, Germany, Israel and some US states, where parolees are granted unique access to rehabilitation and reintegration programs under postrelease supervision. Limited access to essential components of parole programs, such as housing, employment opportunities, and social services—particularly for minorities—can hinder successful reintegration and increase the likelihood of parole violations, ultimately leading to re-incarceration (Cohen & Palmor, 1985; Cohen, 2020).
Using data on 2,361 parole decisions recorded in the Israeli Prison Service (IPS) between 2010 and 2020, this study explores the factors contributing to disparities in parole outcomes, with particular attention to way that biases interact with individual and procedural factors. Drawing on the work of Sabol et al. (2020, p. 21), who note the difficulty of isolating “a single factor responsible for race-specific variations in imprisonment,” this research seeks to uncover the interaction between individual characteristics and procedural dynamics that contribute to these disparities.
The study underscores the importance of reconceptualizing parole not merely as a mechanism of early release, but as a critical tool for advancing justice through equitable access to institutional resources and reentry programs (Glaze & Kaeble, 2014; MacDonald, 1992). Addressing these disparities may generate broader societal benefits, including reductions in incarceration rates, alleviated fiscal pressures, and improvements in prison conditions.
Access to Justice and Parole
AtoJ refers not only to the capacity of all individuals to meaningfully engage with the legal process, including access to equitable legal representation, timely proceedings, and substantive participation in decision making, but also to equality before the courts and the right to a fair hearing (Gutterman, 2022). Prevailing AtoJ discourse concentrates on the provision of legal services, while frequently overlooking the deeper normative dimensions of fairness, equality, and institutional legitimacy (Lucy, 2020). MacDonald (1992) cautions that an exclusive focus on formal access risks concealing entrenched systemic injustices, thereby impeding a comprehensive understanding of how law and justice intersect (Lucy, 2017). Accordingly, meaningful evaluations of AtoJ must extend beyond the availability of legal resources to include their capacity to yield substantively just outcomes. Despite the formal existence of legal information and representation, AtoJ remains constrained by enduring barriers—financial, procedural, and systematic in nature—which continue to marginalize vulnerable populations (Lucy, 2020; MacDonald, 1992). Thus, an effective AtoJ framework must not only ensure the presence of legal mechanisms but also critically interrogate the structural and institutional contexts within which these mechanisms operate. This involves identifying and dismantling deep-rooted inequalities that impede full and equitable participation in legal processes.
The evolution of AtoJ underscores its increasing complexity. As MacDonald (1992) outlines, the 1960s prioritized legal representation and courtroom access; the 1970s emphasized institutional performance and design; the 1980s introduced concerns over outcome equality; and the 1990s expanded the focus to include alternative dispute resolution and public participation. In contemporary scholarship, AtoJ is conceptualized as a multidimensional construct encompassing access to public services, protection from institutional and interpersonal violence, and inclusive participation in civic and legal life (Gutterman, 2022). Within the parole system, these principles require a rigorous assessment of both procedural dynamics—such as the quality of legal representation, the timing of hearings, and bureaucratic inefficiencies—and biases that disproportionately disadvantage minority populations, whether individual or institutional, such as language barriers and limited access to culturally competent advocacy (Danziger et al., 2011; MacDonald, 1992).
Background on the Israeli Parole System
The Israeli parole system offers a salient case study for examining how AtoJ principles can be operationalized in a non-US context. Unlike the US model—which affords parole boards broad discretionary powers (Petersilia, 2003; Travis, 2000)—Israel's approach is characterized by more rigid statutory criteria, as set out in the Conditional Release from Imprisonment Law, 2001 (also known as the Israeli Parole Act). Nearly all inmates considered for parole need to have completed two-thirds of their sentence, participated in a rehabilitation program in prison and have expressed remorse for their crime. They also have the right to legal representation, which is typically provided by a public defender at the request of the inmate or the committee.
The decision to grant parole involves an evaluation of various factors, including expert opinions, behavior during and prior to incarceration, the nature of the offense and recommendations for the parole program. The parole board is composed of a diverse committee, including a criminologist, an education specialist and a representative from the prison system, with a judge making the final decision. Parolees must diligently adhere to all conditions imposed, and failure to do so may result in parole revocation, leading to a return to full incarceration. The decision-making process is designed to balance the rehabilitation of the individual with the protection of society, ensuring that parole is granted based on a thorough and careful examination of all relevant factors.
