Keywords

Performance appraisal (PA) is perhaps the most common human resource management tool used in organizations (Rynes, Gerhart, & Parks, 2005). Scholars and practitioners have argued that employee reactions to performance appraisals are a set of key criteria by which to judge appraisal effectiveness (e.g., Cardy & Dobbins, 1994; Folger, Konovsky, & Cropanzano, 1992; Keeping & Levy, 2000) given the importance of buy-in to the process (Murphy & Cleveland, 1995). Performance appraisals have important implications for organizations in terms of the overall performance management process and the allocation of scarce resources, such as merit increases (Aguinis, 2013). Employee reactions to performance appraisalsFootnote 1 are correlated with job attitudes and subsequent performance (e.g., Anseel, Van Yperen, Janssen, & Duyck, 2011; Jawahar, 2010). Thus, appraisal reactions are essential to appraisal effectiveness (Bernardin & Beatty, 1984; Cardy & Dobbins, 1994; Folger et al., 1992; Pichler, 2012; Pichler, Beenen, & Wood, 2018).

With that said, research has shown that employees are often dissatisfied with the appraisal process (e.g., Taylor, Tracy, Renard, Harrison, & Carroll, 1995). Managers generally dislike giving negative feedback, and employees generally react negatively to such feedback (Brett & Atwater, 2001), which could reduce their subsequent performance. This is a key problem in that the primary purpose of performance appraisal is to give employees feedback so as to improve their performance (Aguinis, 2013; Tziner, Murphy, & Cleveland, 2005). Performance appraisal research has traditionally not focused on ways by which to improve appraisal reactions. Historically, most performance appraisal research focused on the so-called psychometric approach (Folger et al., 1992), that is on reducing rating errors and improving rating accuracy. This led to a science–practice gap in the sense that there is relatively less research on how to improve appraisal reactions—criteria that practitioners are most interested in compared to psychometric features of performance ratings (Balzer & Sulsky, 1990). This is surprising since researchers have been interested in appraisal reactions for some time (e.g., Zander & Gyr, 1955). Indeed, Ilgen, Barnes-Farrell, and McKellin (1993) argued more than 20 years ago that performance appraisal research needed to move beyond rating format and cognitive perspectives so as to better inform theory and practice related to performance appraisal.

There is, however, a growing body of research on appraisal reactions. For instance, Levy and Williams (2004) identified employee reactions as a key outcome of the social context of performance appraisal, which was subsequently tested by Pichler (2012). There have also been several meta-analyses on predictors of appraisal reactions, namely employee participation (Cawley, Keeping, & Levy, 1998), the social context of performance appraisal (Pichler, 2012), and adequate notice in performance appraisal (Pichler et al., 2018). Since acceptance of feedback is important if not necessary for changing behavior, that is to improve one’s performance (Carver & Scheier, 1981), and with this growing stream of research (Levy & Williams, 2004), the time is ripe to review of this literature. Although appraisal reactions might no longer be “neglected criteria” (Murphy & Cleveland, 1995, pg. 310), the literature still lacks cohesion in terms of an integrative view of the cumulative body of evidence—and where the field is headed. The purpose of this chapter, therefore, is to review the existing research on appraisal reactions with the intention of identifying key predictors of appraisal reactions, identifying opportunities for future research on this topic, and drawing implications for managers and organizations in terms of performance appraisal effectiveness.

The next section will provide an overview of appraisal reactions as a set of criteria of the appraisal process. The following sections will review the literature on two key contextual predictors of appraisal reactions, namely due process performance appraisal and the social context of performance appraisal, respectively. Although there have been reviews of the due process metaphor (Levy, Cavanaugh, Frantz, & Borden, 2015) as well as the social context of performance appraisal (Levy & Williams, 2004), there has been no comprehensive review of empirical studies linking due process performance appraisal or the social context to appraisal reactions. The chapter will conclude with a discussion of implications for future research, as well as practical implications for managers and organizations.

Employee Reactions to Performance Appraisal

In terms of performance appraisal effectiveness, Cardy and Dobbins (1994) proposed three key criteria, that is, rater errors, rating accuracy, and qualitative criteria. Given that rater errors and rating accuracy are two of three criteria identified by these authors, this reflects on the historical prominence of the psychometric view of performance appraisal. Appraisal reactions are a component of qualitative criteria, or qualitative outcomes of the appraisal process (Murphy & Cleveland, 1995). One could argue that appraisal reactions are an additional, distinct set of key criteria of performance appraisal. Indeed, Levy and Williams (2004) later replaced qualitative criteria with employee reactions in their model of appraisal effectiveness. Heretofore, scholars have not proposed an operational definition of appraisal reactions and, as such, Pichler (2012) offered the following: individual-level attitudinal evaluations of and responses to the performance appraisal process. Keeping and Levy (2000) provide perhaps the best treatment of the nature of appraisal reactions as appraisal criteria.

