Judicial and Litigant Perceptions in Dutch Court Cases: Perceptions of Outcome Importance Overlap, Perceptions of Procedural Justice Diverge

The current paper aims to provide insight into judges’ perceptions of how fairly they treat litigants and how important case outcomes are to litigants, and whether these perceptions relate to litigants' perceptions of procedural justice and outcome importance. Respondents were litigants involved in bankruptcy, landlord-tenant, and administrative law cases and judges handling these cases at the district court of the Mid-Netherlands. Both litigants and judges indicated outcome importance and procedural justice. Litigants also indicated their trust in judges. Multilevel analyses using hierarchical regression showed a positive association between judicial and litigant perceptions of outcome importance and no significant association between judges' and litigants’ perceptions of procedural justice. This indicates that whereas judges and litigants largely agreed on how important case outcomes were, their views about how fairly judges handled cases diverged. These insights enhance our understanding of linkages and discrepancies between judges' and litigants' views on important aspects of the legal system.


Introduction
The traditional concern of judges, and the goal emphasized in their legal education, is the correct application of the law to particular legal disputes, in such a way that people are provided with a forum where they can obtain justice (Tyler, 2007). The rule of law requires that laws are applied equally, without unjustifiable differentiation. For example, inconsistency between persons can sometimes result in serious breaches of fairness (Leventhal, 1980;Steyn, 1997). One of the key responsibilities associated with being a judge is making decisions that are fair and just, while simultaneously treating litigants in a fair and just manner. The extent to which judges use fair procedures when making decisions is critical to people who are taking their case to court. Treating people fairly has numerous positive effects, which have received considerable attention within the research field of procedural justice (for an overview, see Lind & Tyler, 1988).
Despite the increasing attention for the empirically based perspective on procedural justice in the legal field and legal profession, relatively little attention has been paid to the level of fairness that judges themselves see in their own behavior towards litigants. Although procedural justice has been the subject of several courtroom studies, most of these studies examined fairness perceptions of court litigants and focused on the aspects of the judge's treatment that contributed to these fairness perceptions (Casper et al., 1988;Lind, 1990;Tyler, 1984;Tyler & Huo, 2002). In other words, many studies take the citizens' perspective on procedural justice as their starting point. But what about the judges' perspective on procedural justice? Do judges perceive their own actions during a court hearing as fair? The current study provides an in-depth understanding of how judges view and interpret the fairness of their own behavior.
One of the reasons why we believe it is important to study the judicial perspective on procedural justice is because it tells us something about consistency across different judges and case types. Researchers seeking to explain judicial behavior are often motivated by the search for consistency within the behavior of jurists (Collins, 2008). After all, the principle of consistency implies that like cases are treated alike, both over time and between individuals (Leventhal, 1980;Steyn, 1997). There is, however, no such thing as a general duty for judges to act consistently because that would strike at the heart of their discretion. Legal professionals working in courts have considerable discretion in what they decide and how they reach decisions. Consequently, judges routinely disagree as to how the law should be applied in a given case and as to how it should be interpreted (Collins, 2008). Does the judge's discretion inherently imply a difference in how they treat litigants who appear in different court cases too? If we were to sit in different courtrooms and watch different judges, we may well see that these judges run their courtrooms in different ways (Tyler & Bies, 1990). Our study aims to explore these potential differences between judges by examining how judges perceive their behavior during the court hearing.
Furthermore, judges are, now more than ever, concerned with treating defendants fairly (Burke & Leben, 2007). This raises the question of whether the way legal professionals perceive the way they act in the courtroom corresponds with the way they are perceived by citizens. Is there, indeed, a robust connection between the level of fairness as perceived by the judge and the level of fairness as perceived by the litigant? An important aim of our study, therefore, is to examine whether and how judicial and litigant perceptions of fairness are related.

The Current Research
Taken together, this paper focuses on judicial perceptions of procedural justice. Our main aim is to understand perceptions of the court hearing from the perspective of judges. More specifically, we want to examine how fair judges think they treat litigants. We also take into consideration how important judges consider the outcome for litigants to be. Therefore, we will examine how judges rate their cases on an outcome importance scale and how they rate themselves on a procedural justice scale, and whether these judges differ in their scores on these scales. We expect judges to consistently rate themselves high on procedural justice. After all, it is conceivable that judges strive to treat litigants equally fairly, which takes up such a large part of their training and organizational culture.
Judges normally handle different cases during their court hearings. For example, an average single-judged court hearing in the Netherlands often consists of multiple law cases adjudicated consecutively by the same judge. These law cases are therefore nested within judges, with law cases and the judges who handle these cases defined at separate levels in multilevel analyses (Hox, 2010). In other words, judicial variables in courtroom research can be observed at the case level (i.e., the level of an individual law case) and at the judge level (i.e., the level of a judge who handles several law cases). When we want to know whether judges differ in how they rate themselves and how they rate the cases they handle, it is interesting to examine whether these differences reflect differences between judges and/or differences between the specific cases in which a judge is involved. In this paper, we will use multilevel analyses to examine these issues.
An important reason to use multilevel analyses is that focusing on the perceptions of individual judges and litigants either at the aggregate level of judges or at the level of individual cases would yield an incomplete picture. For instance, procedural justice perceptions of litigants involved in law cases that have all been adjudicated by the same judge may be similar to each other because it is the same judge who is evaluated by the litigants. This raises the question of whether all variation in litigant perceptions of procedural justice can be explained at the case level. Although much can be learned by examining the variation across cases in our variables, it is also necessary to explore whether the fact that these cases have been adjudicated by different judges influences variation in our variables.
The main aim of our multi-level approach is to examine whether judicial perceptions of outcome importance and procedural justice are associated with litigant perceptions of outcome importance and procedural justice. We believe that it is worth considering the empirical linkage between judicial perceptions and litigant perceptions of the same court hearing. After all, one can wonder whether an actor's initiating action, such as fostering justice, is or is not closely related to a receiver's reciprocating response (Cropanzano et al., 2017). We will briefly elaborate on the empirical linkages between judicial perceptions and litigant perceptions that we propose with our hypotheses.
We included outcome importance in our conceptual model because we believe that when outcomes are considered to be more important, litigants will probably pay more attention to how fairly they are treated by important legal authorities (Grootelaar & Van den Bos, 2018;Tyler, 1988). We argue that judges, too, will pay more attention to how fairly they treat litigants when they consider the outcome to be important for these litigants. Judges are experienced and trained professionals who will be able to assess the characteristics of each case. As such, we believe judges are able to reasonably assess what litigants have at stake. Hypothesis 1 therefore proposes that litigant perceptions of outcome importance will be positively associated with judicial perceptions of outcome importance. When litigants perceive their outcome as relatively important, judges will also regard the outcome to be relatively important for the litigant.
We argue that when outcomes are more important for litigants, issues of fair treatment become more important. We expect judges to anticipate this and pay more attention to how fairly they treat litigants for whom they believe there is much at stake. For example, when a judge knows that there is much at stake for a specific litigant, he or she may give this litigant relatively more opportunity to voice his or her opinion on the case during the court hearing. Therefore, Hypothesis 2 proposes that judicial perceptions of outcome importance will be positively related to judicial perceptions of procedural justice. When judges regard the outcome as relatively important for the litigant, they will also perceive their own treatment of the litigant as relatively fair.
The main aim of the current study is to examine whether it is the case that when judges perceive their treatment of the litigants as more fair that the litigants will also perceive the judges' treatment as more fair. In other words, is the degree to which judges perceive their behavior as fair significantly related to how fairly litigants' feel treated by these judges? Hypothesis 3 therefore tests whether judicial perceptions of procedural justice are positively related to litigant perceptions of procedural justice.
The well-known and often-studied positive influence of perceived procedural justice on a variety of people's reactions (for an overview, see Lind & Tyler, 1988) suggests that treating litigants fairly positively affects their evaluations of the legal authorities with whom they interact. Building on this insight, we believe that being treated fairly positively affects how much trust litigants put in judges. Thus, Hypothesis 4 predicts that litigant perceptions of procedural justice will be positively associated with trust in judges. Litigants who perceive higher levels of procedural justice will also perceive higher levels of trust in judges.
In sum, we propose that litigants' perceptions of the importance of their outcomes is related to judges' perceptions of the importance of litigants' outcomes, which in turn influences how fairly the judges treat these litigants. We further propose that the extent to which judges perceive the way they treat litigants as fair is positively related to the way litigants perceive procedural justice, which in turn affects their trust in judges. We will examine these relationships in the context of Dutch courtrooms with litigants involved in three different types of law cases. Conducting our study in this real-life courtroom context means that different kinds of courtroom characteristics (i.e., legal assistance, prior court experience, the judge's gender and litigant demographics) may affect judicial perceptions of the case and the treatment of the litigant involved. For that reason, we will explore the role of these courtroom characteristics for the relationships found.

