Employee Disenchantment: The Development of a New Construct and Measure

This study developed a psychometrically sound measure of disenchantment, which had five factors demonstrating the targeted negative affect experienced as a result of poor treatment at work. The five factors are: Inequity and Unfairness, Broken Promises and the Psychological Contract, Organisational Lying, Hypocrisy and Deceit, Bullying and Mistreatment, and Distrust. Theoretical implications suggest that future research should look to account for employee disenchantment as a specific negative effect when researching Counterwork Behaviour (CWB). Implications and limitations are acknowl-edged.

hantment would represent the transition from being engaged with work and admiring your organisation, to feelings of disappointment, betrayal, and disillusionment from how you have been treated. Disenchantment would therefore need to identify an increase in beliefs regarding these factors, as well as psychological state of embitterment towards the organisation itself.
Disenchanted employees would feel that CWB is prevalent within the organisation (i.e., witnessing abusive management) and have a motivation to no longer adhere to proper work practice (i.e., abandoning normative behaviours due to feelings of disgruntlement and resentment). Disenchantment should account for negative emotions directed towards colleagues, managers, and the organisation.
This would provide a more specific explanation of the psychological process that occurs between job stressor and CWB. Based upon an examination of the literature, this study proposes that there are five main types of disenchantment.

Inequity and Unfairness
Adams' Equity Theory (Adams, 1965) argued that an individual feels fairly treated when they perceive their inputs to be equal to the received outcomes. That is, in exchange for the work they contribute to an organisation, they are equitably compensated through salary, rewards, bonuses, promotions, workload, and development opportunities. Employees' make judgements regarding the fairness by the outcomes they receive (Distributive Justice; Adams, 1965) and the policies and processes used to reach these outcomes (Procedural Justice; Greenberg, 1990;Lind & Tyler, 1988). When an employee feels their input exceeds their outcomes, they feel disenchantment towards their colleagues, managers, and organisation. Evidence shows deprived employees are more likely to be absent and leave their organisation compared to those who are fairly or advantageously treated (Stecher & Rosse, 2005). Meta-analyses have demonstrated how these perceptions have a profound impact on organisational commitment and withdrawal (Colquitt et al., 2007), job and pay satisfaction (Cohen-Charash & Spector, 2001;Roch & Shanock, 2006), and trust in the organisation and management (Aryee et al., 2002).
Within an organisational setting, equity theory also includes social comparisons (Carrell & Dittrich, 1978): employees will evaluate their input/output ratio in relation to the ratio of their colleagues. Employees perceived their organisation and management as fair when there is equitable distribution in praise, workload, bonuses, and promotions amongst the workforce (Forret & Love, 2008). Perceived inequity leads to a breakdown in trust amongst co-workers when individuals feel their co-workers are being treated more favourably, taking advantage of unfair distribution, and being overly rewarded for the smaller amount of effort they contribute (Forret & Love, 2008). Disenchantment towards the organisation will grow as employees feel bitter that loyalty, hard work and productivity have less to do with success than other attributes such as personal characteristics or skill at internal politics.

Broken Promises and the Psychological Contract
The psychological contract has been defined as an employee's belief regarding the reciprocal exchange agreement between employee and employer (Rousseau, 1989). The contract consists of perceptions regarding promises of a future return (i.e., pay for performance), a needed contribution (i.e., organisational input from the employee), and an obligation to make good on said promises (Robinson & Rousseau, 1994;Davis & Rothenstein, 2006); salary increases, promotions, or bonuses in return for the work they contribute (Robinson & Rousseau, 1994).
Psychological contracts therefore are subjective perceptions viewed from the perspective of each individual employee (Caldwell & Hayes, 2007).
When expectations and promises are upheld, employees believe that their managers have behavioural integrity, are trustworthy, and are credible (Davis & Rothenstein, 2006). However, when what is received grossly misaligns with what was promised, employees feel betrayed by the source of the wrongdoing. Previous research has supported this inevitable disenchantment that follows psychological contract breach or violation, with meta-analyses showing resulting decreases in job satisfaction, trust in management, organisational commitment, in-role performance, and increased turnover intentions (Zhao et al., 2007).

