Original Research

Influence of employee perceptions of corporate social responsibility on affective commitment

Amorei van der Westhuizen, Dirk J. Malan

Received: 02 Nov. 2022; Accepted: 03 Feb. 2023; Published: 14 Mar. 2023

Copyright: © 2023. The Author(s). Licensee: AOSIS.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Orientation: The study attempted to establish the role played by different perceived corporate social responsibility (CSR) activities on the organisational commitment of employees.

Research purpose: To investigate the nomological network of variables playing a role in the relationship between Perceived CSR and Affective Commitment.

Motivation for the study: Understanding the contribution of perceived corporate social responsibility to the organisational commitment of employees could inform managerial attempts to facilitate positive employee attitudes.

Research approach/design and methods: The current study utilised structural equation modelling to evaluate the hypothesised multivariate causal relationships in the proposed model. A convenience sample of 172 employees of a consumer goods company completed several questionnaires, including measures of Affective Commitment, Perceived CSR, Organisational Social Identity, Perceived Organisational Justice, Psychological Contract, as well as three cultural value orientations.

Main Findings: The most significant finding of the study was the positive relationship between Perceived CSR and Affective Commitment. The results revealed that Perceived CSR has a positive impact on Perceived Organisational Justice and fulfilment of the Relational Psychological Contract. Both variables impacted on Organisational Social Identity, which was positively related to Affective Commitment.

Practical/managerial implications: Organisations should ensure that employees are made aware of their CSR policies and activities through corporate communication. Organisations should also invest their resources in dealing with social problems that are regarded as important by employees and engage in fair human resource practices.

Contribution/Value-add: Engagement in CSR activities could help retain and attract talented candidates and increase commitment and identification with the organisation.

Keywords: affective commitment; employees; nomological network; perceived social responsibility; South Africa.

Introduction

Orientation

Traditionally, profit maximisation was regarded as the primary objective of any business and its investors. In more recent times, however, other stakeholders and social considerations have become equally important in terms of organisational objectives and economic successes. This development has led to a broader definition of profit, which refers to economic, social, and environmental concerns, the so-called ‘triple-bottom line’ (Morland, 2006). The current focus on Corporate Social Responsibility (CSR) indicates that organisations have realised that, although financial success and profit maximisation for shareholders are essential, the organisation also has a responsibility towards the society and community in which it operates (Carroll & Shabana, 2010).

While the standing of an organisation among its shareholders and the general public is positively influenced by the CSR reputation of the company, the question arises as to whether the Perceived Corporate Social Responsibility of the company also positively influences the employee attitudes of its employees. Competent and motivated employees with favourable employee attitudes, such as Organisational Justice, Organisational Trust, Organisational Commitment and Job Satisfaction, are regarded as essential for several positive organisational outcomes (Iqbal et al., 2012; Judge & Kammeyer-Mueller, 2012). In recent studies, Perceived Corporate Social Responsibility has been positively linked to employee attitudes (Farooq et al., 2014; Gond et al., 2010), such as Organisational Commitment, and more specifically Affective Commitment (Papacharalampous & Papadimitriou, 2021).

A better understanding of the nomological network of variables that directly and indirectly influence the relationship between Perceived Corporate Social Responsibility and Affective Commitment could facilitate attempts to maximise the impact of CSR activities on employee attitudes.

Research purpose

The research question was therefore defined as: What role does employee perceptions of CSR play in explaining variance in Affective Commitment in the South African context, and which other latent variables play a direct or indirect role in the dynamics of this relationship? (Van der Westhuizen, 2019).

The current researchers therefore hypothesised that Pe rceived CSR positively influences Affective Commitment (H1).

Literature overview

Allen and Meyer (1996) developed a multi-dimensional and comprehensive model of Organisation Commitment with three distinguishable components, namely the Affective-, Continuance-, and Normative Commitment components.

Affective Commitment refers to identification with, involvement in, and emotional or affective attachment to the organisation. Continuance Commitment refers to commitment based on the employee’s recognition of the perceived costs associated with leaving the organisation, while Normative Commitment refers to commitment based on a sense of obligation to remain with the organisation. Employees with strong Normative Commitment remain with the organisation because they feel they ought to (Allen & Meyer, 1996).

Schleicher et al. (2011) regard the following as important antecedents of Organisational Commitment, namely job characteristics, aspects of the organisational climate, perceptions of constraints and hindrances versus support, role perceptions, justice and fairness perceptions, and the upholding or breach of the psychological contract. Other studies such as those by Papacharalampous and Papadimitriou (2021) have confirmed the role of individual level variables, like PsyCap.