Sample
Respondents in our sample were litigants and judges involved in the same court hearing at the district court of the Mid-Netherlands. The district court is a relatively large court of first instance. The 175 judges working at the court deal with approximately 170,000 civil law, criminal law, and administrative law cases a year. These cases are assigned to the various divisions by the management board. Within each division, cases are assigned by a coordinating judge, with the assistance of a court clerk. In the Netherlands, cases are assigned to judges based on the kind of procedure, the judge's expertise and skills, judicial continuity and randomness. We chose this district court because we carried out our former field study here and thus had good ties with it. We gained the court's permission to conduct our study there.
Our sample of 38 judges consisted of 9 men (23.7%) and 29 women (76.3%). Ten of these judges (26.3%) adjudicated landlord-tenant cases. Eight of these judges (21.1%) adjudicated bankruptcy cases. Twenty of these judges (52.6%) adjudicated administrative law cases. The judges involved in our study adjudicated an average of 5.4 (SD = 3.7) cases that were included in our study. Three judges adjudicated only 1 case in our study, and the maximum number of cases in our study adjudicated by one judge was 16. Because the team of bankruptcy judges was relatively small, 4 judges involved in bankruptcy cases adjudicated more than 10 cases in our study.
As noted, we selected three types of law cases for our study. In bankruptcy cases (N = 78), litigants were ordered to appear before the district court for a court hearing as they had requested a court decision allowing them access to a legal regime of debt adjustment of natural persons under the Dutch Bankruptcy Act. On the basis of this act, judges can only approve the petition for access to the debt adjustment regime if the judge believes that this individual debtor has entered his or her debts "in good faith" and that this individual debtor can be trusted to successfully fulfill the terms of a debt repayment plan over a period of years. In order to earn this fresh start of debt relief, the individual is required to devote all of their disposable income to the repayment of creditors and to make their best effort to pay off as much of the debt as possible over a threeto-five-year period. Litigants in bankruptcy cases are often assisted by a social worker or a community-care worker.
In landlord-tenant cases (N = 33), litigants were ordered to appear before the district court for a court hearing as their landlord wished to terminate the tenancy agreement and the tenant did not agree. The judge then needed to decide whether there were payment arrears and whether the amount justified termination of the tenancy agreement. In general, it is assumed that there must be a minimum of three months of delay in payment. During the court hearing, the judge investigated the possibilities to settle the dispute between tenant and landlord. Litigants in tenancy cases often defend themselves.
In administrative law cases (N = 96), litigants applied for a judicial review of decisions made by administrative authorities. These cases predominantly concerned social security issues such as social benefits, social support and tax surcharges. The court hearing in these types of cases is often used by the judge to ask questions to both the representative of the administrative authority and the respondent, sometimes represented by a lawyer. The administrative law judge can declare appeals well-founded, unfounded, or inadmissible.
These three types of law cases share important similarities but differ on key points too. These cases all involve an interaction between a single judge and a litigant, often in the presence of a professional such as the representative of an administrative authority. We explicitly chose to include cases in which the litigant had no personal relationship with the other party. Bankruptcy cases, however, are not a classic two-party adversarial process in the way landlord-tenant and administrative law cases are.
In all three types of law cases, judges can decide to give an oral verdict at the end of the court hearing. When they do not give an oral verdict, judges in the bankruptcy court decide cases within eight days after the court hearing has taken place. Subdistrict-court judges decide landlord-tenant cases in a written judgment two to four weeks after the court hearing has taken place. Administrative law judges usually decide upon cases in a written judgment six weeks after the court hearing has taken place. Thus, although we did measure outcome favorability, we refrained from testing associations between outcome favorability, procedural justice and trust in judges. After all, not all judges in our study gave an oral verdict during the court hearing, which resulted in a relatively small number of cases (N = 49) in which outcome favorability was assessed, so we were not able to test those associations in a reliable manner.

