The Dynamic Influence of Personality on Psychological Contract Evaluations: a 2-Study Investigation of Approach/Avoidance Goals and Emotion Regulation Strategies

Abstract

Consistent with recent developments, we question the validity of trait neuroticism (N) and conscientiousness (C) as antecedents of psychological contract breach (PCB) and violation feelings. We investigate how key mediating (i.e., approach/avoidance goals) and moderating (i.e., emotion regulation strategies) mechanisms of the personality-PCB relationship operate over time. In Study 1 (550 observations), state N or C was associated positively with PCB and state N was associated positively with violation feelings. In study 2 (394 observations), state N was positively related to momentary avoidance goals, which in turn were related negatively to PCB and related positively to violation feelings. Moreover, suppression moderated the latter relationship; as suppression increased, the relationship between avoidance goals and violation feelings grew stronger. In contrast, state C was related positively to approach goals, which in turn were related positively to PCB and negatively to violation feelings. We discuss implications for theory and practice.

Employees have a formal written contract as well as a psychological contract (PC) with their employer. A PC emerges when employees engage in a reciprocal exchange relationship with their organization characterized by the exchange of organizational resources (e.g., providing flexible work schedules, career guidance, and/or mentoring) in return for employee contributions (e.g., performing essential duties, supporting organizational objectives; Rousseau, 1989; Rousseau et al., 2018). In the past two decades, organizational scholars from a variety of fields have demonstrated that when the PC is breached (i.e., PCB) feelings of violation are triggered (Armstrong, 2006; DiMatteo, 2013; Rousseau, 2011). These violation feelings may, in turn, prompt negative attitudinal and behavioral reactions such as reduced performance, satisfaction, commitment, increased turnover intentions, and counterproductive acts (for a meta-analysis see Zhao et al., 2007; for a review see Coyle-Shapiro et al., 2019).

Whereas PCB is the cognitive component of the PC that arises from a perceived discrepancy between organizational obligations and actual delivered resources, violation feelings refer to the ensuing negative emotional state (Morrison & Robinson, 1997). The extent to which one experiences PCB or violation feelings varies since both PCB and violation feelings are inherently subjective intra-personal processes (Robinson & Morrison, 2000; Rousseau, 1989). Therefore, PCB is not perceived automatically when an organization fails to uphold its obligations; the likelihood of perceiving a PC-deviation as a PCB thus differs from one employee to another (Rousseau et al., 2018; Schalk & Roe, 2007).

Most of studies in the field have dealt with the above-described consequences of PCB (Coyle-Shapiro et al., 2019; Zhao et al., 2007). As Conway and Briner (2005) suggest, “while many outcomes have been considered, the basic explanation behind the findings of these empirical studies is very simple. When employees perceive the organization to have breached the PC they view their relationship with the organization more negatively and are less likely to do things for the organization” (p. 70). Considering substantive evidence concerning the general potential of the concept of the PC, it is surprising that possible antecedents have received relatively little attention (for a notable exception see Vantilborgh et al., 2016).

The antecedents of PC and PCB remain poorly understood. If PCs are idiosyncratic, personality dimensions should significantly impact them. Successful organizations foster knowledge-sharing PCs, acknowledging the role of employees' personalities. Despite over two decades of research, few studies explore the link between personality and PCs. Existing work (Ho et al., 2004; Nikolaou et al., 2007; Raja et al., 2004; Tallman & Bruning, 2008) offers valuable insights into this connection. Personality, a key predictor of cognitive evaluations and sensitivity to workplace events, influences reactions to positive and negative feelings like violation feelings (Perugini et al., 2003).

Previous research on personality and PCs, such as Orvis et al. (2008), Raja et al. (2004), Restubog et al. (2007), Saeed (2020), and Tallman and Bruning (2008) focused on stable between-person effects. This approach neglected explanatory mechanisms (e.g., approach and avoidance goals) and boundary conditions (e.g., emotion regulation strategies) influencing the state personality and PC relationship. This gap is crucial for understanding and addressing PCB perceptions. Exploring approach/avoidance goals reveals how certain traits influence individuals' responses to work situations, impacting PCB perceptions. Emotion regulation strategies serve as critical boundary conditions moderating the state personality-PC relationship, offering pathways for modification or influence. This holistic perspective aids scholars and practitioners, preventing oversimplifications and addressing relationship complexities. Examining state personality fluctuations contributes to predictive models for employee behavior, enhancing the ability to anticipate responses to workplace conditions. Practically, understanding explanatory mechanisms and boundary conditions informs organizational policies. If avoidance goals link to PCB, goal-setting processes may align with employees' motivations. Emotion regulation insights could guide organizations in fostering supportive emotional climates, reducing PCB likelihood. Managers, aware of state C and N fluctuations influencing PC perceptions, can tailor leadership and communication styles, accordingly, adjusting feedback mechanisms, recognizing stability's importance, or mitigating neuroticism-related stress impacts.

In the current 2-study paper we integrate the literatures on state personality, PCs, and key mechanisms underlying this relationship (i.e., approach/avoidance goals and emotion regulation strategies, e.g., Ferris et al., 2011). Specifically, we develop a framework that addresses three critical issues. First, we sought to elucidate personality-PC relationships by exploring underlying mechanisms, utilizing the approach/avoidance framework (Elliot & Thrash, 2002; Ferris et al., 2011). We argue that momentary approach/avoidance goals mediate within-person personality fluctuations, PCB, and violation feelings intensity. State neuroticism (N) and conscientiousness (C) impact PC-evaluations and affective reactions differently based on their effects on approach versus avoidance goals.

Second, we further differentiate the relationship between approach/avoidance goals, PCB, and the intensity of violation feelings by introducing relevant self-regulatory processes, namely, the emotion regulation strategies of reappraisal and suppression (Gross & John, 2003). We argue that these emotion regulation strategies play an important role in either mitigating or aggravating the relationships between approach/avoidance goals and (1) PCB perceptions and (2) the intensity of violation feelings.

Third, our understanding of the personality-PCB relationship often relies on cross-sectional or longitudinal studies with substantial time lags. Recent developments challenge the static treatment of personality and PCs (Debusscher et al., 2016; Heller et al., 2007; Fleeson, 2012; Horstmann & Ziegler, 2020; Sosnowska et al., 2019, 2020; Griep & Vantilborgh, 2018a, b; Lambert et al., 2003; Rousseau et al., 2018). Advocating for a within-person approach, we propose using more measurement waves to explore PC breach and its dynamics within the same individual over time. This approach aligns with recent theoretical developments in PC research and broader organizational behaviour studies (Chen & Mathieu, 2008; Chen et al., 2011; Hausknecht et al., 2011). By employing time-lagged variables, we aim to understand how personality at time T influences PCB perceptions at time T + 1 within short intervals, emphasizing the temporal evolution of individuals and their interactions with constructs (Curran & Bauer, 2011; Fisher & To, 2012). Such within-person research provides temporal precision, elucidates dynamic phenomena, and offers novel insights about constructs and their relationships (Rousseau et al., 2018) that are not possible with a between-person perspective.

From a practical perspective, focusing on state neuroticism (N) and conscientiousness (C) as predictors of PCB and violation feelings allows organizations to more effectively address downstream impacts on their employees. While many organizations use personality tests in hiring for insights into candidates' traits, preferences, and work styles, our study suggests viewing these tests as ongoing tools to assess how employees continually evaluate their PC. Woods et al. (2013) caution against one-time personality assessments due to personality fluctuations over time and situations. Recognizing the malleable nature of personality can provide organizations with better insight into the dynamic interplay between employees' changing personalities and their PC evaluations.

We tested within-person predictions in two studies. Study 1 employed two daily prompts to examine the direct relationship between state personality fluctuations and (1) PCB, and (2) violation feelings. Study 2 utilized three daily prompts to explore (1) the mediating role of approach/avoidance goals in the personality-PCB relationship, and (2) the moderating impact of emotion regulation strategies on the connection between approach/avoidance goals and the intensity of PCB and violation feelings.

Theoretical Framework and Hypotheses Study 1

From a Between to a Within-Person Perspective on the Personality – PC Relationship

The stability and change of personality have been a controversial issue in personality psychology and a source of heated debate. Organizational and management research has historically treated personality traits as stable and unaffected by organizational influences. Recently, organizational scholars have begun to acknowledge that personality is malleable and have called for incorporating personality change into current research (Li et al., 2014, 2019; Tasselli et al., 2018; Woods et al., 2019). Indeed, ample evidence shows that personality characteristics are not fixed (i.e., the trait between-person differences approach) but rather change over time (i.e., the state within-person approach) (Anusic & Schimmack, 2016; Ferguson, 2010; Specht et al., 2014; Wagner et al., 2020).

Recent theoretical developments and research in the domain of personality emphasize the within-person variability of an individual’s personality (for excellent reviews of within-person variation in personality see Beck & Jackson, 2021; Jayawickreme et al., 2021; Kuper et al., 2021; Rauthmann, 2021). For example, changing factors over time were found to be responsible for 17% of the variance in personality over time (see Anusic & Schimmack, 2016). Moreover, and pertinent to this specific study, the amount of within-person variability in neuroticism and conscientiousness tends to be as large, or even larger, than between-person variability (e.g., Debusscher et al., 2016; Fleeson, 2012; Heller, et al., 2007). Furthermore, neuroticism and conscientiousness showed the strongest link with the PC in previous research, compared to the other three Big Five dimensions (Orvis et al., 2008; Raja et al., 2004). Therefore, we focus on these dimensions in the current study.