Meta-analyses by Schleicher et al. (2011) have revealed various organisational level and individual level outcomes associated with Organisational Commitment, including motivation, job performance, organisational citizenship behaviours, stress and well-being, withdrawal cognitions and behaviours (such as absenteeism, attendance, lateness, intent to leave or search for alternative employment, and turnover). Studies like those conducted by Ha et al. (2014) and Brunetto et al. (2012) have, for instance, confirmed the impact of Organisational Commitment on turnover intention.

All definitions of CSR revolve around the perception that organisations have a responsibility towards society and the environment, over and above making profits for important stakeholders (Carroll & Shabana, 2010). Carrol’s original conceptualisation of the dimensions of CSR includes the economic, legal, ethical, and discretionary (philanthropic) expectations (Carroll & Shabana, 2010).

The economic dimension refers to the viewpoint that none of the stakeholder groups should be exploited to only allow management and shareholders to benefit from profits (Carroll & Shabana, 2010). It also links to the concept of sustainability, where organisations strive to focus on more than mere profits, but also on their products and their impact on the environment. Research indicates that positively Perceived Corporate Social Responsibility strategies can actually increase investment interest in the organisation, the purchase of products and services, as well as building better relationships with established governmental bodies (Valentine & Fleischman, 2007). There is also evidence of a positive relationship between CSR and overall organisational financial performance (Wang & Choi, 2010).

Compliance with laws and regulations forms part of the social contract between organisations and society (Carroll & Shabana, 2010), although organisations are generally not lawfully obliged to implement CSR strategies or activities, but adherence to several legal obligations can reflect the organisation’s standing on CSR. One aspect that is closely linked to the legal dimension of CSR is consideration for the environment, as companies must operate within set legal parameters from an environmental perspective. In a similar vein, organisations have become sensitive to ethical and social concerns in their business, as well as considering the impact of the business on communities (Dahlsrud, 2008).

The discretionary dimension of CSR refers to desirable or voluntary expectations and discretionary or philanthropic responsibilities, that lead to ‘actions that appear to further some social good, beyond the interest of the firm and that which is required by law’ (McWilliams et al., 2006:1). It shows good corporate citizenship when organisations engage in acts or programmes that will promote welfare and goodwill in the society or communities (Carroll & Shabana, 2010; Ferrell et al., 2011).

In the following section various variables that mediate or moderate the relationship between perceived CSR and affective commitment are discussed.

Brammer et al. (2007) established that externally aimed CSR was positively related to Affective Commitment, because of the indirect benefits through Social Identity for employees. Employee perceptions of internally aimed CSR, like training and procedural justice, were also positively related to Organisational Commitment. In their study, female subjects, however, viewed procedural justice as more important than training and externally focused CSR activities. In contrast Ditlev-Simonsen (2015) did not find any gender effects in her Scandinavian study. According to Roudaki and Arslan (2017) their study on employees of Pakistani hotels indicated that the organisational commitment of the employees was strongly influenced by the internally focused economic and externally focused philanthropic aspects of the CSR activities that they were exposed to.

Chang et al. (2013) claim that human resource practices and-systems play an important role in the link between CSR strategies and financial performance. Research on CSR, however, reveals that CSR strategies often do not regard employee expectations as equally important (Aguinis 2011; Chang et al., 2013; Wang & Choi, 2010). Bauman and Skitka (2012) discuss four basic psychological needs that explain how CSR perceptions affect employees’ views and relationship with the organisation. Firstly, both internally and externally aimed CSR activities create a sense of security and safety about material needs, while secondly, a positive Social Identity enhances employees’ self-esteem. Thirdly, feelings of belongingness and support for values can be increased by CSR, whereas lastly, CSR creates deeper meaning and purpose for employees in the work context. Empirical support has been found for various outcomes, such as job satisfaction, trust, as well as work-life satisfaction and identification (Martínez & Del Bosque, 2013; Senasu & Virakul, 2015).

