Psychological empowerment as moderator of the relationship between core self-evaluation and proactive work behaviour

Abstract

This study examined the relationships between core self-evaluation, psychological empowerment, and proactive work behaviour. Oriented by situation strength theory, the study also examined the moderation role of psychological empowerment in core self-evaluation and proactive work behaviour relationships. This study relates personal characteristics of core self-evaluation, the situational variable of psychological empowerment, to organisationally desirable employee outcomes of proactive work behaviour. While core self-evaluation and psychological empowerment were analysed as one-dimensional variables, proactive work behaviour has four dimensions: problem prevention behaviour, individual innovation behaviour, voice behaviour, and taking charge behaviour. Two hundred and fifty-five employees sampled from public and private organisations were used for data analysis. Data were analysed with the PROCESS tool hosted in SPSS regression. The results showed that psychological empowerment was positively associated with proactive work behaviour and moderated the relationship between core self-evaluation and proactive work behaviour. The study contributes to the body of knowledge with the potential for a middle-range theory on psychological empowerment and proactive work behaviour relationship, substantially supports situation strength theory, and provides knowledge on the best combination of core self-evaluation and psychological empowerment for proactive work behaviour.

IMPACT STATEMENT

Human resource strategies, particularly staff selection and training, are heavily influenced by personality. This study looked at the efficiency of a powerful personality model (core self-evaluation) in predicting proactive work behaviour, a type of employee behaviour that significantly impacts organisational productivity since it boosts individual and team performance. Because personal dispositions interact with situational circumstances to affect employee behaviour, this study looked at psychological empowerment in core self-evaluation and proactive work behaviour links as a moderation factor. Three key takeaways for HR professionals emerged from the data. First, proactive work conduct is not supported by critical self-evaluation attributes. Second, to promote proactive job behaviour, an employee must have a low or average level of psychological empowerment. Third, cultivating a low or moderate level of psychological empowerment in employees is required to diminish the organisation’s proactive work behaviour characteristics if the organisation’s selection approach has resulted in individuals with core self-evaluation features.

Keywords:

• Core self-evaluation
• psychological empowerment
• proactive work behaviour
• situation strength theory

REVIEWING EDITOR:

• Jorge Linuesa-Langreo, Universidad de Castilla-La Mancha - Campus de Cuenca, Spain

SUBJECTS:

• Multidisciplinary Psychology; Work & Organizational Psychology; Human Resource Management; Critical Management Studies

Introduction

Employee behaviour has been extensively discussed as an individual enhancer in organisational performance (Jha et al., 2019; Lisbona et al., 2018; Nurjaman et al., 2019; Wadkj & Mos, 2021). The contemporary business world is characterised by uncertainty and severe competition; therefore, organisations need self-initiating employees engaged in anticipatory actions to improve work processes. The results are ultimate desires for societal sustainability. This change-focused process is captured with the phrase ‘proactive work behaviour’ (PWB) in the organisational behaviour literature. Additionally, several studies have been conducted to identify precursors of PWB to understand how stakeholders can effectively promote it in the workplace by designing evidence-based interventions (Boonyarit, 2023).

The existing literature indicates that personality, a comparatively stable set of psychological attributes that differentiate an individual from another, receives much empirical attention along with PWB (Griffin et al., 2020). This trend is expected, as the emerging literature has identified personality as a malleable variable (Baranski et al., 2020; Monestier & Bell, 2020), boosting the use of knowledge in the workplace. However, while several personality models (e.g. Big Five, Big Three, character strengths, HEXECO, proactivity, and core-self evaluation) are available, the search for antecedents of PWB from the domain of personality mainly centred on the proactive personality model (McCormick et al., 2019; Waterwall, 2019) and, to a lesser extent, on the Big Five personality model (Chen et al., 2020; Zuraik, Kelly, & Dyck, 2020). The literature is scant on studies investigating PWB from other personality models, such as core self-evaluation and character strengths. However, empirical evidence exists with some degree of relatedness between the Big Five personality and other personality models, such as the Big Three and the five alternatives (Aziz & Jackson, 2001), with no available empirical confirmation of similarity between the CSE and other personality models. This means the existing knowledge does not offer information on how the core self-evaluation model of personality relates to other personality models; therefore, understanding other personality models could not be justifiably generalised to core self-evaluation. So, how core self-evaluation relates to PWB needs to be investigated. Studying CSE for this understanding is justified as the extant literature substantially references organisationally desirable outcomes of PWB (Akhtar & Ali, 2023; Nurjaman et al., 2019).

Core self-evaluation, which covers traits that depict belief in one’s inner worth and basic competence (Robbins & Judge, 2019), is a practical model of personality that has received substantial attention with many organisational behaviour constructs. Much interest is built on CSE as the big five models are not generally accepted as superior to other models in every situation. For example, studies have shown that character strength provides incremental validity above the Big Five. In addition to the overlap of character strengths and the Big Five, there were significant contributions to positive outcomes by character strengths unaccounted for by the Big Five (Grohman et al., 2017; McGrath et al., 2020). The literature is lacking in studies on the relationship between CSE and PWB. Although some studies addressed the individual elements of CSE as they related to PWB (Chang et al., 2019; Dawwas & Shafig, 2018; Peariasamy et al., 2020; Rasyid & Marta, 2020), when the components traits of CSE are treated separately, they no longer represent the concept of CSE. The search for predictors of PWB in the personality domain needs to be inclusive and appropriately extended to the CSE model. Gestalt principles have been proposed with the confirmation that the whole differs from the sum of aggregated parts (Lemon et al., 2007). The researchers vigorously argued that each dimension of CSE has a rich literature and that CSE is a higher-order concept that should be investigated (Neves & Champion, 2015).