However, this ostensibly neutral structure exists within a sociolegal environment marked by pronounced disparities in sentencing and incarceration rates between Jewish and Arab citizens (for review of relevant studies see Rahav et al., 2015), raising critical questions about the equitable application of parole policies and the susceptibility of the decision-making process to implicit biases and institutional discrimination. In this article we refer to the Jewish-Arab distinction as an ethnic one, however the dividing lines of ethnicity and race as conceptualized in the US case do not carryover straightforwardly to Israel. For a more thorough comparison of the institutional and symbolic bases of exclusion, discrimination and stigmatization in the United States and Israel, see Lamont et al., 2016. Racial distinctions (white vs. black) have historically been the “master status” (Hughes, 1945) of divisions in the United States and the focus of much research on bias in the criminal justice system. In contrast, the Jewish/Arab division is the predominant “master status” in Israel, but it does not fit readily into racial or ethnic categories: Arabs are not considered a distinct race; a large percentage of the Israeli Jewish population have roots in Arabs countries, referring to Jews who immigrated from countries such as Morocco, Egypt, Syria, Tunisia, etc. (Shenhav, 2006); and there are over two million Arab citizens of Israel, or about 20% of the population. (Prisoners without citizenship or permanent residency status are outside the scope of the current study.) The similarities and differences between the Israeli and US context underscore the importance of contextualizing AtoJ efforts within local frameworks of exclusion and power—an argument emphasized by Van Cleve (2016) in her analysis of sentencing outcomes in Chicago.
In 2017, the Israeli Supreme Court issued a ruling to reduce overcrowding and improve prison conditions that underscored the imperative of providing humane living conditions (High Court Decision Number 1892/14). Subsequent policy measures were instituted to fortify the overall conditions within the prison system. The inception of the Short Sentence Release Unit in 2018 aimed to ameliorate prison overcrowding. Concurrently, administrative release criteria were made more conducive to increased parole opportunities by considering the broader dynamics of the prison population and sentence duration (Knesset, 2018).
Factors Affecting Parole Decisions
Within the framework of discretionary parole decisions, a multitude of factors come into play, spanning the period both before and during incarceration. These considerations encompass individual-level variables related to enduring individual characteristics such as ethnic origin and gender, or case-level characteristics like the type and seriousness of the crime for which the individual was incarcerated. In parallel, procedural aspects of the hearing, such as the order in which the case is heard, can have a bearing on the outcome (Danziger et al., 2011). This study contributes to the literature by examining the extent to which ethnic origin, along with a range of individual and procedural factors, is associated with parole outcomes, as well as how these factors operate differently for majority- and minority-group offenders.
Individual-Level Characteristics
Minority Status
In contrast to the vast literature on the role of minority status in other areas of the criminal justice system (Baumgartner et al., 2018; Kurlychek & Johnson, 2019; Tonry, 1995; Walker et al., 2016), relatively little attention has been paid to its role in parole decisions. According to focal concerns theory, biased decision making is rooted in the three central concerns of judicial sentencing decisions: defendant blameworthiness, safety of the community, and practical constraints and consequences (Blumstein et al., 1999; Steffensmeier et al., 1993; Steffensmeier et al., 1998; Ulmer, 2012). Judges might rely on stereotypes or attributions relevant to assessing these concerns. Consequently, if the focal concerns intersect with stereotypes or attributions related to race, ethnicity, gender, and/or age, this can lead to disparities in treatment based on these legally irrelevant characteristics. Similar to widespread stereotypes about African Americans in the United States, stereotypes depicting Arab citizens of Israel as aggressive, dangerous, frightening, and prone to violence are common in Israel (Bar-Tal & Teichman, 2005; Hasson et al., 2017; Karniel & Lavie-Dinur, 2011; Schafferman, 2008). Since parole board decisions in Israel are informed by concerns related to risk and public safety (Walk & Dagan, 2024), to the extent these concerns align with stereotypes about the Arab minority, we can expect biased decision making (Avni et al., 2024).
In the US context, findings about bias in parole decisions have been somewhat mixed. Early studies found that bias did not directly influence parole decisions, though it may have operated indirectly through other criteria such as expert recommendations (Heinz et al., 1976; Scott, 1974). Similarly, Proctor (1999) found no direct effect of race in parole decisions but did find that it was associated with less favorable institutional recommendations, which were in turn highly influential in the final decision. Morgan and Smith (2008) analyzed parole board decisions across multiple jurisdictions and concluded that, although race was not a consistent predictor of parole outcomes, patterns emerged that suggested a possible association in certain contexts. They emphasized that racial disparities may still arise indirectly through institutional practices, localized policies, or broader systemic dynamics, even in the absence of overt racial bias. Young and Pearlman (2022) found direct effects of race in a study of parole among individuals sentenced to life imprisonment, as well as indirect effects where incarcerated African American men received less favorable assessments and recommendations from correctional officials. Finally, Huebner and Bynum (2008) found that African American men spend longer times in prison awaiting parole, though in another study, they found that minority sex offenders were released more quickly than white sex offenders (Huebner & Bynum, 2008).