Organisational Lying, Hypocrisy, and Deceit
Beliefs of organizational hypocrisy are the perceived inconsistency between words, previous actions, and subsequent decisions. Employees begin to see their employer as hypocritical when: formal talks result in loose plans and informal agreements, decisions are made formally, recorded hierarchically, and implemented bureaucratically, and when the organisational actors "do" the opposite of what has been formally agreed upon, what is publicised, or how they instruct others to act (Perez & Robson, 1999). It is a macro-level perception; the corporate model is hypocritical, and this is reflected in what your manager pedals and what employees have to do on a daily basis.
Perceived organisational hypocrisy leaves the employee disenchanted due to the development of negative perception towards the organisational procedures, processes, and management (Wilkerson, 2002). This conceptualisation is related to research on Organisational Cynicism (Dean et al., 1998); a belief that elements relating to the organisation are deceitful and lacking integrity, resulting in disparaging and critical behaviours towards the organisation (Cartwright & Holmes, 2006;Dean et al., 1998). This perception is broad in its application-it can be specific to one element, or applied to all aspects, of the organisation-and is "learned belief" as a result of experience (Andersson, 1996;Andersson & Psychology Bateman, 1997;Reichers et al., 1997). A recent meta-analysis found that organisational cynicism was related to lower organisational commitment, job satisfaction, and job performance, whilst also being related to higher quitting intentions (Chiaburu et al., 2013).

Bullying and Mistreatment
This aspect of employee disenchantment assesses the belief that one's senior staff and colleagues are bullies; they are nasty, uncaring, and abusive to others in their organisation. Despite the majority of organisations having a "zero-tolerance" policy regarding bullying, around 30% of employees feel that they have experienced bullying of some form during their employment (Lutgen-Sandvik et al., 2007).
When employees are bullied, they are left angry, anxious, depressed, distressed, and emotionally exhausted (Zhang & Liao, 2015). The psychological damage is even greater for victims when the source of the bullying stems from their superiors (Einarsen & Raknes, 1997). Reviews of the literature have found that abusive supervision has a significant adverse effect on employee attitudes, well-being, organisational justice perceptions, work-place behaviours, performance, and family-related outcomes (Zhang & Liao, 2015). As their disenchantment grows, employees feel a greater need to retaliate against their supervisors (Mitchell & Ambrose, 2012;Thau & Mitchel, 2010;Zhang & Liao, 2015) and organisation as whole (Bowling & Mitchel, 2011;Zhang & Liao, 2015).

Distrust
The literature on trust in an organisational setting is extensive yet fragmented (McEvily et al., 2003). Researchers have continually introduced divergent conceptualisations of trust, with no agreed-upon definition (Dietz & Hartog, 2006).
As a result, the concept of trust has ended up being blurred amalgam that attempts to encompass cognitive processes, personal dispositions, and individual differences. For disenchantment, trust is distinguishable from trustworthiness (distinctive attributes that stimulate positive expectations, McKnight et al., 1998) and trust propensity (individual differences in the willingness to trust, Lewis & Weigert, 1985). Instead, trust is an attitude of the employee regarding their organisation, management, and co-workers. It is an individual's willingness to be vulnerable to another party when it cannot be monitored or controlled (Mayer et al., 1995;Mayer & Gavin, 2005). Trust also implies positive expectations, with the confidence that the other party will not exploit that vulnerability (Korczynski, 2000).
The importance of an employee having trust in their organisation, management, and co-workers is extensive. Trust is seen as a vital component for organisations that want effective working relationships (Lind, 2001;Tyler & Lind, 1992). A meta-analysis demonstrated how higher levels of trust are associated with improved task performance, citizenship behaviours, risk taking behaviours, and lower levels of counterproductive work behaviour (Colquitt et al., 2007).
However, what happens when organisations, managers, and co-workers break down an employee's willingness to vulnerable? Distrust is argued to represent a lack of confidence in others, concerns that they will intentionally do you harm, and the belief that they do not care for your welfare (Govier, 1994). When the trust is gone, employees become unwilling to engage with their management (Holland et al., 2015). They are left angry, bitter, and with a desire to leave their organisation for good (Alder et al., 2006). One of the central tenants to distrust is suspicion (Deutsch, 1958); to "actively entertain multiple, possible rival, hypotheses about the motives and genuineness of a person's behaviour" (Fein & Hilton, 1994: p. 168). Distrust builds disenchantment by leaving employees suspicious, bitter, and vigilant of the perceptually conniving environment they work within.

