Peer Justice and the Individual

Abstract

Peer justice refers to the perceived fairness of the treatment an employee receives from coworkers. Although research has found peer justice to be an important predictor of work attitudes and behavior, prior work has only conceptualized the construct at the unit or group level. This limitation can be problematic because peer justice theories sometimes involve within-person effects. In response, we propose and test an individual level model of peer justice. We then test the model with 100 employees from a luxury tourist resort. Respondents answered a questionnaire on ten consecutive working days, resulting in 1000 daily diary recordings. Within-person variance was significant, not only for daily peer (70%) and supervisor (59%) justice but also for the outcomes (52–70%), thus justifying the study of intraindividual associations. Using growth models, the results showed significant links from daily peer justice to daily changes in overall job satisfaction, in-role performance, and helping coworkers. Each of these effects went beyond the daily treatment received from the supervisor.

Introduction

Organizational justice is a key component of work-team effectiveness. When team members behave fairly toward one another, they are likely to have better work attitudes, achieve higher performance, and perform more organizational citizenship behaviors (OCB, see Colquitt et al., 2002; Li et al., 2015). These findings are especially important as the modern workplace continues to evolve. Ambrose and Schminke (2001) observed that modern business organizations have become flatter, with horizontal peer relationships playing as important a role as vertical supervisory authority. Likewise, Shapiro and Kirkman (1999) proposed and found that justice was important in self-managed work teams where formal management may not be available. Observations such as these have given rise to the concept of peer justice. Investigations of peer justice explore how coworkers within the same unit or work team judge the fairness they receive from their peers (Li & Cropanzano, 2009; Molina et al., 2016). For example, coworkers may perceive that their peers share information and treat them with courtesy, thus enhancing peer justice. The unit or work team can also make decisions that influence the degree to which its members perceive fair treatment from peers. When teammates behave fairly, they are likely to have more effective units and, as discussed below, more comfortable work environments (Li et al., 2013). Despite the importance of these contributions, progress in the study of peer justice has been limited. Based on social exchange approaches to justice that consider the existence of different justice sources (i.e., multifoci), we argue that this work faces a series of conceptual challenges that have limited the utility of peer justice as a scientific construct (Lavelle et al., 2007; Rupp & Cropanzano, 2002; Rupp & Paddock, 2010).

First, peer justice has usually been operationalized as a unit-level construct. This is atypical because it would be reasonable to first investigate peer justice at the individual level before aggregating to the unit level, as was historically done with individual-level experiences and justice climate (Martínez-Tur & Moliner, 2017). However, for some reason, researchers have focused almost exclusively on the unit level and relatively neglected relevant individual-level approaches. It is possible that the focus on the unit level is related to the interest in organizational climates and their importance to organizational behavior (Naumann & Bennett, 2000; Whitman et al., 2012). The review by Li and Cropanzano (2009) exemplifies this very well. Although their theoretical foundations of peer justice were often based on the study of individual experiences derived from the treatment received by peers, they concentrated on peer justice as a climate and paid attention to the shared perception at the unit level.

Second, because peer justice has been considered a relatively stable aspect of a team’s climate, little is known about how it might change over time (cf., Jones & Skarlicki, 2013). However, if peer justice is considered at the individual level of analysis, one would expect a great deal of instability (Rupp & Paddock, 2010). For example, one day an employee may feel that he/she is treated fairly by his/her peers (e.g., with respect and dignity), but the next day he/she may feel that he/she is being treated rudely. Other research has found considerable individual variation in justice perceptions (Loi et al., 2009). Holtz and Harold (2009) reported that some of the variance in supervisor justice was within-person (i.e., employees’ perceptions of their supervisors’ behaviors changed over time). If peer justice can be treated as an individual level perception, it would also be expected to fluctuate within each employee. Furthermore, these fluctuations, like those found by Holtz and Harold (2009), are significant in their own right because they would be expected to predict work outcomes such as daily job performance.

Third, peer justice only pertains to peers. However, other sources of fairness are important. In fact, seminal works on sources of justice (e.g., Cropanzano et al., 2001) have argued that employees are involved in fairness processes and social exchanges with distinguishable actors. Especially relevant is treatment by supervisors. Given that perceptions of supervisor and individual peer justice are likely to be correlated, it is important to consider each beyond the effect of the other (Lavelle, McMahan, & Harris, 2009a, 2009b; Lavelle et al., 2015). Justice models based on social exchanges also raise a fourth concern: a range of important work criteria are impacted by individual level peer justice. These criteria include work attitudes, such as job satisfaction, stress, such as emotional exhaustion, and work behaviors, such as performance and OCB. Below, we consider each of these challenges in greater detail. We begin by briefly describing different approaches to peer justice research. We then turn our attention to peer justice, proposing an individual level approach to its investigation and identifying conceptual weaknesses in current thinking.

Fairness at Work: Individual- and Unit-Level Perspectives

Within the organizational sciences, “fairness” is understood as the way individuals perceive they have been treated by others, usually by those in decision-making positions. It is concerned with aspects of the work environment that are expected to influence fairness perceptions, as well as the consequences of these perceptions (Colquitt et al., 2013; Colquitt & Zipay, 2015). Prior research has shown that fairness can be studied as either an individual level or unit level perspective. In the organizational justice literature, the former perspective is much more extensive. In fact, justice research has evolved over time from the individual level exclusively to a multilevel perspective where the unit level has also acquired relevance (Martínez-Tur & Moliner, 2017). However, as described below, the unit level perspective is more common in studies of coworker fairness. The present study will seek to address this imbalance.

Individual level Approaches to Coworker Justice

Research has identified three or four broad attributes of the work environment that promote fairness (Colquitt & Zipay, 2015). Distributive justice is related to the way outcomes are allocated to individuals. Outcomes are viewed as fair if they conform to accepted distribution rules, such as equity, equality, and need. Procedural justice refers to the fairness of the allocation process, including elements such as consistency, freedom from bias, and voice, among others. Interactional justice involves the fairness of interpersonal transactions. It is sometimes subdivided into informational justice, providing explanations and rationales for decisions, and interpersonal justice, the dignity and respect with which a person is treated (Colquitt et al., 2013; Colquitt & Zipay, 2015). These different antecedents, in turn, combine to produce overall justice (a holistic judgment of justice), which is more proximally related to workplace outcomes (Ambrose et al., 2015).