In order to understand relationships between appraisal reactions and their predictors, it is important to briefly explain each of these reactions (see also Keeping & Levy, 2000; Levy & Williams, 2004). Accuracy is an individual’s perception that their rating accurately reflects their objective performance. Fairness is a measure of an individual’s perception that their appraisal review was fair overall. Motivation to improve is a measure of an individual’s intention to improve their performance subsequent to the review. Satisfaction is a measure of how satisfied an individual is overall with the appraisal review. Utility is a measure of the extent to which an individual felt that they received useful feedback. Measures of appraisal-specific organizational justice are also included in studies of appraisal reactions. Performance appraisal procedural and distributive justice are generally adapted measures to reflect the process and outcome fairness of the appraisal review. Although not included in the Keeping and Levy (2000) study, researchers have more recently begun to include measures of interactional justice, that is, perceptions of fairness of interpersonal treatment (Colquitt, 2001), as it relates to the appraisal review (e.g., Elicker, 2000).

The purpose of the Keeping and Levy (2000) study was to test the measurement properties of scales used to operationalize appraisal reactions. They found that existing scales generally performed well, which is to say that they represented distinct constructs when subjected to confirmatory factor analysis. In this connection, they also found that appraisal satisfaction, which is general in nature, could be effectively separated into session and system satisfaction. The former reflects satisfaction with the appraisal review session itself; the latter reflects satisfaction with the overall performance appraisal process as a system. It is important for future research to distinguish between these constructs since they are distinct conceptually and differentially related to correlates (Cawley et al., 1998; Keeping & Levy, 2000). Keeping and Levy (2000) also found that appraisal reactions represented a higher-order latent factor; thus, it is important that researchers test for this higher-order factor in studies where multiple criteria are included and, when supported and when possible given sample size and power, include this latent factor in substantive analyses, that is, in latent variable structural equation modeling.

Due Process Performance Appraisal

Now that the reader has been presented with information about appraisal reactions, the different types of appraisal reactions that are typically measured in existing research, and the nature of their measurement properties, it is important to consider predictors of appraisal reactions. When it comes to predictors of appraisal reactions, scholars have generally looked to the so-called performance appraisal context. Murphy and Cleveland (1995) defined performance appraisal context as “a heterogeneous mix of factors, ranging from the social and legal system in which the organization exists to the climate and culture within the organization” (pg. 31). Thus, the appraisal context is broad and could include a wide range of factors. Two aspects of appraisal context that are relatively prominent in the literature are due process performance appraisal (Folger et al., 1992; Levy et al., 2015) and the social context of performance appraisal (Levy & Williams, 2004; Pichler, 2012). Scholars have offered a number of models of performance appraisal context, but they do not necessarily consider the role of appraisal reactions as related to context (e.g., Erdogan, 2002; Murphy & Cleveland, 1995).

The following sections will consider the role of due process and social context as they are related to one another and to appraisal reactions. Due process performance appraisal is discussed prior to the social context because due process contains elements of broader organizational features that may influence social-contextual variables.

The Due Process Metaphor

Taylor et al. (1995) noted that performance appraisal is “one of the great paradoxes of effective human resource management” in that the intention is to give feedback so as to improve performance, yet this is often unrealized due, in part, to negative appraisal reactions and, in connection, a lack of due process. Folger et al. (1992) developed a due process model of performance appraisal based on due process of the law. They did so in response or in contrast to what they call the “test metaphor” of performance appraisal, which is different vernacular for the psychometric approach to performance appraisal described in the introduction of this chapter. Folger et al. (1992) noted that the test metaphor relies on several tentatively held assumptions, namely that work performance can be measured reliably and validly; that raters can judge performance accurately; and that there is some ultimate criterion of performance against which ratee performance can be evaluated. Because these are tenuous and criticized assumptions in the performance appraisal and job performance literatures, researchers have, as described above, redefined appraisal effectiveness.

The due process model of performance appraisal (Folger et al., 1992) is based on principles of due process of law, which are tied to the Fifth and Fourteenth amendments to the United States Constitution. These amendments are meant to increase fairness in legal proceedings in that they prescribe adequate notice (the publication and dissemination of laws), a fair hearing (the accused has a right to present evidence on his or her behalf and have a voice in proceedings), and judgment based on evidence (decisions are free from bias). As applied to the performance appraisal context, this means that the appraisal process should include adequate notice (e.g., knowledge of performance standards and frequent feedback as to performance), a fair hearing (e.g., participation in the appraisal), and judgment based on evidence (e.g., performance is evaluated based on job-relevant factors and the supervisor is unbiased). Of course, some of these aspects may be questionable themselves, such as the extent to which supervisors are unbiased, but this provides a useful framework for understanding first, appraisal fairness, and second, key contextual predictors of appraisal reactions.

Most research that has included variables related to due process has done so without explicitly referencing due process or the model developed by Folger et al. (1992), with some important exceptions (e.g., Pichler et al., 2018; Taylor et al., 1995; Taylor, Masterson, Renard, & Tracy, 1998). That said, other key studies in the performance appraisal literature support the importance of due process. For instance, in his seminal study Greenberg (1986) proposed factors that represent performance appraisal fairness, namely employee input prior to the appraisal (part of adequate notice), rater familiarity with critical ratee performance incidents (part of fair hearing), and the consistent application of performance standards (part of judgment based on evidence). There are a number of other studies that have included one or more dimensions of due process as they related to appraisal reactions, which will be discussed below.