Because we are taking a within-person perspective of personality, we do not focus on stable traits (McCrae & Costa, 1992) but rather on the more volatile personality states (Fleeson, 2007). These states are conceptualized as, “momentary enactments that have the same affective, behavioral, and cognitive content as their corresponding traits” (Fleeson, 2012, p. 52). More specifically, state neuroticism (state N) encompasses the momentary level of anxiety, distress, and impulsivity a person has, whereas state conscientiousness (state C) represents a person’s momentary level of orderliness, dutifulness, and achievement striving. Underwriting the superior predictive power of said personality states, Hoff et al. (2021) demonstrated that fluctuations in neuroticism and conscientiousness had stronger effects on career outcomes compared to their stable trait level counterparts. Next, research on within-person fluctuations in personality states has already been applied in various work-related studies, including studies on job performance (Debusscher et al., 2016, 2017; Huang & Ryan, 2011), learning transfer (Huang & Bramble, 2016), mood and job satisfaction (Judge & Ilies, 2002), and work motivation (Judge et al., 2014).

As the static nature of personality has become an element of debate, so too has the static nature of PC research and theory (e.g., Griep & Vantilborgh, 2018a, b; Lambert et al., 2003; Rousseau et al., 2018). New theoretical developments have been advanced in the PC field (e.g., Griep et al., 2019; Hansen & Griep, 2016; Tomprou et al., 2015; Rousseau et al., 2018) to epitomize a more dynamic perspective on the PC, thereby recognizing that intra-individual differences exist in PCB perceptions and violation feelings. Following these re-conceptualizations of personality and PC from the static between-person perspective to the dynamic within-person perspective, we will first explore the relationship between momentary fluctuations in State N and C in relation to PCB and violation feelings.

The Relationship Between State Neuroticism and PCB and Violation Feelings

The role of neuroticism in the PC literature is typically limited to the relationship between neuroticism (high or low neuroticism as a between-person difference) and the type of PC resources (i.e., relational or transactional) employees will focus on (e.g., Hassan et al., 2020; Raja et al., 2004; Vantilborgh et al., 2013). The consensus is that employees who are highly neurotic will focus on transactional, more extrinsic, PC resources such as salary or health care benefits (Nikolaou et al., 2007; Tallman & Bruning, 2008). Despite numerous calls in the literature for more research on the role of personality in the exploration of PCB and its outcomes (e.g., Coyle-Shapiro et al., 2019; Zhao et al., 2007), there are very few empirical studies to date.

Raja et al. (2004) found that high neuroticism was positively related to PCB. Furthermore, Tallman and Bruning (2008) found that neuroticism was positively related to perceptions of organizational obligations to fulfill employee needs for growth, as well as perceptions of these obligations being breached by the organization. Next, Agarwal (2017) and Jafri (2014) found that neuroticism was positively related to PCB. People high on neuroticism are generally anxious, tense and more impulsive, which would support the argument that highly neurotic people are highly likely to scrutinize their environment for cues hinting towards potential mishaps such as PCB. Erstwhile, due to their underlying anxious and tense nature, highly neurotic people are also more likely to experience pronounced emotional reactions following negative workplace events such as PCB. In line with this assumption, Ho et al. (2004) reported stronger negative emotional responses to broken employer promises (antecedents of PC obligations; Rousseau et al., 2018) for respondents who scored high on neuroticism. In sum, it thus seems that neuroticism relates positively to PCB and triggers stronger negative reactions to PCB. Given that state personality mimics trait personality (e.g., Debusscher et al., 2014, 2016; Fleeson, 2012), we hypothesize the following:

• Hypothesis 1: Being momentarily high in state N relates positively to (a) perceiving PCB and (b) the intensity of violation feelings.

The Relationship Between State Conscientiousness, PCB, and Violation Feelings

As with neuroticism, the PC literature generally focuses on conscientiousness in relation to employees’ preference for specific PC resources (e.g., Metz et al., 2016; Ntalianis et al., 2015; Raja et al., 2004; Vantilborgh et al., 2013). For instance, Ntalianis et al. (2015) and Raja et al. (2004) found a positive relationship between conscientiousness and a preference for intrinsic PC resources, as well as the formation of a relational PC characterized by more socio-emotional elements such as social support and approachable supervisors. Other studies have focused on the employer, rather than the employee, and found that managers and small company owners who are high in conscientiousness perceive an increased obligation to provide a relational PC to their employees (Metz et al., 2016; Ntalianis et al., 2015).

Turning to PCB, Raja et al. (2004) reported that people who were higher in conscientiousness reported less PCB. More recently, several scholars (Agarwal, 2017; Jafri, 2014; Saeed, 2020) found that conscientiousness was negatively related to PCB. Indeed, highly conscientious employees would be less likely to perceive breach because organizations are less likely to breach the PCs of such hard-working and high performing employees. Furthermore, conscientiousness has been found to relate negatively to anger and, more broadly, negative emotional reactions (e.g., violation feelings; Jensen-Campbell et al., 2007). This finding was further supported by Orvis et al. (2008) who found that employees lower in conscientious reacted more negatively following PCB, and by Jensen et al. (2010) who found that conscientiousness moderated the relationship between PCB and the enactment of production deviance. We therefore expect that employees who are momentarily high in conscientiousness will be less likely to perceive PCB and exhibit less intense emotional reactions (i.e., less violation feelings) following PCB, just as those who are higher in trait conscientiousness (Debusscher et al., 2014, 2016; Fleeson, 2012).

• Hypothesis 2: Being momentarily high in state C relates negatively to (a) perceiving PCB and (b) the intensity of violation feelings.

Method Study 1

Procedure

To establish initial empirical evidence for the relationship between state N and C, and PCB and violation feelings, we conducted an experience sampling study among Belgian employees from sixteen different organizational sectors. We translated all survey items into Dutch and had three colleagues back-translate the items into English. Inconsistencies between the translation and back-translation were discussed and resolved. Respondents were recruited via e-mail and asked to complete a single general survey prior to completing two daily prompts for five consecutive days. We used an Experience Sampling Methodology (ESM) because we were interested in the dynamic relationship—that is, the study of within-person processes as they unfold over time—between state personality and PC evaluations. An ESM design is ideally suited to account for the temporal context underlying respondents’ affective states, behaviors, and cognitions in their everyday work environment (e.g., Beal, 2011; Fisher & To, 2012). We sent the morning prompt at a random time between 10.30AM and 11.30AM and the afternoon prompt at a random time between 3.30PM and 4.30PM. Respondents were required to respond to the morning prompt before 2.30PM (i.e., as such both prompts were at least one hour apart) and to respond to the afternoon prompt before 10PM (i.e., to avoid potential distorting effects of work detachment). Each prompt received an electronic time stamp. We coded responses as missing data when they failed to (timely) complete the survey. A similar prompting logic was used by Debusscher et al. (2014) in their study on the effects of state neuroticism on momentary task performance. Each prompt took approximately 2 min to complete, and respondents were not financially compensated for their time; details follow.

Participants

We contacted 88 Belgian employees, of whom 70 completed the general online survey and took part in the ESM study (response rate of 79.55%). Our data have a multilevel nature, meaning that the unit of analysis equals ‘prompts’ rather than ‘respondents’ (Conway & Briner, 2002), resulting in an effective sample size of 550 observations (70 respondents x average of 7.86 responses per individual). Our sample of 70 respondents and 550 observations has sufficient power to provide an accurate estimate of standard errors and fixed effects (Browne & Draper, 2000; Maas & Hox, 2005). Our respondents were, on average, 39.60 years old (SD = 12.22), 51.5% were female, 89.40% obtained a higher educational degree, 86.40% had a full-time contract, and the average company tenure was 13.06 years (SD = 13.83). We conducted logistic regression analyses to estimate differences between our final sample and dropouts during the experience sampling prompts.¹ Dropout could only be explained by respondents’ age and tenure; older (β = −0.64, SE = 0.007, p < 0.01) and more tenured (β = −0.68, SE = 0.008, p < 0.01) respondents were less likely to drop out.

Measures²

General Survey Measures

We used the general online survey to collect demographic information on respondents’ age (in years), gender (female or male), educational background (highest level of formal education), employment status (full- or part-time), company tenure (in years), and the level of obligated resources, which was assessed to confirm that the PC resources were relevant to this sample. Respondents rated the extent to which they believed that their employer was obligated to provide them with each of 20 items on a 5-point scale (1 = “minimally or not at all”, 5 = “to a very large extent”; for a similar approach see Montes & Irving, 2008). These items represented the most widely studied types of resources: transactional and relational, as well as ideological resources. In this study we also included ideological resources because previous studies (e.g., Vantilborgh et al., 2013) demonstrated that respondents may also perceived this type of PC resources as part of their PC. The scores for the transactional (α = 0.84), relational (α = 0.84), and ideological (α = 0.96) obligations ranged from 3.92 to 4.37, 4.03 to 4.81, and 3.88 to 4.25, respectively, indicating that these resources were all relevant for this sample.

Experience Sampling Measures (Prompts)

For the ESM, we used shortened scales to ensure a reasonable length and to avoid endangering the compliance of respondents (Beal, 2011). In addition, we counterbalanced scales to rule out potential order effects in the results (Fisher & To, 2012). Further, to clarify and reinforce the timing focus of respondents, all items were worded such that they included “at the present moment” for the morning prompts, and “since the previous report” for the afternoon prompts.

State N and state C were assessed—during the morning prompt—with Saucier’s (1994) Mini-Markers scale. For state N, this scale contains six positive (e.g., relaxed, unenvious) and two negative adjectives (e.g., temperamental, moody). For state C, there are four positive (e.g., practical, organized) and four negative adjectives (e.g., sloppy, inefficient). We asked respondents to indicate how accurately the adjectives described them at the present moment (i.e., since their day had started) on a 9-point Likert scale ranging from (1) “extremely inaccurate” to (9) “extremely accurate”. We estimated level-specific reliability using the multilevel confirmatory factor analysis approach advocated by Geldhof et al. (2014). The within-person omega reliability coefficient for state N (ω = 0.68) and state C (ω = 0.69) were significant and satisfactory (Geldhof et al., 2014).