To date, there has been no in-depth examination of how race and ethnicity interact with both individual and procedural factors in parole release decisions (though see Morgan & Smith 2008). This study aims to determine whether ethnicity is associated with access to parole and whether other individual-level factors also play a role, while also considering procedural factors contributing to these disparities.
Criminal Background
Criminal background refers to the prior incarceration and criminal history of the offender. Proctor (1999) found that criminal background significantly influenced parole decisions, along with criminal history and the prisoner's education level. According to Caplan (2007, p. 18), criminal history stands out as “one of the most influential factors affecting parole release for parole-eligible inmates.” Moreover, compared to systems with a minimum incarceration period, in systems without a stipulated minimum incarceration period, as in Israel, studies have noted that parole board members place greater emphasis on factors related to punishment, such as the severity of the crime and the inmate's prior criminal history, when deciding on parole release (Carroll & Burke, 1990). The present research examines whether prior incarceration is a significant factor in parole release decisions and whether its association with parole outcomes differs between minority and majority groups.
Time Served
In a recent study by Vîlcică (2018), it was found that the duration of incarceration failed to demonstrate a statistically significant correlation with parole decisions after controlling for other factors. In contrast, Proctor (1999) found that the proportion of the sentence served significantly influences parole decisions: for every 25% increase in the length of the sentence served, the odds of being granted a full parole board hearing decreased by a corresponding 25%. This indicates a direct and proportional relationship between the length of time served and the likelihood of receiving a parole board hearing.
Offense Type
A study conducted in the United States examined various determinants influencing parole outcomes, underscoring the pivotal role played by the nature of the offense for which an inmate was incarcerated. The research found that the type of crime committed was most strongly associated with parole outcomes, with inmates convicted of serious and violent offenses more likely to be denied parole compared to those incarcerated for less severe crimes (Turpin-Petrosino, 1999). Vîlcică (2018) examined the role of offense characteristics in explaining parole decisions, finding that the type of offense was a major predictor of release and revealing a reluctance within parole boards to grant release to individuals serving sentences for offenses widely perceived as heinous crimes. Even in the presence of factors indicating an inmate's readiness for reintegration, such as successful institutional programming and good behavior, the research indicated that individuals incarcerated for sex offenses encountered notable challenges in gaining early release. The present study examines how offense type is associated with parole outcomes across minority and majority groups.
Institutional Behavior and Risk Assessments
A 2013 study by Mooney and Daffern investigated the impact of offender behavior on parole outcomes among 146 violent offenders, highlighting key predictors such as aggressive disciplinary incidents, Violence Risk Scale (VRS) scores, and recommendations by Community Corrections Officers (CCOs). Building on this framework, the present research examines whether aggressive incidents, VRS scores, and CCO recommendations are associated with parole decisions and vary between minority and majority groups.
Remorse and Rehabilitation
Bronnimann (2020) examined the role of remorse within the context of parole hearings, illuminating its implications for the criminal justice system. However, there are inherent challenges in evaluating remorse and acknowledging responsibility for crime. Issues of sincerity, cultural disparities, and the subjective nature of emotions underscore the complexity of this endeavor. To address these challenges, the current study draws upon data from treatment programs within prisons, serving as evidence of inmates’ willingness to take responsibility and engage in emotional processes. (In Israel, the expression of remorse is regarded as a crucial and integral component of the rehabilitation process, significantly influencing its effectiveness and outcomes.) Acknowledging the importance of treatment program participation and willingness to take responsibility (Lucy, 2020; MacDonald, 1992), this study examines the association between participation in prison rehabilitation programs with equitable access to parole.
Procedural Factors
Timing and Scheduling
Danziger et al. (2011) found that extraneous factors—such as the timing of parole decisions—are associated with parole outcomes, despite their limited relevance to legal merit. Specifically, cases scheduled earlier in the day were more likely to result in parole approval, potentially due to decision fatigue prompting a preference for the status quo. While Danziger et al. (2011) focused solely on the time and order of the hearing, the present study expands this line of inquiry by enlarging the range of relevant procedural factors and by examining whether the relevance of hearing timing varies according to the ethnic background of the incarcerated individual. Moving beyond legal formalism (Leiter, 2005), this study offers a more nuanced account of the parole decision-making process.
Postponements
Timor and Elisha (2021) highlight the adverse implications of delays in parole hearings, noting that postponements—whether procedural or technical—can undermine the rehabilitative aims of conditional release and broader principles of justice. Such delays may erode inmates’ trust in the parole system, disrupt therapeutic continuity, reduce motivation, and even lead to case denial (Timor & Elisha, 2021). This study builds on these insights by using quantitative data to examine whether prior postponements are associated with parole outcomes, with a specific focus on disparities between minority and majority groups.