Research Procedure
Judges involved in bankruptcy cases, landlord-tenant cases, and administrative law cases (i.e., the three types of law cases that were selected for this research) were asked to participate in our study and were informed that data would be reported anonymously and in aggregate only. The judges involved in the 207 cases of this study all agreed to participate, resulting in a 100% response rate. Of the 207 questionnaires, in the end a total of 199 were filled out by the judges (96.1%). Eight questionnaires were not filled out because the relevant judge was ill, had been replaced, or had forgotten to fill out the questionnaire. The pre-hearing questionnaire was filled out prior to the court hearing and asked judges what they thought was at stake for the litigants involved. The post-hearing questionnaire was filled out when litigants left the courtroom, and measured perceptions of procedural justice, that is, the degree of procedural justice judges thought that they had given to litigants.
Litigants who were scheduled to appear at a court hearing in bankruptcy cases, landlord-tenant cases, and administrative law cases between February 2 and June 14, 2017 were approached by the first author while they were waiting in the hallway of the court building for their court hearing to begin. Litigants involved in these cases were asked to participate in a study about their courtroom experiences and were informed that their identities would remain anonymous, that data would be reported in aggregate only, that the study was conducted independently of the court, and that only researchers at the university responsible for conducting the study would have access to the data. All in all, of the 286 litigants approached, 207 agreed to participate, resulting in a 72.4% response rate.
The method of data collection in our study was such that collecting data among the litigants was independent from the data collection among the judges. It was only after data among both types of respondents were collected that we assessed whether there was a match between a judge and a litigant being involved in the same court case. We also approached all possible litigants and judges and asked them to participate in our study. It is of course possible that, due to practical circumstances in the courthouse, we may have missed some potential litigant respondents, but this constituted the procedure of data collection.
The pre-hearing questionnaire was filled out prior to the court hearing and asked respondents what they had at stake. The post-hearing questionnaire was filled out when respondents left the courtroom after they had appeared before the judge and measured how fairly respondents felt that they had been treated by the judge and how much trust they had in judges. Respondents were also asked for personal information, including age, gender, income, and educational level. We report all measurements in our study, so we note that we used 18 items in the post-hearing questionnaire that measured other reactions, such as litigants' willingness to accept the court's decision and litigants' trust in the Dutch judiciary. These items were measured after the variables reported here, were included for exploratory purposes, and did not affect the findings reported here. 1 All questionnaires were conducted in Dutch and the stimulus materials are available on request.
1 Conceptually, decision acceptance and trust in the legal system are different variables, that cannot be easily combined together. Furthermore, in our study decision acceptance was assessed only among those litigants who already knew the decision of the judge when filling out the post-hearing questionnaire (N = 60). Both decision acceptance and trust in the legal system were moderately to highly correlated with litigant procedural justice (respectively, r = .62 and r = .38), and with trust in judges (respectively, r = .57 and r = .57). When trust in the legal system is included in the model instead of trust in judges, the results are similar, that is, only litigant procedural justice contributed to trust in the legal system (judge level, b = .59, SE = .13, p < .001; case level, b = .44, SE = .09, p < .001).
After filling out the questionnaires, respondents were informed that they could give their e-mail address if they wanted to be informed of the results of our study. One month after we completed the final analyses, we debriefed these respondents by sending them an e-mail summarizing our results.

Compliance with Ethical Standards
We declare that we have no conflict of interest in our research that involved human respondents. Before taking part in our study, potential respondents were asked for their informed consent: All respondents were informed that participation in our study was voluntary. Furthermore, judges were informed that data would be reported anonymously and in aggregate only. Litigants were informed that their identities would remain anonymous, that data would be reported in aggregate only, that the study was conducted independently of the court, and that only researchers at the university responsible for conducting the study would have access to the data. It was only after receiving this information and respondents agreed to participate in the study that data collection actually started. All procedures performed in our study adhere to the tenets of the ethical principles of the American Psychological Association and the Declaration of Helsinki.

Litigant Perceptions of Outcome Importance
To assess what litigants had at stake before entering the court room, we constructed a 4-item scale with items inspired by Brockner (2010) and Tyler (1987) consisting of the following statements: "The outcome in this case is very important to me," "There is a lot at stake in this case for me," "My financial well-being depends on the outcome of this case," and "The outcome is important for me in order to move on with my life." All responses in our study were measured using 7-point Likerttype scales (1 = strongly disagree, 7 = strongly agree). Cronbach's alpha was high (α = 0.89). The items were averaged to yield a scale for litigants' perceptions of outcome importance with higher scores indicating more at stake for the litigant. There was no significant variance at the judge level for litigant perceptions of outcome importance, the intraclass correlation ICC = 0.04.

Judicial Perceptions of Outcome Importance
We asked judges to assess what they thought that litigants had at stake before entering the courtroom. Inspired by Flynn and Brockner (2003), who also included the giver's and receiver's perspective in their study, we used the same items as presented to litigants and rewrote them for judges. We constructed a 4-item scale consisting of the following statements: "The outcome in this case is very important to the litigant," "There is a lot at stake in this case for the litigant," "The litigant's financial well-being depends on the outcome in this case," and "The outcome is important for the litigant in order to move on with his or her life." All responses were again measured using 7-point Likert-type scales (1 = strongly disagree, 7 = strongly agree). Cronbach's alpha was high both within and between judges (α w = 0.90; α b = 0.88). The items were averaged to yield a scale for judicial perceptions of outcome importance with higher scores indicating more at stake for the litigant. The intraclass correlation ICC = 0.46, which means that almost half of the variance in judicial perceptions of outcome importance was systematic variance between judges.
We note that, although we use the overarching term "litigant" here, we ensured that each judge received a tailor-made questionnaire in which the right terminology for their specific type of law case was used. Litigants in landlord-tenant cases are summoned by the landlord to appear in court and are therefore called "defendants." In bankruptcy cases, litigants themselves request access to a legal regime of debt adjustment and are therefore called "applicants." In administrative law cases, litigants themselves apply for judicial review of decisions made by administrative authorities and are therefore called "claimants."

Litigant Perceptions of Procedural Justice
Our measure of litigants' perceptions of procedural justice was based on earlier literature (Grootelaar & Van den Bos, 2018;Lind & Tyler, 1988;Lind et al., 1993;Van den Bos et al., 2014) and asked respondents to indicate to what extent they agreed with the following 11 statements: "I was treated in a fair manner," "I was treated in a polite manner," "The judge was impartial," "I was able to voice my opinions," "My opinion was seriously listened to," "I was treated in a just manner," "I was treated with respect," "The judge has carefully studied my case," "The judge who handled my case was competent," "I believe the judge has treated me in the same way as others," and "The judge who handled my case was professional." Cronbach's alpha showed the procedural justice items had strong internal consistency (α = 0.95). The items were averaged to construct a scale for litigants' perceptions of procedural justice with higher scores indicating more positive litigant evaluations of procedural justice. There was no significant variance at the judge level for litigant perceptions of procedural justice, the intraclass correlation ICC = 0.13.

Judicial Perceptions of Procedural Justice
We asked judges about the degree of procedural justice given to litigants during the court hearing. Our measure of judicial perceptions of procedural justice largely reflected the litigant's perception of procedural justice measure: we rewrote the perceived procedural justice items to match the judge's perspective (Flynn & Brockner, 2003). We explicitly instructed judges that these items concerned the way they behaved themselves and that these items did not concern how they thought that their behavior was perceived by litigants. We asked the judges to indicate the extent to which they agreed with the following 11 statements: "I have the feeling that I treated the litigant in a fair manner," "I treated the litigant in a polite manner," "I have the impression that I acted impartially," "I made sure that the litigant could voice his or her opinions," "I seriously listened to the litigant's opinion," "I believe that I treated the litigant in a just manner," "I treated the litigant with respect," "I carefully studied the litigant's case," "I acted in a competent way," "I believe that I treated the litigant in the same way as others," and "I acted in a professional way." Cronbach's alpha showed that the procedural justice items had strong internal consistency both within and between judges (α w = 0.86; α b = 0.95; see Appendix for more information). The items were averaged to construct a scale for judicial perceptions of procedural justice with higher scores indicating more positive judicial perceptions of procedural justice. The intraclass correlation ICC = 0.67, which means that two thirds of the variance in judicial perceptions of outcome importance was systematic variance between judges.