Those studies that have explicitly used the due process framework to study appraisal reactions have found that appraisal reactions are more favorable, even when performance ratings are lower, under conditions of due process (Taylor et al., 1995). Research has also shown that relationship quality is higher and managers are less likely to deliberately bias ratings under conditions of due process (Taylor et al., 1998). This is one important tie in the literature between due process and the social context of performance appraisal. Moreover, different aspects of due process have been found to have multiplicative effects: Pichler et al. (2018) found that the relationship between feedback frequency and appraisal reactions was enhanced when knowledge of performance standards was higher. This line of research supports the overarching argument of Folger et al. (1992) and their due process metaphor, which was that the process of performance appraisal was as important—if not more important—than measurement aspects of performance appraisals.

Adequate Notice

The basic principle of adequate notice in terms of performance appraisal is that employees should be aware of the standards to which they will be held accountable at the time of the review, that is, that they’ve been adequately notified of performance standards. This also requires that employees be given feedback throughout the review cycle so that they can understand whether their performance is meeting those standards. A number of studies have measured relationships between adequate notice and appraisal reactions.

Knowledge of Performance Standards

Although scholars have tended to use different operationalizations, several studies have measured relationships between knowledge of performance standards, such as perceived system knowledge (Levy & Williams, 1998; Williams & Levy, 2000) and appraisal reactions. Knowledge of performance standards is important because without it, employees likely do not know how they will be evaluated or the behaviors and results for which they will be held accountable. Empirical findings tend to be consistent, that is, researchers have consistently found positive relationships between knowledge of performance standards and reactions such as procedural justice (Levy & Williams, 1998), distributive justice (Inderrieden, Keaveny, & Allen, 1988), fairness perceptions (Evans & McShane, 1988), and system satisfaction (Inderrieden et al., 1988; Williams & Levy, 2000). This suggests that providing employees with knowledge of performance standards may help to improve appraisal reactions. Indeed, Pichler et al. (2018) found a meta-analytic correlation of Mr = 0.49, p < 0.05 (k = 19, N = 5445) between knowledge of performance standards and appraisal reactions.

Feedback Frequency

Feedback frequency is another aspect of adequate notice, which has received relatively more attention in the empirical literature (see Pichler et al., 2018). A number of scholars have suggested that frequent feedback is an important aspect of the appraisal process (e.g., Ilgen, Fisher, & Taylor, 1979; Kinicki, Prussia, Wu, & Mckee-Ryan, 2004; Landy, Barnes, & Murphy, 1978). In fact, research on feedback frequency dates back to at least the 1950s (e.g., Zander & Gyr, 1955). Frequent feedback is important so that employees can make adjustments to their behavior throughout the review cycle, that is, prior to the review. Without frequent feedback, employees are more likely to feel surprised during the review because they hadn’t been apprised heretofore that there was a performance issue. This is, of course, a key problem with appraisal reviews: Employees feel surprised by their evaluation at the time of the appraisal review.

The results of this line of research are mixed. Researchers have found positive relationships between feedback frequency and appraisal reactions, such as procedural and distributive justice (Erdogan, Kraimer, & Liden, 2001; Inderrieden et al., 1988; Inderrieden, Allen, & Keavey, 2004), but have reported mixed results when it comes to other outcomes, such as appraisal accuracy and motivation to improve (e.g., Kinicki et al., 2004), composite measures of appraisal reactions (Klein & Snell, 1994; Landy et al., 1978), and appraisal fairness (Evans & McShane, 1988). It is important for future research to examine the reasons why there might be differential relationships between feedback frequency and different appraisal criteria. That said, Pichler et al. (2018) found a meta-analytic correlation of Mr = 0.45, p < 0.05 (k = 24, N = 5227) between feedback frequency and appraisal reactions despite the mixed results in the primary literature, which suggests that frequent feedback is important as related to appraisal reactions.

Fair Hearing

The most studied aspect of the fair hearing is employee participation in the performance appraisal review. Participation in the review has been operationalized in a number of ways, including the amount of time the ratee talked, goal-setting, self-appraisal, as well as instrumental and value-expressive voice (see Cawley et al., 1998).

Time Talked

The amount of time the ratee was able to talk during the review may not seem like a particularly substantive measure of appraisal participation, but it has been established as a predictor of appraisal reactions, one that is distinct from other reactions, such as voice (see Greller, 1975; Meinecke, Lehmann-Willenbrock, & Kauffeld, 2017).

Goal-Setting

Goal-setting is important in the sense that it allows for participation in the review, as well as an opportunity for the employee to set goals for the next appraisal period (Nemeroff & Wexley, 1979). Moreover, the relationship between goal-setting and behavioral change has been well documented for decades (e.g., Kay, Meyer, & French, 1965). Goal-setting research has found that setting goals leads to more positive attitudes and performance (e.g., Locke & Latham, 1990), and thus would suggest that setting goals during the appraisal review could lead to more favorable appraisal reactions, especially since performance standards should be clearer. Although several studies have measured goal setting during the review process as related to appraisal reactions, there is very little research tying key propositions of goal-setting theory, such as the importance of setting difficult goals and allowing for frequent feedback, to appraisal reactions. It could be useful for performance appraisal researchers to more fully leverage goal-setting theory and research in studies of appraisal reactions.