Various studies have indicated that the relationship between perceived CSR and affective commitment is not necessarily a direct one. Mueller et al. (2012), for instance, have established the fact that Perceived CSR was positively related to Affective Commitment, after controlling for job satisfaction and unemployment rates. Farooq et al. (2014) confirmed that perceived CSR was a strong predictor of Affective Organisational Commitment, but that it was mediated by Organisational Social Identity and Organisational Trust. In a similar vein, Bouraoui et al. (2019) have confirmed the relationship between CSR and affective commitment, but found that perceived organisational support (POS), organisational identification, and perceived person–organisation fit mediated the relationship. In a later study (Bouraoui et al., 2020), the authors found that the relationship between person–organisation fit and perceived CSR was moderated by the need to belong. In the case of individuals with a high need to belong the impact of CSR was less salient. Ahmad et al. (2020) conducted a study on 736 academics from Pakistani universities and found support for the mediating role of organisational trust in the relationship between perceived CSR and affective commitment.

A study by Rego et al. (2007) indicated that employees who perceive their company as being responsible in terms of CSR and caring about well-being and values tend to develop Affective Commitment and Normative Commitment. Ditlev-Simonsen (2015) has conducted a Scandinavian study involving four companies and found that perceived CSR was a significant predictor of affective commitment, although POS proved to be a stronger predictor. Some theoreticians, however, might argue that POS could be seen as the result of internally focused CSR.

Social Identity is regarded as a key variable in the relationship between the two constructs (Farooq et al., 2014; Gond et al., 2010). Social Identity Theory suggests that employees have a need to identify with and feel part of the organisation they are working for to support and strengthen their self-esteem (Gond et al., 2010). Once an individual identifies with an organisation, they will also search for signals that will further enhance favourable views of the organisation (Burke & Stets, 2000). The current researchers hypothesised (H2) that organisational social identity mediates the relationship between perceived CSR and affective commitment (Van der Westhuizen, 2019).

Previous research has established that certain Cultural Values have indicated a significant moderating effect on the relationship between Perceived CSR and Organisational Commitment (Hofman & Newman, 2013; Mueller et al., 2012). For the purposes of the current study, the emphasis will thus be on identifying which cultural dimensions are relevant to the relationship between Perceived CSR and Organisational Commitment. Mueller et al. (2012), utilising the GLOBE Project values, found that employees’ CSR perceptions were influenced by a Humane orientation, Institutional collectivism, In-group collectivism, Future orientation, and Power distance. The authors further indicate that Cultural Values are substantial moderators of the effect of employees’ CSR perceptions on affective Organisational Commitment. Hofman and Newman (2013) established that Masculinity moderated the relationship between employee perceptions of CSR and Organisational Commitment, especially when there were positive perceptions of CSR towards the society.

Waldman et al. (2006) examined the effect of Institutional-Level Collectivism, In-group Collectivism, and Power Distance on CSR perceptions of managers. It was established that managers from cultures characterised by Institutional Collectivism valued all three dimensions of CSR as identified in the study.

Thanetsunthorn (2015) reviewed several studies that examined the role of national culture on CSR. The study, which was conducted in Asia and Europe, found that a national culture that is characterised by a higher Power index, Individualism, and Masculinity, was less concerned about CSR related to social aspects, including employees, the community, and environment. It was further established that managers from countries with greater Power Distance devalue the aspects of CSR.

Having considered the different studies that have researched the effect of Cultural Values on CSR and Organisational Commitment, it is hypothesised that Collectivism (H3), Power Distance (H4), and Masculinity (H5) dimensions will have a moderating effect on the relationship between Perceived CSR and Organisational Commitment (Van der Westhuizen, 2019).

Studies have indicated a positive relationship between employee perceptions of CSR and Organisational Justice. More specifically, it was established that procedural justice is a predictor of Organisational Commitment, although other types of justice are also regarded as significant. Gond et al. (2012) indicate that employee perceptions regarding CSR can influence perceptions of justice at an individual or collective level. Individually, perceptions regarding CSR can impact how employees perceive fairness when CSR strategies are aimed at, for example, improving working conditions, establishing fair wages, and promoting fair policies. Gond et al. (2012) argue that third-party judgements can also be formed, where employees evaluate how others are treated by the organisation.

Similarly, Rupp et al. (2013) investigated deontic and distributive justice in their study on Perceived Corporate Social Responsibility. Deontic justice theory developed around moral and ethical norms, where individuals do not only react to the treatment they themselves receive (first-party justice – distributive justice), but also show reaction to how others are treated (third-party justice). Rupp et al. (2013) note that, although research on third-party justice has been mainly concerned with studies conducted on co-workers, the same results are expected in terms of external stakeholders. Once again, the notion of supporting the community, environmental concerns, and other type activities associated with CSR is linked to concerns about the self and others. Moon et al. (2014) also established that Perceived CSR has a positive influence on Organisational Justice, while Organisational Justice influences Affective Commitment.