Various researchers have argued that justice—understood as an individual level construct—can be meaningfully applied to employees working collaboratively in teams (e.g., Colquitt et al., 2002; Cropanzano & Schminke, 2001; Rupp & Paddock, 2010). Evidence appears to support this contention. Branscombe and her colleagues (2002) classified student participants into high status or low status groups. They were then treated either respectfully or disrespectfully. When treated respectfully (high interactional justice), low status individuals were more willing to work hard on behalf of their group. According to the authors, respect from coworkers indicates that even low status individuals have dignity (see also, Spears et al., 2006).

Based on social exchange theory (for a general review, see Cropanzano & Mitchell, 2005), research on multifoci justice (Lavelle et al., 2007; Rupp & Cropanzano, 2002) emphasizes sources of justice. Social exchange theory maintains that individuals reciprocate the good and bad treatment they encounter at work. In doing so, they form close, social exchange relationships with those who treat them justly. However, these relationships are less likely to be formed with parties who behave unfairly. Central to these models is the concept that justice and injustice can come from various sources or foci, such as supervisors, customers, or the organization as a whole (Lavelle et al., 2015). Within social exchange theory, “coworkers” (Cropanzano et al., 2001, p. 184; Lavelle et al., 2007, p. 844) are an important, although understudied, source of fairness, distinguishable from supervisors’ fairness (Rupp et al., 2014; Rupp & Paddock, 2010).

Although limited, evidence is consistent with these ideas. Donovan et al. (1998) found that employees distinguished between the interpersonal treatment they received from coworkers and the treatment they received from their supervisors (cf., Fulmer & Gelfand, 2012). Similarly, Bishop et al. (2000; 2005) observed that employees distinguish the support they receive from their teammates from the support they receive from their organization. These two sources of support are related to team commitment and organizational commitment, respectively. Lavelle et al. (2009a, 2009b, study 1) reported multifoci effects in a study of 106 workers at a medical clinic. Specifically, commitment to the workgroup was a better predictor of organizational citizenship behaviors directed toward individual co-workers (OCBI), whereas commitment to the organization was a better predictor of organizational citizenship behaviors directed toward the business as a whole (OCBO). In their second study, Lavelle et al. (2009a, 2009b) further reported that when workers believed that they were treated fairly by their teammates, they reported higher workgroup commitment and performed more citizenship behaviors that benefited their peers.

In short, the aforementioned social exchange research suggests that the treatment workers receive from their teammates can be an important source of fairness (Cropanzano et al., 2001; Lavelle et al., 2007; 2015; Rupp & Paddock, 2010). Judgements of coworker justice are distinct from those involving supervisors, and they subsequently impact attitudes and behavior (Brandscome et al., 2002; Donovan et al., 1998; Lavelle, McMahan, & Harris, 2009a, 2009b). Despite these arguments based on social exchange theory, as well as the supporting findings, most of the research on teammates has been conducted at the unit level of analysis, as we describe below.

Peer Justice

Peer justice has been conceptualized as “a shared perception regarding how individuals who work together within the same unit and who do not have formal authority over each other judge the fairness of their peers” (Cropanzano et al., 2011, p. 568). The nomenclature can be confusing because previous names for the construct include “intra-unit justice climate” (Li & Cropanzano, 2009, p. 564) and “peer justice climate” (Haider et al., 2020, p. 2). Thus, peer justice was understood to be a unit level construct consisting of overall justice perceptions held by group members and about group members. Paralleling the predictions made by social exchange theory, unit level perceptions of peer justice should be distinct from unit level perceptions of justice climate (see Li et al., 2013; Molina et al., 2015; 2016). Research on unit level peer justice is promising (Li et al., 2015; Wang, 2020). For example, when unit level peer justice is high, work groups show higher performance and more OCB (Cropanzano et al., 2011; Molina et al., 2015), and these findings are stronger when team members have a level of autonomy in achieving common goals (Bosch-Sijtsema et al., 2011; Magpili & Pazos, 2018).

Using Social Exchange Theory to Critique Peer Justice

Following from social exchange theory, which has focused on distinguishable justice sources or foci when employees establish social exchanges (Cropanzano et al., 2001; Lavelle et al., 2007; 2015), we argue that peers are an individual level source of justice, though not the only source. That is, there are different meaningful actors involved in justice, including both peers and authorities (i.e., supervisor). We also recognize that the group level has produced significant advances in peer justice research (Li & Cropanzano, 2009; Li et al., 2015). However, the individual level approach has been neglected. This neglect, in turn, has created four theoretical challenges, which we discuss next.

Theoretical Challenge #1: Conceptual Arguments for Individual Level Peer Justice

Peer justice is typically said to operate at the group level. Some of the proposed theoretical mechanisms would tend to create agreement, such as social information processing (Li et al., 2015; Whitman et al., 2012) and social contagion (Degoey, 2000). However, social exchange theory approaches tend to be based on individual reactions to the treatment workers receive from others (Lavelle et al., 2007; Lavelle, McMahan, & Harris, 2009a, 2009b). This suggests that peer justice may operate through individual-level mechanisms that could lead to dissensus among teammates. For example, in the first published investigation on the topic, Cropanzano et al. (2011) argued that unit-level peer justice is enhanced by creating stronger relationships among team members. Based on social exchange theory, peer justice would be expected to promote reciprocal relationships of trust by building mutually supportive behaviors (Cropanzano & Mitchell, 2005). Social exchange theory has usually been studied at the individual level of analysis. Relying on these psychological processes suggests that unit-level peer justice, at least partially, has individual-level effects. Arguably, there is a mismatch between these individual-level theoretical mechanisms and the unit-level justice among peers. These mechanisms suggest that peer justice should be exhibited in individual employees, and at least a few studies have successfully shown this (Branscombe et al., 2002; Lavelle, McMahan, & Harris, 2009a, 2009b).

Theoretical Challenge #2: Within-Person Variability and Its Consequences

Considering peer justice at the individual level of analyses has implications for its stability. In this regard, peer justice effects have been examined over periods ranging from one to two months (Cropanzano et al., 2011; Li et al., 2013). Peer justice is stable to the extent that the mean levels of aggregated perceptions are similar across time periods. Some individuals have higher levels, whereas others have lower levels. Therefore, people who are treated well (high justice) should have better outcomes than those who are treated poorly (low justice). We have learned a lot from between-subjects research. However, despite the benefits of this earlier work, the effects of within-person variability have probably been understated (cf., McCormick et al., 2020).