Legal Representation
Legal assistance plays a vital role in supporting reintegration after incarceration (Fajriando & Sujatmiko, 2021). This study investigates the association between legal representation and parole outcomes, with particular attention to differences between minority and majority prisoners. Prior research underscores that prisoners often face intersecting disadvantages, such as mental illness, substance abuse, poverty, and limited education which interact to impede access to justice (Grunseit et al., 2008). Given these challenges, legal representation is essential for protecting human rights and advancing equitable rehabilitation. Understanding whether legal advocacy may operate differently for majority and minority groups is critical for ensuring fair parole decisions and promoting equal access to justice (Lucy, 2020; MacDonald, 1992).
Data and Methods
The dataset used in this analysis builds upon Danziger et al. (2011), which primarily focused on the timing of parole board discussions. However, the present dataset offers enhanced depth by incorporating a broader range of variables. It includes 2,361 parole decisions and captures personal and procedural attributes of both the offenders and the parole committees, alongside a binary outcome indicating whether parole was granted. Key variables include the ethnic background of prisoners, gender, prior incarceration history, length of incarceration, type of offense, behavioral risk, and willingness to take responsibility, as reflected in participation in treatment programs. Additional variables reflect procedural factors related to the hearing: the gender of the judge, sequencing of cases during hearings, session timing, delays, duration of parole deliberations, and legal representation.
The data were extracted from parole case files by a legal professional who was unaware of the research questions; all personal identifiers were removed prior to dataset compilation. The records span parole decisions made between 2010 and 2020 and are drawn from national-level data representing parole hearings conducted throughout Israel.
The variable for ethnic origin distinguishes Arab Israeli prisoners, classified as minority (=1), from Jewish prisoners, categorized as the majority (=0). The sample consists entirely of Israeli citizens; it therefore excludes security prisoners, who are ineligible for parole. Arab prisoners are identified based on their religion, with over 98% being Muslim, representing about 18% of Israel's population. Only adult prisoners aged 18 and older are included in this study.
Of the total sample, 2,204 were men and 157 were women, including 134 Jewish and 23 Arab women. Notably, 96% of minority inmates were men, and only 4% were women. Gender was coded dichotomously, with men coded as 1 and women as 0. The number of prior incarcerations is a count of prior incarcerations plus the current one. It reflects how many times the individual has been imprisoned, regardless of the duration or nature of the offenses, and serves as an indicator of prior involvement with the criminal justice system. Offense type is categorized into six distinct groups according to the prison authority's classification of the inmate's primary conviction: traffic-related offenses, property-related offenses (e.g., theft or burglary), drug-related offenses (e.g., possession or trafficking), violent offenses (e.g., assault or robbery), sexual offenses, and homicide-related offenses (including manslaughter and attempted homicide). This classification reflects the most severe crime attributed to each offender as documented in their official case file. Violent offenses are the reference category. Behavioral risk is assessed based on aggressive disciplinary incidents, the total score on the VRS, and recommendations by the CCO. Considering these factors, bad behavior is coded as 1, while good behavior is coded as 0. Willingness to take responsibility is determined by participation in prison treatment programs, as involvement in these programs reflects and often requires an expression of accountability or remorse. This is coded as 1 for participation and 0 for nonparticipation. The variable for judge's gender equals 1 for male judges and 0 for female judges.
For procedural factors, “sequence” is a continuous variable that refers to the order in which the case was heard during the day (1, 2, 3, …, n), while “session timing” divides the day into three sessions: before breakfast break (excluded category), after breakfast but before lunch (session 1), and after lunch (session 2). The length of parole discussion measures the duration of the discussion about the inmate in minutes. Technical delays of the case indicate whether there have been postponements, with 1 for delays and 0 for no delays. Legal representation is coded dichotomously, indicating whether the prisoner was represented by a lawyer (1 = represented, 0 = not represented). In accordance with Israel's Public Defense Act (1995), all individuals eligible for parole are entitled to legal counsel. This variable captures the presence of legal representation during the hearing—whether public or private.
Logistic regression is used to assess factors associated with parole outcomes, including personal (e.g., ethnicity, gender, criminal history, prior incarceration, time served, offense type, behavioral risk, and willingness to take responsibility) and procedural variables (e.g., judge's gender, case sequence, session timing, discussion length, prior delays, and legal representation). Analyses are conducted for all inmates and separately for Arab and Jewish inmates in order to identify group-specific patterns. Descriptive statistics are provided by ethnicity and parole outcome, and diagnostic tests reveal no significant multicollinearity among predictors.