Having considered the discussion above, it is hypothesised (H6) that Perceived Organisational Justice will mediate the relationship between employee perceptions regarding CSR and Affective Organisational Commitment (Van der Westhuizen, 2019).

De Roeck et al. (2014) have found empirical support for the relationship between Perceived Organisational Justice and Organisational Social Identity. In a study conducted by Fuchs and Edwards (2012), it was also found that perceptions of fair treatment in an organisation encourage employee behaviours in change management interventions, which are partially mediated by Organisational Social Identity.

It is therefore hypothesised (H7) that Perceived Organisational Justice and Organisational Social Identity will be positively related (Van der Westhuizen, 2019).

The Psychological Contract can be viewed as the mutual beliefs and expectations concerning the exchange relationship between the individual and the organisation (Guest, 2004). Lee and Liu (2009) found that the Psychological Contract has a significant influence on Organisational Social Identity and work attitudes, like work devotion and work satisfaction. According to Rousseau (2011), fulfilment of the Psychological Contract can lead to increased Organisational Commitment. On the other hand, Zhao et al. (2007) found that a breach of the Psychological Contract has a significant impact on almost all outcomes, including mistrust, perceived violation, job satisfaction, Organisational Commitment, turnover intentions, Organisational citizenship behaviour (QCB), and task performance.

O’Donohue and Nelson (2009) argue that social values and beliefs, which include a willingness to behave ethically and take responsibility for social and environmental consequences, form part of the Psychological Contract. When the organisation’s action is perceived as incongruent with ethical values and beliefs, and the perceived ethical climate of the organisation, the actions are likely to be perceived as a breach of the psychological contract. It is therefore expected that employees form certain expectations regarding CSR, which include perceptions regarding ethical and social beliefs and values. Their expectations should be met to fulfil the agreement and influence Affective Commitment.

De Meuse et al. (2001) distinguish between two types of psychological contract, namely the transactional versus relational contract. While the transactional contract deals with specific short-term monetary remuneration for certain services, the relational contract focuses on the socio-emotive interaction between the employer and the employee (Lee & Liu, 2009; Rousseau, 2004). It therefore seems appropriate to focus on the Relational Contract in the context of the present study, which is concerned with Affective Commitment and not external monetary rewards. Affective Commitment can also be seen as the emotional aspect of Organisational Commitment. Therefore, it is hypothesised (H8) that the Relational Contract plays a mediating role in the relationship between Perceived CSR and Affective Commitment (Van der Westhuizen, 2019).

Lee and Liu (2009) found that Psychological Contract fulfilment influences Organisational Social Identity. It is subsequently hypothesised (H9) that the Relational Contract will be positively related to Organisational Social Identity (Van der Westhuizen, 2019).

Rodwell and Gulyas (2013) established that fulfilment of the Psychological Contract enhances perceptions of justice. Kickul et al. (2001) established a three-way interaction, in that anti-citizenship behaviour were higher following a contract breach when employee perceptions regarding justice were low. Epitropaki (2013), also empirically evaluated the relationship between breach of the Psychological Contract and Organisational Social Identity and found that procedural justice climate and leadership also impacted the relationship. The results thus suggest that justice perceptions and perceptions regarding the Psychological Contract are related.

Considering the goals of the current study, and the perspectives on Overall Justice perceptions, it is hypothesised (H10) that the Relational Contract will influence the perceptions of Organisational Justice of employees (Van der Westhuizen, 2019).

Based on the literature review and the different relationships between Organisational Commitment, Perceived Corporate Social Responsibility, Organisational Social Identity, Cultural Values, Perceived Organisational Justice, and the Psychological Contract, a theoretical model was conceptualised. The model depicted in Figure 1 thus reflects the proposed linkages between the different constructs.

FIGURE 1: Structural model representing the hypothesised relationships between the selected constructs.

Research design

Research approach

The present study used an ex post facto correlational design, with no manipulation of variables, to determine the relationships between the different constructs. The data for the study were collected at a single point in time by means of a survey to evaluate and empirically assess the proposed relationships between the latent variables, which defines it as a cross-sectional survey study.

A self-compiled questionnaire consisting of a selection of measurement instruments for each latent variable from the structural model depicted in Figure 1 was used to operationalise the model.