PCB was measured—during the afternoon prompt—by presenting the list of 20 commonly used PC resources and asking participants to indicate whether (yes or no) their organization had breached obligations for one or more of the PC resources (Griep et al., 2016; Solinger et al., 2016). Thus, we assessed PCB globally (Robinson & Rousseau, 1994) as opposed to assessing various facets of PCB to keep the survey length at a reasonable level while at the same time maximizing the probability of capturing PCB incidents. That is, we likely could not have captured many episodes of PCB within a single day if we only inquired about a select few PC resources. When respondents indicated that at least one PC resource was breached, PCB was coded as one (73 out of 550 prompts). When respondents indicated that no PC resources were breached, PCB was coded as zero (Griep et al., 2016; Solinger et al., 2016). Note that this approach aligns with theoretical arguments by Schalk and Roe (2007) stating that one will only report a PCB (i.e., indicate “yes” to the dichotomous single item) once a deviation from the organizational obligations exceeds one’s personal threshold. In all other cases, one will not report a PCB (i.e., indicate “no” to the dichotomous single item). This single dichotomous measure of PCB has been used in previous studies and was found to correlate significantly and in the expected direction with Robinson and Morrison’s (2000) traditional PCB measure (e.g., Griep et al., 2016), confirming its validity (Fisher & To, 2012).

Feelings of violation were measured—during the afternoon prompt—with a single item (i.e., “To what extent did the breach of this (these) obligation(s) have a negative emotional effect on you during the past day”). Responses ranged from (1) “minimally or not at all” to (5) “to a very great extent” (Solinger et al., 2016). Although, all respondents were presented with this item, those who did not perceive a breach event indicated that the item was not applicable to their situation. Hence, these respondents received a code of zero on violation feelings (Griep et al., 2016). Note that this aligns with the theoretical arguments that one cannot experience violation feelings in the absence of a preceding PCB (Morrison & Robinson, 1997). In previous studies, this single item correlated significantly and in the expected direction with global measures of negative affectivity (e.g., Griep et al., 2016), confirming its construct validity (Fisher & To, 2012).

Analytic Strategy

In this study, we were interested in predicting perceptions of PCB and the intensity of violation feelings as a function of momentary fluctuations in state N and state C. To address this, we combined the information from the PCB measure (i.e., yes or no response format) and the violation feelings measure (i.e., continuous response option) into a single variable containing a binary and a count part. The binary part consisted of 0’s (i.e., “no” response) and 1’s (i.e., “yes” response) indicating the absence or the presence of PCB, respectively. In line with the theoretical reasoning behind a logistic probability, this binary part can be used to estimate the extent to which one perceives a PCB. The count part is a continuous variable that represents the intensity of violation feelings. This count part contains information on both the intensity of violation feelings (i.e., the continuous variable), as well as the presence of a preceding perception of PCB (i.e., the binary part). To model such a binary and count part, a zero-inflated Poisson regression (ZIP) model is required (Lambert, 1992).

A ZIP model is based on a zero-inflated probability distribution—cf. normal distribution—that allows for a high prevalence of zero-valued observations (i.e., no PCB and no violation feelings), as well as for the positively skewed distribution of violation feelings (i.e., zero is the most common observation, followed by a decreasing number of observations when moving along the Likert-scale) to be taken into account when analysing the data (see also Conway & Briner, 2002; Griep et al., 2016; Vantilborgh et al., 2013). A ZIP model assumes that with probability p the only possible observation is zero (i.e., no PCB and no violation feelings), whereas the probability 1 – p is a Poisson random variable representing the intensity of violation feelings. In addition, we can predict the probability p of the perfect, zero state (i.e., absence of PCB and violation feelings), as well as the mean number of violation feelings in the imperfect state (i.e., presence of preceding PCB) based on other variables (i.e., momentary fluctuations in state N or state C). Note that the characteristics of this ZIP model thus account for the fact that violation feelings are an emotional state resulting from a preceding perception of PCB (Morrison & Robinson, 1997).

We predicted everyone’s likelihood to perceive PCB and the intensity of the experienced violation feelings during the afternoon (time T2) based on that same individual’s momentary fluctuations in state N and state C scores during the morning (time T1). State N and state C scores were allowed to correlate, thereby acknowledging that one could, for example, score high on both states during the morning prompt. Because these data have a 2-level nested structure (i.e., morning and afternoon prompt nested within individuals), we estimated intra-class correlation coefficients (ICC) of state N, state C, PCB, and violation feelings to assess the need for a multilevel model (Hox, 2010). Results indicated that a considerable proportion of the variance in these variables (ICC values are 0.43, 0.43, 0.27, and 0.36, respectively) could be attributed to within-person differences.

Consequently, we used person-mean centering when estimating a 2-level ZIP regression model (Lee et al., 2006; Maas & Hox, 2005) because person-mean centering generally yield more accurate variance estimates of within-person effects than grand-mean centering or no centering when interested in within-person effects (see Raudenbush & Bryk, 2002; Wang & Maxwell, 2015). We conducted all analysis in Mplus version 7.1 (Muthén & Muthén, 2012) and used the MLR estimator. At this point, we would like to point out that we did not include any control variables at the between-person level (e.g., age, gender, education), because (a) these do not directly influence within-person effects and (b) we focused on within-person as opposed to between-person effects.

Results

(Multilevel) Confirmatory Factor Analysis (CFA)

We performed a CFA at the between-person level to ensure relevance of the transactional, relational, and ideological items for this sample. We used Dyer et al. (2005) conventional standards to assess model fit and compared competing models using loglikelihood ratio tests. The theorized 3-factor model fit the data well (RMSEA = 0.07, CFI = 0.97, TLI = 0.96, SRMR = 0.08), with each PC item loading significantly, and in the expected direction, onto its respective latent factor, with one factor making up the transactional resources, another factor making up the relational resources, and a final factor making up the ideological resources (Table 1). Alternative model A (∆χ ² = 33.52, ∆df = 2, p < 0.001), alternative model B (∆χ² = 76.65, ∆df = 2, p < 0.001), alternative model C (∆χ² = 82.10, ∆df = 2, p < 0.001), and alternative model D (∆χ² = 140.26, ∆df = 3, p < 0.001) fit the data significantly worse. Hence, we are confident that the transactional, relational, and ideological items can be used to assess our respondents’ perceptions of obligated PC resources in the current sample.

Table 1 Results from (multilevel) confirmatory factor analyses (Study 1)

Next, we tested whether state N, state C, PCB, and violation feelings could be empirically distinguished from each other using a series of multilevel CFAs. We compared a theoretical model containing four first-order latent factors to three alternative models (see Table 1). The theorized 4-factor model fit the data well, with each item loading significantly, and in the expected direction, onto its respective latent factor (Table 1). Although alternative model B fit equally well to the data as the theoretical model (∆χ² = 1.24, ∆df = 1, p = 0.27), it had to be rejected because not all items loaded significantly on their latent factor, and several fit indices did not reach the suggested cut-off values. Alternative model A (∆χ² = 374.24, ∆df = 3, p < 0.001), and alternative model C (∆χ² = 646.62, ∆df = 4, p < 0.001) fit the data significantly worse. Hence, our 4-factor theoretical model (RMSEA = 0.07, CFI = 0.95, TLI = 0.92, SRMR = 0.07) guided hypotheses testing.

Descriptive Results

Table 2 provides an overview of the means, standard deviations, and correlations at both the between-person and within-person level of the variables under study. In both cases we accounted for the conditional relationship between violation feelings and PCB by (1) computing the correlations between violation feelings and all other variables on a subset of the data (i.e., those data points representing PCB), and (2) computing no correlation between violation feelings and PCB as this would result in an artificially inflated correlation.

Table 2 Means, Standard deviations, and correlations among the focal variables (Study 1)

Hypothesis Testing

Figure 1a and b show the standardized estimated paths in the 2-level ZIP regression model. As can be seen in Fig. 1a, our results indicate that being momentarily high in state N or state C during the morning relates positively to the likelihood of perceiving PCB during the afternoon. These findings support Hypothesis 1a, whereas they contradict Hypothesis 2a. In addition, being momentarily high in state N during the morning relates positively to the intensity of violation feelings during the afternoon (Fig. 1b). These findings support Hypothesis 1b. No significant relationship was found between state C and the intensity of violation feelings, failing to support Hypothesis 2b.

Fig. 1

a (left pane) and b (right pane): Standardized estimated paths in the 2-level ZIP regression model. Notes. * p < .05. ** p < .01. *** p < .001. (N = 550). Results indicate change in each variable by controlling for the auto-correlation at the previous moment in time

Sensitivity Analysis

Research suggests that employees high on trait N tend to form transactional PCs, whereas employees high on trait C tend to form relational PCs (e.g., Metz et al., 2016; Ntalianis et al., 2015; Raja et al., 2004; Vantilborgh et al., 2013). As such, we created a composite score for each PC type (i.e., transactional, relational, ideological) and tested for potential differential effects on the outcomes under study (multi-group equivalence test; Byrne & van de Vijver, 2010). We compared a model containing separate regression parameters for each PC type (i.e., configural 2-level mixture ZIP regression model) with a model in which these regression parameters were constrained to be equal for all PC types (i.e., constrained 2-level mixture ZIP regression model). Results indicated that the proposed relationships in our hypothesized model did not differ significantly among the three (i.e., transactional, relational, and ideological) PC types [χ² (10, N = 70) = 1.27, p = 0.99]. This implies that momentary fluctuations in state N and state C have a similar influence on the likelihood to perceive a PCB and the intensity of violation feelings irrespective of the PC type.