Findings
The overall parole approval rate for the sample was 44.26%. While 44.96% of majority prisoners were granted parole, the approval rate for minority prisoners was slightly lower at 42.65%. Delving further into the demographic breakdown, 69.41% of the sample consists of majority prisoners, with 70.52% of those granted parole falling within this group. In contrast, 30.58% of the sample comprises minority prisoners, and 29.47% of parolees belong to this ethnic category. Both minority and majority groups have a nearly equal proportion of prisoners granted parole relative to their representation in the total sample (Table 1).
Disparities in Parole Granting.
Descriptive statistics of the sample are presented in Table 2. Chi-square tests (for categorical variables) and t-tests (for continuous variables) indicated significant differences between Arab and Jewish inmates across several parole-related factors. As the majority group, Jews were significantly less likely to be assessed as exhibiting risky behavior, experience technical delays and have legal representation. In contrast, Arabs, as the minority group, experienced significantly fewer technical delays. No significant differences were found between the groups regarding criminal background, prison time, offense type (traffic, property, drug, violent, or sexual), rehabilitation plan implementation or hearing duration.
Descriptive Statistics for All Prisoners, Arabs and Jewish.
Table 3 presents logistic regression results on factors associated with parole decisions for all inmates and separately for minority and majority groups. Ethnicity, prior incarcerations, time served, traffic violations, drug offenses, sexual offenses, murder-related charges, risk behavior, judge's gender, hearing timing, delays, and legal representation show significant associations with the odds of release. More prior offenses and longer imprisonment are linked to lower parole odds. Traffic, drug, and murder offenses are associated with higher odds of release compared to violent offenses, while sexual offenses decrease parole odds. Misconduct reduces parole likelihood; cases heard earlier and by female judges are more likely to result in parole (Danziger et al., 2011). Delays are linked to lower parole odds; legal representation increases them.
Logistic Regression of Granting Parole on Personal and Procedural Factors.
†p < .10, *p < .05, **p < .01; aViolent Offenders are reference category; bSession 0 is reference category.
Prior incarcerations and prison time show stronger negative associations for minority prisoners. Traffic and drug offenses are more strongly associated with parole for majority prisoners, while murder convictions are more associated with parole among minority prisoners. Sexual offenses reduce parole odds mainly for majority prisoners. High-risk behavior is more negatively associated with parole among minority prisoners.
Procedural factors, including hearing sequence and delays, have weaker associations with parole for minority prisoners. Postponements significantly reduce parole odds for the full sample and majority group, but not for minority prisoners. Although legal representation has a higher odds ratio among minority prisoners (Exp(B) = 3.276) compared to majority prisoners (Exp(B) = 1.470), the association is stronger and more consistent for majority prisoners based on the overall results.
In the parole decision analysis presented in Table 4, an interaction variable is introduced to investigate the influence of prior incarcerations for each type of offense. This methodology aims to discern whether previous incarcerations may exert different effects for various offense categories in the parole decision-making process. Table 4 reveals that the presence of prior offenses decreases the odds of early release especially for inmates convicted of traffic offenses. While traffic offenses are generally associated with higher odds of parole, a history of prior incarcerations notably linked to reduced odds of parole in this category is compared to other offense categories. These findings offer valuable insights into parole decision making, suggesting that the influence of criminal history is more pronounced in certain offense contexts.
Interaction Variable Between Prior Incarcerations and Type of Offense.
†p < .10, *p < .05, **p < .01; aViolent Offenders are reference category; bSession 0 is reference category.
Table 5 presents logistic regression results for the analysis of parole decisions, introducing an interaction term to assess how actual prison time influences parole outcomes across different offense types. This approach examines whether imprisonment duration is differently associated with parole likelihood depending on the offense. The results indicate that traffic offenses are associated with lower odds of parole among minorities, while drug-related offenses are linked to higher odds among majority prisoners. Although longer imprisonment generally reduces parole likelihood, an exception is observed for drug offenses, where imprisonment time appears to increase parole odds, possibly reflecting the positive role of extended rehabilitation programs delivered while incarcerated (Haviv & Hasisi, 2019).
Interaction Variable Between Actual Prison Time and Each Type of Offense.
†p < .10, *p < .05, **p < .01; aViolent Offenders are reference category; bSession 0 is reference category.