According to Rupp and Paddock (2010), within-person variability is important when investigating different justice sources or foci, as recommended by social exchange theory. Each employee is likely to encounter distinct justice events that largely depend on the individuals with whom they are interacting. These distinct justice events are then combined to form perceptions of multifoci justice. Rupp and Paddock’s model is important because it predicts both individual level effects (not everyone has to have the same experiences) and shared, unit level effects (because individual judgements are socially aggregated). Thus, an individual does not generally have a perfectly stable or “set” level of justice perceptions. Instead, perceptions ebb and flow as a consequence of workplace events. Several studies suggest that these within-person justice effects are important, although the evidence for peer justice is limited. For example, Holtz and Harold (2009) assessed overall organizational justice and overall supervisor justice in three waves separated by about 1 month. For organizational justice, 24% of the variance was within-person and for supervisor justice, this figure rose to 29%. More recently, Koopman et al. (2020) conducted two experience-sampling methodology (ESM) studies, both of which found, among other things, that individual justice perceptions varied over time.

Work by Matta et al. (2020) further suggests that justice variability is an important antecedent in its own right. For example, Matta et al. (2017, Study 1) found that research participants who received variable treatment from coworkers, sometimes fair and sometimes unfair, manifested greater stress than did even those who were treated consistently unfairly. If this holds for stress, then similar effects might be observed for other criterion variables (Lavelle et al., 2007). However, because peer justice has not been treated as an individual level construct, the effects of variability within workers has received insufficient attention, which suggests that we may have missed some important consequences of peer justice.

Theoretical Challenge #3: Supervisor Justice and Peer Justice

The social exchange literature argues that both supervisors and peers can be sources of fairness (Lavelle et al., 2015). For example, Yang and Diefendorff (2009) conducted a daily diary study of 231 employees in various occupations. They found that unjust treatment by a supervisor or coworker could incrementally create daily negative emotions. When these emotions occurred, they drove the employees to perform counterproductive work behaviors, especially those who were low in agreeableness and conscientiousness. Given that supervisory actions can impact the justice perceptions of subordinates (e.g., Judge et al., 2006; Loi et al., 2009), this approach requires a control variable at the same level of analysis. Specifically, it is necessary to control for the treatment a target individual receives from the supervisor. Consequently, to demonstrate the effect of peer actions toward an individual coworker, it would be worthwhile to control for the effect of supervisory actions toward that same coworker. We address this in the present study by accounting for supervisor justice in all of our analyses.

Theoretical Challenge #4: Criterion Variables

Based on the work of Lavelle et al. (2015) and Rupp and Paddock (2010), we argue that individual employees benefit from being treated favorably by their coworkers, and these benefits account for variance beyond the treatment they receive from their supervisors. We also propose that this contribution of peer justice exists in the daily episodic social exchanges between employees and their coworkers. That is, daily peer justice leads to daily attitudes, feelings, and behaviors. To clarify this, we return to processes involved in individual social exchanges throughout the day.

We expect changes in peer justice over time to play a meaningful role in job satisfaction. According to the fairness heuristic theory, employees use justice perceptions to guide their attitudes (Cropanzano et al., 2001). Employees rely on (in)justice perceptions to interpret the quality of the social exchange they have with the organization and its agents, responding with better or worse attitudes in the workplace. Based on this argument, Loi et al. (2009) proposed and found positive associations between daily fair treatment by the supervisor and daily job satisfaction. According to Loi and colleagues, this is congruent with the argument that heuristics can be episodic and based on brief but relevant justice information (Lind, 2001). Throughout the day, the employee may interact several times with the supervisor. The treatment received by the employee is easy to remember, impacting his/her daily attitude at work. Accordingly, employees use daily supervisor justice information to assess their daily job satisfaction in the workplace (Loi et al., 2009). We expand this rationale to peers as a justice source. In their daily social exchanges with coworkers, employees use their perceptions about the treatment they receive from peers to guide their daily job satisfaction. That is, the daily treatment received by coworkers is relevant and easy to remember. Therefore, on days when employees are treated fairly in their episodic daily interactions with coworkers, they are more likely to be satisfied. Because coworkers are a meaningful source of justice, we expect daily peer justice to have a significant relationship with daily job satisfaction, beyond the daily treatment from the supervisor.

• Hypothesis 1. Changes in daily justice from peers are positively related to changes in daily job satisfaction, after controlling for the effect of daily supervisor justice.

Regarding emotional exhaustion, generally speaking, lack of justice is seen as a relevant stressor that is predictive of burnout (e.g., Maslach & Leiter, 2008). That is, organizational justice is seen as a gesture of goodwill that transmits the existence of more socio-emotional resources to the employee, reducing emotional exhaustion (Cole et al., 2010). Building on this argument, we argue that the episodic treatment by coworkers also becomes a brief but relevant daily heuristic for employees, providing information that is interpreted in terms of the absence/presence of socio-emotional resources, which exacerbates/reduces daily emotional exhaustion. On days when employees are treated unfairly in their daily interactions with coworkers, they are more likely to experience emotional exhaustion than on days when they perceive fair treatment from peers. Considering the existence of different meaningful sources of justice, the contribution of daily peer justice to daily emotional exhaustion should remain significant once daily supervisor justice is controlled for.

• Hypothesis 2. Changes in daily justice from peers are negatively related to changes in daily emotional exhaustion, after controlling for the effect of daily supervisor justice.

Whereas hypotheses 1–2 deal with employees’ attitudes and feelings, our next two predictions are related to their behavior. According to social exchange theory, people reciprocate the fair treatment they receive from others. As a result, justice can build strong social exchange relationships among coworkers (Cropanzano & Mitchell, 2005). These close relationships have open-ended obligations, which mean that people can provide support without fear of exploitation. Because we are examining fairness from peers, we suspect that peer justice will boost helping behavior toward coworkers (Lavelle et al., 2007). When people are in social exchange relationships, they are more willing to make an effort to help the other party, for example, by increasing job performance and engaging in helpful citizenship behaviors (i.e., OCB, see Masterson et al., 2000; Rupp & Cropanzano, 2002), even if there is no automatic and direct payoff.

Recently, using a between-subject design, Fortin et al. (2020) found some initial evidence for peer justice that is congruent with this reciprocity rationale based on social exchange. They argued that employees who receive fair treatment from coworkers reciprocate by showing better performance. Using a within-subject design, we transfer this argument to episodic exchanges, emphasizing coworkers as a relevant source of justice in the workplace. That is, employees reciprocate daily fair treatment received from coworkers by improving their daily performance. In other words, on days when employees receives fair treatment from coworkers, their performance improves. By contrast, on days when this treatment is less fair, performance suffers. Assuming the aforementioned multisource justice approach, we also examine the extent to which the daily relationship between peer justice and performance is meaningful beyond the daily fair treatment of supervisors.

• Hypothesis 3. Changes in daily justice from peers are positively related to changes in in-role performance, after controlling for the effect of daily supervisor justice.