Trust in Judges
We assessed litigants' trust in judges as directly and precisely as possible by asking them to indicate their level of agreement with the following six statements: "I have confidence in this judge," "This judge is someone I trust," "I find this judge reliable," "I do not trust this judge," "I am confident that the judge has taken the right decision," and "I have the feeling that I cannot trust this judge." If necessary, items were reverse scored. Higher scores on the scale reflect a higher degree of trust in judges. The items demonstrated strong internal consistency (α = 0.90). The items were averaged to yield a scale for trust in judges. There was no significant variance at the judge level for trust in judges, the intraclass correlation ICC = 0.10.
We also assessed two types of background variables. First, we measured whether or not people had legal assistance and former court experience. Second, we assessed the demographic characteristics of the litigants. The litigants were asked to indicate their gender, age, education, and income.

Multilevel Analyses
The litigants who participated in our study were involved in 207 law cases, which were adjudicated by 38 judges. The number of cases per judge varied from 1 to 16 (M = 5.4, SD = 3.7). This means that the cases in our research were nested within judges. As a consequence, our observations at case level may not be completely independent. For example, it is likely that litigants whose cases were adjudicated by the same judge are more similar in their perception of procedural justice than litigants whose cases were adjudicated by a different judge. Moreover, judges are likely to differ from each other in how they perceive their cases and their own behavior in these cases. Computing intraclass correlations (ICCs) of both judicial and litigant perceptions revealed that part of the variance of our variables is indeed explained by the nested structure of our data.
To do justice to this nested structure, we used a multilevel approach. In particular, using multilevel analyses we tested our hypotheses in two ways: at the judge level (averaged over cases), and at case level (within judges). These analyses were then used to explore the linkage between judicial perceptions and litigant perceptions both at case level and at judge level. We tested our conceptual model using a hierarchical regression in the multilevel analysis, in which we entered the variables step-by-step, starting with the first variable in the model (litigant perceptions of outcome importance) and working towards the end (litigant procedural justice). Technical details of our multilevel analyses can be found in the Appendix to this paper. We note that, due to occasional missing values, the number of cases vary across our analyses from 154 to 176 cases.

Descriptive Statistics and Correlations
Descriptive statistics and correlations for our main variables averaged at the judge level (N = 38) are presented in Table 1. This table shows that both judges and litigants perceived both the importance of the law case and the level of procedural justice as relatively high on scales from 1 to 7. The highest values were found for how judges perceived the procedural justice in treating their litigants (M = 6.14, SD = 0.46). The relatively low standard deviation shows that judges did not differ strongly on the procedural justice scale. That is, none of the judges scored lower than 5 on our 7-point Likert scales, and 68.4% of the judges rated themselves a 6 or higher on the 7-point procedural justice scale. Thus, judges perceived the way they treated litigants as relatively fair, and none of the judges indicated that they treated litigants as relatively unfair.
Judges who perceived the outcomes of litigants as relatively important were also more likely to perceive the way they treated litigants as fair (r = 0.33, p = 0.045). Furthermore, Table 1 shows that the perceptions of judges and litigants of the outcome importance of the case were related (r = 0.38, p = 0.018). Furthermore, judges who perceived outcomes of litigants as relatively important were also perceived as more procedurally fair by litigants (r = 0.35, p = 0.033) and these litigants trusted their judges more (r = 0.34, p = 0.035). There also was a strong relationship between litigants' perception of procedural justice and their trust in judges (r = 0.80, p < 0.001), indicating that judges with litigants who perceived high procedural justice also were more trusted by these litigants.

Relationships Between Judicial and Litigant Perceptions
Because the most important aim of our study was to examine whether and how judicial and litigant perceptions of fairness are related, we performed multilevel analyses to explore the linkage between judicial perceptions and litigant perceptions both at case level and at judge level. The data were analyzed using multilevel analyses in SPSS 24, with the Mixed Model procedure and using Maximum Likelihood estimation. The results of these analyses are summarized in Table 2. As noted earlier, technical details of the multilevel analyses that we conducted are presented in the Appendix. Consistent with Hypothesis 1, the first column of Table 2 shows that judges thought the outcome was more important for a litigant when litigants indeed found the outcome of their cases to be more important (judge level, b = 0.54, p = 0.004) and as the litigant in this specific case found the outcome more important than other litigants of this judge (case level, b = 0.11, p = 0.013). This suggests that judges' assessments of how important outcomes are for their litigants correspond significantly with the outcome importance as assessed by the litigants.
In line with Hypothesis 2, the second column of Table 2 shows that judges rated their own behavior as more procedurally fair as they rated the outcomes of their cases on average as more important (judge level, b = 0.19, p = 0.037) and as they perceived the outcome for a specific litigant as especially important (case level, b = 0.08, p = 0.027). This suggests that judges indicate to act more fairly when the stakes are high for a litigant.
As the third column of Table 2 shows, the extent to which judges considered themselves to be acting procedurally fair was not significantly related to how fair litigants perceived they had been treated by the judge. This was found at both the aggregate judge level (b = -0.29, p = 0.227) and the individual case level (b = 0.35, p = 0.192). Thus, in terms of Hypothesis 3, our multilevel analyses show that judicial perceptions of procedural justice are not related in statistically significant ways to litigant perceptions of procedural justice. This indicates that how procedurally fair judges believe to act is statistically unrelated to how this is perceived by the litigant. This is an important finding to which we will come back in our Discussion.
Consistent with Hypothesis 4, the fourth column shows that litigants were more likely to trust their judge when their case was treated by a judge who on average was regarded as procedurally fair by litigants (judge level, b = 0.76, p < 0.001) and when they regarded the procedural justice in their case as higher than other litigants of this judge (case level, b = 0.74, p < 0.001). This shows that litigants' trust in judges is strongly related to the procedural justice they perceive.

Exploring the Role of Contextual Factors
We exploratively tested the role of contextual factors in our conceptual model. That is, we tested whether litigants' age, gender, educational level, income level, legal assistance, and prior court experience, and the judge's gender influenced our main variables by adding these context variables to the multilevel regressions as presented in Table 2.
Adding the context variables to the multilevel regression of litigant perceptions of outcome importance revealed a statistically significant contribution of type of law case, F(2, 146) = 6.64, p = 0.002, where litigants involved in landlord-tenant cases (b = 0.76, SE = 0.36, p = 0.034) and bankruptcy cases (b = 0.89, SE = 0.25, p < 0.001) considered the outcome more important than in administrative law cases (the reference group). Adding the context variables to the multilevel regressions of judicial perceptions of outcome importance, judicial perceptions of procedural justice, and litigant perceptions of procedural justice did not yield any significant relationships of these context variables with the main variables. However, adding the context variables to the regression of trust in judges revealed a statistically significant association between the judge's gender and trust in judges (b = 0.42, SE = 0.15, p = 0.007), showing that litigants involved in cases adjudicated by a male judge were more likely to trust judges than those involved in cases adjudicated by a female judge. Moreover, prior court experience contributed to the multilevel regression of trust in judges (b = 0.25, SE = 0.12, p = 0.036. Litigants trusted judges more when this was their first court experience than when they had prior court experience.
Adding the context variables did not meaningfully change the main results shown in Table 2, with one exception: Case level judicial outcome importance no longer significantly contributed to higher judicial procedural justice after controlling for these context variables. All other results in Table 2 remain essentially the same. Details of these multilevel analyses including the context variables are shown in the Appendix, Table 7.