Research setting

Different organisations were approached to participate in the current study, but only one organisation allowed participation. The organisation that participated in the study is a multi-national, fast-moving consumer goods (FMCG) company. The company was appropriate for inclusion in the study given its international exposure (being a large multi-national company that employs individuals from several cultural orientations), as well as the fact that the company implements CSR strategies, which may not be the case for smaller businesses. Only the South African subsidiary company was included in the study. Prior to the commencement of the data-collection procedure, permission was obtained from the participating organisation. Approximately 400 questionnaires were distributed among employees (excluding senior management), of which 172 questionnaires were completed. The response rate was thus approximately 43%.

Research participants

The sample can be described as a convenience non-probability sample. The researcher attempted to include participants from across the different organisational levels and the cultural diversity of South Africa. The majority of the eventual participants were white (38%), followed by African (35%), Coloured (21%), Indian or Asian individuals (6%). In total 56% of the sample were male, while 44% were female participants. The majority held at least a Grade 12 level of education or higher (86%), and 70% of the participants were aged between 21 and 40 years. The participants were mostly non-managerial individuals (58%).

Measuring instruments

None of the questionnaires utilised in the research are regarded as psychological acts, as defined by the Health Professions Council of South Africa (HPCSA).

Organisational commitment: Subscale Affective commitment

In order to operationalise Affective Commitment, an eight-item subscale of the Organisational Commitment scale, developed by Allen and Meyer (1990), was utilised. Some items were negatively phrased, and others reverse scored to reduce response bias. Higher scores suggest a higher commitment towards the organisation. The subscale includes items such as ‘I would be happy to spend the rest of my career at this organisation’. The authors report strong factorial validity and a reliability coefficient of 0.80 for the total scale.

Perceived corporate social responsibility

Perceived Corporate Social Responsibility was measured with a scale adapted from Maignan and Ferrell (2001). The scale consists of four components, namely the economical, legal, ethical, and discretionary components of CSR that served as indicator variables.

The total Perceived Corporate Social Responsibility scale consists of 29 items. The economic dimension, the legal dimension, and the ethical dimension are assessed by seven items, while the discretionary or philanthropic dimension is evaluated by eight items. The scale includes items like ‘the managers of this organisation try to comply with the law’.

Higher scores on each of the dimensions indicate that employees have more favourable perceptions regarding the organisation’s CSR policies, while low scores on the dimensions indicate that employees have less favourable perceptions regarding the organisation’s stand in terms of the aspects of CSR.

Research evidence indicates that the Cronbach’s Alphas are satisfactory for the legal (0.86), ethical (0.88), philanthropic (0.90), and economic responsibilities (0.87) (Maignan & Ferrell, 2001).

Organisational social identity

Organisational Social Identity was measured by a ‘six-item scale developed by Mael and Ashfort (1992)’. High scores on the scale indicate that the employees identify positively with the organisation. The scale is theoretically strongly founded on Social Identity Theory. The scale includes items such as ‘I feel strong ties with my organisation’. Research on the measure reports satisfactory reliabilities, ranging from 0.80 to 0.89 across different samples (Mael & Ashfort, 1992).

Cultural values

Cultural Values were measured utilising the Cultural Value Scale (CVSCALE), a 26-item scale of Individual Cultural Values, developed by Yoo et al. (2011). The CVSCALE was adapted from Hofstede’s cultural value dimensions with the specific aim of utilising it on the individual level. Only 15 items from the CVSCALE were utilised in the study, five items related to Power Distance, six items related to Collectivism, and four items related to Masculinity. High scores on each scale indicated that the employee identifies strongly with a particular cultural dimension. The resulting scale includes items such as ‘Group success is more important than individual success’. The scale was validated across several countries and the observed Cronbach’s alphas were high with Power Distance between 0.79 and 0.84, Collectivism between 0.72 and 0.78, and for Masculinity an alpha of 0.71 (Yoo et al., 2011).

Perceived organisational justice

In order to operationalise Perceived Organisational Justice, the Ambrose and Schminke (2009) adaptation of the Perceived Overall Justice (POJ) scale was used. According to the authors, Perceived Organisational Justice is based on individual personal justice experiences, as well as the general fairness of the organisation. The scale assesses overall fairness in terms of individual personal justice experiences, as well as the general fairness of the organisation. Higher scores on the scale indicate that employees perceive the organisation to be fair and thus perceive justice as being high. Items include statements such as ‘overall, I am fairly treated by my organisation’. The authors report an extensive validation process and an observed Cronbach Alpha of 0.93.