Self-Appraisal

Scholars have argued that allowing employees to self-appraise should increase their participation in the review and thus improve their appraisal reactions (Burke, Weitzel, & Weir, 1978). That said, there are inconsistent empirical relationships between self-appraisal and appraisal reactions (Cawley et al., 1998; Korsgaard, Roberson, & Rymph, 1998). There are a variety of explanations for these mixed findings.

It could be that self-appraisals increase disagreement during the review given leniency of self-ratings. Korsgaard et al. (1998) found that self-appraisals were positively related to appraisal reactions when ratees were trained to self-appraise and were trained to be assertive. Another key explanation is that self-appraisals can backfire so to speak if they aren’t really leveraged during the review. In fact, when self-appraisals do not influence ratings, they can lead to negative appraisal reactions (Inderrieden et al., 2004). In contrast, when self-appraisals are used as the basis of the performance appraisal discussion (Bassett & Meyer, 1968) or in the performance rating decision process, they are positively related to appraisal reactions (DeGregorio & Fisher, 1988). The cumulative research on self-appraisals is mixed and this could be due to the moderating roles of ratee decision control and rater–ratee disagreement. It is important for future research to try to tease apart the reasons why self-appraisals are sometimes positively, and other times negatively related to appraisal reactions.

Voice

The above constructs, namely goal-setting, self-appraisal, and especially time talked, are relatively objective measures of appraisal participation. Researchers have also measured employee perceptions of two forms of voice: value-expressive (otherwise known as voice for the sake of voice) and instrumental (otherwise known as decision control) (e.g., Korsgaard & Roberson, 1995). A key finding of the Cawley et al. (1998) meta-analysis was that value-expressive voice was more strongly related to appraisal reactions than decision control. These results suggest that simply allowing employees to express themselves, for example, through active listening, can improve appraisal reactions, more so than affording the employee a sense of control over the appraisal process. At first blush, this may seem surprising in the sense that individuals want to have some influence in processes that affect them and their livelihoods. On the other hands, procedural justice theories, such as the group-value model (Lind & Tyler, 1988), suggest that being a valued member of the group is paramount to individuals, and that a lack of ability to express oneself is especially detrimental to employee attitudes. That said, Cawley et al. (1998) noted that this finding was based on a limited number of studies and should be revisited.

This is an important area for additional research because other primary studies have found conflicting results. For instance, Suh (1992) found that instrumental voice was related to procedural justice whereas value-expressive voice was not. The author suggested that the importance of having voice in the process was not supported, whereas having influence was supported. This contrasts with the conclusion of Cawley et al. (1998). Bonness and Macan (2006) found that when self-appraisals were considered during the appraisal review, appraisal reactions were more favorable as compared to when employees were allowed to self-appraise but their perspective was not considered. The authors concluded that instrumental voice was more important to the participants in their study than value-expressive voice. One-third variable to consider here is the favorability of the individual’s performance feedback. It could be the case that instrumental voice (value-expressive voice) is more important when one’s feedback is relatively negative (positive). Researchers should continue to investigate and compare relationships between different types of voice and appraisal reactions, and the moderating role of other variables, such as rating favorability.

Supervisor Job Knowledge

Another aspect of a fair hearing in the performance appraisal context is supervisor knowledge of the employee’s job and job-related performance. Supervisor job knowledge should, all else equal, lead to a more accurate and fairer evaluation, and thus to more favorable reactions. A number of studies have investigated relationships between supervisor job knowledge (e.g., the extent to which one’s supervisor “Became thoroughly familiar with your performance,” Folger & Konovsky, 1989) and appraisal reactions. Results tend to suggest that supervisor job knowledge is positively related to appraisal reactions. For instance, studies have found that job knowledge is positively related to distributive justice (Folger & Konovsky, 1989), appraisal fairness (Giles, Findley, & Field, 1997), motivation to improve (Kinicki et al., 2004), and composite measures of appraisal reactions (Landy et al., 1978). This suggests that it is important for managers to give employees the impression that they are familiar with the employee’s job, for example, by expressing knowledge about the job description, as well as the employee’s performance, for example, by reiterating and discussing critical performance incidents.

Judgment Based on Evidence

The judgment based on evidence dimension of the due process metaphor posits that decisions should be based on objective performance-related information. In this connection, supervisors should be neutral and performance ratings should be based on job-relevant factors, a valid rating instrument, and appealable.

Supervisor Lack of Bias

Research has consistently found positive relationships between employee perceptions of supervisor lack of bias and appraisal reactions, including perceptions of fairness (Kavanagh, Benson, & Brown, 2007), motivation to improve (Kinicki et al., 2004), and composite reactions (Kleiman, Biderman, & Faley, 1987). There is very little research on why this is the case, however. These relationships could be due to how employees are treated during the review, such as more interpersonally fair (Colquitt, 2001), or simply because of the perception that the supervisor is unbiased. It could also be the case that perceptions of supervisor bias are related to other variables, such as social support or trust, which are then related to appraisal reactions. Indeed, since trust is comprised of perceptions of benevolence (Mayer, Davis, & Schoorman, 1995), it would seem that perceptions of supervisor lack of bias should be tied to perceptions of supervisor trust. As is mentioned below, there is very little research connecting due process performance appraisal to the social context of performance appraisal; this is one such opportunity.