Discussion
Our research makes several useful contributions, in particular to the study of procedural justice in the legal domain. First, we added a judicial perspective on procedural justice to the dominant "litigant only or predominantly" perspective. Second, we examined linkages between judicial perceptions of procedural justice and litigant perceptions of procedural justice by testing these empirical relationships with multilevel analyses. Third, our research revealed how factors related to the courtroom specific context of our study influenced our main variables.
Perhaps the main contribution of the multilevel approach to judicial and litigant perspectives adopted here was that judges' perceptions of outcome importance hang together significantly with litigants' perceptions of outcome importance. In contrast, judges' perceptions of procedural justice were not significantly associated with litigants' perceptions. This indicates that how fairly judges think they handle their cases does not correspond with litigants' assessment of procedural justice in these cases. In what follows we examine the implications of what we have found. We also discuss the limitations as these suggest avenues for future research.

Judicial Perceptions of Justice
With this paper, we aimed to understand judicial perceptions of the justice system. More specifically, we aimed to examine how judges rated themselves on how fair they treated litigants, and how they estimated what litigants had at stake. Furthermore, we examined which part of the variance in these variables could be explained by the fact that judges rated themselves differently on our measures, and which part of the variance could be explained by the fact that these judges were involved in different cases.
Our results first of all show that judges were relatively consistent in rating themselves on both the outcome importance scale and procedural justice scale. For example, none of the judges indicated that they treated litigants relatively unfairly, and the majority rated themselves a 6 or higher on both the 7-point outcome importance and procedural justice scale.
We found that judges systematically differed in how important they considered the litigants' outcome, although more than half of variance in judicial perceptions of outcome importance is between cases within judges. In other words, how important judges considered the outcome to be for litigants for a substantial part reflected the differences between the specific cases.
Interestingly, we obtained different results with regard to how our judges rated themselves on the judicial perceptions of procedural justice scale. That is, we found that the majority of variance in judicial perceptions of procedural justice is between judges and not between cases within judges. This means that the variance in judicial perceptions of procedural justice largely reflects the differences between judges in how they rated themselves on average on fairness rather than the differences between the specific cases within these judges. In other words, some judges rated their own behavior higher in terms of procedural justice than other judges, regardless of the law case in which they were involved.
In short, how judges rated their cases on the outcome importance scale largely reflected the differences between specific cases and not between judges. In contrast, how judges rated themselves on the procedural justice scale is mostly due to how fair they perceive their own behavior on average, regardless of the specific law case in which they are involved. If this difference would be robust in future research, how can we explain it?
One starting point to understand this potential difference is the reasonable assumption that judges strive to treat each litigant equally fairly. After all, this is something courts and judges in courts typically aspire to (Ellsworth & Mauro, 1998). Building on this assumption it is understandable that judges consistently rated themselves relatively high on the procedural justice scale. Anything else would suggest that judges deliberately view themselves as treating some litigants less fairly than others. It is at the same time understandable that judges find it easier to differentiate between cases on the outcome importance scale. After all, some litigants do consider their case as more important than others, but one would assume that all litigants consider it important to be treated fairly. Thus, judges' professional role conception, and their key responsibility to treat litigants in a fair and just manner may explain why they consistently rated themselves relatively high on the procedural justice scale.
We note here that examining the resemblance of cases at the judge level (i.e., the ICC) and describing how consistent judges scored on our justice measure is one way to examine the principle of consistency. Of course, consistency is a broad concept in the study of law that can be examined in various ways. The principle of consistency can also be studied by examining, for example, variability in judicial decision making and judicial sentencing behavior. Future research could study this issue, for instance by assessing consistency between judges in complex legal cases that involve more than one judge per case. Future research could also examine consistency in judicial sentencing over time. These avenues for future research may be implications of the research we put forward here. We hope that by providing insight into the question of how consistently judges rated themselves on our measures, and by examining this issue both at the level of specific cases and at the level of the judge, we are contributing insights valued by legal scholars.
Differences between judges in how they rate their own procedural justice behavior may only partly be related to their actual behavior (as evidenced by the absence of a relationship with the procedural justice ratings by litigants). Individual differences between judges in how they interpreted and used the scale, in modesty or selfknowledge, or in differences in impression management or experienced social desirability may also have influenced the observed differences between judges in judicial procedural justice.

Linking Judicial Perceptions to Litigant Perceptions
Procedural justice research has largely taken a one-sided perspective on the dynamics of social relationships by focusing on one actor's role only (Korsgaard et al., 1988;Luo, 2005). Procedural justice studies in the legal field mostly used only one data source, such as subjective ratings of court litigants (Beier et al., 2014). In contrast, the current study took a two-sided perspective: we used a multisource approach to analyze the relationship between judicial perceptions and litigant perceptions of justice.
With our multilevel approach we linked the perceptions of litigants to their judicial counterparts. We proposed that judicial procedural justice perceptions were influenced by judicial outcome importance perceptions, which in turn were related to litigant outcome importance perceptions. Indeed, we found that when litigants perceived their outcome as more important, judges also tended to perceive litigants' outcome as more important. We further found that judges rated themselves higher on the procedural justice scale when they perceived the litigant's outcome to be more important.
Furthermore, on the basis of the procedural justice literature it perhaps would be expected that judicial procedural justice perceptions would be related to litigant perceptions of procedural justice. We did not observe this relationship between judicial and litigant perceptions of procedural justice in our study, whereas outcome concerns of judges and litigants were significantly related to each other. Thus, how fair judges perceived their own behavior was not related to how fair litigants perceived these judges.
Recognizing that interpreting null effects can be a tricky thing, we argue that this is an interesting observation, as it suggests that apparently judges' evaluations of how fairly they as trained legal professionals treated litigants is not reflected by litigants' evaluations of the same interaction.
This absence of relationship between judicial and litigant procedural justice perceptions is especially interesting given the fact that we did observe in our study a significant relationship between litigant perceptions of outcome importance and judicial perceptions of outcome importance. This suggests that judges can reasonably estimate what is at stake for the litigants involved. Judges, however, systematically differed in the level at which they rated themselves on the judicial perceptions of procedural justice scale, regardless of the specific law case in which they were involved, whereas the majority of the variance in litigant perceptions of procedural justice was explained between cases within judges. This may be an important reason why judicial and litigant perceptions of procedural justice were unrelated in our study.
We further emphasize that individual perceptions of procedural justice are the product of a complex and nuanced psychological process and not just the outcome of a onetime application of the procedural justice criteria (Rupp et al., 2017). In other words, the fact that judges rated themselves relatively high on procedural justice criteria and that litigants perceived the way they were treated as relatively fair does not mean that these two perceptions of the same interaction are empirically correlated concepts.
Furthermore, the perceptual focus of litigants and judges may be different, which could perhaps even create two psychologically different realities. It seems possible that judges' psychological distance from the case affects their perceptions of the court hearing in that they perceive the hearing as more decontextualized and abstract than litigants do (Liberman & Trope, 1998;Liberman et al., 2007;Trope & Liberman, 2000). For example, attorneys view court procedures as fairer than members of the public because the procedures are not as critical for them as they are for the public (Rottman, 2007). Instead, attorneys and judges have been found to pay more attention to outcome concerns (Heuer, 2005;Rottman, 2005; see also Heuer et al., 2007). After all, judges are trained to focus primarily on the relevant legal issues and to provide fair outcomes (Burke & Leben, 2007).