Psychological contract

The scale used to measure Psychological Contract was originally developed by Millward and Hopkins (1998) and adapted by Raja et al. (2004). Both these forms are often used in research pertaining to the Psychological Contract. The initial 31-item Psychological Contract Scale developed by Millward and Hopkins (1998) was reduced to 18 items in Raja et al.’s (2004) study. This shorter version includes nine items related to the Transactional Contract, and nine items related to Relational Contract. Only the nine items of the Relational Contract were utilised in the current study. The Relational Contract focuses on the interplay between economic and internal emotional factors, and suggests that employees are more willing to work, help co-workers and support organisational change. The subscale includes statements such as ‘I expect to grow in this organisation’ (Lee & Liu, 2009; Rousseau, 2004).

Lower scores on this scale suggest that employees experience a breach of the Psychological Contract, whereas higher scores indicate that employees experience the organisation as fulfilling and meeting their expectations and agreements. Lee and Liu (2009) provided evidence of convergent validity and an Alpha for Relational Contracts of 0.79.

Research procedure and ethical considerations

The researcher obtained organisational approval prior to commencing with the data-collection procedure. The researcher provided the management team of the selected FMCG company with an overview of the study, and an agreement was reached in terms of distribution of the questionnaires to ensure that it did not interfere with the company’s operations. The invitation to participate was distributed by the Human Resource (HR) business partner and participants who elected to participate in the study did so by means of the landing page of university’s survey site. The researcher communicated the purpose of the study in the documentation and the participants were informed that participation in the study was entirely voluntary. Informed consent was explained, and participants were further made aware of the fact that they will remain anonymous and that their responses will be regarded as confidential, as only aggregate data will be made public in the study.

Participants could provide their email addresses in order to receive personal feedback of a summary of the research results and findings None of the participants, however, provided their email addresses in order to receive feedback of the study results, and therefore all participants remained anonymous. An arrangement was made to provide feedback to the participating organisation by presenting the results obtained in the study to the management team.

Ethical considerations

Ethical clearance to conduct this study was obtained from the Stellenbosch University Social, Behavioural and Education Research Ethics Committee (RED: SBE). (No SU-HSD-004868).

Results

Statistical analysis

For the purposes of this study, the imputation method (Statistica) was selected to resolve missing values.

In order to investigate the hypotheses formulated and depicted in the structural model in Figure 1 values, and path coefficients, Structural Equation Modelling (SEM), using the Partial Least Squares (PLS) technique were used. This technique is referred to as PLS SEM (Hair Jr et al., 2017) and is recommended for testing and validating exploratory models and can be used for the prediction of variance in variables in the theoretical model (Hair Jr et al., 2017; Henseler et al., 2009).

The Cronbach’s alpha coefficient was used to evaluate the reliabilities of the subscales. According to Tabachnick and Fidell (2013), Cronbach’s alphas of 0.70 or higher can be regarded as adequate. In terms of evaluating the measurement model, composite reliability, average variance extracted (AVE), discriminant validity (using the Hetrotrait-Monotrait [HTMT] Ratio), and outer loadings were evaluated. For the structural model, possible multicollinearity (by calculating the variance inflation factor [VIF]), R-square values, and path coefficients were evaluated (Hair Jr et al., 2017; Henseler et al., 2009).

Main findings

Reliability analysis was performed on the items for each scale or subscale in the questionnaire. A summary of the psychometric properties of the scales and subscales is provided in Table 1. The number of items, means, standard deviations, standardised Cronbach’s alphas, and average inter-item correlations are provided for all subscales.

The Cronbach alphas for all scales were above 0.70 (ranging from 0.74 to 0.90), which suggests that the scales showed good internal reliability. The average inter-item correlations further showed that the item correlations (0.29 to 0.61) fell in the acceptable ranges, which indicates the degree of consistency between the items, but also that they were not too highly correlated. According to Pope (2009), inter-item correlation values below 0.2 are considered weak, values between 0.2 and 0.39 are considered good, and values exceeding evident 0.4 can be considered very good.

Evaluation and interpretation of the measurement model

In order to determine the reliabilities of each latent variable for the measurement model, the composite reliability and AVE were evaluated. All the composite reliabilities of the latent variables were above 0.70. It can therefore be concluded that the reliabilities of all the latent variables were satisfactory (Bryman & Bell, 2011; Hair Jr et al., 2017).

The AVE values for most of the latent variables were close to or above 0.50, which indicates that these constructs explained more than or approximately 50% of the variance in indicators (Alarcón & Sánchez, 2015; Hair Jr et al., 2017). In the case of Perceived CSR: Economic Citizenship, however, it appears that more variance remains in the error variance (unexplained variance) than the variance explained by the construct (0.40).