Discussion

We found that momentary levels of state N and state C were positively related to perceptions of PCB, thus indicating that being momentarily high on either state N and state C seems to result in an increased tendency to notice and attend to discrepancies between perceived employer obligations and actual delivered resources. Although momentary levels of state N and state C were both positively related to the likelihood of perceiving PCB, our results demonstrated that the intensity of violation feelings only increased as a function of momentary levels of state N. This seems to indicate that only when an individual is momentarily high in state N, will that person attend to, and react negatively to, negative stimuli during the monitoring process, resulting in heightened violation feelings (Ferris et al., 2011; Perugini et al., 2003). However, give the extant findings regarding trait conscientiousness and PCB, we sought to re-examine this relationship in a second study.

Study 1 was associated with several shortcomings, which we sought to overcome in Study 2. First, we included additional variables to further elucidate the mechanisms underlying the personality-PC relationship, akin to studying its boundary conditions. More specifically, we build on the approach/avoidance framework (Elliot & Thrash, 2002; Ferris et al., 2011), to explicitly test two potential mediators (i.e., approach and avoidance goals) of the relationship between momentary fluctuations in state N and state C, and perceptions of PCB or violation feelings. Furthermore, we draw on two prolific emotion regulation strategies, namely reappraisal and suppression (Gross & John, 2003), because the extent to which an employee relies on these strategies is likely to influence the strength of the avoidance goal – PC evaluation (i.e., PCB and violation feelings) relationship. Second, we included different measures of PCB and violation feelings to enable validation and replication of the findings we obtained using single items in Study 1. By doing so, we were able to conceptually replicate the proposed relationships across different measurement instruments. We believe that such a conceptual replication is warranted because there are various operationalizations of PCB and violation feelings available in the literature. Our measure of PCB was dichotomous, global, and did not refer to the extent to which an employee received each of the PC resources relative to the extent to which that employee perceived these PC resources to be obligated by the employer.

To overcome these limitations, Study 2 deployed a direct comparison measure of PCB to assess the intensity of PCB. On a related matter, although Morrison and Robinson (1997) conceptualized violation feelings as a mixture of negative emotions, our operationalization of violation feelings in Study 1 did not reflect a wide selection of negative emotions. Therefore, in Study 2, we used a measure of job-related negative affect to capture this mixture of negative emotions. Finally, given that Study 1 was conducted among a sample of Belgian employees, Study 2 relied on a sample of US employees from diverse organizations to increase the generalizability of the results.

Theoretical Framework and Hypotheses Study 2

The Mediating Role of the Approach/Avoidance Goals

Generally, people tend to approach positive end states or to move away from negative end states (Higgins, 1997). Indeed, research has shown that individuals with an approach motivation tend to be highly sensitive to positive information as they search for, and attempt to procure, valued outcomes. In contrast, individuals with an avoidance motivation are highly sensitive to negative information as they search for, and try to avoid, negative outcomes (Elliot, 1999). Research suggests that these motivational tendencies align with aspects of personality to affect attitudes and behaviors (e.g., Elliot & Thrash, 2002; Ferris et al., 2011, 2013; Larsen & Ketelaar, 1991). For example, Elliot and Thrash (2002) showed that neuroticism was strongly associated with avoidance. Like distinctions made about similar constructs (e.g., behavioral activation system vs. behavioral inhibition system; Gray, 1970), approach/avoidance motivation reflects a biological sensitivity to positive/negative stimuli; this motivation influences vigilance for corresponding positive/negative cues in the environment and influences positive/negative affective reactions and behavior (Elliot & Thrash, 2002). Thus, the approach/avoidance framework (Elliot & Thrash, 2002; Ferris et al., 2011) is a useful starting place for explaining relations between personality and PC experiences (i.e., PCB and violation feelings).

Approach and avoidance motivation are linked tightly to the structure of personality; both approach/avoidance motivation and personality consist of a biological stable dispositional component and a more volatile social-cognitive component (Elliot & Thrash, 2002). For approach/avoidance motivation, researchers differentiate between general approach or avoidance temperaments or tendencies (i.e., between-person fluctuations) and approach or avoidance goals (i.e., within-person fluctuations) (Elliot & Thrash, 2002). Approach and avoidance temperaments are innate between-person differences in sensitivity and reactivity toward positive and negative stimuli. Specifically, an approach temperament involves a sensitivity toward positive stimuli, whereas avoidance temperament encompasses a sensitivity toward negative stimuli (Elliot & Thrash, 2002). In addition, approach and avoidance temperaments are characterized by perceptual vigilance, affective reactivity, and a behavioral predisposition to these positive (and respectively negative) stimuli (Elliot & Thrash, 2002). Although these temperaments account for how behavior is instigated (i.e., they are an expression of biological approach/avoidance motivation), the within-person approach/avoidance goals account for the direction of the behavior and how and when it is displayed (Elliot & Thrash, 2002). In other words, these approach/avoidance goals have a strong self-regulatory function guiding the more general approach/avoidance tendencies. This implies that two employees can have the same approach/avoidance temperaments, while having different approach/avoidance goals, thereby also differing in their attitudes and behaviors.

A similar subdivision into a dispositional and a social-cognitive component is observed in personality wherein personality traits account for the between-person dispositional aspect and personality states are part of the within-person social-cognitive component. In this article, we focus on the within-person social-cognitive component (i.e., the momentary level) of the approach/avoidance framework and within-person personality fluctuations as our study conceptualizes personality and PC experiences as dynamic within-person constructs. We expect an employee’s approach/avoidance goals—which are volatile in nature—to be influenced by the individual’s momentary personality states. In other words, we expect the relationship between state personality (i.e., state N and state C) and PCB and intensity of violation feelings to be mediated by approach/avoidance goals.

Avoidance Goals—Mediating the Relationship Between State N and PCB.

In line with the approach/avoidance framework (Ferris et al., 2011) we argue that when employees are momentarily high on state N, they will be motivated by momentary avoidance goals and consequently will focus more on negative stimuli in the work environment. This heightened momentary avoidance motivation implies that employees are likely to adopt avoidance goals (see Elliot & Thrash, 2002), leading them to vigilantly monitor the extent to which the organization fails to deliver on its obligations (Rousseau et al., 2018; Schalk & Roe, 2007). Such individuals are motivated to do so to prevent negative outcomes from occurring (i.e., avoid losing a desired resource to prevent doing badly; Elliot, 1999). Because of this momentary increase in monitoring, employees are more likely to notice and attend to negative stimuli, such as inconsistencies in the PC (Morrison & Robinson, 1997). In other words, an employee who is momentarily high in state N will be more likely to perceive PCB because of an increased tendency to notice and attend to negative cues pointing towards a discrepancy between the initially obligated resources and the delivered resources (Adler & Obstfeld, 2007; Öhman et al., 2001). These theoretical arguments lead to the following hypothesis:

• Hypothesis 1a: Being momentarily high in state N relates positively to momentary avoidance goals, which in turn relate positively to perceiving PCB.

Approach Goals—Mediating the Relationship Between State C and PCB

Employees who are momentarily high in state C are strongly driven by approach goals and thus are highly attentive to positive stimuli during the monitoring of cues and actions (Zweig & Webster, 2004). Unlike employees who are momentarily high in state N, individuals high in state C scan their environment for positive information. This implies that employees who are momentarily high in state C monitor (Huang & Ryan, 2011) the extent to which an organization delivers valued resources—i.e., their goal is not to identify instances in which their PC has been breached. As such, individuals high in state C will be less likely to perceive PCB because of their predominant slant to monitoring for and procuring positive outcomes (i.e., striving to attain a valued resource to achieve hopes and dreams; Elliot, 1999). In addition, Rousseau et al. (2018) argue that when one perceives or focuses on positive emotions—being characteristic of someone with approach goals—one will be less likely to notice PCB or, if significant enough to attract attention, to dismiss such discrepancies. This brings us to the following hypothesis:

• Hypothesis 1b: Being momentarily high in state C relates positively to momentary approach goals, which in turn relates negatively to perceiving PCB.

Avoidance Goals—Mediating the Relationship Between State N and Violation Feelings

As previously argued, we expect employees who are momentarily high in state N to possess momentary avoidance goals, meaning that they will focus more on negative stimuli and will show stronger reactions as a result (Ferris et al., 2011; Perugini et al., 2003). This heightened affective reactivity toward negative stimuli for people high in avoidance motivation (avoidance goals) is also acknowledged by Elliot and Thrash (2002) and is consistent with findings of Suls and Martin (2005). Suls and Martin examined the relationship between trait neuroticism and emotional reactivity toward negative events at the between-persons level. They found that stable inter-individual differences in trait N relate to stronger reactions to negative stimuli (i.e., PCB). Given that trait and state N result in similar reactions (Fleeson, 2012), we expect that, when PCB is perceived by individuals who are momentarily high in state N, their high momentary avoidance goals will result in particularly strong violation feelings (i.e., a collection of negative emotions). Consequently, we hypothesize:

• Hypothesis 2a: Being momentarily high in State N relates positively to momentary avoidance goals, which in turn relate positively to the intensity of violation feelings.

Approach Goals—Mediating the Relationship Between State C and Violation Feelings

Drawing again on the approach/avoidance framework (e.g., Ferris et al., 2011), we expect that employees who are momentarily high in state C will be characterized by momentary approach goals, which in turn result in less intense violation feelings (Ferris et al., 2011; Perugini et al., 2003). This reasoning is based on findings that high momentary state C is linked to high momentary approach motivation, which involves adopting more approach goals (Ferris et al., 2011; Zweig & Webster, 2004). At the core of this, momentary approach goals are strongly focused on positive, instead of negative, stimuli. In turn, this positive outlook (e.g., ability to delay gratification, focus on success and achievement) is expected to attenuate the negative reactions that may follow from perceptions of PCB. Therefore, violation feelings are likely to be less intensely felt by individuals with a momentarily high in state C. As such, we hypothesize:

• Hypothesis 2b: Being momentarily high in state C relates positively to momentary approach goals, which in turn relates negatively to the intensity of violation feelings.