Valid Rating Instrument

There has not been extensive research connecting appraisal reactions to the validity of performance ratings. That said, there is a relatively robust literature on differences between evaluation formats (e.g., Borman, 1979; Tziner & Kopelman, 2002). There is also a literature connecting appraisal reactions to different rating formats. Consistent with the concept of judgment based on evidence, scholars have argued that because behavior observation scales at least ostensibly require raters to observe and document performance, they should be perceived as fairer as compared to other formats, such as graphic rating scales (Tharenou, 1995). That is, behavior observation scales should be perceived as a more valid rating format so to speak by ratees. Research has supported this proposition, and has shown that, compared to graphic rating scales, behavior observation scales are related to higher levels of appraisal satisfaction as well as subsequent performance (Tziner, Kopelman, & Joanis, 1997; Tziner, Kopelman, & Livneh, 1993).

Job-Relevant Factors

Folger et al. (1992) posited that to the extent performance evaluations are based on information that is job-relevant, perceptions of fairness should be higher. Again, the notion here is that more job-relevant information signals that the evaluation is more accurate and less biased. For instance, Nathan, Mohrman, and Milliman (1991) developed a scale that measured “the extent to which the actual evaluation was based on results achieved, job-related behaviors, skills and abilities and predetermined goals” (pg. 358). Research has shown that perceptions that a performance evaluation was based on job-relevant factors are positively related to procedural justice (Cobb, Vest, & Hills, 1997; Erdogan et al., 2001). There is very little research otherwise, and it would be useful to better understand the reasons why job-relevant factors are related to perceptions of justice, and if job-relevant factors are related to other appraisal reactions.

Appeals Process

Having an appeals process in place so that employees can appeal their performance evaluations to an authority other than their line manager may be important in connection to appraisal reactions. This is consistent with procedural justice theory, which suggests that having resource over decisions that affect them should increase perceptions that the process was fair (Greenberg, 1987). Several studies have measured relationships between having an appeals process and appraisal reactions, and results have been consistent in that appeals processes are positively related to appraisal reactions (Cobb et al., 1997; Giles et al., 1997). This suggests that organizations should consider implementing appeals processes, for example, where the next-level manager or someone from human resources reviews the evaluation. Some organizations have implemented performance rating calibration, where managers come together to discuss their evaluations of employees so as to try to ensure consistency and accuracy in ratings (Lombardi, 2011). That said, there is unfortunately little research on appeals processes in performance appraisal, the outcomes of these processes, or how these processes are actually implemented in organizations.

The Social Context of Performance Appraisal

Another key aspect of the appraisal context, in addition to due process, is the social system within which the appraisal process is embedded (e.g., Erdogan, 2002; Levy & Williams, 2004; Murphy & Cleveland, 1995). In fact, some researchers have argued that the social context is the most important aspect of performance appraisal (Russell & Goode, 1988). What is important to note here is that there is a link between due process and social context in the sense that for due process to be effective, it should be linked to leader–member interactions and communications (Levy et al., 2015), although there has been relatively little research that has examine due process and social-contextual variables simultaneously.

The Social-Contextual Domain

When it comes to social context, researchers have studied a number of different variables ranging from interpersonal affect (Varma, Pichler, & Srinivas, 2005), to rater–ratee similarity (Ferris, Judge, Rowland, & Fitzgibbons, 1994; Pichler, Varma, & Petty, 2008), and influence tactics (Dulebohn & Ferris, 1999). Levy and Williams (2004) provided an excellent review of the role of the social context in performance appraisal based on a review of the performance appraisal literature from 1995 to 2003. They identified distal and proximal predictors of rater–ratee behavior (performance ratings, rater/ratee attitudinal, behavioral and cognitive reactions, and perceptions of justice). Distal variables, which in their model are also related to proximal variables, include environmental (e.g., legal and economic conditions), as well as organization (organizational culture and HR strategies) factors. Proximal variables were separated into process (e.g., supervisor–subordinate relationships) and structural (e.g., the appraisal system and performance standards) features. Their model is notably comprehensive, and was influential not only in organizing research to that point, but in providing directions for future research. Their model is also consistent with subsequent research, which has shown that feedback climate (Steelman, Levy, & Snell, 2004) and justice climate (Pichler, Varma, Michel, Levy, & Budwar, 2016) are important predictors of supervisor–subordinate leader–member exchange and appraisal reactions.

Leader–Member Relationship Quality

Although appraisal reactions play prominently in the model developed by Levy and Williams (2004), not all of the social context studies preceding or following the publication of their model also focused on appraisal reactions. Of those that have, scholars have measured constructs such as leader–member exchange (Dulebohn & Ferris, 1999; Erdogan, 2002), trust (Korsgaard & Roberson, 1995; Levy & Williams, 2004), supervisor support (Giles et al., 1997), and supervisor satisfaction (Nathan et al., 1991) as predictors of appraisal reactions. Based on leader–member exchange theory, Pichler (2012) argued that each of these constructs could be conceptualized as aspects of relationship quality. That is, high-quality leader–member exchanges are, by definition, characterized by higher levels of trust and support (e.g., Liden, Sparrowe, & Wayne, 1997). Based on this premise, he treated each of the variables mentioned immediately above as aspects of relationship quality; meta-analyzed relationships between these variables and appraisal reactions (as a composite); and tested three competing models of the role of relationship quality as related to appraisal reactions using meta-analytic structural equation modeling.