Discussion
The access to justice movement, a central aspect of wider legal reforms across many jurisdictions, is rooted in the principle that all individuals should enjoy equality before the law (Letto-Vanamo, 2005). However, ensuring that this ideal is consistently realized in practice, particularly within discretionary decision-making bodies such as parole committees, remains a complex challenge. This study investigates whether principles of equal and fair access to justice are upheld within parole committee systems, with a particular focus on differences between minority and majority prisoners in Israel in being granted parole. Without taking any other factors into consideration, Jewish and Arab Israeli inmates have similar chances of being paroled. However, after controlling for a range of individual and procedural factors, Arab Israeli inmates are significantly less likely to be granted parole. Moreover, the factors most strongly associated with parole outcomes differ between minority and majority groups. For minority prisoners, individual-level characteristics such as prior incarceration history, time served in prison, and assessed risk behavior are more strongly associated with parole outcomes, all associated with lower odds of early release. In contrast, among prisoners of the majority group, procedural factors—including the timing of parole hearings and delays in the hearing process—show a more substantial association with outcomes. These patterns are evident in the regression analyses, where variables related to procedural factors reach statistical significance among Jewish prisoners but display less consistent associations among Arab prisoners. These findings raise important questions regarding the operation of parole systems and the ways in which ostensibly neutral processes may differentially affect distinct population groups.
For prisoners of the majority group, cases were more likely to result in release when the presiding judge was a woman and when the hearing occurred earlier in the day. Delays in case proceedings were associated with substantially lower odds of release for majority-group prisoners. However, for minority prisoners, delays did not show a significant association with parole outcomes. These patterns suggest that procedural dynamics within parole hearings can meaningfully shape outcomes (Danziger et al., 2011), but it also points to potential disparities in how procedural advantages translate across inmates from different ethnic groups.
Legal representation was significantly associated with higher odds of parole release for both groups, though the strength of the association differed: it helped minority prisoners more than inmates from the majority group. These findings underscore the centrality of legal representation in parole outcomes, aligning with prior research emphasizing the critical role of effective advocacy within legal proceedings (Dancig-Rosenberg et al., 2011). One possible explanation for this stronger association among minority prisoners is that legal representation may play a particularly important role in assisting individuals who face systemic and structural disadvantages. Minority prisoners are more likely to encounter language barriers, limited familiarity with legal norms and procedures, and lower levels of trust in the justice system. For these individuals, legal counsel serves not only as an advocate but also as a mediator who helps them understand and access a complex and unfamiliar system. In this way, representation facilitates accessibility and engagement with the legal process, which may be especially crucial for individuals who lack the tools or confidence to effectively present their case on their own.
However, the quality and accessibility of representation may differ substantially between groups, and such differences could impact the effectiveness of legal advocacy. Future research should continue to examine whether disparities in parole outcomes are driven not only by the presence of legal representation, but also by its quality and its responsiveness to the specific needs of marginalized populations who are disproportionately affected by systemic barriers.
The findings suggest that group disparities in parole outcomes may be influenced by multiple factors, including ethnic and familiarity biases, variations in offense types, and the structural dynamics of decision-making processes within parole hearings. These factors may contribute to unequal treatment across racial and ethnic groups within the criminal justice system. Our research supports existing studies showing that racial and ethnic biases remain prevalent, even as some disparities, particularly between Black and White offenders, have decreased in the United States since 2000 (Sabol et al., 2020), with the exception of the COVID-19 pandemic period (Klein et al., 2023). However, concerns about unequal treatment suggest that racial and ethnic disparities in the criminal justice system remain a significant and unresolved issue.
In the Israeli context, Avni et al. (2024) examine how racial and ethnic dynamics influence prosecutorial decision making, particularly regarding property crimes like burglary. Their research suggests that prosecutorial decisions are shaped not only by the individual characteristics of suspects but also by broader societal perceptions, especially when stereotypes intersect with focal concerns (Steffensmeier et al., 1993, 1998; Ulmer, 2012) and perceptions of group threat. They argue that these perceptions contribute to disproportionately punitive outcomes for Arab suspects, even when the evidentiary conditions are comparable to those for Jewish suspects. The intersection of ethnic bias and perceived group threat becomes particularly pronounced when the victim is Jewish, further highlighting the impact of sociopolitical narratives on criminal justice outcomes. Similar dynamics likely shape decision making in other areas of the criminal justice system, such as parole decisions. The finding that Arab Israeli inmates are less likely to be granted parole suggests that stereotypes about them align with the focal concerns of judges in granting parole. Furthermore, our results show that the association between risk behaviors, which are related to concerns about public safety, and lower odds of parole approval is significantly stronger for Arab Israeli inmates. In the context of minority prisoners, heightened perceptions of risk, influenced by implicit biases and societal preconceptions, may contribute to more restrictive parole decisions.