The CSE is a personality model that comprises four traits: self-efficacy, self-esteem, locus of control, and emotional stability. Although widely associated with enduring, stable, and consistent characteristics, situational factors influence personality manifestations. This understanding is vigorously pursued in studies investigating intervening variables in the relationship between personality traits and behaviour (Beckert & Ziegele, 2020; Geuens et al., 2020). The outcome of such studies is laudable as they provide insight into the possibility of meaningful changes in individual personality (Roberts et al., 2017). Eto and Watanabe (2014) noted that the CSE literature is mature enough and that researchers should examine moderating and mediating effects and uncover why CSE predicts its outcomes. A literature review indicated that (Eto & Watanabe, 2014) believed the issue had not been substantially addressed. This study examined PE as a moderator in the CSE and PWB relationship. PE is the energy embedded in the belief that tasks contribute to some larger purpose (Colquitt et al., 2019). It is a situational variable that appears steadily in studies that examine the interaction of personality and situation with behaviour (Aydogmus et al., 2015; Yildiz et al., 2017). This study is driven by a lack of empirical research on the relationship between the CSE model of personality and PWB and scholarly calls to provide understanding regarding the mechanism involved in the relationship. Consequently, this study investigated the relationship between CSE and PWB and the moderating role of psychological empowerment in the relationship.

Theoretical review and hypotheses development

Core self-evaluation

Judge et al. (1998) proposed the concept of core self-evaluations as a high-order aggregated construct of self-esteem, self-efficacy, internal locus of control, and emotional stability (Sharma & Misra, 2017). According to Farčić et al., (2020), it represents the essential assessment of individuals’ self-worth and capabilities. Although generalised self-efficacy estimates one’s ability to cope with various situations, self-esteem is a general assessment of one’s self-worth. Emotional stability is the ability to maintain a low level of neuroticism. The locus of control is the individual, believing that they manipulate events in their life (Cristofaro et al., 2020). An individual could be high or low in self-core evaluation. Characteristics of high-core self-evaluation individuals include self-confidence, self-worth, competence, and acceptance of oneself in various situations (Chen et al., 2023). In the literature, core self-evaluation is reported to have organisationally desirable direct and indirect relationships with several employee behaviours. For instance, CSE presents a direct influence on job turnover intention (Akosile & Ekemen, 2022), depression (Chen et al., 2023), job satisfaction (Sutanti & Sandroto, 2021), employee engagement (Ma et al., 2022) and moderated the association of demands with work-family enrichment (Jain & Nair, 2020).

Proactive work behaviour (PWB)

Proactive work behaviour refers to an employee’s self-initiated efforts to achieve progress, predict, deal with problems, and capture opportunities (Chen et al., 2019; Martínez-Córcoles & Zhu, 2020; Otto et al., 2019). It is individual employees’ self-initiated, anticipatory, future-oriented, and change-inducing behaviour. Several models exist in PWB (e.g. Parker & Collins, 2010; Spychala & Sonnentag, 2011). This study adopted the Parker and Collins (2010) proactive work behaviour model of four factors (individual innovation behaviour, problem prevention behaviour, voice behaviour, and taking charge behaviour). The model is substantially inclusive.

Individual innovation behaviour (IIB) involves creating and implementing ideas (Kawasaki, 2019). Employee innovation implies a notion of employees related to introducing novel processes, products, markets, or mixtures into the organisation (Elidemir et al., 2020). Problem prevention behaviour, the least researched among the dimensions of PWB (Nguyen et al., 2020), refers to behaviour related to the prevention of the recurrence of problems or challenges at work (Kawasaki, 2019). Voice behaviour (VB) is a proactive behaviour that highlights challenges in work situations to improve and not criticise (Jha et al., 2019). Voice behaviour entails suggestions to enhance organisational processes or prohibitive, which involves concern about the likely wearisome policies, practices, and work behaviour (Um-e-Rubbab & Naqvi, 2020). Taking charge behaviour (TCB) is intended to change how work tasks are executed (Kawasaki, 2019). Taking charge is the positive effort of employees to initiate self-improvement, improve organisational procedures, and support purposeful organisational changes (Zeng et al., 2020). TCB reveals faulty or misdirected procedures that undermine and prevent organisations from adapting to changing environments, hindering organisations’ effectiveness (Homberg et al., 2019).

Core self-evaluation and proactive work behaviour

The CSE is a composite construction with characteristics that can positively impact the PWB. The construct reflects the individuals’ belief that they can control their success (Chen et al., 2016). For instance, employees who feel more self-confident about a particular task will tend to perceive higher levels of expectancy and, therefore, be more likely to choose to exert high levels of effort (Colquitt et al., 2019). Individuals with high emotional stability can adapt to unexpected or changing workplace demands (Robbins & Judge, 2019). Self-efficacy is related to this type of feedback variable on self-competence (Hao et al., 2019). Therefore, self-efficacious individuals tend to develop a strong commitment to the activities that interest them.