The results of these meta-analyses were both expected and somewhat surprising. First, Pichler (2012) found that the relationship quality variables were related to appraisal reactions at about the same magnitude (Mr = 0.58 to 0.60), which was expected based on leader–member exchange theory. Second, a composite of these relationship quality variables was notably strongly related to appraisal reactions (Mr = 0.66); the magnitude of this relationship was not expected. This relationship was noticeably larger than other predictors, such as the favorability of one’s performance rating and the amount of employee participation in the review (see Cawley et al., 1998). These bivariate results suggest that the social context of performance appraisal, at least when it comes to relationship quality specifically, is particularly important to employees in terms of how they react to their appraisal. That said, this is a difficult inference to make based on bivariate relationships alone.

To further test this notion, Pichler (2012) tested three competing structural models with three sets of predictors of appraisal reactions, that is, relationship quality, appraisal participation, and rating favorability. Meta-analytic structural equation modeling is particularly useful for clarifying research findings and theory-testing (e.g., Fassina, Jones, & Uggerslev, 2008). In the direct effects model, all predictors were treated as antecedents. In the instrumental model, appraisal participation and rating favorability were treated as endogenous variables; there was no direct path between relationship quality and appraisal reactions. The relationship-driven model was the same as the instrumental model, except a direct path from relationship quality to appraisal reactions was added. Results indicated that the third model was the only one that fit the data well, which suggests that when it comes to contextual predictors of appraisal reactions, relationship quality is paramount. To probe the boundary conditions of this finding, the author tested rating favorability as a potential moderator of the relationship quality–appraisal reactions relationship, but the results were nonsignificant. This suggests that relationship quality is important when it comes to appraisal reactions, regardless of one’s performance rating, which was not expected.

The Reciprocal Nature of LMX and a Model of Appraisal Reactions

It is important to note here that previous models of the performance appraisal process have positioned leader–member exchange as an outcome of the appraisal process (Masterson, Lewis, Goldman, & Taylor, 2000). Levy and Williams (2004) position leader–member relations as a proximal process variable in their model of the social context of performance appraisal. The literature on the social context of performance appraisal has yet to fully develop the role of leader–member exchange in the overall performance appraisal process. Inderrieden et al. (1988) noted that performance appraisal research should examine the appraisal as a process, not as a single event. They suggested studying how interactions between raters and ratees prior to the appraisal impact appraisal reactions. This is consistent with the model of the appraisal process developed by Klein, Snell, and Wexley (1987), in which they proposed the process is conceptualized as inputs (the appraisal context, e.g., leader–member exchange), throughputs (the appraisal session, e.g., participation), and outputs (reactions).

Cumulatively, this research suggests that contextual variables (e.g., organizational culture, due process characteristics, and justice climate) are related to leader–member exchange, which is related to key variables in the appraisal review (e.g., appraisal participation), which are in turn related to appraisal reactions, subsequent job attitudes and performance, as well as leader–member exchange (see Fig. 5.1 for an illustration). In other words, I propose a dynamic process whereby the appraisal context is related to leader–member exchange; leader–member exchange is subsequently related to variables that occur during the appraisal review; variables in the appraisal review are subsequently related to appraisal reactions; and appraisal reactions are subsequently related to job attitudes, leader–member exchange, and feedback climate.

Fig. 5.1
figure 1figure 1

Longitudinal model of appraisal context, appraisal reactions, and appraisal outcomes

Full size image

To elaborate, when the appraisal context is more favorable, for example, when employees are aware that their review can be appealed (due process) or when employees receive more regular, ongoing feedback (feedback climate), this should foster conditions in which leader–member exchanges should be higher on average (Folger et al., 1992). Differences in leader–member exchange between subordinates should be related to how supervisors treat subordinates during the review, for example, leaders should be more likely to encourage participation from subordinates with whom they have higher quality exchanges. More favorable treatment during the review should be related to more favorable reactions (Meinecke et al., 2017; Pichler, 2012), and more favorable reactions should be positively related to more distal outcomes, such as job attitudes and performance (Anseel et al., 2011; Jawahar, 2010) and LMX (Masterson et al., 2000). This model integrates previous theory (Folger et al., 1992) and conceptual models of the appraisal process (e.g., Inderrieden et al., 1988; Klein et al., 1987; Levy & Williams, 2004) with empirical research on appraisal process and outcomes (e.g., Anseel et al., 2011; Masterson et al., 2000; Pichler, 2012).

One important note as related to this model is that many if not most studies of appraisal reactions have been cross-sectional. That is, they have measured characteristics of the appraisal context, such as relationship quality, at the same time as appraisal reactions. It is important that future research separate measurements of contextual variables and appraisal reactions (Fig. 5.1).