While the focal concerns perspective has emphasized the role of negative stereotypes, familiarity bias may also play a role. Familiarity bias refers to the cognitive tendency to view familiar information as less risky and unfamiliar information as more threatening (De Vries et al., 2017). Grounded in psychological mechanisms that provide comfort and security, familiarity bias leads judges to favor familiar information and perceive risks associated with unfamiliar cases more acutely. In Israel, 91% of judges in all Israeli courts are Jewish, while 9% are non-Jewish (Knesset Research and Information Center 2023). Similarly, most parole judges in Israel are Jewish. Familiarity bias may affect how judges assess various factors, leading to higher parole denial rates for prisoners who are less familiar to them, such as those from minority groups. In these cases, personal factors like criminal history, time served, and risk behavior may be scrutinized more rigorously, contributing to higher denial rates for prisoners who are less familiar.
Individual-level factors, such as prior incarcerations, time served and risk behavior, are more salient for Arab Israeli prisoners, leading to lower odds of parole compared to Jewish inmates. In contrast, procedural factors have on the whole a larger influence on decisions involving inmates from the majority group. This suggests that, for minority prisoners, personal risk factors—especially those perceived through the lenses of negative stereotypes—play a more prominent role, whereas for majority prisoners, procedural factors may hold more weight when they disrupt automatic thought processes in familiar situations (De Vries et al., 2017; Kahneman, 2011; Kahneman & Chajczyk, 1983; Kahneman et al., 1982).
Exploring the decision-making processes in parole hearings is essential, as they represent a critical juncture for intervention before an offender's reintegration into society. Understanding how judges assess different types of offenders is crucial for ensuring fairness and equality in parole decisions, particularly in determining which individuals are most appropriate for early release. Such an approach not only contributes to the administration of justice (MacDonald, 1992) but also optimizes the criminal justice system by promoting resource efficiency, reducing costs, and enabling rehabilitation programs to be tailored more effectively to the specific needs of each offender.
Type of Offense
The classification of offenses into six categories—traffic, property, drug, violent, sex, and murder offenses—based on prison assessments, revealed that, compared to violent offenses, traffic and drug offenses were more likely to result in early release by the parole committee, particularly among Jewish prisoners. Understanding how the nature of the offense influences parole decisions allows for a more comprehensive approach to addressing systemic inequalities that disproportionately impact marginalized populations. This exploration not only promotes fairer parole determinations but also provides critical insights into mitigating biases that affect decision making across diverse offender groups.
Drug offenses encompass individuals struggling with addiction, a condition often seen as chronic and difficult to treat (McLellan, 2017; Sarkar et al., 2015). Addiction is characterized by a cycle of use, misuse, and abuse, with individuals exhibiting chronic, progressive, and relapsing behaviors, alongside denial and continued use despite negative consequences (Chamberlain et al., 2017). Similarly, traffic violators have been described as impulsive individuals with a pattern of reckless driving (Martí-Belda et al., 2019). Both substance abusers and traffic violators share common traits as habitual lawbreakers. Judges, along with rehabilitation systems, may adopt a more lenient approach toward these offenders when it is their first incarceration, offering opportunities for parole and rehabilitation. Despite the repetitive nature of their actions, there may be an underlying belief that the impulsivity associated with substance abuse and traffic violations can be moderated, particularly through treatment. However, this perception appears to be more prevalent among offenders from the majority group, suggesting that treatment opportunities for this group are more common. In contrast, for minority offenders, linguistic and cultural barriers, along with systemic obstacles, may contribute to lower odds of parole, potentially influencing how their cases are perceived and handled by the justice system.
Arab inmates convicted of homicide-related crimes were more likely to be paroled compared to Jewish inmates. In contrast, Jewish traffic and drugs offenders were generally more likely to receive parole than Arab inmates convicted of similar crimes. However, for sex crimes, Jewish offenders were less likely to receive parole compared to their Arab counterparts. Studies show that individuals who justify violence are more likely to engage in criminal behavior (Kishi et al., 2018; Walters, 1995, 2012). This tendency may explain differences in violent behavior across groups, as noted in the Institute for National Security Studies report on Arab violence (Kurtz & Shavit, 2023). These patterns suggest parole committees may assess crimes differently based on the offender's ethnic or social background, contributing to disparities in parole decisions. This underscores the need for program-specific approaches tailored to each offense category, as some offenses, such as substance abuse and drug-related crimes, are more prevalent among the majority group, whereas offenses involving weapons are more frequently associated with minority offenders. Although minority groups are more likely to be granted parole for severe offenses like murder, this does not translate to a higher overall parole likelihood. The severity of the crimes may lead parole boards to exercise more caution, reducing the overall chances of release for minority offenders despite higher parole rates for specific serious offenses.