• Hypothesis 4. Changes in daily justice from peers are positively related to changes in helping coworkers, after controlling for the effect of daily supervisor justice.

The Present Study

Given these concerns, we designed a study that (a) is carried out at the individual level of analysis, (b) provides a within-person examination of justice effects, and (c) controls for supervisor effects at that same level. To meet these strict criteria, we conducted a daily diary study of about 100 hotel service workers, which allowed us to monitor the daily treatment that each employee received from their coworkers and supervisor. Thus, we were able to test all of our peer justice effects beyond the impact of supervisor justice.

The present study makes several important contributions to the literature. First, we extend the concept of peer justice by examining whether it can be treated as an individual level variable. To do so, we control for the effects of supervisor justice. We focus on a variety of outcomes to examine the predictive power of individual peer justice beyond supervisor justice, instead of examining multifoci effects through the match between sources and outcomes. Second, our findings add to the emerging literature on fluctuations in justice over time by considering the daily effects of just and unjust treatment. We argue that variability in treatment is an important predictor in its own right. Third, we provide evidence that individual level peer justice is related to important workplace outcomes such as satisfaction, emotional exhaustion, performance, and OCB.

Method

Procedure and Participants

Employees of a luxury tourist resort located in the Middle East participated in the study. To be included in the study, employees had to interact every day with a supervisor and coworkers. A total of 150 employees received an invitation to participate, along with an informed consent document to sign. The research team also informed employees about the content of the questionnaire. Although participants did not necessarily respond to the questionnaires during the same 10-day time frame, each individual had to answer the questions on ten consecutive working days. More specifically, they had to answer at the end of each workday. The questionnaires were administered on paper. Participants were explicitly informed that they had to answer individually. One of the authors was at the resort to coordinate the procedure and collect the data and informed consent forms.

This researcher created a WhatsApp group with participants as members. This electronic communication system helped to maintain the participants’ commitment over time. It also allowed the researcher to provide daily reminders. Because this communication system was only used for reminders, it did not affect the anonymity of the responses. As an added precaution, the WhatsApp group was created as a broadcast list to keep participants from knowing who else from their workplace was participating. A total of 100 employees (66.67%) consented to participate in the study, resulting in 1000 daily diary recordings in all. The participants’ mean age was 33.74 (SD = 9.77), with an age range between 19 and 55 years. Gender was almost evenly divided, given that 53% were men. Employees who participated in the study were working in 10 different departments of the organization: Executive Office, Engineering, Finance, Human Resources, Sales, Marketing & Catering, Food & Beverage, Rooms Division, Revenue, Security, and Recreation and Spa. In all these departments, a group of employees interacted in order to perform tasks that required coordination. Thus, peer justice was meaningful to participants. To ensure the anonymity of their responses and group together the ten questionnaires filled out by each employee, participants created an individual code and wrote it on the questionnaire every day.

Measures

One of the challenges of diary studies is to maintain employees’ participation over time while asking them to answer the same questions every day. To do so, we followed the methodological recommendations for diary studies. We combined short forms and single items (see Ohly et al., 2010), allowing participants to complete the questionnaire in less than 5 min (see Reis & Gable, 2000). Unless otherwise noted, all the constructs were measured using seven-point Likert-type scales anchored by 1 = strongly disagree and 7 = strongly agree. All surveys were provided in the English language.

Individual Peer and Supervisor Justice

In assessing individual peer justice, we took two considerations from previous research into account. First, Li et al., (2013) and Molina et al., (2016) argued that peer justice can be represented as a general, second-order factor that includes the different subtypes of justice (e.g., distributive, procedural, etc.). Second, Fortin et al., (2020) found that people use different criteria to evaluate coworkers and supervisors (see also, Lavelle et al., 2007). These distinct criteria, in turn, could produce different structures among the subfactors. Given these ongoing considerations, we followed prior research and measured individual peer justice and supervisor justice as overall constructs. Specifically, we employed Ambrose and Schminke’s (2009) five-item scale. The wording of this instrument was slightly modified to test our theoretical ideas. For supervisor justice, the items referred to the supervisor as a source of justice (e.g., “Overall, today I was treated fairly by my supervisor”). For peer justice, the items referred to how each respondent was treated by their coworkers (e.g., “In general, today the treatment I received from my colleagues was fair”).

We conducted confirmatory factor analysis (CFA) to test the discriminant validity between peer and supervisor justice. MLR was the estimation method used. We tested two nested competing models: (1) a one-factor model where individual peer and supervisor justice items were combined into a general factor; and (2) a two-factor model that identified the items on the two separate scales. The proposed two-factor model for the variables measured in time 1 showed adequate goodness of fit indices (χ2 = 58.249, df = 34, p = 0.0059; RMSEA = 0.084; CFI = 0.955; TLI = 0.940), whereas the one-factor model showed a non-satisfactory fit (χ2 = 241.780, df = 35, p = 0.0000; RMSEA = 0.243; CFI = 0.616; TLI = 0.506). The chi-square difference between the two-factor and one-factor models at time 1 (χ2 _diff = 26.796; df _diff = 1, p < 0.01) was statistically significant, indicating that the two-factor model was the best fitting model. We observed the same trend at all the time points. Alpha coefficients ranged from .87 to .95 and from .69 to .92 for supervisor and peer justice, respectively. We centered both justice measures at the person mean.

Criterion Measures: Job Satisfaction, Emotional Exhaustion, Job Performance, and Helping Behaviors

We selected central single items to measure our outcomes. That is, we chose the questions that referred more directly to the construct. For job satisfaction, we followed the typical overall format (Price, 1997) to measure this attitude (“Today I found enjoyment in my job”). Employees responded on a five-point Likert-type scale ranging from 1 = strongly disagree to 5 = strongly agree. Although multi-item scales are preferred for psychometric reasons, research suggests that single-item measures of job satisfaction show reasonable validity (Nagy, 2002; Wanous et al., 1997). We also measured emotional exhaustion by adapting one of the items from the revised Maslach Burnout Inventory (Schaufeli et al., 2002) that focuses on emotional exhaustion symptoms (“Today I have felt emotionally drained”). As in the case of job satisfaction, evidence suggests that single-item measures of emotional exhaustion can be useful in work settings (West et al., 2009).

In keeping with our focus on diurnal activities and feelings, our measures of job performance and helping behaviors explicitly referred to that specific day. Job performance was assessed with a single item, based on previous research (e.g., Goodman & Svyantek, 1999). We chose an item that clearly represents in-role performance—“Today I performed my job well by carrying out my tasks as expected.” Helping coworkers, a type of citizenship behavior, was assessed with a single item—“Today I voluntarily helped my colleagues when they had too much work to do.” Although self-reports of OCB have the risk of some leniency bias, meta-analytic work by Carpenter et al. (2014) suggests that such measures are useful for research purposes.