The Moderating Role of Emotion Regulation Strategies

It is widely accepted that not all discrepancies between obligations and actual delivered resources are perceived as PCB and not all instances of PCB result in strong feelings of violation (e.g., Rousseau et al., 2018). One factor that may contribute to these inconsistencies is the tendency for people to engage in emotion regulation strategies. In the current research, we explored the moderating role of emotion regulation strategies to bolster understanding of the self-regulatory processes affecting the state personality-PC relationship. To explain these relations, we focus on two prolific emotion regulation strategies, namely reappraisal and suppression (Gross & John, 2003). We chose these strategies because one (i.e., reappraisal) is more cognitive in nature (therefore, relevant to PCB perceptions) and the other (i.e., suppression) is more affective in nature (therefore, relevant to violation feelings). Note that for the purposes of exploring the moderating effect of emotion regulation strategies, we focus solely on the relationship of avoidance goals with PC experiences (i.e., PCB and violation feelings). We do so because emotion regulation strategies specifically play a pivotal role in coping with negative stimuli (Gross & John, 2003), which are characteristic of employees whose behavior is momentarily driven by avoidance goals (Ferris et al., 2011). Employees whose behavior is driven by approach goals experience more positive effects (Ferris et al., 2011), resulting in the fact that there is no need to reappraise or suppress emotional reactions because one’s emotional reaction is already positive.

Emotions unfold over time from initial evaluation of emotion cues to full-on response tendencies (Gross, 2001). Reappraisal is an antecedent-focused emotion regulation strategy. Such strategies are engaged after the initial emotion cues but before full emotion activation and before behavior has been affected (Gross & John, 2003). Reappraisal reflects a type of cognitive change that helps individuals (re)construct a potential emotion-eliciting situation in such a way that it alters its full emotional impact. For example, upon noticing a discrepancy between organizational obligations and actual resources, an individual may reappraise the negative situation as one that is merely temporary in nature as a means by which to manage potentially strong negative emotions. In the context of PC processes, reappraisal would occur after a discrepancy between obligations and delivered resources is perceived (cuing negative affect) but before the ensuing violation feelings. In contrast, suppression is a response-focused emotion regulation strategy involving the inhibition of ongoing emotion-expressive behaviors (Gross & John, 2003). Suppression occurs after the full effect of emotion is activated to reduce the experience of negative emotion. In the context of PC processes, suppression would occur amid violation feelings to dissipate this collection of negative emotions. These emotion regulation strategies have a stable genetic component, but also show long-term (John & Gross, 2007) and short-term (Gross & Thompson, 2007) fluctuations. Given our interest in within-person fluctuations, we focus on short-term fluctuations in reappraisal and suppression.

Given the cognitive nature of reappraisal, we expect this emotion regulation strategy to influence the cognitive component of PC experiences, namely PCB perceptions. More specifically, we anticipate that reappraisal will moderate the relationship between avoidance goals and perceptions of PCB. Empirical evidence suggests that the reappraisal of negative stimuli is associated with a reduction of experienced negative emotion (Goldin et al., 2008), and more importantly with attempts to down-regulate emotional stimuli by reframing it (Ochsner et al., 2004). In terms of the PC, this suggests that employees who notice an inconsistency (because of their heightened momentary level of state N and avoidance goals), may reappraise this inconsistency as an unintended or otherwise innocuous discrepancy between organizational obligations and actual delivered resources. Consequently, they will be less likely to perceive this inconsistency as PCB.

• Hypothesis 3a: The use of a reappraisal strategy will moderate the positive relationship between avoidance goals and PCB in such a way that the relationship between avoidance goals and PCB will be less positive when the use of a reappraisal strategy is high.

The effect of suppression will become apparent when considering the emotion-related part of PC experiences (i.e., violation feelings), as suppression is a typical response-focused emotion regulation strategy that is mainly emotional in nature. For employees high in momentary N—who are driven by avoidance goals—we expect the positive relationship between these avoidance goals and violation feelings to be stronger when they apply a suppression strategy. Empirical evidence indeed indicates that suppressing emotions drains a lot of energy and resources, which relates positively to self-regulatory failure (Muraven et al., 1998), ruminative thoughts (Martin & Tesser, 1989), aggression (Baumeister, 1997), activation of the amygdala, and negative affect (Luan Phan et al., 2005). Based on a review of the emotion regulation and aggression literatures, Roberton et al. (2012) concluded that individuals who suppress negative emotions can experience future negative affective consequences. Just as suppressing the thought of a white polar bear can paradoxically increase the frequency of that thought, the use of suppression can increase the experience of negative emotions (Gross & John, 2003; John & Gross, 2004). Therefore, we hypothesize the following:

• Hypothesis 3b: The use of a suppression strategy will moderate the positive relationship between avoidance goals and violation feelings in such a way that the relationship between avoidance goals and violation feelings will be more positive when the use of a suppression strategy is high.

Method Study 2

Procedure

In Study 2, we tested whether approach/avoidance goals mediated the relationship between momentary fluctuations in state N or state C and PCB (Hypotheses 1a and 1b), and the intensity of violation feelings (Hypotheses 2a and 2b). Moreover, we included emotion regulation strategies (i.e., reappraisal and suppression) to further elucidate the mechanism underlying the personality-PC relationship (Hypotheses 3a and 3b). We recruited US respondents to complete a single general online survey and subsequent (three) daily prompts for ten consecutive workdays. We sent the first daily prompt at a random time between 3.00PM and 3.30PM, the second daily prompt at a random time between 5.00PM and 5.30PM, and the last daily prompt at a random time between 8.00PM and 8.30PM. Respondents were required to respond to the first prompt before 5.00PM, the second prompt before 8.00PM, and the last prompt before 11.00PM. As in Study 1, each prompt received an electronic time stamp, and responses were coded as missing data when respondents failed to (timely) complete the survey. For similar reasons as in Study 1, we decided to use an ESM design.

Participants

Of the 78 invited, 55 respondents completed the general survey (response rate = 70.51%) and 51 respondents completed the three prompts each day (response rate = 65.38%). The effective sample size included 394 observations (51 respondents x average of 7.73 responses per individual). As in Study 1, this indicates that our sample had sufficient power to provide an accurate estimate of standard errors and fixed effects (Browne & Draper, 2000; Maas & Hox, 2005). Our respondents were, on average, 47.94 years old (SD = 9.88), 43.10% were female, 52.90% obtained a higher educational degree, 80.40% had a permanent full-time job, 35.30% had managerial responsibilities, and the average company tenure was 11.94 years (SD = 9.09). Results from a logistic regression analysis revealed that attrition was not explained by any of the demographic or other variables under study.

Measures³

General survey Measures

We again used the general survey to collect demographic information and to assess the level of obligated PC resources. Level of obligated resources were assessed for a similar reason as in Study 1. Respondents rated the extent to which they believed that their employer was obligated to provide them with each of 12 items on a 7-point scale (1 = “minimally or not at all”, 7 = “to a very large extent”). These 12 items represent a wide set of commonly studied transactional and relational PC resources. In this study we choose to select only the two most commonly studied PC types rather than also include the less commonly studied ideological PC type because the sensitivity analysis of Study 1 indicated that the proposed relationships in our hypothesized model did not differ significantly among the three (i.e., transactional, relational, and ideological) PC types; meaning that state N and state C have a similar influence on the likelihood to perceive a PCB and the intensity of violation feelings irrespective of the PC type. The scores for the transactional (α = 0.82) and relational (α = 0.78) PC resources ranged from 4.40 to 5.91, and 4.35 to 6.31, respectively. As expected, the theorized 2-factor PC resource model (RMSEA = 0.08, CFI = 0.90, TLI = 0.86, SRMR = 0.08; Dyer et al., 2005) fit the data significantly better than an alternative single resource model (∆χ² = 3.36, ∆df = 1, p < 0.001).

Experience Sampling Measures (Prompts)

State-N and state-C were assessed—during the first daily prompt—with the same instrument used in Study 1 (Saucier, 1994). The within-person omega reliability coefficient for state N (ω = 0.74) and state C (ω = 0.74) were significant and satisfactory (Geldhof et al., 2014).

Approach/Avoidance goals were measured—during the second daily prompt—using the Behavioral Avoidance (BIS) and Approach (BAS) System survey (Carver & White, 1994). The BIS scale contains seven items (e.g., “I would get pretty “worked up” if I thought something unpleasant was going to happen”). The BAS scale contains 13 items (e.g., “I felt I would often act on the spur of the moment”). Respondents rated all items on a 5-point Likert scale ranging from (1) “strongly disagree” to (5) “strongly agree”. The within-person omega reliability coefficient for BIS (ω = 0.62) and BAS (ω = 0.85) were significant and satisfactory (Geldhof et al., 2014).

Reappraisal and suppression were measured—during the second daily prompt—using the Emotion Regulation Questionnaire (ERQ; Gross & John, 2003). The ERQ contains six items for reappraisal (e.g., “I made myself think about a stressful situation in a way that helped me stay calm”) and four items for suppression (e.g., “I controlled my emotions by not expressing them”). Respondents rated all items on a 7-point Likert scale ranging from (1) “strongly disagree” to (7) “strongly agree”. The within-person omega reliability coefficient for reappraisal (ω = 0.82) and suppression (ω = 0.67) were significant and satisfactory (Geldhof et al., 2014).

PCB was measured—during the third daily prompt—using a direct comparison approach (Montes & Irving, 2008; Turnley & Feldman, 1999). We presented respondents with the same list of 12 common transactional and relational PC items as in the general survey and asked them to think about the past day (i.e., on the job today) when indicating to what extent they received each of the PC resources compared to the extent to which each was perceived obligated to them. Respondents rated all items on a 5-point Likert scale ranging from (1) “received much less than obligated” to (5) “received much more than obligated”. The within-person omega reliability coefficient for PCB (ω = 0.78) was significant and satisfactory (Geldhof et al., 2014).