The Courtroom Context of Our Study
The courtroom context of our study conducted in single-judged court hearings in the Netherlands enabled us to examine on an exploratory basis whether several courtroom-specific characteristics in our study influenced our main variables. A remarkable result of our study is that the judge's gender played an important role. Litigants involved in cases adjudicated by a male judge were more likely to trust judges than those involved in cases adjudicated by a female judge. Furthermore, litigants trusted judges more when they had no prior experience in the court room.

Implications
From a theoretical standpoint, examining the relationship between judicial and litigant procedural justice assessed at the same measurement moment can provide a direction for building models of fair treatment that depart from the dominant models of treatment fairness, which often incorporate only litigant perceptions. One interpretation of our data is that how judges reflect on their own actions and how fair litigants perceive the same actions do not always correspond. This is interesting because it involves two individuals who evaluate the same court hearing independently of each other. Of course, procedural justice is in the eye of the beholder ) but that does not mean that there are two separate versions of the same reality. After all, our research showed that judicial perceptions and litigant perceptions are related when it comes to outcome importance. This means that to develop a full and more detailed picture of how these perceptions are related, additional research will be needed. We believe that the joint examination of judicial and litigant perspectives of justice may help to refine research on procedural justice in the legal field.
Gaining insight into the differences between lay and expert justice perceptions also has practical importance because these differences point to areas in which expert-designed procedures might fail to meet with approval on the part of lay people using the procedure . After all, possible differences between judges and litigants in how they perceive procedural justice to be enacted in legal cases can have important implications for the operation of court procedures and for innovations to existing procedures. Future research can follow up on these findings and may want to zoom in on the question of what judges and litigants consider to be the most important procedural justice components (such as voice, due consideration, neutrality) and look for differences and similarities in their answers.
The fact that we conducted our study in a real-life courtroom context enabled us to explore the role of contextual factors such as legal assistance and type of law case on the judicial perceptions examined. Although it may not come as a surprise that educational level, income, and type of law case are related variables, including contextual factors yielded some interesting results as well. For example, we found the judge's gender and prior court experience to be related to litigants' trust in judges. These findings may deserve more attention in further work. Future research may benefit from correlational research, despite its limitations, by including these contextual factors and examining their influence on the relationships studied.
We also note that while our focus is on the courts, there is a lot of research on procedural justice and related concepts in other areas of the criminal justice system. Future studies that appeal to that broader literature may increase the relevance of the pattern of findings we reported here (see also Shook et al., 2021).
We also note here that although our present study focused on consistency between judges and litigants, what different judges do, and what various litigants experience, consistency is of course only one aspect of procedural justice (see, e.g., Leventhal, 1980). In fact, theoretically, a judge could be consistently low in procedural justice across cases. Thus, consistency should not be equated with procedural justice (or vice versa). Future research could focus on exploring the relationship between the disentanglement of the relationship between procedural justice and consistency (see also Van den Bos et al., 1996).
Given that a large part of our work focused on perspective divergence between litigants and judges as a focal phenomenon, future research on this issue could profit from psychological theorizing on such divergence and its cognitive and motivational underpinnings. For example, the classic actor-observer bias (Jones & Nisbett, 1972) and more recent work on behavior explanations (e.g., Malle et al., 2007) may be helpful for explaining why procedural justice perceptions diverge more strongly than outcome importance perceptions. Our predictions now focused on what seemed reasonable expectations from what we know on the basis of the literature on perceived procedural justice and outcome importance. The attribution literature may be useful to understand the patterns of results we obtained in our study.
We hasten to note that, of course, we want to prevent the generation of hypotheses after results are known (Kerr, 1998), and this is one important reason why our predictions were quite modest and focused on exploring litigants' and judicial perceptions of procedural justice and outcome importance, and their associations with constructs such as trust. We find it important to honor this partially explorative quality of our study.
This stated, we note explicitly that future studies are needed to explore the implications of the study presented here. This includes the role that attributions play in justice judgments (e.g., Weiner, 1985) and responses that follow once those judgments are made (e.g., Blount, 1995). For example, Malle et al. (2007) note that impression management plays an important role in behavior explanations, and we think managing your impression how you acted and whether you look good may play differential roles among (some) judges and (some) litigants. Paying more attention to what goes on inside the judicial mind may be relevant in this respect (see, e.g., Guthrie et al., 2001Guthrie et al., , 2007.