Plan of Analyses

To examine our hypotheses, we used growth modeling (Duncan et al., 2013), computing hierarchical linear models with SPSS 24 (see Heck et al., 2013). This approach makes it possible to measure model change over time using longitudinal data. Growth modeling allowed us to examine whether changes in individual peer justice across the days led to changes in the reactions and behavior of workers. More specifically, we computed four models to check whether changes in peer justice over time were related to changes in the outcome variables (overall job satisfaction, emotional exhaustion, in-role performance, and helping coworkers), after controlling for changes in supervisor justice. To test the models, we considered the nested structure of the data: level 1 (occasions) and level 2 (individuals). We used restricted maximum likelihood (REML) as the estimation method, and -2 Restricted Log Likelihood (-2RLL) values were used to compute chi-square differences among the models (Δχ²). We also controlled for age and gender (0 = women; 1 = men) because previous studies observed that these variables may influence the way employees react to organizational justice (e.g., Sweeney & McFarlin, 1997; Tenhiälä et al., 2013).

Results

Preliminary Results

As discussed earlier, we argue that perceptions of fairness fluctuate within individuals as they encounter better or worse treatment from peers and supervisors. Although this is consistent with prior theoretical thinking, peer justice has not typically been operationalized in this fashion. Consequently, it is necessary to check for between- and within-person variance in all our daily variables (Table 1). Without significant within-person variation, our perspective would not be supported.

Table 1 Variance components of the daily measures

As anticipated, we confirmed that between- (σ²_τ00) and within-person (σ²) variance was statistically significant (p < 0.01) for both justice from supervisors and justice from peers. Additionally, we estimated the percentage of total variance due to differences across individuals [σ²_τ00 / (σ²_τ00 + σ²)] and the percentage of total variance produced by within-person differences over time [σ² / (σ² + σ²_τ00)]. Our results indicated that 41% (supervisor justice) and 30% (peer justice) of the total variance was attributable to differences between individuals, and, consequently, 59% and 70% of the variance was based on daily variability over time (within-person variance). In sum, the within-person effects were stronger than the between-person effects, suggesting that prior research has overlooked an important aspect of peer justice.

Fluctuating levels of fairness perceptions are unlikely to predict stable work outcomes. Therefore, it is also necessary to examine the diurnal variability in the criterion variables. As with the predictors, both within-person variance over time and between-person variance were statistically significant for all the outcomes. The percentage of total variance due to within-person differences over time ranged from 52% (emotional exhaustion) to 70% (helping coworkers). Therefore, the study of intraindividual associations is justified. Once again, at least half of the variability was within-person, which is consistent with the arguments presented above.

Descriptive statistics and aggregated correlations among the main study variables are presented in Table 2. As expected, positive and significant links were found from individual peer justice to job satisfaction, in-role performance, and helping coworkers. There was a small, but significant, positive relationship between peer justice and emotional exhaustion. There was also a positive and significant correlation between peer justice and supervisor justice. In addition, we tested the within-individual relationship between peer and supervisor justice using growth modeling. Results showed that changes in individual peer justice were related to changes in supervisor justice (γ20 = .45, p < .01), and changes in supervisor justice were related to changes in peer justice (γ20 = .25, p < .01).

Table 2 Means, standard deviations, and correlations between daily study variables

Hypothesis Testing

Table 3 summarizes the results of the models testing the relationships between daily peer justice and daily outcomes, focusing on within-individual variance in the outcome variables rather than the overall variance. We examined whether intra-individual daily changes (or fluctuations) in individual peer justice perceptions are related to intra-individual daily changes in the outcomes, after controlling for intra-individual daily changes in the treatment received from the supervisor (supervisor justice). The expression “change as a predictor of change” does not apply to the control variables (sex and age). The variables whose coefficients indicate the effects of “change as a predictor of change” are daily supervisor justice (γ20) and daily peer justice (γ30). Below, we consider each of our four hypotheses. As we will see, most of our findings supported them.

• Hypothesis 1 stated that fluctuations in justice from peers would be positively and significantly associated with daily changes in overall job satisfaction. This was observed in our data (γ₃₀ = .25, p < .01), which means that positive slopes in peer justice are associated with positive slopes in overall job satisfaction. That is, the greater the positive change (increase) in peer justice over time, the greater the positive change (increase) in overall job satisfaction over time. As Table 3 indicates, this relationship was found after controlling for daily supervisor justice, which also showed a positive and significant coefficient (γ₂₀ = .28, p < .01). When introduced in the model, daily peer justice predicted an additional 4.2% of the variance in job satisfaction, which was also statistically significant (Δχ² = 34.89; Δdf_diff = 1; p < .01).

• Hypothesis 2 stated that fluctuations in individual peer justice would be negatively and significantly related to daily emotional exhaustion. Results in Table 3 did not support hypothesis 2 because the coefficient was not statistically significant (γ₃₀ = − .15, p > .05). This result was found after controlling for daily supervisor justice, which showed a negative and significant coefficient (γ20 = − .31, p < .01). When we tested the additional variance explained by daily peer justice (when introduced in the model), we found that it was small (0.4%) and non-significant (Δχ² = .03; Δdf_diff = 1; p > .05). Consequently, we conclude that we did not find support for our second prediction.

• Hypothesis 3 stated that fluctuations in individual peer justice would be positively related to in-role job performance, and this proved to be the case (γ₃₀ = .27, p < .01). Thus, positive slopes in peer justice are associated with positive slopes in in-role job performance. That is, the greater the positive change (increase) in peer justice over time, the greater the positive change (increase) in in-role job performance over time. This relationship was found after controlling for daily supervisor justice, which also showed a positive and significant coefficient (γ₂₀ = .21, p < .01). Fluctuations in justice from peers predicted an additional 3.5% of the variance in daily in-role performance, which was statistically significant (Δχ² = 28.01; Δdf_diff = 1; p < .01).

• Hypothesis 4 indicated that daily justice from peers would be positively related to fluctuations in helping coworkers. This effect was supported (γ₃₀ = .60, p < .01), which means that positive slopes in peer justice are associated with positive slopes in helping coworkers. That is, the greater the positive change (increase) in individual peer justice over time, the greater the positive change (increase) in helping coworkers over time. This relationship was found after controlling for daily supervisor justice, which also showed a positive and significant coefficient (γ20 = .09, p < .05). Daily peer justice explained an additional 9% of the variance in helping coworkers daily, which was statistically significant (Δχ² = 86.77; Δdf_diff = 1; p < .01).