Testing the model presented in Fig. 5.1 is challenging in at least two key ways: First, this would require a relatively sophisticated multilevel research design with a relatively large sample size, and second, this would require longitudinal research. There are examples of multilevel (e.g., Pichler et al., 2016) and longitudinal (e.g., Nathan et al., 1991) research on appraisal reactions, but there has yet to be a study combining multilevel and longitudinal designs. This seems particularly important, however, since various models of the appraisal process include both organization-level (e.g., organization culture) and lower-level (e.g., appraisal reactions) variables. Moreover, models of the appraisal process position social contextual variables, such as relationship quality, at different stages in the appraisal process (e.g., Erdogan, 2002; Klein et al., 1987; Levy & Williams, 2004). Thus, a multilevel longitudinal model seems like the direction in which appraisal reaction research should be headed.

Discussion

Performance appraisal is one of the most common human resources tools used in organizations. The key purpose of performance appraisal is to give employees feedback so as to improve their performance. One key problem is that feedback acceptance is important if not necessary for performance improvement, and many managers and employees react negatively to appraisal reviews. In this sense, employee reactions to performance appraisals are a key set of criteria by which appraisals should be evaluated. In fact, some scholars have suggested that appraisal reactions are the most important outcome of appraisals. There is a growing literature on appraisal reactions (e.g., Levy & Williams, 2004; Pichler et al., 2016; Pichler et al., 2018), and a growing literature on performance appraisal context (e.g., Levy et al., 2015; Levy & Williams, 2004 ; Pichler, 2012). The overarching purpose of this chapter, therefore, was to review the literature on appraisal reactions with a focus on how two key contextual factors, namely due process performance appraisal (Folger et al., 1992; Levy et al., 2015) and the social context of performance appraisal (Levy & Williams, 2004; Pichler, 2012), are interrelated—and how they are related to appraisal reactions. This review culminated in a multilevel longitudinal model linking appraisal context to appraisal reactions and ultimately to more distal appraisal outcomes, such as subsequent job performance.

Implications for Future Research

In addition to providing a review of the literature on appraisal reactions, a related purpose was to highlight important gaps in the literature and directions for future research. When it comes to measuring appraisal reactions, there are some standardized measures (e.g., Greller, 1975), and it seems that the measures used in the literature conform to a priori factor structures (Keeping & Levy, 2000). Many studies measure multiple appraisal reactions, for example, fairness and satisfaction, and other studies measure a composite of appraisal reactions (see Cawley et al., 1998; Pichler, 2012). Still others use original measures developed for the purposes of a given primary study. It is important that researchers 1) use consistent measures of appraisal reactions, and 2) when possible, test for a higher-order latent appraisal reactions factor (Keeping & Levy, 2000) and include this in substantive hypothesis testing. In this way, findings will be more comparable across studies.

In terms of appraisal context, a number of studies have linked due process performance appraisal to appraisal reactions. This literature has generally supported a key proposition of the due process metaphor developed by Folger et al. (1992), which is that with increased process fairness should come more favorable reactions to appraisal reviews. That said, most studies that have measured due process variables have not done so explicitly, that is they have not referenced the due process framework, with some important exceptions (e.g., Pichler et al., 2018; Taylor et al., 1995, 1998). It seems important that researchers more consistently leverage this framework and use the same variables across studies. For instance, when it comes to knowledge of performance standards, since no variable was developed and validated to measure this aspect of due process per se, researchers might consider using the validated perceived system knowledge measure by Williams and Levy (1992). The more consistently researchers use the same variables to operationalize due process constructs, the easier it will be to cumulate research evidence as to performance appraisal effectiveness and to draw statistically valid conclusions.

Although there are generally consistently positive relationships between most due process variables and appraisal reactions, one important exception is the relationship between feedback frequency (a dimension of adequate notice, Folger et al., 1992) and appraisal reactions. This is surprising because scholars and practitioners have long argued for more frequent feedback as a way to improve the appraisal process. After all, one reason appraisal reactions are often negative is because employees are surprised by their evaluation. More frequent feedback should reduce this surprise because, at least in theory, employees will have a better sense of where they stand going into the review. Pichler et al. (2018) found a multiplicative relationship between feedback frequency and knowledge of performance standards; they concluded that frequent feedback is more effective when employees have knowledge of performance standards. That said, it is important for future research to further investigate why frequent feedback is inconsistently related to appraisal reactions, and to examine why or under what conditions there might be differential relationships between feedback frequency and different appraisal reaction criteria.

The cumulative evidence as to the relationship between self-appraisals and appraisal reactions is also mixed. What is interesting about this is that few studies have attempted to investigate why—or, put differently, under what conditions self-appraisals are positively related to appraisal reactions. Many organizations encourage, and some even require, that employees self-evaluate prior to the appraisal review, ostensibly so as to increase fairness in the sense that employees have a voice in the process. That said, it is likely that managers vary widely in the extent to which they allow for and actually listen to the employee’s self-evaluation during the review. Moreover, self-appraisals might be related to disagreement during the review due to self-inflated ratings. It is important that future research more fully investigate the role of self-appraisals in the overall appraisal process and, more specifically, as related to appraisal reactions.