Several related empirical studies strongly confirm the various theoretical understandings in the preceding discussion. For example, CSE correlated positively with PWB (Effong & Akpan, 2020) and innovative work behaviour (Purba & Paundra, 2018). Self-esteem mediates between team trust and voice behaviour (Chang et al., 2019). Self-efficacy leads to greater personal initiative (Lisbona et al., 2018) and moderates the effect of organisational politics on proactive behaviour (Rasyid & Marta, 2020). It is also related to entrepreneurs’ innovative behaviour (Mielniczuk & Laguna, 2020) and mediated gender-voice association (Eibl et al., 2020). The high internal locus of control increased individual innovation (Dawwas & Shafig, 2018), impacted creativity (Liu et al., 2019), and predicted innovative work behaviour (Töre & Yolal, 2017). Therefore, it is hypothesised that:

H1: There is a positive influence of (1a) core self-evaluation on proactive work behaviour, (1b) core self-evaluation on problem prevention behaviour, (1c) core self-evaluation on individual innovation behaviour, (1d) core self-evaluation on voice behaviour, and (1e) core self-evaluation on taking charge behaviour

Psychological empowerment and PWB

Psychological empowerment refers to the belief of employees in the degree to which they influence their work environment, their ability, the significance of their job, and their perceived autonomy in their work (Robbins & Judge, 2019). Psychological empowerment is a four-dimensional variable covering meaning, competence, self-determination, and impact. Meaning represents the value of the work to expectations. Competence indicates the ability to execute tasks skillfully. Self-determination implies individual autonomy and responsibility to decide on the method, pace, and effort when completing tasks. Moreover, impact refers to how a person can influence outcomes at work (Colquitt et al., 2019; Robbins & Judge, 2019). Employees who feel psychologically empowered would see themselves as influential in their workplaces and be inspired to be more independent, proactive, and show initiative (Kustanto et al., 2020; Prabowo et al., 2018). In several studies, PE has been directly associated with PWB and its various aspects. For example, PE related to PWB (Kawasaki, 2019), innovative work behaviour (Wang et al., 2021), voice behaviour (Wei et al., 2020), the autonomy of the work method, and independence of the work schedule independently affect innovative behaviour (Takaishi et al., 2019). Perceived work meaningfulness predicted voice behaviour (Chen et al., 2018). More so, previous literature has identified psychological empowerment as an important driver of proactive performance at the individual level (Boonyarit, 2023). Therefore, it is hypothesised that:

H2 There is a positive influence of (2a) psychological empowerment on proactive work behaviour, (2b) psychological empowerment on problem prevention behaviour, (2c) psychological empowerment on individual innovation behaviour, (2d) psychological empowerment on voice behaviour, and (2e) psychological empowerment on taking charge behaviour.

Psychological empowerment as moderator in the relationship between CSE and PWB

Situation-strength theory gives impetus to the proposed plausibility of PE as a moderator in the CSE relationship with PWB. Situation-strength theory suggests that personality translates into behaviour depending on the situation’s strength (Robbins & Judge, 2019). It is essentially about how a situation has clear behavioural expectations, incentives, or directives that blur individual differences (Colquitt et al., 2019). This theory proposes that the strength of a given situation constrains or compresses the variance of the dependent variable. It is an interactionist perspective to the understanding and explanation of behaviour. The interactionist perspective posits that factors related to the individual, such as personality traits and the situation, influence behaviour. PE is an attitude that could lead to identification, individual satisfaction, and feelings and cognition that can alter the relationship between personality traits and behaviour. Some studies have offered confirmation of the several elements of situation strength theory. For example, a climate of innovation, flexibility, and transformational leadership moderated the association between proactive employee personality and proactive behaviour (McCormick et al., 2019). Likewise, the autonomy of work criteria interacts with a proactive personality to effect innovative behaviour (Takaishi et al., 2019). The characteristic of the job strengthens the relationship between proactive and innovative work behaviour (Nurjaman et al., 2019).

Several studies have examined the intervening role of PE in PWB nomological networks, and the emerging results point toward confirmation. For instance, PE mediated the relationship between empowering leadership and proactive behaviour (Ul Haq et al., 2019), authentic leadership and employees’ proactive behaviour (Zhang et al., 2018), task significance, and proactive work behaviour (Kawasaki, 2019), transactional leadership and innovative work behaviour (Kustanto et al., 2020). Employee perceptions of meaningful work mediated the relationship between servant leaders and innovative work behaviour (Cai et al., 2018), transformational leadership, and employee innovation (Pradhan & Jena, 2019). Employee perceptions of autonomy moderated the relationship between servant leaders and innovative work behaviour (Cai et al., 2018). Therefore, it is hypothesised that:

H3: PE moderates the relationship between (3a) core self-evaluation and proactive work behaviour, (3b) core self-evaluation and problem prevention behaviour, (3c) core self-evaluation and individual innovation behaviour, (3d) core self-evaluation and voice behaviour, and (3e) core self-evaluation and taking charge behaviour.

Figure 1 presents the conceptual framework of this study. The framework, including its dimensions, represents the relationship between CSE, PE, and PWB. In this framework, CSE and PE were proposed to have a recursive association with PWB. The effect only flows from CSE and PE to PWB, not reverse. Additionally, the model depicts PE as a moderator in the CSE and PWB relationship.