Table 3 Relations between daily peer justice and daily reactions

Testing the Reverse Direction

In this study, we are assuming that changes in justice from peers are related to subsequent changes in each of the outcomes (e.g., the greater the positive change—increase—in peer justice, the greater the positive change—increase—in job satisfaction). However, alternative causality would also be possible, whereby changes in criterion variables would be related to subsequent changes in peer justice. We ran alternative models that tested these competing causal relationships. Results showed that changes in emotional exhaustion were not related to changes in peer justice (γ₃₀ = − .02, p > .05). Although changes in job satisfaction (γ₃₀ = .18, p < .01), in-role performance (γ₃₀ = .13, p < .01), and helping coworkers (γ₃₀ = .16, p < .01) were significantly and positively associated with changes in individual peer justice, the strength of the relationships was always higher when the antecedent variable was peer justice than when we tested the reverse causal relationship, thus supporting the hypothesized model.

Auxiliary Analyses

To get a better understanding of justice and its relationship with outcomes, we carried out three additional analyses. First, we tested the specific contribution of supervisor justice after controlling for individual peer justice. Second, we explored the role of peer justice variability in predicting each outcome’s level. Third, we examined the interaction between supervisor justice and peer justice.

Revisiting Supervisor Justice

Traditionally, organizational justice research has paid more attention to formal authority figures, such as supervisors, and less to coworkers. We sought to remedy this imbalance here. Supervisors are important sources of daily fairness (and unfairness). Table 2 shows that supervisor justice is correlated with most of our criterion variables, albeit with somewhat smaller relationships than those observed for peer justice. To get a better sense of these effects, we examined the unique contribution of daily supervisor justice when controlling for daily justice from peers. The additional variance explained by daily supervisor justice was statistically significant for overall job satisfaction (9.5%; Δχ² = 81.63, Δdf_diff = 1; p < .01), emotional exhaustion (2.0%; Δχ² = 18.70; Δdf_diff = 1; p < .01), and in-role performance (3.2%; Δχ² = 30.71, Δdf_diff = 1; p < .01). However, the capacity to predict helping coworkers was small and non-significant (0.3%; Δχ² = − .37; Δdf_diff = 1; p > .05). Consequently, supervisor justice predicted three outcome variables beyond the effects of individual peer justice—job satisfaction, emotional exhaustion, and in-role performance. Peer justice, as we have seen, predicted three outcomes beyond the effects of supervisor justice—job satisfaction, job performance, and helping coworkers. Consistent with the work of Lavelle and his colleagues (2007), we can conclude that both sources of justice are important for understanding work outcomes.

Peer Justice Variability

As Matta et al. (2017) observed, it is possible that variability in peer justice is related to each outcome. Due to the uncertainty this instability could create, variability in treatment by coworkers could have a negative role beyond the level of peer justice. To explore this proposal, we carried out one regression analysis for each outcome, considering the peer justice level (operationalized as the average peer justice perception at the ten time measurements) and variability (operationalized as the standard deviation in each individual’s peer justice perceptions over time) as predictors, whereas the outcome level was the criterion variable (operationalized as the average of the ten time measurements for each outcome). There are empirical precedents for this approach in the investigation of variability (Eid & Diener, 1999; Kernis et al., 1993; Matta et al., 2017, Study 2), and it has been recommended in previous literature on comparison of dispersion indices in multilevel research (Roberson, et al., 2007). Variability in justice from peers only had a significant relationship with helping coworkers but not with the other criteria. Variability in peer justice contributed to reducing helping behaviors (β = − .29; p < .01), whereas the level of peer justice was positively related to this criterion (β = .40; p < .01) (controlling for age and sex). Thus, our findings found a smaller role for individual peer justice variability than Matta and his colleagues found for justice climate variability.

The Interaction Between Supervisor Justice and Individual Peer Justice

We analyzed the interaction between supervisor justice and peer justice at each measurement time. The interaction was not statistically significant on any of the ten assessment days. Consequently, it appears that, although both are important, individual peer justice and supervisor justice contribute independent main effects. We did not predict or find that the impact of one source of justice changed across levels of the other.

Discussion

Previous research has demonstrated that peer justice is a relatively stable aspect of climate, and it describes the justice that members of a work group generally exhibit. This previous research has conceptualized peer justice at the unit or group level (Cropanzano et al., 2007; Molina et al., 2015). By contrast, individuals’ experience of justice from peers has been neglected, which is surprising because social exchange approaches to justice predict these effects. Consequently, the unit level of analysis is not the only way to understand peer justice effects, and sometimes it is not even the best. To reinforce this individual perspective, we revisit the theoretical challenges we examined earlier.

First, and perhaps most fundamentally, our findings indicate that this individual level conceptualization is meaningful. We found that daily peer justice predicted daily changes in three out of four outcomes—overall job satisfaction, in-role performance, and helping coworkers, though not emotional exhaustion. Second, building on this first point, we found that peer justice perceptions exhibit a large amount of variability over time. More specifically, we found that most of the variance in peer justice—about 70%—was due to within-person variation. This means that prior work that examined unit level effects may have substantially underestimated the impact of peer justice. Parenthetically, it is worth mentioning that almost 60% of the variance in supervisor justice was also within-person. Justice perceptions seem to fluctuate over time, and these changes meaningfully predict responses to the work environment (see also, Holtz & Harold, 2009; Koopman et al., 2020; Loi et al., 2009). Third, we also observed that peer justice, operationalized at the individual level, continued to predict satisfaction, performance, and helping, beyond the effects of supervisor justice. Consistent with social exchange theory (Cropanzano et al., 2001; Lavelle et al., 2015), individual peer justice and supervisor justice were distinguishable constructs, with each contributing incremental predictive power beyond the contribution of the other. In short, peer justice, although important, did not tell the whole story.

Our individual perspective offers a complementary view to previous research on peer justice. We pay special attention to (a) individual perceptions that (b) fluctuate over time. This represents a change from a wide-angle photo to a movie that captures close-ups and makes it possible to observe justice from a different perspective. Our results suggest that peer justice is not only an aspect of organizational climate, which would suggest a unit level of analysis and the accompanying stability. In addition, studies of peer justice should also consider the possibility that it exists within individuals and may change over time. Although this does not negate previous research, it does suggest an extensive reconceptualization of the peer justice construct.