Limitations
Doing research in this relatively under researched subfield of judicial perceptions of procedural justice in the legal context inherently brings along limitations. First of all, it was a challenge to figure out how to precisely measure procedural justice from a judicial point of view. Because we were not aware of other studies that examined how legal authorities assess the fairness of their own actions, we decided to develop our own judicial perceptions of procedural justice scale: we adjusted the frequently used litigant perceptions of procedural justice items to the specific courtroom context of our study, taking the judge's perspective as a reference point and reflecting the litigants' items (Flynn & Brockner, 2003). Of course, self-report questionnaires are one of the most widely used methods of collecting data in a real-life research context. As with all subjective perceptions used in social science studies that work with self-reporting, our results may be limited by a reference bias indicating that the judges in our study used different standards of comparison.
Furthermore, it could be argued that judges may have answered our questionnaires in a socially acceptable manner. After all, our results indeed show that judges rated themselves relatively high on the judicial perceptions of procedural justice scale (M = 6.15, SD = 0.46) and none of the judges scored lower than 5 on our 7-point Likert scales. Judges are expected to behave fairly and to be concerned about being fair, and it is their job to see that fair procedures are followed . It is therefore not surprising that the judges in our study unanimously rated themselves high on the procedural justice scale. This is an important issue for further research on authorities' perceptions of procedural justice.
The primary focus of our study was both judicial perceptions of justice and their linkage with litigant perceptions of procedural justice and outcome importance. Although we did measure outcome favorability, we refrained from testing associations between outcome favorability, procedural justice, and trust in judges. Because not all judges in our study gave an oral verdict during the court hearing, which resulted in a low number of valid cases for outcome favorability (N = 49), we were not able to test those associations in a reliable manner. The majority of the respondents thus rated their perceptions of procedural justice without knowing the outcomes they came for. Although knowing the outcome may influence the way people judge the fairness of the procedure, research has shown that prior knowledge about the outcome does not necessarily change the weight litigants place on their procedural justice perceptions when forming judgments about trust (Grootelaar & Van den Bos, 2018). Future research may focus more specifically on the role of outcome favorability in this research context, using more robust outcome information.
In addition, we acknowledge that our measure of outcome importance concerned the subjective perceptions that judges and litigants had prior to the court hearing and did not involve the monetary value at stake for the litigants. We assume, however, that the litigants involved in these three types of law cases have high financial stakes because these cases involve issues of debt adjustment in bankruptcy cases, payment arrears in landlord-tenant cases, and social benefits in administrative law cases. Furthermore, the court hearing itself may influence the appreciation of the outcome importance, especially as perceived by the judges. Measuring outcome importance beforehand rather than after the hearing therefore limits the relationships between judicial outcome importance on the one hand and both the litigant's perception of outcome importance and judicial procedural justice on the other. We consider this separation in measurements as a strength of our design, which counteracts our hypotheses that judicial outcome importance is positively related to the litigant's perception of outcome importance and judicial procedural justice.
Finally, we note explicitly that our sample cannot be regarded as a random or representative sample of all law cases in the Netherlands as this sample consists of only three types of law cases in one specific court district. Nevertheless, given the procedure followed in which all applicable litigants were contacted over a substantial period of time, and given the high response rate among both litigants (72%) and judges (100%), we believe selection is only a minor issue in this study. Future studies among different types of law cases, different court districts, and different countries could provide insight into the extent to which our results can be generalized to the perceptions of judges and litigants in general.

Coda
This paper addresses an important and relatively under researched set of questions about court hearings: how do we understand the way judges and litigants perceive the justice system? By delving into judicial perceptions of justice and linking them to litigant perceptions, we extended the literature on procedural justice, which mainly takes a "litigant only" perspective. One of the most notable implications of the findings reported here is that they offer empirical support for the notion that litigants' and judges' evaluations of the same court hearing are, while different, to some extent related. This paper shows the importance of conducting multilevel analyses in a law and society context. In short, the findings presented here have revealed the relevance of broadening the central focus in procedural justice research on differences between litigants and cases with a focus on the systematic differences that may exist between judges who adjudicate these cases. In spite of the limitations of the current research, we think it is safe to conclude that this paper helps us to better understand litigants' and judges' perspectives of procedural justice and the linkages and discrepancies between these perspectives.