Violation feelings were measured—during the third daily prompt—using the 10-item job-related affective well-being scale (JAWS; Van Katwyk et al., 2000). We asked our respondents to rate the extent to which they felt each of the emotions (e.g., anger, discouraged, furious, frightened) during the past day on a 7-point Likert scale ranging from (1) “minimally or not at all” to (7) “to a very great extent”. The within-person omega reliability coefficient for violation feelings (ω = 0.73) was significant and satisfactory (Geldhof et al., 2014).

Analytic Strategy

In line with the recommendations of Edwards and Lambert (2007), we simultaneously tested moderation and mediation effects. The mediation effects were tested by means of the product-of-coefficients approach (Preacher & Hayes, 2008) and their significance was scrutinized by means of 95% Monte Carlo Confidence Intervals (95% CI; Preacher & Selig, 2012). We linked the regression coefficients of the type of emotion regulation strategy (second daily prompt), the type of approach or avoidance goals (second daily prompt), and the interaction terms (reappraisal X avoidance goal BIS; suppression X avoidance goal BIS) to PCB (third daily prompt) and violation feelings (third daily prompt). Finally, we tested time-lagged mediation effects (i.e., predictor during the first daily prompt to mediator during the second daily prompt, and mediator during the second daily prompt to outcomes during the third daily prompt) by linking momentary fluctuations in state N and state C to the intensity of PCB and violation feelings via approach and avoidance goals.

The moderation effects were tested by including an interaction between (1) the avoidance goal BIS (second daily prompt) and the reappraisal strategy (second daily prompt), and (2) the avoidance goal BIS (second daily prompt) and the suppression strategy (second daily prompt). To further interpret these multilevel moderation effects, we used the region of significance approach or the Johnson-Neyman technique (Johnson & Neyman, 1936; Preacher et al., 2006) instead of the traditional simple slopes’ method. The Johnson-Neyman technique is preferred because it identifies the full range of the moderator—instead of the mean and standard deviations—for which the interaction is significant (i.e., all values of the emotion regulation strategies where the 95% confidence bands do not include zero). While the upper dashed line in such plots indicates the 2.5% upper region boundaries of significance, the lower dashed line indicates the 2.5% lower region boundaries of significance. The solid line in between the confidence bands represents the size and the direction of the relationship between the independent and the dependent variable for different values of the moderator.

As in Study 1 our data had a 3-level nested structure (i.e., prompts nested within days nested within individuals), we estimated ICCs of state N, state C, approach goals (BAS), avoidance goals (BIS), reappraisal, suppression, PCB, and violation feelings. The largest proportion of the variance in these variables (ICCs are 0.17, 0.14, 0.09, 0.09, 0.11, 0.09, 0.16, and 0.10, respectively) could be attributed to within-person differences. Hence, we used person-mean centering and estimated a 3-level moderated mediation model because person-mean centering generally yield more accurate variance estimates of within-person effects than grand-mean centering or no centering when interested in within-person effects (see Raudenbush & Bryk, 2002; Wang & Maxwell, 2015). We conducted all analysis in Mplus version 7.1 (Muthén & Muthén, 2012). For similar reasons as in Study 1, we did not include any control variables.

Results

Multilevel CFA

We tested whether state N, state C, approach goals (BAS), avoidance goals (BIS), reappraisal, suppression, PCB, and violation feelings could be empirically distinguished from each other. To this end, we performed a series of multilevel CFAs on the within-person covariance matrix. As in Study 1, we used Dyer et al. (2005) conventional standards to assess model fit and compared competing models using loglikelihood ratio tests. Based on our theoretical expectations and the sensitivity analysis of Study 1 (i.e., indicating a single PCB factor) we compared a theoretical model containing eight first-order latent factors to four alternative models (see Table 3). Alternative model A (Δχ²(9) = 90.97, p < 0.001), alternative model B (Δχ²(9) = 94.66, p < 0.001), alternative model C (Δχ²(9) = 44.70, p < 0.001), and alternative model D (Δχ²(9) = 566.81, p < 0.001) fit significantly worse to the data than the theory-based model. Hence, our 8-factor theoretical model (RMSEA = 0.04, CFI = 0.97, TLI = 0.96, SRMR = 0.07) guided hypotheses testing.

Table 3 Results from multilevel confirmatory factor analyses (Study 2)

Descriptive Results

Table 4 provides an overview of the means, standard deviations, zero-order (i.e., between-person) and person-centered (i.e., within-person) correlations among the study variables.

Table 4 Means, standard deviations, and correlations among the focal variables (Study 2)

Hypothesis Testing

As a first step, we assessed whether a full or partial moderated mediation model fit the data best. When comparing the model fit criteria and BIC values, the 3-level partial moderated mediation model (RMSEA = 0.001, CFI = 1.00, TLI = 1.00, SRMR = 0.03, BIC = 617.44) yielded a better fit to the data compared to the 3-level full moderated mediation model (RMSEA = 0.12, CFI = 0.74, TLI = 0.35, SRMR = 0.06, BIC = 670.02). Hence, Fig. 2 displays the results of the 3-level moderated mediation model.

Fig. 2

Standardized estimated paths in the 3-level moderated mediation model. Notes. *: p < .05. **: p < .01. ***: p < .001. Dotted lines indicate non-significant relationships. Double arrowed lines represent correlations. Due to the nature of the direct comparison measure of PCB, response options ranging from (1) “received much less than obligated” to (5) “received much more than obligated”, a negative estimate indicates more intense perceptions of PCB, whereas a positive estimate indicates less intense perceptions of PCB

Next, we replicated and extended the findings of Study 1. That is, being momentary high in state N and state C in the morning was positively related to the likelihood of perceiving PCB in the evening. Additionally, being momentary high in state N in the morning was positively related to the intensity of violation feelings in the evening, whereas being momentary high in state C in the morning was negatively related to the intensity of violation feelings in the evening. Second, our results indicated that being momentary high in state N during the first daily prompt in the morning was positively related to the momentary avoidance goal BIS in the afternoon (R² = 0.08, p < 0.001), which in turn was positively related to the intensity of PCB in the evening (R² = 0.24, p < 0.001) and the intensity of violation feelings in the evening (R² = 0.10, p < 0.001). In contrast, being momentary high in state C during the first daily prompt in the morning was positively related to the momentary approach goal BAS in the afternoon (R² = 0.12, p < 0.001), which in turn was negatively related to the intensity of PCB in the evening and the intensity of violation feelings in the evening. Finally, we found a significant time-lagged indirect effect of momentary levels of state N on the intensity of PCB (95%CI = [−0.004; −0.001]),⁴ as well as on the intensity of violation feelings (95%CI = [0.001; 0.002]) via the momentary avoidance goal BIS. Moreover, we found a significant time-lagged indirect effect of momentary levels of state C on the intensity of PCB (95%CI = [0.004; 0.007]), as well as on the intensity of violation feelings (95%CI = [−0.014; −0.003]) via the momentary approach goal BAS. These results support Hypothesis 1a, 1b, 2a, and 2b.

Turning to the moderation results, our findings indicated that reappraisal in the afternoon did not significantly moderate the relationship between the avoidance goal BIS in the afternoon and the intensity of PCB in the evening. Hence, no support was found for Hypothesis 3a. In contrast, suppression in the afternoon did significantly moderate the relationship between the avoidance goal BIS in the afternoon and the intensity of violation feelings in the evening. The simple slopes in the Johnson-Neyman plot were significant outside the −0.61 to 0.29 region, as indicated by the dotted lines in Fig. 3. Interpreting these values considering the minimum (i.e., -2.40) and maximum (i.e., 2.55) person-mean centred value of suppression, suggests that when suppression is low (below -0.61; N_observations = 50 or 12.69%; N_{respondents =} 28 or 54.90%) or high (above 0.29; N_observations = 126 or 31.98%; N_{respondents =} 41 or 80.39%) the relationship between the avoidance goal BIS and the intensity of violation feelings is negative or positive, respectively. This finding supports Hypothesis 3b.

Fig. 3

Johnson-Neyman plot for the moderating role of suppression on the relationship between the avoidance goal BIS and the intensity of violation feelings. Suppression significantly moderates the relationship between the avoidance goal BIS and the intensity of violation feelings for any value on the left-hand and right-hand side of the vertical dotted lines

Sensitivity Analysis

Research has demonstrated that effects of receiving more than what the organization was obligated to provide versus less than what the organization was obligated to provide, may be linear in some cases (e.g., transactional resources), yet curvilinear in others (e.g., relational resources) (e.g., Irving & Montes, 2009; Montes & Irving, 2008; Lambert et al., 2003). As such, we scale-centered our PCB scale so that negative values represented under-fulfillment and positive values represented over-fulfillment and tested a curvilinear effect of PCB in the above-described 3-level moderated mediation model. Although the fit indices reached the recommended cut-off values (RMSEA = 0.02, CFI = 0.99, TLI = 0.98, SRMR = 0.03), this model fit the data worse (BIC = 1121.58) than the general linear discrepancy PCB 3-level moderated mediation model (BIC = 615.59).

Because some scholars (e.g., Ntalianis et al., 2015; Raja et al., 2004) have found that employees high on trait N are more likely to form transactional PCs, whereas employees high on trait C tend to form relational PCs, we acknowledged the possibility that the effects may depend on the type of PC resource. Hence, we conducted two additional sensitivity analyses. First, we included a separate transactional and relational linear PCB component. Although the fit indices reached the recommended cut-off values (RMSEA = 0.03, CFI = 0.99, TLI = 0.99, SRMR = 0.02), this model fit the data worse (BIC = 667.37) than the general linear discrepancy PCB 3-level moderated mediation model (BIC = 615.59). Second, we included a separate transactional and relational curvilinear PCB component and found that several fit indices did not reach the recommended cut-off values (RMSEA = 0.10, CFI = 0.89, TLI = 0.79, SRMR = 0.09), indicating that this model did not fit the data. In conclusion, these sensitivity analyses suggest that (1) receiving more than obligated is positive, whereas receiving less than obligated is negative (i.e., a linear discrepancy PCB effect), and (2) the type of PC resource (i.e., transactional, or relational) does not influence our results.