The Current Paper
It has hypothesised that disenchantment is the negative emotion felt towards work, representing disaffection resulting from five key areas: Inequity, Broken Promises, Organisational Lying, Bullying, and Distrust. The main aim of this paper is to demonstrate the dimensionality of disenchantment, and the instrumental role it plays in predicting CWBs.
This paper attempts to define, measure, and prove the instrumentality of disenchantment. This study will compile a pool of items that will measure disenchantment by reviewing literature relating to the five key sources. It will then reduce the item pool down in order to focus on the structure, reliability, and concurrent validity of a psychometric tool for disenchantment.
After reviewing the literature relating to the five key disenchantment-drivers, a pool of 75 items was generated. This study examined the psychometric properties of the 75 items in order to reduce it down to a set that could parsimoniously measure disenchantment, whilst maintaining the possibility of multiple, inter-related factors. After generating a set of items, statistical indices were used to select items that created a measure of disenchantment. The psychometric properties of this measure were compared against pre-existing measures in order to verify its construct validity.

Item-Generation
Based upon literature review and the specified domain of employee disenchantment, items were created to reflect the aforementioned five possible elements of disenchantment: Organisational Lying (e.g. "There is a wide gulf between what my organisation says about itself and what really occurs"); Perceived Inequity (e.g. "My manager has favourite employees within the organisation"); Bullying and Mistreatment (e.g. "I am often ignored at work"); Distrust (e.g. "In this organisation, managers trust their a great deal"); and Broken Promises (e.g. "I always have to work more hours than I am paid for to fulfil my tasks"). The 75-items were discussed with other psychologists in the Clinical, Organisational, and Personality disciples. The items were inspected, clarified, and re-worked in order to reduce social desirability, ambiguity, and duplication.

Participants
In total, 372 individuals participated in this study. Of these, 166 were male, 204 were female, and 2 indicated that they would rather not identify their gender. The participants were all American and in full-time employment. The mean age of the sample was 36.9 years (SD = 11.45), ranging from 20 to 78 years.

Employee Disenchantment
A measure of employee disenchantment was generated based on items generated from the literature. The measure contained 75 items, 34 of which were reverse scored. Participants were presented with statements about their perceptions towards the organisation they worked for. Participants rated upon a 10-point Likert scale the extent to which they agreed with the statement (1 = "Strongly Disagree" to 10 = "Strongly Agree").

Psychological Contract Breach Scale
Psychological Contract Breach (PCB) was measured using Robinson and Bennett (1995) measure. This measure consisted of nine items, measuring an individual's perceptions of psychological contract breach and feelings of contract violation. Participants indicate the extent to which they agree with the statement upon a 10-point Likert scale (1 = "Strongly Disagree" to 10 = "Strongly Agree"). A high score on this scale indicates increased feelings of psychological contract breach and violation. The internal consistency for this measure was very high (α = .97), which is similar to levels seen in previous studies (α = .92).

Corporate Hypocrisy
This study utilised Philippe et al.'s (2005) measure of corporate hypocrisy. A high score on this scale is indicative of having a positive perception of the corporation's culture. The scale is made up of three factors that relate to perceptions of corporate hypocrisy: Perceived Managerial Actions, Perceived Culture, and Perceived Reward Systems. This scale comprises of 40 items, 11 of which are reverse scored. Participants respond upon a 10-point Likert scale, indicating the extent to which they agree or disagree with a statement (1 = "Strongly Disagree" to 10 = "Strongly Agree"). Previous research has demonstrated that these factors have inter-correlations ranging from .57 to .69, as well as having sufficient internal validity (α = .92, α = .91, and α = .73 respectively; Philippe et al., 2005).

Organisational Trust
This study used Huff and Kelly's (2003) measure of organisational trust. This scale assesses four elements of organisational trust: Propensity to Trust, Propensity to Distrust, Internal Trust, and External Trust. As the first two elements focus on individual differences rather than cultural perceptions, only Internal Trust and External Trust were used for this study. Each element had four items, where participants rated upon a 10-point Likert scale the extent to which they agreed or disagreed (1 = "Strongly Disagree" to 10 = "Strongly Agree"). Internal trust has been defined as perceptions towards the climate of trust within an organisation, focusing on the intent and behaviours of employees (Shockley-Zalabak et al., 2000). External Trust focuses more on inter-organisational trust, assessing the extent to which employee's perceive their organisation to trust partner or co-operating firms.