Figure 1. Conceptual Framework.

Methods

Sample and design

Two hundred and fifty participants were drawn from public and private organisations in Nigeria’s Delta State. The sampled organisations covered public and private educational institutions and health facilities. The participants comprise 52% men and 48% women; 77% hold certificates below the first degree, 23% hold certificates above the first degree, 61% from public-owned organisations, and 39% from private-owned organisations. The mean age of the participants was 33.9 years (SD, 9.87; range, 40 years). The sample included all employee categories (junior, senior, management, and non-management staff members), as this improves ecological validity. The sample size sufficiently satisfied the sample-to-item ratio (5-1) and sample-to-variable ratio (20-to-1). This approach to sample size determination is gaining broad recommendation and acceptance in the literature, which proves its validity (Memon et al., 2020; Wu et al., 2023). The research design was cross-sectional, and a convenience sampling procedure was used to select the participants. Convenience sampling is the dominant sampling technique in social science research (Winton & Sabol, 2022) and the most used in psychology (Scholtz, 2021).

The study is cross-sectional as the data were collected from the participants at a point in time during the convenience sampling procedure because the participants were sampled based on availability. Both cross-sectional design and convenience sample are standard features of behavioural research literature (Andrade, 2021; Scholtz, 2021), which points to their acceptability. Specifically, researchers (e.g. Spector, 2019) noted the value and adequacy of cross-sectional design in organisational behaviour studies. A convenience sampling technique was applied as several organisations were sampled for the study. More so, the various organisations sampled lacked sampling frames, which ruled out probability sampling. Since data were collected with self-report measures, several procedural controls were embedded in the design to reduce the likelihood of common-method variance (CMV). For instance, to create a physical gap and cut the flow of thought from one variable to the other, the three substantive variables of this study (CSE, PE, and PWB) were presented to the respondents on different sheets of paper with varied introduction notes. The questionnaire informed the participants of their anonymity and confidentiality of the information provided, which aimed to reduce the apprehension of respondents about being evaluated and, therefore, improve their honesty in response (Rodríguez-Ardura & Meseguer-Artola, 2020; Ruiz-Palomino et al., 2021).

A total of 302 surveys in the English language were distributed to participants at their work sites. In all the organisations sampled, the questionnaires were distributed with the help of administrative staff members. Within approximately four weeks, 274 completed questionnaires were received. Physical examination of the received questionnaires showed that 24 respondents did not provide sufficient information, including missing values and outliers that invalidate their study questionnaires. Consequently, analyses were performed on data provided by 250 participants. It gives a response rate of 90.72% and is considered satisfactory (Luiten et al., 2020).

Measures

A 12-item one-dimensional scale (Judge et al., 2003) was used to measure CSE. The CSE scale measures one factor that embedded the concepts of generalised self-efficacy, locus of control, emotional stability and self-esteem. The scale has been validated and used across cultures and demographics with reports of good psychometric properties (Chen et al., 2023; Cristofaro et al., 2020; Manichev et al., 2022). The sample items on the scale are ‘I am confident that I will get the success I deserve in life and ‘I am capable of coping with most of my problems’. A higher average score indicates a higher CSE for the individual. Spritzer (1995)’s 12-item scale was used to measure PE. The scale addresses the four dimensions of PE: meaning, competence, self-determination, and impact. Each dimension has three items but is analysed in this study as a composite variable in the present study. The scale has translation in several languages (e.g. Hancer, 2005; Uner & Turan, 2010). Several studies have adopted the scale (e.g. Akosile & Ekemen, 2022; Boonyarit, 2023; Khusanova et al., 2019; Nwanzu et al., 2023; Sutanti & Sandroto, 2021). Moreover, satisfactory psychometric properties were reported. The sample items on the scale are ‘my job activities are personally meaningful to me; ‘I have significant autonomy in determining how I do my job’.

The dependent variable of this study is proactive work behaviour. Parker and Collins (2010) 13-item scale measured proactive work behaviour with four dimensions: problem prevention behaviour (3-items), individual innovation behaviour (3-items), voice behaviour (4-items), and taking charge behaviour (3-items). It has good psychometric properties (Kawasaki, 2019). Sample items on the scale are ‘I frequently promote and champion ideas to others?’; ‘I frequently generate creative ideas’.

Core self-evaluation and PWB were measured on a five-point Likert scale (strongly agree 5, agree 4, undecided 3, disagree, 2. and strongly disagree 1). Similarly, PE was measured on a seven-point Likert scale (strongly disagree 1, disagree 2, somewhat disagree 3, neither disagree nor agree 4, somewhat agree 5, agree 6, strongly agree 7). Data analysis was based on an individual level; for all scales, scores were calculated by averaging participants’ responses to items. Analyses were conducted with the Process Micro add-on to IBM SPSS version 28.

Control variables

Any variable that affects a presumed influence and its outcome is a common cause or confounding variable or the ‘third variable problem’ (Keith, 2019). In a research study, a control variable is a variable that is kept constant. Such variables are controlled or held constant even though they are not of interest to the study’s goal but can influence the study findings due to their relationship with the focal variables (Allen, 2017; Frey, 2021). In social science literature, the phenomenon of the control variable is intensely discussed, recommended and applied (Atinc et al., 2012; Becker et al., 2016; Nielsen & Raswant, 2018; York, 2018). Therefore, in this study, gender and age were analysed as control variables, as substantial studies implicated them as common causes of CSE and PWB (see, e.g. Bianco et al., 2020; Ogihara, 2020; Zuraik, Kelly, & Perkins, 2020), and as analysed in several related studies (e.g. Akosile & Ekemen, 2022; Choeni et al., 2023). More so, sociocultural theory (Eagly & Wood, 2012) explains the influence of gender and age on behaviour in social settings.