Finally, because, based on the works of Restubog et al. (2015) and Bordia et al. (2008), it could be argued that suppression and reappraisal could also moderate the relationship between PCB and violation feelings,⁵ we conducted an additional sensitivity analysis in which suppression and reappraisal moderated the relationship between PCB and violation feelings rather than the a priori hypothesized relationships. We found that several fit indices did not reach the recommended cut-off values (RMSEA = 0.06, CFI = 0.70, TLI = 0.44, SRMR = 0.09), indicating that this model did not fit the data. This sensitivity analysis, in combination with the results from the a priori hypothesized model, suggest that suppression and reappraisal are more likely to moderate the relationship between avoidance/approach goals and violation feelings rather than the relationship between PCB and violation feelings.

Discussion

In general, the results of Study 2 supported our hypotheses. Being momentarily high in state N related positively to avoidance goals (BIS), which in turn related positively to perceptions of PCB and the intensity of violation feelings. In addition, our conditional indirect effect supported the idea that the use of suppression would aggravate the already positive relationship between avoidance goals (BIS) and the intensity of experienced violation feelings. That is, being momentarily high in state N was positively related to the intensity of violation feelings via avoidance goals (BIS), but only when one relies on suppression as a self-regulatory mechanism to manage emotions. In contrast, our results did not support the idea that the use of reappraisal would mitigate the positive relationship between avoidance goals (BIS) and perceptions of PCB. Failure to show a significant moderation effect could be explained by the work of Gross (2002), and Gross and John (2003) as these scholars stated that the use of reappraisal may only be positive once the employee has had time to reflect upon the situation. In this respect, our time lag might have been too short to pick up this process, and hence the positive effect of reappraisal.

Next, our results indicated that being momentarily high in state C related positively to approach goals (BAS), which in turn, related negatively to perceptions of PCB and the intensity of violation feelings. Furthermore, it should be noted that the direct relationships between state N and state C, and perceptions of PCB and the intensity of violation feelings are largely in line with the findings obtained in Study 1; providing additional support for the conclusions offered in Study 1.

It is noteworthy that we found a linear discrepancy effect of PCB. This aligns with the results obtained by Irving and Montes (2009), Montes and Irving (2008), and Lambert et al. (2003) with regard to transactional resources but contrasts their findings pertaining to relational resources. However, the time-lags of ten to twelve weeks between subsequent measures in these studies, might explain why they found curvilinear effects for relational resources. According to the principle of peak-end rule (Kahneman, 2000), reflecting over a longer period of time is likely to result in recollection of extreme events of over-fulfillment of relational resources (i.e., getting too many opportunities for skill development); this may subsequently trigger violation feelings or negative attitudes and behaviors toward the organization because the over-fulfillment of relational resources is considered taxing on one’s abilities. In contrast, in a short period of time, such as a day, an employee is far less likely to have encountered such extreme incidences of over-fulfillment, causing the employee to rely more on the average of all experiences across the day. As a result, a linear discrepancy effect is most logic in the short-term, whereas curvilinear effects might arise over the long-term.

Finally, we found that our relationships under study were not influenced by the nature of the resources, despite previous arguments for such differences by Raja et al. (2004). Despite the differences in preference for transactional (trait N) or relational (trait C) PCs at the stable between-person level, an employee seems to monitor the extent to which an organizational obligation deviates from the actual delivered resource, irrespectively from the type of resource. This indicates that, at the daily level, the preference for a specific type of PC matters far less compared to the actual accumulation of resources; a finding that can be explained by the theoretical arguments that individuals strive to retain, protect, and accumulate as many personal and environmental resources as possible (Hobfoll, 2001).

General Discussion

We aimed to understand the personality–PC relationship from a dynamic perspective by investigating the direct relationship between momentary fluctuations in state N and state C, and the likelihood to perceive a PCB and the intensity of experienced violation feelings (Study 1). In addition, we focused on the mediating role of momentary approach (BAS) and avoidance (BIS) goals, as well as on the role of emotion regulation strategies as a crucial moderator of the personality–PC relationship (Study 2). Across both studies we expand on previous findings concerning the stable trait personality–PC relationship by highlighting the importance of momentary approach (BAS) and avoidance (BIS) goals in the relationship between an employee’s momentary level of state C or state N, and perceptions of PCB and the intensity of experienced violation feelings, respectively. Moreover, we extended these previous findings even further by highlighting the importance of self-regulatory processes, namely suppression, in an employee’s process of experiencing violation feelings. That is, the more an employee, who is momentarily high on state N, relies on suppression, the more intense the positive relationship between momentary avoidance (BIS) goals and the intensity of experienced violation feelings.

Theoretical Implications

This study contributes to the PC literature in several ways. First, our results suggest that the likelihood of perceiving PCB increases as a function of one’s momentarily state N and state C levels. While the former was expected based on previous empirical work, the latter was not expected as previous work found a negative relationship between conscientiousness and PCB (Raja et al., 2004). The heightened vigilant monitoring because of high momentarily levels in both states could possibly explain our findings. Heightened momentarily levels of state N results in increased monitoring for things that could possibly go wrong whereas heightened momentary levels of state C make employees more attentive to positive stimuli. Irrespective of the underlying motivation, this increased monitoring results in employees noticing and attending to cues pointing towards a discrepancy between promised and delivered resources (Adler & Obstfeld, 2007; Öhman et al., 2001), which in turn lead to an increased likelihood to perceive a PCB.

The second contribution is that the intensity of violation feelings increases as a function of one’s momentary N-state level. This finding aligns well with the approach/avoidance framework; a higher momentarily level of state N is related to increased avoidance motivation, causing employees to focus more on negative stimuli, and thus reacting more strongly to them (Ferris et al., 2011). Contrary to our hypothesis, we did not find a significant negative relationship between one’s momentary level of state C and feelings of violation. A possible explanation for the lack of this effect might be found in the complex relationship between the sensitivity towards positive and negative stimuli. Specifically, we argue that being momentarily high in state C makes employees focus more on positive cues (Ferris et al., 2011). However, this does not imply that employees do not perceive any negative cues or do not experience any violation feelings. Being momentarily high in state C might result in a heightened focus on positive stimuli that reduces—but does not significantly buffer against—the intensity of violation feelings. In other words, this finding seems to suggest that the heightened focus on positive stimuli and the reduced focus on negative stimuli are orthogonal; one can experience a heightened focus on positive stimuli without necessarily experiencing a reduced focus on negative stimuli.

Finally, our findings indeed highlight the importance of suppression in relation to violation feelings. Specifically, our findings underline the importance of suppression as a response-focused emotion regulation strategy (Gross & John, 2003) which further strengthens the positive relationship between momentary avoidance goals of employees who are momentary high on state N and violation feelings. It seems that the use of a suppression emotion regulation strategy drains a lot of energy and resources, which in turn leads to an increased experience of negative emotions (i.e., violation feelings) further down the road. These findings add to the existing literature in which a plethora of studies found that the use of suppression was positively related to self-regulatory failure (Muraven et al., 1998), ruminative thoughts (Martin & Tesser, 1989), aggression (Baumeister, 1997), activation of the amygdala, and increasing negative affect (Luan Phan et al., 2005). However, in contrast to this, we found no support for the potential buffering effect of reappraisal in the relationship between momentary avoidance goals and the likelihood to perceive a PCB at a later point in time. The lack of support for this relationship might be explained by the fact that employees who are momentary high on state N do not reappraise inconsistencies in their PC as unintended (i.e., incongruence; Morrison & Robinson, 1997) but rather perceive this inconsistency as an inability or unwillingness from the employer to deliver an obligation (i.e., reneging; Morrison & Robinson, 1997).

Limitations

Notwithstanding the methodological and theoretical contributions provided by these two studies, our research has limitations that deserve further attention. First, our data were collected using repeated measurement surveys in which the mediator and moderator variables were collected at the same point in time (all other variables were collected at different points in time). Although this approach was required to investigate the 3-level moderated mediation model, it might raise concerns about common method variance (Podsakoff et al., 2012). However, by collecting our independent (morning), mediator and moderator (afternoon), and dependent (evening) variables separated in time, we reduced risk owing to common method variance. In addition, we presented all scales in a random order, both within and between blocks. Finally, Siemsen et al. (2010) argued that common method bias cannot explain or distort interaction effects. Hence, the significant interaction in Study 2 helps to strengthen our argument that the observed results are more likely a function of the studied constructs than of methodological artifacts.

A second limitation concerns the self-report nature of our repeated measures, which might raise questions concerning social desirability (Podsakoff et al., 2012). However, we guaranteed confidentiality and relied on discretionary participation to minimize the risks owing to social desirability effects. Additionally, it is quite impossible to implement other-rated measures of state N, state C, and PC evaluations as these concepts are inherently idiosyncratic in nature and therefore hard (if not impossible) to observe by others. Hence, we relied on self-reports to assess the concepts under study but took steps to assure anonymity and a sense of privacy to lower self-report biases (for a similar approach see Berry et al., 2012).

A third limitation concerns the directionality versus causality of our obtained findings. We were able to infer directionality of our effects as measures of state N and state C preceded measures of avoidance and approach goals, and emotion regulation strategies, which in turn preceded measures of PCB and violation feelings (Albers & Kratochwill, 2010). Hence, we can confidently state that momentary fluctuations in personality states influenced momentary fluctuations in avoidance and approach goals, which in turn influenced PC evaluations, rather than the other way around. However, our design did not allow us to infer causality. Future research would benefit from using an experimental design in which state N, state C, and emotion regulation strategies are manipulated and their effects on PCB and subsequent violation feelings are measured in a controlled setting.