Procedural Factors
The findings also underscore the significant role of procedural factors—such as the judge's gender, the timing of hearings, delays in decisions, and legal representation—in shaping parole outcomes, particularly among majority prisoners. These procedural variables seem to have a more pronounced effect on parole decisions for majority-group offenders. This aligns with the notion that cases involving familiar offenders, characterized by predictable patterns or common offenses, may lead judges to rely on heuristic judgments, rather than conducting a detailed evaluation of each individual case, as part of a more automatic decision-making process (Danziger et al., 2011; De Vries et al., 2017; Kahneman, 2011). Under such conditions, it is to be expected that exogenous factors will play a role. For instance, the timing of hearings reveals a significant association, with higher odds of parole being granted at the start of the day or session. This suggests that decisions made later in the day may be influenced by cognitive fatigue, leading judges to make quicker, less demanding judgments (Danziger et al., 2011). Such a pattern implies that these decisions may be more routine and less thoroughly scrutinized, particularly for offenders from social groups that are more familiar to the judges. Conversely, when evaluating unfamiliar cases involving minority inmates, judges may adopt a less automatic decision-making process (Kahneman, 2011). This shift often involves taking risks more seriously and paying closer attention to personal factors and risk behaviors. The heightened scrutiny in these cases reflects a more careful evaluation, indicating that unfamiliarity with the offender prompts a more deliberate, nuanced approach to the parole decision.
Postponements in parole decision making can arise from factors like administrative inefficiencies, scheduling conflicts, or the need for additional documentation, such as updated risk assessments or psychological evaluations. These postponements often lead to delayed hearings, introducing procedural challenges that complicate the case. Issues like outdated records, incomplete documentation, or the expiration of relevant legal files hinder the parole board's ability to make timely and informed decisions. The results indicate that postponements generally reduce the likelihood of parole approval, with a notably stronger impact on the majority group. For majority prisoners, postponements add complexity to cases typically perceived as straightforward. Parole boards, initially seeing these cases as familiar and aligning with common offense patterns, may adopt a more cautious approach as the process becomes more complicated. This increased caution results in a higher likelihood of parole denial for majority prisoners, as procedural postponements heighten hesitations in granting parole.
Legal representation plays a critical role in shaping parole outcomes and promoting fairness within the criminal justice system (Casper, 1978; Chaara & Falisse, 2022; MacDonald, 1992). We find that offenders with legal counsel are consistently more likely to be granted parole, with legal aid serving as a vital safeguard against procedural biases and systemic obstacles—especially for minority offenders. Legal representation facilitates access to rehabilitative programs, reduces recidivism, and helps overcome linguistic and cultural barriers that disproportionately affect minority and marginalized groups (Shoham et al., 2024). Additionally, it enhances perceptions of procedural justice, particularly among minority and marginalized inmates, fostering greater trust in the legal process and the broader justice system (Chamberlain et al., 2017; Kishi et al., 2018). These insights highlight the need for systemic efforts to expand and improve legal support as a means of promoting justice and reducing disparities in parole outcomes.
Limitations and Future Directions
This study acknowledges several limitations, particularly regarding the parameters considered—both personal and procedural factors—which may not fully capture the complexity of parole decisions. For instance, while the study focuses on adults aged 18 and older, age was not explicitly examined as a factor as those data were unavailable. Additionally, the small proportion of female prisoners compared to their male counterparts results in a dataset where most of the data pertains to male prisoners. The predominance of majority judges also raises questions about whether minority judges might make different decisions for minority prisoners, potentially highlighting the role of familiarity bias. Furthermore, due to the small number of cases in certain subcategories, offense types were grouped together, which may obscure important distinctions within these categories. Nevertheless, the study seeks to reflect the primary offense categories as classified by prison intake assessment. These classifications are intended to categorize offenders based on the nature of their crimes, which in turn helps to guide decisions regarding treatment, rehabilitation, and the monitoring of inmates throughout their incarceration. The categorization process is designed to ensure that offenders receive appropriate interventions tailored to the severity and type of their offense and to facilitate their rehabilitation and reintegration into society upon release. Another significant limitation is the absence of data on socioeconomic status (SES), which could intersect with access to legal representation and influence parole outcomes. Individuals from lower SES backgrounds often experience systemic disadvantages within the criminal justice system, including limited access to private legal counsel and rehabilitative resources. Future research should integrate SES indicators—such as income, education level and employment status—potentially through linkage with tax registry or census data, in order to better understand how socioeconomic inequality influences parole decisions. Despite the limitations of this study, the findings highlight the critical need for a comprehensive understanding of how both personal and procedural factors interact in shaping parole decisions differently for minority- and majority-group inmates. Providing equitable access to legal representation and ensuring a fair, transparent decision-making process are crucial mechanisms for minimizing bias and ensuring that parole decisions reflect not only individual circumstances but also systemic fairness and that they advance access to justice.
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.