Ethical consideration

The Research Ethics Committee of the Faculty of Social Sciences of Delta State University, Nigeria, approved the study. Participation was voluntary, and the information about the respondents was treated confidentially.

Results

Measurement model

The Harman single-factor test and correlation matrix statistical procedures were used as diagnostic tools to assess common-method variance (CMV) bias in the data. Harman’s single-factor test revealed that all factors explained 75.85% of the total variance, while the first factor explained 23.42% of the variance. The variance explained by the first factor did not represent up to 50% of the total variance, indicating that there is not a significant amount of CMV in the data. From the correlation matrix procedure, the correlations between the variables studied were substantially below 0.9, which means CMV is also not an issue with the data (Martínez-Córcoles & Zhu, 2020; Rodríguez-Ardura & Meseguer-Artola, 2020).

Internal consistency reliability measures were performed through Cronbach’s alpha. The observed coefficient alpha is presented in Table 1, with alpha between 0.74 and 0.94, indicating satisfactory internal consistency reliability for each scale (Field, 2018). Face and content validity was achieved by adopting existing scales (Mirjana et al., 2018). Construct validity was tested with convergent and discriminant validity, an estimate of the extent to which variance in the measure reflects variance in the underlying construct (Jha et al., 2019). The observed Cronbach alpha statistics supported the convergence validity of the scales (Field, 2018). Discriminant validity with factor structure (cross-loading) was tested. The items’ loading in their construct was higher than their cross-loading, indicating acceptable discriminant validity. Such an item-loading pattern substantiates that the items belonged solely to their factors (Makhijaa & Akbarb, 2019). Based on theory and empirical work, age and sex were proposed as confounding factors in the relationship between CSE and PWB. However, the only notable difference between the pre-and post-introduction of the control variables in the analysis was the change from a non-significant negative relationship between CSE and PWB to a significant negative relationship.

Table 1. Mean, standard deviation, alpha, and zero-order correlation of the variables studied.

				Α	1	2	3	4	5	6	7	8
		x¯	SD
1	Gender	1.50	0.50		1
2	Age	33.49	9.87		−0.26**	1
3	CSE	2.66	0.58	0.85	−0.07	0.01	1
4	PE	5.58	1.00	0.94	−0.01	−0.01	−0.16*	1
5	PWB	3.67	0.62	0.74	−0.15*	−0.004	−0.13*	0.31**	1
6	PPB	3.87	0.71	0.74	−0.02	−0.02	−0.19**	0.22**	0.49**	1
7	IIB	3.57	0.73	0.90	−0.08	−0.01	−0.12	0.34**	0.61**	0.40**	1
8	VB	3.73	0.73	0.82	0.01	−0.09	−0.22**	0.31**	0.61**	0.36**	0.50**	1
9	TCB	3.73	0.82	0.85	−0.08	−0.12	−0.16**	0.33**	0.64**	0.31**	0.44**	0.52**

Note: CSE = Core-self-evaluation; PE = Psychological empowerment; PWB = Proactive work behaviour; PPB = Problem prevention behaviour; IIB = Individual innovation behaviour; VB = Voice Behaviour; TCB = Taking charge behaviour; ** < 0.015; * < 0.05 (two-tailed).

Descriptive statistics

Table 1 shows the means, standard deviations, coefficient alpha, and zero-order correlation coefficients for the variables studied. The observed means could be interpreted to be moderate when judged from a 5-point and 7-point Likert scale format.

Hypotheses testing

Hypothesis testing was conducted with the PROCESS tool (Hayes, 2018), the most appropriate approach to moderation and mediation (Field, 2018). Table 2 (Hypotheses 1a, 2a, and 3a) shows that CSE negatively and statistically predicts PWB, while PE positively and significantly predicts PWB and significantly moderates the influence of CSE on PWB, β = 0.26, 95% CI [0.15, 0.38], t = 4.46, p < 0.001. CSE and PE jointly account for 22% of the variance in PWB. Independently, CSE accounts for less than 1% variance in PWB, while PE accounts for 10% variance in PWB, as indicated in the squared part correlations.

Table 2. CSE, PE, and PWB relationships.

	B	Se	t	P
CSE	−0.13 [−0.25, 0.00]	0.06	−1.05	0.05
PE	0.24 [0.16, 0.32]	0.04	56.04	0.001
CSE × PE	0.26 [0.15, 0.38]	0.04	4.46	0.001
	R = 0.47; R^2 = 0.22; PC^2 (CSE) = 0.01, PC^2 (PE) = 0.10

Note: CSE = Core-self-evaluation; PE = Psychological empowerment; PC² =Part correlation squared.