A final limitation concerns the inclusion of only state N and C as two of the Big Five personality dimensions. However, it could be argued that other personality states could affect either the likelihood to perceive PCB and/or the intensity of violation feelings. For example, being momentarily high in state extraversion could be positively related to the likelihood of perceiving PCB and the intensity of violation feelings because when one is momentarily high in state extraversion, this person tends to exhibit dominant behavior and expressiveness during their interactions with others (McCrae & Costa, 1987), tends to be notably attentive to how their organizations fulfilled their obligations within the PC (Raja et al., 2004), displays a proactive nature (Crant & Bateman, 2000), and demonstrates a keen interest in personal development (Antonacopoulou, 2000), career advancement, and job satisfaction (Seibert et al., 2001). As a corollary, individuals who are momentarily high in state extraversion, possibly due to their heightened vigilance and information-seeking tendencies, consistently evaluate the extent to which their organizations meet their obligations. In doing so, even minor disparities between what was promised and what is being delivered by their organization could be seen as a PCB and trigger more intense violation feelings.

Being momentarily high in state agreeableness is another personality state which could be negatively related to the likelihood of perceiving PCB and the intensity of violation feelings because when one is momentarily high in state agreeableness, this person tends to be cooperative, trusting, conflict-avoidant, considerate and understanding of others and their actions, and concerned for co-operation and social harmony (Costa & McCrae, 1992; Graziano & Eisenberg, 1997). As such, it seems reasonable to assume that when one is momentarily high in state agreeableness, one is less likely to focus on the discrepancy between was promised and what is being delivered by their organization and more likely to be concerned about social cohesion in the workplace and thus less likely to perceive a PCB and/or intense violation feelings.

Suggestions for Future Research

As our 2-study paper found that momentary fluctuations in state N and state C influence perceptions of PCB, an important avenue for future research lies in identifying the extent to which one perceives a discrepancy from obligated resources as a PCB. That is, according to the work of Schalk and Roe (2007), one will only notice and attend to a discrepancy once it surpasses one’s personal zone of acceptance (i.e., an individual level of tolerance toward deviations from organizational obligations). Beyond the influence of momentary fluctuations in state N and state C, this personal zone of acceptance potentially influences the extent to which one perceives PCB: when one has a narrower personal zone of acceptance, one will be more likely to interpret a discrepancy as PCB. In contrast, when one has a broader personal zone of acceptance, one will be less likely to interpret a discrepancy as PCB. The extent to which one perceives a discrepancy as PCB (i.e., latitude of the personal zone of acceptance) could be investigated by means of an ESM design and spline regression analyses. In spline regression analysis, the slope of the regression line can change for different ranges of an independent variable. The point where the slope of regression line changes, represents one’s personal zone of acceptance at which one shift from “not perceiving a discrepancy as PCB” to “perceiving a discrepancy as PCB” (Seber & Wild, 2003).

Practical Implications

Our findings have important practical implications that may assist practitioners in their efforts to reduce PCB perceptions and violation feelings. First, that personality states fluctuate over the span of a day based on the situational characteristics an employee is faced with, suggests that personality can be altered actively. One possible way to achieve this is by changing employee job demands. Debusscher et al. (2016) showed that job demands, such as work pressure and task complexity, are positively associated with state N. In a similar way, Minbashian et al. (2010) found that changes in job demands, such as the difficulty and urgency of the task, influence state C. A first practical implication then is to reduce work pressure and task complexity to decrease momentary levels of state N. Our results indicated that lower levels of state N are associated with a lower likelihood to perceive PCB and less intense violation feelings.

Acknowledging that the current economic climate is faced with numerous challenges, such as high work pressure and constant change, altering job demands is not always possible. For this reason, a second practical recommendation pertains to developing and coaching employees to better cope with job demands, hence lowering momentary levels of state N (e.g., Debusscher et al., 2016). If employees can actively lower their state N levels, we expect, based on our results, a reduced likelihood to perceive PCB and less intense violation feelings. We thus propose that leaders explicitly provide emotional and practical support and appreciation for their subordinates. These interventions provide an employee with more opportunities to cope with stressful situations resulting from high job demands and may, therefore, alleviate the stressful impact of said demands (Bakker & Demerouti, 2007; Karasek, 1998). In a similar vein, expressing appreciation not only helps employees to do their work more effectively but also improves the quality of the relationship between leader and subordinates (Bakker & Demerouti, 2007).

In terms of state C, we caution against attempting to alter one’s momentary levels of state C. The reason for this being that momentary levels of state C only increase the likelihood to perceive PCB, while it does not influence the intensity of violation feelings. Zhao et al. (2007) provided meta-analytical evidence for the role of violation feelings underlying attitudinal and behavioral outcomes. Hence, we expect few adverse effects on attitudinal and behavioral outcomes when one is momentarily high in state C. In addition, lowering momentary state C levels—solely in an attempt to reduce the likelihood of PCB perceptions—would be negatively related to (1) an employee’s tendency to inhibit and constrain impulse-related behaviors (Kunisato et al., 2011), (2) the capacity of the working memory and one’s executive functions (Kunisato et al., 2011), (3) general performance at work (Goodwin & Friedman, 2006), and (4) problem-solving effectiveness (Jundt et al., 2015) As such, we would recommend against intervening in terms of momentary levels of state C.

Next, our results indicate that the use of a suppression strategy further strengthens the positive relationship between avoidance goals and violation feelings for employees who score momentary high on state N. Consequently, it would be advisable for employees, when they experience neuroticism in the moment, to employ strategies to not engage in suppression. This is especially important as previous studies have indicated that the use of suppression might decrease outward expressions of emotion but not the inner emotional experience. In other words, the use of suppression doesn’t make the negative emotional experience go away, it just reduces the outward display of it (Gross, 2002). Furthermore, the use of suppression has been found to be associated with—in addition to the experience of more severe negative emotions down the road (i.e., increased violation feelings)—increase anxiety, depression, and stress-related complaints.

Therefore, we advise people who experience momentary high state N to engage in interventions with the objective of reducing their desire to suppress their emotional experience. In this regard, mindfulness-based interventions have been proposed (Cullen, 2011; Kabat-Zinn, 2003). In a seminal article, Cullen (2011, p. 188) summarized the formal practices of mindfulness-based interventions as follows, “mindful movement (gentle hatha yoga with an emphasis on mindful awareness of the body), the body scan (designed to systematically, region by region, cultivate awareness of the body—the first foundation of mindfulness—without the tensing and relaxing of muscle groups associated with progressive relaxation), and sitting meditation (awareness of the breath and systematic widening the field of awareness to include all four foundations of mindfulness: awareness of the body, feeling tone, mental states and mental contents)”. Evidence suggests that engaging in one or more of these mindfulness-based practices functions to reduce negative emotional experiences (Baer, 2003), and thus might be a useful strategy to engage in for people who are momentarily high on state N to prevent a further increase in violation feelings.

Finally, more generally, it is important to highlight the pivotal role of human resource (HR) management strategies and organizational policies for the prevention of PCB. An effective HR strategy, along with well-defined organizational policies, can significantly contribute to minimizing perceptions of PCB and thus preventing violation feelings from arising. In this regard, an essential element of any successful HR management strategy is clear and transparent communication. HR professionals must ensure that employees have a comprehensive understanding of their roles, responsibilities, and the organization's values and culture. HR professionals should also facilitate regular and honest conversations with employees about their job roles, performance expectations, career progression, and any changes within the organization. Regular and open communication channels help employees align their expectations with the organization's objectives, it fosters trust and reduces the likelihood of misunderstand and thus the risk of PCB (Guest & Conway, 2002).

Relatedly, encouraging employee involvement in decision-making processes that directly affect them, can have a significant impact on preventing PCB. Employees who have a say in matters that affect them directly are more likely to feel a sense of ownership and alignment with the organization's goals. Second, HR professionals should establish and enforce fair and consistent HR policies and procedures, including equitable compensation practices, transparent promotion criteria, and consistent disciplinary measures. When employees perceive that the organization treats everyone fairly, they are less likely to feel that their PC has been breached (Cassar & Buttigieg, 2015).

HR management strategies should actively promote opportunities for employee development and growth. This includes providing training, mentoring, and coaching programs. When employees see a clear path for skill development and career advancement, they are more likely to remain engaged and feel that the organization is fulfilling its end of the PC. Relatedly recognizing and rewarding employees for their contributions is a vital component of preventing PCB and violation feelings. HR can implement performance-based recognition programs and provide regular feedback to acknowledge and appreciate employees' efforts. Feeling valued and appreciated reinforces a positive PC (De Hauw & De Vos, 2010). Finally, because not all PCBs are preventable by having a clear HR management strategy and organizational policies in place, HR professionals should establish effective mechanisms for addressing conflicts and grievances. Having a fair and accessible process for resolving issues ensures that employees feel heard and that their concerns are being taken seriously, reducing the risk of PCB due to unresolved problems. In conclusion, by prioritizing clear communication, fairness, employee development, recognition, and involvement, HR can create a positive workplace culture where employees feel that their PC is fulfilled.

Conclusion

Our findings indicate that the relationship between momentary fluctuations in an employee’s levels of state C and state N, and one’s PC evaluations are far more complex than originally believed. The mediating role of approach/avoidance goals and the moderating role of emotion regulation strategies highlight the self-regulatory processes that affect the state personality-PC evaluations relationship. Our present 2-study paper underscores the importance of taking an intra-individual perspective in the study of personality as an antecedent of the PC. We are hopeful that our findings, along with the advanced methodologies used in these studies, will stimulate scholarly attention and novel avenues of research.