Theoretical Implications and Future Research Needs

Empirical investigations of organizational justice have a lot in common with other organizational behavior research, particularly because they share an implicit operationalization of justice as being somewhat stable and fixed. Of course, theorists understand that this is not the case, although research methodology lags somewhat behind conceptual thinking (Jones & Skarlicki, 2013). Given this situation, we highlight new research directions suggested by our theoretical model.

The Importance of Time

Our findings suggest that time deserves specific attention. We suspect that some aspects of justice tend to be more stable (e.g., salary), whereas others are less stable. For instance, our results corroborate proposals that fairness in daily encounters is highly sensitive to the temporal dynamic (Bies, 2001, 2005; Mikula et al., 1990). That is, justice perceptions associated with interpersonal interactions change from one working day to another. This temporal fluctuation is not restricted to the relationship with the supervisor (Ferris et al., 2012) because it also occurs in interactions with coworkers. The explicit consideration of time provides a more complete view of employees’ justice perceptions while avoiding a limited static vision of justice and the context where it takes place. We encourage future researchers to incorporate time into their theoretical thinking about justice and other constructs.

Individual Peer Justice and Supervisor Justice

Our findings were consistent with social exchange theory, successfully distinguishing the construct of peer justice, or treatment from coworkers, from the construct of supervisor justice, or treatment from supervisors. Apart from separating these two constructs, we obtained preliminary indications that they are differentially related to some criteria. For example, we observed that daily peer justice showed a prominent role in predicting daily fluctuations in helping coworkers (the variability in peer justice perceptions also predicted the average level of helping behavior), whereas daily supervisor justice was not related to this outcome. This is especially important because helping coworkers is relevant in enhancing organizational effectiveness (e.g., Halbesleben & Wheeler, 2015; Love & Forret, 2008). The high capacity of daily peer justice to predict changes (and the average level over time) in helping supports social exchange theory (Lavelle et al., 2007), which is also congruent with the multi-foci approach to justice (Rupp & Cropanzano, 2002; Rupp & Paddock, 2010). In their daily interactions with others, employees are able to capture different sources of justice, producing differentiated responses. Accordingly, employees are especially motivated to reciprocate and direct their attitudes and behaviors toward the specific source of justice. Fluctuations (and the average level) in the help employees give to their coworkers depend to a large extent on the treatment they receive from these coworkers each day.

However, the findings did not always favor individual peer justice. Daily supervisor justice was associated with variance in daily emotional exhaustion, whereas daily peer justice was unable to predict additional significant variance. This null finding was unexpected and should be viewed as a limitation. A tentative explanation could be related to the severity of unfairness as a stressor. It is possible that changes in emotional exhaustion only occur in relatively extreme situations. Exclusion by coworkers, hindering the satisfaction of belongingness needs (Baumeister & Leary, 1995; Herr et al., 2018), might only be significant when bullying emerges as a highly stressful and unjust offense from coworkers (Maarit & Vartia, 2001; Okechukwu et al., 2014). This type of situation is not likely to be frequent in our study sample or in the period of time considered, but it is an interesting topic for future research on daily peer justice. A related possibility may have to do with the chronic nature of stress. Daily injustice, which by definition is temporary, may be painful, but chronic injustice, which continues over a long period of time, may have stronger effects on employees’ emotional exhaustion. This is an intriguing possibility because it suggests that between-person effects may be stronger for some criterion variables, whereas within-person effects may be stronger for others. Of course, this possibility awaits future investigations.

In sum, we have considered a number of relevant outcomes in examining whether peer justice is able to predict additional variance beyond supervisor justice. Now that our findings have confirmed both the predictive power of peer justice and its initial differentiation from supervisor justice (mainly because of the relationship with helping coworkers), future research can propose and test target-specific hypotheses. Delving deeper, it is possible to suggest the existence of unique outcomes for peers as a source of justice. For instance, satisfaction with coworkers can be a specific outcome associated with peer justice. This will provide a richer picture of specific peer justice outcomes.

Antecedents of Individual Peer Justice

Our study treated daily peer justice as an antecedent to various important workplace outcomes. This was reasonable, given that we needed to establish the usefulness of our approach. However, it also comes with a cost. At present, we know relatively little about the antecedents of daily peer justice. We seek to predict the variability in individual peer justice ratings, but some of the antecedents to these fluctuations are likely to be variable themselves. For example, if we propose that respectful treatment boosts individual peer justice reports (cf., Bies, 1987), a proper theory will need to explain why respectful treatment changes, or at least why individual perceptions of respectful treatment change. These fluctuations, in other words, should be built into multiple stages of a causal theory. Researchers have only just begun to consider dynamic theories of justice, and much more research is needed (Holtz & Harold, 2009). We have taken an initial step by examining the directionality of the relationships between individual peer justice fluctuations, on the one hand, and changes in relevant constructs (job satisfaction, emotional exhaustion, in-role performance, and helping coworkers), on the other hand. Our results supported peer justice changes as an antecedent, but more efforts are needed to examine what causes fluctuations in justice from peers over time.

Weighting Treatment from Peers

In order to test our conceptual model, we prompted respondents to indicate how they had been treated. We then looked at the relationship between these reports and various criterion variables. Although our results are promising, they suggest that future researchers should take a step back in the causal model and examine how individual peer justice perceptions are formed. Justice researchers assume that certain events occur that prompt workers to make judgments of fairness or unfairness (Fortin et al., 2020). The nature of these events should be better articulated, particularly in the context of group work.

At least four issues will require closer examination— the magnitude of unfairness, the number of unfair events, the source of these events, and the type of justice. Not all justice events would be expected to have the same impact because some may have smaller consequences and others larger. For instance, highly negative events associated with daily peer justice could transform the way employees perceive the typical treatment among coworkers in the organization. The number of negative events will also have to be considered, at least as they relate to positive occurrences. They will probably be best understood in terms of offsetting positive events. Baumeister et al. (2001) found that it takes multiple positive events to allay the harm caused by even a single negative occurrence. Thus, we would expect that unfair treatment may cause employees to ignore, or at least not adequately weigh, the just events in their work lives. Third, the behavior of some individuals is likely to be more impactful than the behavior of others. For example, just treatment from a high-status teammate may create a greater perception of fairness than treatment from lower status individuals. Fourth, the type of justice can play a relevant role. Colquitt et al. (2001) observed that the effects of organizational justice depend on the type of justice. Distributive justice had a stronger relationship with personal outcomes (e.g., salary satisfaction), whereas procedural justice showed stronger connections with system outcomes (e.g., organizational commitment). In addition, interpersonal justice was especially relevant in predicting agent outcomes (e.g., citizenship behaviors oriented toward the supervisor). This rationale can also be applied to peer justice. Although overall justice is the most direct precursor of worker reactions and behaviors, the consideration of different types of justice could also contribute to a better understanding of peer justice and its specific effects on coworkers. It would be reasonable to expect, for instance, that individual peer procedural justice would show a stronger link to unit or team commitment, whereas peer interpersonal justice would have a stronger relationship with helping coworkers.