Conditional effect statistics show that when PE was high (+1 SD), there was a non-significant relationship between CSE and PWB (β = 0.13, t = 1.54, p > 0.13), and when PE was low (−1 SD), there was a significant relationship between CSE and the PWB, (β = −0.39, t −4.39, p < 0.001). Figure 2a shows the interaction effect of CSE and PE on PWB. The difference between the squared multiple correlations (0.11) of the model of ‘main effect only’ and the squared multiple correlations (0.22) of the interaction model produced a semi-partial squared correlation of 0.11, which indicates the strength of the interaction effect. The interaction of CSE and PE accounted for 11% of the variance in PWB. Since the moderation statistics were significant, a simple slope analysis was performed. A simple slope was performed to locate PWB in the CSE within high and low PE.

Figure 2. (a) Interaction effect of core self-evaluation and psychological empowerment on proactive work behaviour. (b) Effect of the interaction of core self-evaluation and psychological empowerment on individual innovative behaviour. (c) Effect of the interaction between core self-evaluation and psychological well-being on voice behaviour. (d) Effect of the interaction of core self-evaluation and psychological empowerment on taking charge behaviour.

Table 3 (Hypotheses 1b, 2b, and 3b) shows that CSE has a significant negative influence on PPB. CSE and PE jointly account for 8% of the variance in PPB. Independently, CSE accounts for a 2% variance in PPB, while PE accounts for a 3% variance in PPB, as indicated in the squared part correlations. PE significantly positively influences PPB but did not moderate the relationship between CSE and PPB, β = 0.06, 95% CI [−0.09, 0.21], t = 0.76, p > 0.45.

Table 3. CSE, PE, and PPB relationships.

	β	Se	t	P
CSE	−0.20 [−0.37, −0.03]	.09	−2.32	0.02
PE	0.15 [0.04, 0.25]	0.05	2.80	0.01
CSE × PE	0.06 [−0.09, 0.21]	.08	0.76	0.45
	R = 0.27; R^2 = 0.08; PC^2 (CSE) = 0.02, PC^2 (PE) = 0.03

Note: CSE = Core-self-evaluation; PE = Psychological empowerment; PC² = Part correlation squared.

Table 4 (Hypotheses 1c, 2c, and 3c) shows that CSE has a negative but insignificant predictive influence on IIB. PE has a positive and significant predictive influence on IIB and moderates the relationship between CSE and IIB, β = 0.37, 95% CI [0.23, 0.51], t = 5.20, p < 0.001. CSE and PE jointly accounted for 22% of the variance in IIB. Independently, CSE accounts for less than 1% variance in IIB, while PE accounts for 10% variance in IIB, as indicated in the squared part correlations. Conditional effect statistics show that when PE was high (+1 SD), there was a significant relationship between CSE and IIB (β = 0.27, t = 2.56, p < 0.01), and that when PE was low (−1 SD), there was also a significant relationship between CSE and IIB, (β = −0.46, t −4.35, p < 0.001). The difference between the squared multiple correlations (0.12) of the ‘main effect only’ model and the squared multiple correlations (0.22) of the interaction model produced a semi-partially squared correlation of 0.10, indicating the strength of the interaction effect. The CSE and the PE interaction accounted for 10% of the variance in IIB. Since the moderation statistics were significant, a simple slope analysis was performed. A simple slope was performed to locate IIB within high and low PE in the CSE. Figure 2b shows the interaction effect of CSE and PE on IIB.

Table 4. CSE, PE, and IIB Relationship.

	β	Se	t	P
CSE	−0.10 [−0.25, 0.06]	0.08	−1.25	0.21
PE	0.30 [0.20, 0.39]	0.05	6.12	0.001
CSE × PE	0.37 [0.23, 0.51]	0.07	5.20	0.001
	R = 0.47; R^2 = 0.22; PC^2, (CSE) = 0.004, PC^2 (PE) = 0.10

Note: CSE = Core-self-evaluation; PE = Psychological empowerment; PC² = Part correlation squared.

Table 5 (Hypotheses 1d, 2d, and 3d) shows that CSE has a negative and significant predictive influence on VB. CSE and PE jointly account for 43% of the variance in VB. Independently, CSE accounts for a 3% variance in VB, while PE accounts for a 6% variance in VB, as indicated in the squared-part correlations. PE has a positive and significant predictive influence on VB and moderates the relationship between CSE and VB, β = 0.26, 95% CI [0.10, 0.42], t = 3.14, p < 0.001.

Table 5. CSE, PE, and VB relationship.

	β	Se	t	P
CSE	−0.20 [−0.38, −0.02]	0.09	−2.22	0.03
PE	0.31 [0.20, 0.42]	0.06	2.45	0.001
CSE × PE	0.26 [0.10, 0.42]	0.08	3.14	0.001
	R = 0.44, R^2 = 0.19, PC^2, (CSE) = 0.03, PC^2 (PE) = 0.06

Note: CSE = Core-self-evaluation; PE = Psychological empowerment; PC² =Part correlation squared.

Conditional effect statistics show that when PE was high (+1 SD), there was no significant relationship between core self-evaluation and VB (β = 0.03, t = 0.26, p > 0.79), and when PE was low (−1 SD), there was also a significant relationship between core self-evaluation and VB, (β = −0.51, t −4.68, p < 0.001). The difference between the squared multiple correlations (0.12) of the ‘main effect only’ model and the squared multiple correlations (0.19) of the interaction model produced a semi-partial squared correlation of 0.07, which indicates the strength of the interaction effect. The interaction of CSE and PE accounted for 10% of the variance in the VB. Since the moderation statistics were significant, a simple slope analysis was performed. A simple slope was performed to locate VB in CSE within high and low PE. Figure 2c shows the interaction effect of CSE and PE on VB.