Discriminant validity is established when an indicator’s loading on the assigned construct is higher than all of its cross-loadings with other constructs. It refers to the extent to which one variable or construct is truly distinct from others (Hair Jr et al., 2017) or the degree to which measures of different traits are unrelated (Alarcón & Sánchez, 2015). An HTMT ratio value of smaller than 1 shows that the true correlations between the two constructs should differ. In the current study, all the HTMT ratios indicated that all constructs did in fact achieve discriminant validity. The fact that the subscales of Perceived CSR did not achieve discriminant validity when compared to one another is to be expected, but they all achieved discriminant validity when compared to the other constructs. It can therefore be concluded that the constructs meet the required discriminant validity levels.

The evaluation of the outer loadings indicated that most of the outer loadings were found to be statistically significant. In the case of the Collectivism subscale, however, the outer loadings were not statistically significant, although the p-values for the items remained below the 0.05 critical value. One can thus conclude that in terms of the Collectivism subscale, the statistical evidence does not provide a clear indication of the reliability of the outer loadings.

Evaluation and interpretation of the structural model

Multicollinearity, R-square values, as well as the main- and moderating effect path coefficients were evaluated. In order to test for multicollinearity, the VIF coefficients were calculated. The VIF coefficients indicate the correlation between predictors in a regression analysis. A value of 5 normally signifies further investigation, while a value of 10 indicates serious multicollinearity (Hair Jr et al., 2017; Henseler et al., 2009). All the VIF coefficients in the study were within limits.

The R-square value determines the amount of variance in the endogenous variables which can be explained by the remaining variables in the model. According to Hair Jr et al. (2017), R-square values of 0.20 and higher are considered high in behavioural studies, as it indicates the predictive accuracy depending on the complexity of the model. The R-square value for Affective Commitment was 0.66, which signifies that the total model accounts for 66% of the variance observed in Affective Commitment. The R-square value for Organisational Social Identity was 0.40; for Perceived Organisational Justice 0.36, and for Relational Contract it was 0.24. These values are an indication that there are possibly other variables that could have had an impact on the endogenous variables in the current study.

The path coefficients indicated that the paths between Perceived CSR and its four sub-dimensions were found to be significant, which suggests that the subdimensions measured the construct. The main effect hypothesis findings are discussed further.

The hypothesised relationship between Perceived CSR and Affective Commitment (Hypothesis 1) was established as being significant. The results of the study confirm that perceptions regarding CSR have a positive effect on the Affective Commitment of employees in general.

The relationship between Perceived CSR and Organisational Social Identity was established as not significant. However, the relationship between Organisational Social Identity and Affective Commitment was found to be significant. It thus appears that Organisational Social Identity positively influences Affective Commitment, but that Perceived CSR does not affect Organisational Social Identity. These results therefore do not support the mediation hypothesis (Hypothesis 2) and it can be concluded that no mediation took place.

The relationship between PCSR and Perceived Organisational Justice was significant. However, the relationship between Perceived Organisational Justice and Affective Commitment was not significant. It thus appears that Perceived CSR influences perceptions regarding Organisational Justice, but that Perceived Organisational Justice does not have an impact on Affective Commitment. Hence, the results do not support the mediation hypothesis (Hypothesis 6) and it can be concluded that no mediation took place. The results are consistent with previous research regarding the relationship between Perceived CSR and Perceived Organisational Justice but in contrast with previous research that suggested a significant relationship between Perceived Organisational Justice and Affective Commitment.

The relationship between Perceived Organisational Justice and Organisational Social Identity (Hypothesis 7) was found to be significant with a path coefficient of 0.27. The results indicate that positive perceptions regarding Organisational Justice will have a positive impact on Organisational Social Identity.

The relationship between Perceived CSR and Relational Contract was established as significant while the relationship between the Relational Contract and Affective Commitment, however, was not significant. It thus appears that Perceived CSR does influence the Relational Contract, but that Relational Contract does not influence Affective Commitment. The results therefore do not support the mediation hypothesis and it can be concluded that no mediation took place (Hypothesis 8).

The relationship between Relational Contract and Organisational Social Identity was significant (Hypothesis 9) with a path coefficient of 0.40. The results suggest that the Relational Contract will positively affect an individual’s Organisational Social Identity.

The hypothesised relationship between Relational Contract and Perceived Organisational Justice was significant (Hypothesis 10) with a PLS path coefficient of 0.18. The results suggest that the Relational Contract positively influences Perceived Organisational Justice.