Stable and Unstable Predictors

Qualifying our earlier point, we did not find that justice only fluctuates within individuals. Some parts of justice are variable, but others are not. Understanding the relationship between unit level peer justice, which tends toward stability, and daily peer justice, which tends toward change, will require attention. For example, it is possible that a strong climate for peer justice buffers employees from the vicissitudes of work life. Individuals might be more willing to forgive ill-treatment if it occurs against a stable backdrop of justice. Of course, one might also make the opposite prediction. When people are generally treated fairly, even a slight injustice might be more noticeable against a generally positive background. Some general conceptual efforts have been devoted to understanding the relationship between the more and less stable components of justice (Jones & Skarlicki, 2013), but specific studies will be needed once the meaningfulness of peer justice variability over time has been confirmed.

Unit level Peer Justice Reconsidered

The purpose of this investigation was to consider peer justice at the individual level of analysis. To do so, we assessed how each employee felt they were being treated by their coworkers. This approach was successful, given that we found considerable within-person variability and were able to predict various criterion variables. However, our approach raises another question. We did not find, nor would we be inclined to argue, that unit level peer justice is an invalid concept. Instead, we believe that our approach complements and extends prior work but does not contradict it. If our analysis is correct, future research should measure both constructs together, examining both individual and unit level effects. This would allow investigators to consider whether the two types of peer justice work together.¹ For example, it is possible that employees feel guilty when they are being treated more fairly than their peers. Likewise, they might be resigned to poorer treatment if it is evenly shared throughout the group. We caution that these comments are speculative, given that there is very little research available that considers these cross-level interactions. However, this would seem to be a promising area for future study.

Implications for Practice

Our findings also have implications for the way research has informed practice. Researchers have often used a static perspective to understand key aspects of organizational behavior. Many variables, including peer justice and its outcomes, are implicitly considered constant. This approach is probably more manageable (e.g., implementing an annual performance evaluation), but it ignores elements of organizational life that can show great activity behind a semblance of stability. Justice perceptions may be roiling under the surface of a seemingly tranquil organization. This mismatch between the stable and the unstable may be most prevalent when employees feel disempowered and unable to share their feelings. Management may become aware of feelings of injustice after they have built up and hardened over time. Therefore, it might be useful to build safe workplaces where individuals can openly discuss and resolve their daily concerns.

More generally, there is a lack of attention paid to the temporal evolution of the variables and how to stimulate the positive aspects and prevent the negative ones. This omission is especially likely in the case of individual peer justice. Supervisors and managers sometimes do not pay attention to daily peer relations (often informal) because they think bonds among coworkers are not their responsibility. However, a good relationship between coworkers is necessary in order to have a social lubricant that enhances the effectiveness of the organization (Love & Forret, 2008). Implementing actions to avoid unfairness among coworkers and maintain daily peer justice over time is one way to achieve this. Likewise, there may be times when individual peer justice is especially important. Based on our earlier comments, we suspect that it is especially relevant during times of organizational change, crises, important conflicts, labor negotiations, and so on. During these periods, the treatment employees receive from coworkers is likely to be affected and can significantly influence their job attitudes and performance.

Limitations

Finally, we acknowledge that all studies, including ours, have certain limitations. Now that daily peer justice appears to be a useful concept, future research should address these issues. First, all our measures were collected via self-report. This single method could inflate some of the relationships obtained. Fortunately, there is some reason for optimism. Our variables showed differential relationships with various constructs, as in the case where individual peer justice predicted helping, but supervisor justice did not. In addition, there were predicted incremental effects, for example, when daily peer justice accounted for criterion variance beyond the effects of supervisor justice. Even if the obtained relationships were somewhat inflated, a general method factor would probably not have produced all the support for the present hypotheses. Moreover, it is worth mentioning that self- and supervisor-reports may converge on some measures, such as OCB (Carpenter et al., 2014). Second, our study assumes that changes in daily peer justice are positively related to changes in the outcome variables (job satisfaction, emotional exhaustion, in-role performance, helping coworkers). However, there are alternative causal possibilities (e.g., changes in the outcomes lead to changes in individual peer justice). Although our design does not allow us to establish definitive causal relationships, the results obtained for the reverse causal direction indicate that the hypothesized model seems more plausible. A third issue had to do with the length of our scales. Some of our measures were as short as a single item. This is common in diary studies, where respondents may experience fatigue with long scales (Ohly et al., 2010). Although short measures can decrease reliability, most of the predictions were supported, suggesting that these measures, if anything, led to more conservative findings. Fourth, we did not consider possible variability in the degree to which employees interact with each other. Given the sector (hospitality) of the company, interaction with colleagues is necessary to perform the daily work. Therefore, individual peer justice is a meaningful construct. However, it is possible that variability exists in the frequency and intensity of the interactions with colleagues, thus influencing daily peer justice and its effects. Investigating this variability in future studies could provide a richer view of individual peer justice. Fifth, our study concentrated on the employee’s perception of justice as an individual, reporting the treatment received from the other group members (direct consensus model). Although this model is meaningful, employees can also experience peer justice as members of a group (referent shift consensus model). These two types of perceptions seem different (the individual’s perception of the treatment received from the rest of the group vs. the experience, as an individual, of the way the group members treat each other), and comparing them could be a fruitful approach in future studies. Finally, our data were collected in a single organization from one country. Thus, the extent to which our findings can be generalized remains an open question. Given the generally strong theoretical arguments for our hypotheses, we are optimistic. However, we strongly caution that external validity is an issue that must be addressed in additional research with independent samples. One interesting area where generalization could be examined is related to the differentiation of peer justice variance into within- vs. between-individual components. Although we found significant between-individual variance, it is reasonable to argue that strong climates in organizations and teams (and therefore shared perceptions among members about the organizational environment) would reduce between-individual variance and facilitate a common evolution of peer justice perceptions over time.

Conclusion

All in all, we believe that this study provides evidence that justice from peers is meaningful from an individual perspective. Employees differ in their perceptions of the treatment they receive from coworkers, and these perceptions fluctuate from one day to another. In addition, daily peer justice is able to predict daily job satisfaction and performance indicators beyond the daily treatment employees receive from their supervisors. We present a first step in understanding individual level peer justice perceptions as a changing judgment inextricably linked to time.