Table 6 (Hypotheses 1e, 2e, and 3e) shows that CSE has a negative and significant predictive influence on TCB. CSE and PE jointly accounted for a 16% variance in TCB. Independently, CSE accounts for a 1% variance in PWB, while PE accounts for a 9% variance in PWB, as indicated in the squared part correlations. PE has a positive and significant predictive influence on TCB and moderates the relationship between CSE and TCB, β = 0.24, 95% CI [0.08, 0.40], t = 2.92, p < 0.001.

Table 6. CSE, PE, and TCB relationship.

	β	Se	T	P
CSE	−0.17 [−0.34, 0.00]	0.05	−1.99	0.05
PE	0.30 [0.19, 0.41]	0.05	5.54	0.001
CSE × PE	0.24 [0.08, 0.40]	0.08	2.92	0.001
	R = 0.40, R^2 = 0.16, PC^2 (CSE) = 0.01, PC2 (PE) = 0.09

Note: CSE = Core-self Evaluation; PE = Psychological Empowerment; PC² =Part Correlation Squared.

Conditional effect statistics show that when PE was high (+1 SD), there was a non-significant relationship between CSE and TCB (β = 0.05, t = 0.49, p > 0.67), and that when PE was low (−1 SD), there was a significant relationship between CSE and TCB, (β = −0.46, t −0.3.71, p < 0.001). The difference between the squared multiple correlations (0.12) of the model of ‘main effect only’ and the squared multiple correlations (0.16) of the interaction model produced a semi-partially squared correlation of 0.04, which indicates the strength of the interaction effect. The CSE and the PE interaction accounted for 4% of the TCB variance. Since the moderation statistics were significant, a simple slope analysis was performed. A simple slope was performed to locate the TCB in the CSE within high and low PE. Figure 2d shows the interaction effect of CSE and PE on TCB.

Discussion

This research was carried out to ascertain the effect of CSE and PE on PWB and the moderation role of PE in the core self-evaluation and PWB relationships. Hypotheses 1a, 2a, and 3a were at the composite level of the analysis. However, while the first two showed statistically significant results in the predicted direction, the last yielded negative and statistically significant results against the prediction. Specifically, PE positively and significantly predicted PWB and moderated the relationship between CSE and PWB. Hypotheses 1b through 1e, 2b through 2e, and 3b through 3e were at the dimensional analysis level regarding PWB. The results of five of the 10-dimensional hypotheses were in the predicted direction. PE positively and significantly predicted PWB, PPB, IIB, VB, and TCB. The direct positive effect of PE on PWB and its dimensions, as obtained in this study, is consistent with the existing literature, as several studies (Kawasaki, 2019; Takaishi et al., 2019; Wang et al., 2021; Wei et al., 2020) have observed and reported similar results.

A plausible explanation for the reported direction of the relationship between PE and PWB is that PE comprises competence, autonomy, and impact embedded with the potential to drive PWB. A sense of competence and autonomy is needed to express a behaviour change. However, the negative and significant effect of core self-evaluation on proactive work behaviour, problem prevention behaviour, voice behaviour, and taking charge behaviour contradicted the predictions of this study and is critical as it indicates that personality could have a damaging effect on PWB. An explanation for the above direction of finding is that some aspects of CSE, such as self-worth, could interfere with the willingness to take risks associated with the initiation of change, which is what PWB is all about. PE moderated the effect of CSE on PWB, IIB, VB, and TCB. The slope revealed that the impact of CSE on PWB, IIB, VB, and TCB was more substantial when PE was high than when it was low. These findings align with what is available in the related empirical literature (e.g. Fajrianthi, 2019; Kawasaki, 2019; Kustanto et al., 2020; Pradhan & Jena, 2019). The result also agrees with the literature on situation strength theory (Alikaj et al., 2020; Takaishi et al., 2019). The moderation effect was so strong that the non-significant negative and significant negative relationships between CSE and PWB (and the dimensions) turned significantly positive and moderated. This finding indicated the strong influence of the situational variables, which PE represents, on personality traits.

A plausible account of the moderating effect of PE in the relationship between CSE and PWB is that PE elements, such as autonomy, provide an environment that activates CSE. However, Hypothesis 3b was not supported. PE did not moderate the relationship between CSE and PPB. It differs from expectations and leads to a call for studies on the interrelatedness of the four dimensions of PWB.

Contributions

PWB is a higher-order construct that embeds some well-established constructs. This research initiated a dimensional analysis of the study of PWB within the framework of PWB. As theoretical contributions, this study opens an investigation that brings CSE, PE, and PWB and introduces a literature model. Its findings serve as a reference point for subsequent tests of the model. More so, very few studies that examined CSE and PWB only approached the latter as a composite (Nurjaman et al., 2019) or examined only a dimension of PWB (Purba & Paundra, 2018). In this study, PWB was analysed as a composite and dimensional variable. Likewise, studies on PE and proactivity at the workplace limit the analysis of innovativeness (Messmann, 2023), a dimension of PWB. Thus, this study provided a comprehensive and more in-depth understanding of the relationship between CSE, PE, and PWB. In CSE, PE and PWB relationship studies, PE was usually treated as a mediating variable (Zhang et al., 2018). However, this study examined PE as a moderator, thereby contributing to a broader understanding of PE roles in various relationships of PWB.