There has been very little research, relatively speaking, on other aspects of due process, especially supervisor neutrality or lack of bias and appraisal reactions, that is, part of the judgment based on evidence dimension of due process (Folger et al., 1992). This is interesting because employee perceptions of supervisors are closely tied to a variety of important variables, such as employee attitudes, rater–ratee relationship quality, and employee performance. It would seem that perceptions of supervisor bias could be not only a key predictor of appraisal reactions, but of attitudes and behavior following the review (Fig. 5.1). It would also be interesting to better understand the factors that predict employee perceptions of the extent to which supervisors are biased during the review. For instance, exchange quality going into the review is a likely predictor of employee perceptions of supervisor neutrality (Fig. 5.1). Again, this could be one way to connect due process performance appraisal to the social context of performance appraisal. Moreover, research has shown that relationship quality is higher and managers are less likely to deliberately bias performance ratings under conditions of due process (Taylor et al., 1998). Thus, it may be the case that due process characteristics, for example, high levels of knowledge of performance standards, might lead to higher levels of exchange quality and, hence, to perceived supervisor neutrality (Fig. 5.1).

Practical Implications

The literature on due process has shown that appraisal reactions are more favorable under conditions of due process even when performance ratings are low (Taylor et al., 1995). The literature on the social context of performance appraisal has also found that appraisal reactions are more favorably when rater–ratee relationship quality is high, even when ratings are low (Pichler, 2012). Taken together, these findings suggest that employees might not be as concerned about performance ratings as managers think. Surely performance ratings are important—they directly represent the manager’s view of the employee’s performance and are often tied to compensation and promotion opportunities (Aguinis, 2013). That said, the existing evidence suggests that other factors in the appraisal context and review may be more important. This is consistent with the key finding of the Cawley et al. (1998) meta-analysis, that is, that value-expressive voice is more important than decision control. It seems that, all else equal, employees favor being heard than having influence over the appraisal process; they also seem to favor feeling that their supervisor is supportive and trustworthy over favorable performance ratings.

These findings have important implications for managers and organizations. Organizational leaders should consider implementing due process characteristics, at least those that are feasible and are consistent with the organization’s culture. Some characteristics, such as knowledge of performance standards, might be easier to implement than others, for example, an appeals process. Disseminating employee handbooks, providing socialization programs, and developing ways by which managers and employees can become familiar with their job descriptions are processes that most large organizations already do, all of which might reduce employee surprise during the appraisal review. Implementing an appeals process, on the other hand, could be time- and resource-intensive. After all, this means that higher-level managers or human resources professionals make time to review contested performance evaluations. With that said, the existing evidence suggests that appeals processes are important to employees and to how they react to performance appraisals.

Another implication is that training managers to develop relatively high-quality relationships with employees is important. Research suggests that employee perceptions of relationship quality with their managers are more important as related to appraisal reactions than performance ratings or appraisal participation. This isn’t to say that appraisal participation is unimportant; research suggests the opposite (Cawley et al., 1998). What this does suggest is that training and development around developing trusting and supportive relationships is not only beneficial to employee health (e.g., Kossek, Pichler, Bodner, & Hammer, 2011), but that this can improve performance appraisal effectiveness. This type of training may not seem like an obvious way to improve performance appraisal, but if the goal is to increase feedback acceptance, and hence employee job performance, manager–employee relationships are key. This goes hand in hand with the findings from Cawley et al. (1998): Hearing an employee out during the review through, for instance, active listening seems more effective than giving them the impression that they’ve influenced the process or the manager’s decision. This is also consistent with from Meinecke et al. (2017), which documented that relationship-oriented statements during the appraisal interview are related to employee participation and appraisal satisfaction (from both manager and employee perspectives).

Finally, it seems that supervisor job knowledge is related to appraisal reactions. Most previous performance appraisal training focused on the psychometric approach, that is, training raters to mitigate rating errors and provide more accurate ratings. This type of training certainly has its place in organizations. That said, there isn’t much evidence that would support the proposition that this leads to more favorable appraisal reactions, and hence to appraisal effectiveness. Managers might consider focusing on becoming as familiar as possible with their employee’s job descriptions and discussing their work activities on a regular basis—and then reiterating this information during the appraisal review. This might seem like a simple solution, and perhaps it is—but we’ve probably all heard the complaint that employees feel that their manager doesn’t fully understand their role or their objective performance in that role. This may be especially the case when the manager in question hasn’t actually performed the same role in the past. Under these circumstances, it seems especially important for the manager to provide the employee with a clear sense that he or she understands the ins and outs of the work role, its requirements, and challenges to success. This is consistent with models of the appraisal process that have highlighted the importance of rater credibility in reactions to performance appraisal (Folger et al., 1992; Ilgen et al., 1979).

Conclusion

Employee reactions to performance appraisals, which include perceptions of appraisal accuracy, fairness, and satisfaction, for instance, are a key set of criteria by which appraisal effectiveness is evaluated. This is, in part, because acceptance of appraisal feedback is important if not necessary for performance appraisal to achieve its key purpose—to give employees feedback so that they can improve their performance. Due process performance appraisal and the social context of performance appraisal are two key predictors of appraisal reactions. There has been very little research integrating these two aspects of appraisal context, despite the notion that due process should influence the social context of appraisal (Folger et al., 1992). Thus, the research gaps identified in this review, and the concomitant model (Fig. 5.1), are meant to be guides for future research. Managers and organizations can improve appraisal effectiveness by developing high-quality relationships with employees and implementing due process characteristics in performance appraisal systems.