Organisational Justice
Organisational Justice assesses four types of justice that an employee can experience in the workplace (Colquitt, 2001)

Procedure
The participants were recruited through Amazon Mechanical Turk (MTurk).
MTurk is an online market that recruits workers to participate in research and surveys. Researchers have found that MTurk yields data that is at least as reliable as other traditional recruitment methodologies, benefiting from marginally greater diversity than standard Internet surveys. The researchers specified that the sample should be in full-time employment due to the nature of the survey.
The five questionnaires were hosted on MTurk, for which workers were compensated roughly $1 for successfully completing all of the questionnaires.

Analysis
The dataset was organised and cleaned using SPSS 22.0. This software was also used to conduct the Exploratory Factor Analyses (EFA) and correlations that will be discussed within the results section. The descriptive statistics and response patterns of the participants were analysed in order to identify potential "careless" responding that could skew the results. A composite, post-hoc careless response score was generated based upon the work of Gee and Furnham (In Review). Details of this procedure can be found in the Appendices. After the analysis, 33 respondents were removed from the original sample due to careless responding, leaving 372 participants.
In regard to Structural Equation Modelling (SEM), the Lavaan package (Rosseel, 2012;version .5 -20) in R (version 3.3.0) was used. SEM utilises a confirmatory approach in order to assess the structural interrelations and interactions between variables within the phenomenon, using theory to shape models that attempt to explain variance in the data. As data was not normally distributed, maximum likelihood with robust standard errors was used for parameter estimation. As there is no consensus within the literature as to which measure of goodness of fit is best, researchers have advised to use multiple tests. The main indices that will be examined are RMSEA, where values of .08 -.05 represent adequate fit, and lower than .05 represent excellent fit. Comparative fit index (CFI) was also used, where values greater than .95 are considered an excellent fit of the data. Finally, the Tucker-Lewis Index was assessed, where values over .90 are good fits of the data.

Item-Level Analysis of Disenchantment
An item-level analysis was conducted on the 75 questions for disenchantment. The inter-item correlations of each 15-item disenchantment factor was analysed in order to reduce the number of questions in the measure, as well as to identify the most appropriate questions. Items were selected based on their strong correlations with other items in the factor. Nine items from each element were selected, leaving 45 items for exploratory factor analysis (EFA).

Exploratory Factor Analysis
Prior to EFA, disenchantment scores were evaluated for normality. A Shapiro-Wilk test of normality showed that scores of Disenchantment and its five elements significantly differed from normality (p < .001). This result supports what was expected; Disenchantment is supposed to represent an uncommon phenomenon, and therefore it should have a positively skewed distribution. Based on the significant non-normality of the data, previous researchers have suggested that a Principal Axis factor analysis is the most appropriate method of EFA.
Furthermore, due to the nature of the elements that comprise Disenchantment, it was hypothesised that any emerging factors would not be truly orthogonal. Whilst five elements were specified to comprise employee disenchantment, they were not hypothesised to be wholly independent from each other. Therefore, an oblique rotation was used, specifically Direct Oblimin. A cut-off point of .30 was applied to ensure that only significantly high loadings were con-sidered.
The Principal Axis EFA generated a four-factor solution with component Eigenvalues greater than 1. This solution accumulatively explained 74.1% of the variance. The pattern matrix of the secondary EFA can be seen in Table 1.

Confirmatory Factor Analysis (CFA)
Based upon the results of the EFA, SEM was used to analyse, confirm, and refine the factor layout of disenchantment. Two competing models were analysed in order to compare the best fitting structure of disenchantment. which implies that the model deviates from the structure of the data. However, it has been noted in previous research that this statistic value can be artificially in-flated by large samples, resulting in a rejection of the model. Therefore, for the rest of the analysis, other absolute fit indices will also be utilised. Overall, these indices indicated that model 1 represented a good fit of the data: CFI = .94; TLI = .93; RMSEA = .053 (lower 90% confidence interval = .051; upper 90% confidence interval = .056). Model 1 can be seen in Figure 1.
Model 2 analysed a theoretical expansion of the EFA. The Distrust & Unfairness factor drew from two sets of research within the literature; namely Perceived Inequity and Distrust. Whilst EFA of this one factor alone did not partition it down any further, the factor was split in the SEM to see if it purported a better fit. Organisational Lying, Respect/Bullying, and Broken Promises were inputted in the same manner as seen in model 1. The items that loaded onto Distrust & Unfairness were instead represented by two latent variables: Perceived Inequity (made of nine observed variables, being the nine Perceived Inequity questions from the literature) and Distrust (made of nine observed variables, being the nine Distrust questions from the literature). As with model 1, Organisational Lying, Respect/Bullying, Broken Promises, Distrust, and Perceived Inequity, were correlated with each other, with these yielding significant relationships.
Both models represented a good fit of the data, so comparative statistics were employed to test for significant differences between the models. It is usually that comparing the χ 2 statistics of the models in order to assess for significance. Details of this can be found in the Appendix. Model 1 was found to be a significantly worse fit of the data compared to Model 2 (χ 2 (4) = 162.05; p < .001).
Model 2 was thus selected as the final factor-representation of disenchantment, representing a simpler and more parsimonious representation of the concept. Thus, CFA determined that employee disenchantment is best represented through five inter-related factors: Organisational Lying, Respect/Bullying, Perceived Inequity, Distrust, and Broken Promises. Table 2 shows the descriptive statistics of employee disenchantment scores.