It was observed that PE positively influences PWB and that the former also plays a moderation role in the CSE–PWB relationship. These findings started and provided an understanding that could help guide middle-range theorising on PE and PWB links. The results that PE moderated the relationship between CSE and PWB updated the existing literature that has offered support and confirmation of the interactionist perspective on the relationship between personality traits, situational factors, and behaviour. More specifically, the findings on the moderation hypotheses substantially offered support to the situation-strength theory.

Several practical implications can be deduced from the results of this study. PE positively influences PWB and its dimensions. PWB is a well-cherished employee activity, as both theories and empirical findings have established its association with the organisation’s effective functioning (Al-Makhadmah et al., 2020; Vu, 2020). This finding indicated that organisational professionals should create a PE environment to improve employee PWB. Transformational leadership is well documented to empower employees psychologically (Abdulrab et al., 2017; Prabowo et al., 2018). In this study, PE was a decisive moderator of the effect of CSE on PWB. The negative relationships between CSE and PWB were tuned to a significantly positive relationship when PE interacted with CSE. This finding implies that although the CSE of an employee could adversely interfere with PWB, that negative influence can be fully attenuated by psychological empowerment. Therefore, to build PWB in employees, organisations should be more concerned with providing a favourable work climate for the employees than dispositional factors.

The finding that CSE has a significant negative influence on PWB as a composite suggests to HR professionals that CSE may not be a factor to consider in personnel selection and training exercises when PWB in the organisation is of interest. The caveat is that although PE is a moderator in the CSE and PWB relationship, organisational practitioners should be aware of the moderation effect that occurs when PE is low or moderate. Therefore, it should be manipulated at the appropriate levels in the employees.

Limitations and suggestions for future research

The sole adoption of cross-sectional data and self-reported measures is a significant limitation of this study. Although some procedural control measures were implemented in the study design and some statistical checks were carried out, the chance of reverse causal influence and common-method bias remains. A combination of self and ‘other’ (e.g. supervisors, peers) is suggested for future research on data collection. Although several controls were embedded in the design to check common-method variance, the chance of such bias can not be ruled entirely on. Moreover, common-method bias has been noted to undesirably impact research results and lead to incorrect interpretations and conclusions. Therefore, It is recommended that in addition to procedural control of common method variance, more stringent statistical tests such as the marker variable technique (Crick, 2023) be implemented in future studies. The moderation model in this study comprises a moderator, and such a simple model is often not a satisfactory account of reality. Consequently, future studies should develop a more comprehensive model incorporating many moderators through theoretical and empirical work.

PE is widely conceptualised as a multi-dimensional variable. However, this understanding was not reflected in this study, as the variable was analysed solely as a composite. Future related studies should incorporate composite and dimensional analysis. Similarly, future studies should examine mediation in the CSE–PWB relationship. The finding of a direct test of the relationship between CSE and PWB was contrary to the prediction. The lack of studies on the variables requires more research to understand the relationship between variables for an informed conclusion.

The measures adopted for independent and dependent variables are established scales with wide usage and acceptance. However, these scales were developed in an environment different from the present study location. More so, the various studies on reliability and the validity of the scales were also conducted in settings alien to this study. There is no evidence of exploratory factor analysis and confirmatory factor analysis studies on the scales in the present research location. Therefore, these scales’ factor structures, reliabilities and validities in the present research location are not yet firmly established. Therefore, rigorous studies should test the usefulness of these scales across wide locations.

Conclusion

The present study examined the relationship between core self-evaluation, psychological empowerment and proactive work behaviour. Data analysis at the composite level revealed that psychological empowerment was positively associated with proactive work behaviour and moderated the relationship between core self-evaluation and proactive work behaviour. Broadly, the dimensional analysis yielded similar results. Therefore, it is concluded that (a) enhancing psychological empowerment enhances proactive work behaviour, (b) personality traits (e.g. core self-evaluation) need to interact with situational factors (e.g. psychological empowerment) to have a positive influence on behaviour (e.g. proactive work behaviour), (c) interaction of core self-evaluation and psychological empowerment has a substantial effect on proactive work behaviour, and (d) low level of psychological empowerment is needed to make the increasing level of core self-evaluation to have a positive influence on proactive work behaviour. Organisational practitioners have the task of appropriately aligning the relationship between core self-evaluation and psychological empowerment for enhanced proactive work behaviour.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data from this manuscript is available upon request from the authors.

Additional information

Notes on contributors

Chiyem Lucky Nwanzu

Chiyem Lucky Nwanzu is a PhD holder in Industrial/Organizational Psychology, a Senior Lecturer, and Head of the Psychology Department at Delta State University, Nigeria. His research interests are workplace behaviour, organisational sustainability and effectiveness.

Sunday Samson Babalola

Sunday Samson Babalola is a Professor of Human Resource Management at Walter Sisulu University, South Africa. He is also a South African Rated Scientist with over 20 years of teaching, research, and community engagement in higher educational institutions. He focuses on applying psychological principles and processes to workplace behaviour/attitude, emphasising human capital management and organisational behaviour. Babalola’s research benefited from a solid background in quantitative analysis, with working knowledge in qualitative analysis.