Appendix
In this appendix we present the details of the multilevel analyses that we conducted. Each table in the appendix shows the results for one of the dependent variables in our conceptual model. In our model, all independent variables were divided into a judge level variable and a case level variable (Enders & Tofighi, 2007). The judge level variables were the mean values for all cases adjudicated by a specific judge and were grand-mean centered, that is, the overall mean was subtracted in such a way that the average across all judges was 0. The case level variables consisted of the remaining variance of the original variables within each judge and were group mean centered (here: judge mean centered), that is, the mean value for the judge was subtracted from the original variable (Enders & Tofighi, 2007). This means that the judge level variables and the case level variables were completely unrelated (r = 0.00), and all variance of the original variables was represented in the sum of the judge level and case level variables.
We tested our conceptual model using a hierarchical regression in the multilevel analysis, in which we entered the variables step-by-step, starting with the first variable in the model (litigant perceptions of outcome importance) and working towards the end (litigant perceptions in procedural justice). First, for each variable in the study, the Intraclass Correlation (ICC) was computed by estimating the random variance in the intercept, i.e., the between-judge variance in the variable (Model 1). In Model 2, the litigants' perceptions of outcome importance scale (both at judge level and at case level) was entered into the regression. In Model 3, the judicial perceptions of outcome importance scale (both at judge level and case level) was entered into the regression. In Model 4, the judicial perceptions of procedural justice scale (both at judge level and case level) was entered into the regression. In Model 5, the litigants' perceptions of procedural justice scale (both at judge level and at case level) was entered into the regression. Only when the fit of the model improved by an additional step in the regression was the contribution of the added predictors considered. In all analyses full information maximum likelihood estimation was used, and random slopes were restricted to zero.
The Intraclass Correlation (ICC) is a measure of the extent to which different cases of a single judge resemble each other and indicates the proportion of the variance explained by the nested structure of our data. A high ICC shows that there are differences in the level of a variable between judges, and a low ICC shows that there are differences in the level of a variable between cases, within judges.
The ICCs for our judicial variables were significant. For judicial perceptions of outcome importance, the ICC was 0.46, showing that 46% of the variance in outcome importance as perceived by the judge systematically varied between judges. The ICC for judicial perceptions of procedural justice was 0.67, which means that 67% of the variance in how judges perceived procedural justice varied systematically between judges. This means that the majority of the variance in judicial perceptions of procedural justice is between judges, and to a lesser extent between cases within judges.
On the other hand, ICCs for the variables measured among litigants were not significant, which means that litigants did not differ strongly in their perceptions depending on the judge who adjudicated their case. For litigant perceptions of procedural justice the ICC was 0.13, which means that 13% of the variance in litigant perceptions of procedural justice varied systematically between judges. For litigant perceptions of outcome importance, the ICC was 0.04, which means that 4% of the variance in outcome importance varied systematically between judges. For trust in judges, the ICC was 0.10, which means that 10% of the variance in trust in judges varied systematically between judges.
The ICCs of both judicial perceptions and litigant perceptions thus revealed that, to different degrees, part of the variance of our variables is explained by the nested structure of our data. To do justice to this nested structure, we divided all independent variables into a judge level variable and a case level variable. The judge level variables contain the differences between judges, and the case level variables contain the differences between specific cases within judges. Table 3 shows the results for judicial perceptions of outcome importance. We found a statistically significant relationship between judge level litigant perceptions of outcome importance and judicial perceptions of outcome importance (b = 0.54, p = 0.004), indicating that judges considered the litigants' outcomes as more important when the litigants in their cases on average considered their outcomes as more important. In other words, if a judge's caseload consisted of three cases with litigants involved, the average perception of these three litigants was related to the judicial perceptions of outcome importance. Thus, on an aggregate or judge level, we found that litigants and judges correspond in their view on outcome importance.
We further found a statistically significant relationship between case level litigant perceptions of outcome importance and judicial perceptions of outcome importance (b = 0.11, p = 0.013). This indicated that when litigants perceived their outcome in the specific law case as more important, judges also tended to perceive the litigants' outcome as more important. In other words, both the average outcome importance in their cases and the outcome importance in a specific case as perceived by the litigants influenced how important judges rated their cases on the judicial perceptions of outcome importance scale. These findings support Hypothesis 1. Outcome importance as perceived by litigants was positively associated with how important judges thought the outcome would be for the litigant. When litigants perceived their outcome as relatively important, judges also regarded the outcome as relatively important for the litigant. Table 4 shows the results for judicial perceptions of procedural justice. Neither case level litigant perceptions of outcome importance nor judge level litigant perceptions of outcome importance in Model 2 contributed to the regression of judicial perceptions of procedural justice. Adding judicial perceptions of outcome importance in Model 3 showed that both case level judicial perceptions of outcome importance (b = 0.08, p = 0.027) and judge level judicial perceptions of outcome importance (b = 0.19, p = 0.037) were significantly related to judicial perceptions of procedural justice. The positive relationship between case level judicial perceptions of outcome importance and judicial perceptions of procedural justice indicates that judges regarded themselves as being especially fair if they regarded the outcome as more important for a specific litigant. In addition, the judge level relationship with judicial perceptions of outcome importance shows that judges who regarded litigants' outcomes on average as more important perceived themselves as treating litigants more fairly. In other words, the fact that judges rated their behavior as relatively high on the judicial perceptions of procedural justice scale is not only influenced by how important they perceive the outcome for a specific litigant, but also by the fact that some judges perceive litigants' outcomes on average as important. These findings are consistent with Hypothesis 2. Judicial perceptions of outcome importance were positively related to judicial perceptions of procedural justice. When judges regarded the outcome as relatively important for the litigant, they also perceived their own treatment of the litigant as relatively fair. Full information maximum likelihood estimation was used, and random slopes were restricted to zero. Table 5 shows the results for litigant perceptions of procedural justice. Case level litigant perceptions of outcome importance contributed significantly to the regression of litigant perceptions of procedural justice in Model 2 (b = 0.30, p < 0.001), Model 3 (b = 0.32, p < 0.001) and Model 4 (b = 0.31, p < 0.001). These findings indicate that litigants who perceived their outcome in the specific case as more important were more likely to perceive their treatment as fair. Adding judicial perceptions of outcome importance in Model 3 showed that neither case level nor judge level judicial perceptions of outcome importance were significantly related to litigant perceptions of procedural justice. After adding judicial perceptions of procedural justice in Model 4, judge level judicial perceptions of outcome importance were significantly related to litigant perceptions of procedural justice (b = 0.34, p = 0.031). This indicates that judges who perceived the outcome on average to be more important for litigants were more likely to be perceived as fair by litigants. Adding judicial perceptions of procedural justice in Model 4 showed no  statistically significant relationship between judicial perceptions of procedural justice and litigant perceptions of procedural justice. Apparently, whether judges considered themselves as acting particularly procedurally fairly or not in a specific case was unrelated to how fair litigants perceived they had been treated by the judge. This means that we did not find support for Hypothesis 3 that judicial perceptions of procedural justice would be positively related to litigant perceptions of procedural justice. Litigant perceptions of procedural justice could not be explained by how fair judges estimated their own behavior. Table 6 shows the results for trust in judges. Case level litigant perceptions of outcome importance in Model 2 contributed significantly to the regression of trust in judges (b = 0.15, p = 0.030), indicating that litigants who perceive their outcome in the specific case as more important are more likely to trust judges. Judge level litigant perceptions of outcome importance in Model 2 also contributed significantly to the regression of trust in judges (b = 0.31, p = 0.038), indicating that judges with cases that on average are considered to be more important by litigants are more trusted by these litigants. This indicates that litigants who perceived their outcome as more important were more likely to trust judges.
Adding judicial perceptions of outcome importance in Model 3 showed that judge level judicial perceptions of outcome importance was significantly related to trust in judges (b = 0.28, p = 0.043). This finding indicates that judges who on average considered litigants' outcomes to be more important were more likely to be trusted by litigants. This relationship remained significant when judicial perceptions of procedural justice was added in Model 4 (b = 0.35, p = 0.023).
Adding litigant perceptions of procedural justice in Model 5 shows a statistically significant relationship between case level litigant perceptions of procedural justice and trust in judges (b = 0.74, p < 0.001). Litigants who perceived higher levels of procedural justice in their specific case also stated that they had higher levels of trust in the judge who handled their case. We also found a statistically significant relationship between judge level litigant perceptions of procedural justice and trust in judges (b = 0.76, p < 0.001). This indicates that judges who on average are perceived as procedurally fair by litigants are more trusted by these litigants. These findings are consistent with our Hypothesis 4. Litigant perceptions of procedural justice were positively associated with trust in judges. After controlling for litigant perceptions of procedural justice, the relationships of litigant perceptions of outcome importance and judicial perceptions of outcome importance with trust in judges were no longer significant.

Multilevel Reliability
The two judicial variables had substantial variation at the judge level, which is why multilevel analyses were necessary. This also means that reliability of these variables has to be assessed both at the case level (level 1, α w ), and at the judge level (level 2, α b ). Following the procedure described by Geldhof et al. (2014), we Social Justice Research (2022) 35:243-274 Table 6 Multilevel regression analysis of trust in judges Unstandardized regression weights are presented. Full information maximum likelihood estimation was used, and random slopes were restricted to zero *p < .05; ** p < .01; *** p computed the multilevel reliability with multilevel structural equation modeling in R using the Lavaan-package (version 0.6-9). For judicial perceptions of outcome importance (4 items) the model converged, and resulted in reliability estimates at the case level of α w = 0.90 and at the judge level α b = 0.88. However, for judicial perceptions of procedural justice, the model assessing the variance-covariance matrix at both levels did not converge due to the large number of items in the scale (11 items). This large number is especially problematic given the limited number of judges and cases. We therefore constructed 4 indicators of judicial perceptions of procedural justice, each consisting of the mean score of 3 or 2 items. In this way, all 11 items were included. This model converged, and resulted in reliability estimates at the case level (α w = 0.86) and at the judge level (α b = 0.95). Finally, we note that it is not possible to consider the judicial procedural justice scale as a ordinal scale as it is based on the mean value of 11 items. We inspected the distribution of this scale in detail, both at the judge level and at the case level. Despite the high mean value of 6.15, the distribution largely resembled a normal distribution. At the case level (N = 199: skewness = −0.115, SE = 0.172; kurtosis = 0.011, SE = 0.343), at the judge level (N = 38: skewness = 0.168, SE = 0.383; kurtosis = −0.205, SE = 0.750; please note that kurtosis in SPSS has value 0 for a normal distribution rather than value 3). This means that the Maximum Likelihood estimation, which in itself is rather robust against small deviations of normality, is suitable for our analysis.

Influence of Context Variables
In Table 7, details of the influence of context variables on the main analyses are shown.
Social Justice Research (2022) 35:243-274 Table 7 The influence of context variables in the multilevel regressions Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.