Descriptive Statistics
Disenchantment was calculated by summing scores for the five factors. The mean score for disenchantment was 176.23 for the 45 items, with scores ranging from 45 to 441 (SD = 94.0). One-way ANOVAs were run in order to assess if there were any differences across self-identified gender for total disenchantment and the five individual factors. No significant differences were noted in the overall model, or in Tukey post-hoc comparisons.
Differences in disenchantment according to age was analysed in two ways.

Discussion
This study defined and examined the psychometric properties of employee disenchantment; a targeted negative affect felt towards an employee's colleagues, superiors, and organisation. Whereas previous research has looked at the role of negative effect, this is the first to attempt to conceptualise negative effect through five inter-related, directed emotional responses. Thorough analysis revealed that the best representation of employee disenchantment was indeed through this five-factor model, comprising of Organisational Lying, Perceived Inequity, Bullying, Distrust, and Broken Promises. Analysis of the psychometric properties of disenchantment revealed it to have both high internal and concurrent validity.

What Is Employee Disenchantment?
The results of the two studies in this paper demonstrate the importance of measuring disenchantment as a work-based phenomenon. Disenchantment represents a novel approach to understanding the negative affect that can manifest in employees, focusing on negative emotion that is targeted towards the source of the employee's stress. Disenchantment is a psychological phenomenon that is accumulated from five psychometrically inter-related, negative perceptions: Organisational Lying, Perceived Inequity, Respect/Bullying, Distrust, and Broken Promises. Each construct identifies attitudinal shifts, representing resentment and embitterment towards the employee's colleagues, supervisors, and organisation. Whilst initial EFA revealed a four-factor model of disenchantment, CFA in both studies indicated that a five-factor model was a significantly better representation of the data. Based upon reviews of the literature, employee disenchantment was to represent the psychological phenomenon underlying a range of negative affect. As an overall phenomenon, it represents the negative affect that emerges within the employee in response to specific job stressors. This negative affect extends previous research by demonstrating that the affect is directed towards the source of job-stressor, namely the employee's colleagues, management, and the organisation. The results suggest that it would be meaningful for theorists and organisations to extend their view job-stressor related negative affect by conceptualising these emotions as specific, targeted, and focused.

Who Is More Prone to Disenchantment?
The two studies assessed whether certain demographic variables or job types were more likely to elicit disenchantment. Disenchantment did not differ by gender or age across both studies. Age was re-categorised into three distinct generations (Baby Boomers, Generation X, and Generation Y) but still no differences were noted. This result suggests negative perceptions towards organisations, management, and colleagues occur at a consistent rate, and for similar reasons, across different age groups. This supports previous meta-analyses that found generational differences in work-related attitudes are modest to non-existent (Costanza et al., 2012). Despite no relationships regarding overall disenchantment, distrust was significantly correlated with age in Study 2. This suggests that older workers are more likely to feel distrusting towards the organisation and its members than younger workers.

Limitations
This study is not without its limitations. The cross-sectional nature of this study only provides a snapshot into how disenchantment facilitates the explanation of employee deviance. Cross-sectional study design has been criticised with regards to predicting related outcomes for not being able to fully verify causality (e.g., Wright et al., 2003;Meier & Spector, 2013). Future research into disenchantment would need to assess the longitudinal impact this negative affect has on the proclivity to harm an organisation or its employees.
Furthermore, the data in this study was collected solely through self-report measures. Future research should aim to address this limitation by collecting observational or behavioural data. Other-report data on disenchantment and CWBs would provide more insight into the relationship between these variables.