In addition to the above, the study evaluated whether Cultural Values (Collectivism, Power Distance, and Masculinity) had a moderating effect on the relationship between Perceived CSR and Affective Commitment. The findings indicated that none of the Cultural Values had any effect on the relationship (Hypotheses 3, 4, 5). It should, however, be borne in mind that these previous studies mainly focused on the cultural value differences between Western and Asian employees, which may thus not translate to the South African environment in a similar manner.

The path coefficients indicating the strength and significance of the hypothesised relationships are indicated in the structural model in Figure 2. The figure includes the main effects, as well as the hypothesised moderating effects.

FIGURE 2: Structural Model with Path Coefficients.

Discussion

Outline of the results

The purpose of the study was to answer the research initiating questions of: ‘What role does employee perceptions of Corporate Social Responsibility play in explaining variance on Affective Commitment in the South African context, and which other latent variables play a role in the dynamics of this relationship’.

The hypothesised relationships between all four direct effect hypotheses were found to be significant. None of the mediation or moderation hypothesis were, however, supported.

The direct relationship between Perceived CSR and Affective Commitment was found to be significant, which shows that positive employee perceptions regarding CSR will have a significant positive effect on Affective Commitment. This finding confirms previous research results, similar to that of Farooq et al. (2014), which determined that employee perceptions regarding CSR are a strong predictor of Affective Organisational Commitment. Both Mueller et al. (2012) and Bouraoui et al. (2019) also found evidence that employee perceptions of CSR have a major impact on Affective Commitment. Even externally focused CSR was positively related to Affective Commitment. Organisational Social Identity, however, did not mediate the relationship between Perceived CSR and Affective Commitment, which contradicts previous findings (De Roeck et al., 2014; Farooq et al., 2014). Organisational Social Identity did have a strong significant impact on Affective Commitment, although Perceived CSR did not have a significant impact on Organisational Social Identity. Contrary to the expectation that the employee would evaluate the organisation in a positive light, the findings indicate that Perceived CSR did not create a social organisational identification. Perceived Organisational Justice and fulfilment of the Psychological Contract did have an impact on the way employees identified with the FMCG company.

Perceived CSR had a significant effect on Perceived Organisational Justice, which suggests that positive employee perceptions regarding CSR will increase the justice and fairness perceptions of employees regarding their organisation, which is consistent with previous research findings (Moon, et al., 2014; Rupp, 2011; Rupp et al., 2013). This finding suggests that not only will positive perceptions of CSR lead to an increase in Affective Commitment, but the perception of the company as being a fair employer will also be enhanced. However, Perceived Organisational Justice did not directly influence Affective Commitment, in contrast to some of the above research findings.

It was furthermore established that the positive Perceived Organisational Justice increases Organisational Social Identity leading to higher Affective Commitment, which is consistent with previous empirical findings (Moon, et al., 2014; Rupp et al., 2013). Taken together these findings provide valuable insights as it suggests that, despite the finding that Perceived CSR does not directly increase Organisational Social Identity, Perceived CSR impacts Justice Perceptions, which in turn enhances Organisational Social Identity. These findings thus demonstrate the complex nomological network of variables and how the different variables can directly and indirectly impact on one another.

Similarly, the findings suggested that positive employee perceptions regarding CSR will lead to fulfilment of the Relational Contract, but the mediation hypothesis was not supported because of the relationship between the Relational Contract and Affective Commitment being non-significant, which is not compatible with the previous research findings in this regard (Lee & Lui, 2009; Rousseau, 2004). As a result of the positive relationship between the Relational Contract and Organisational Social Identity, Affective Commitment will, however, be indirectly influenced.

Managerial implications

The results of the study have significant implications for CSR strategies. The positive relationship between Perceived CSR and Affective Commitment suggests that organisations should ensure that employees are made aware of their CSR policies and activities through corporate communication (Brammer et al., 2007; De Roeck et al., 2014). Furthermore, involvement in CSR activities (from an economic, legal, ethical, and discretionary perspective), which may include being involved with identified areas of need in the community, improving employee welfare or producing quality products that comply with legal requirements, will result in favourable CSR perceptions among employees. Organisations have realised that, although financial success and profit maximisation for shareholders are essential, the organisation also has a responsibility towards the society and community in which it operates (Carroll & Shabana, 2010). Organisations should also invest their resources in dealing with social problems that are regarded as important by employees. The Human Resource Division could furthermore focus on internal CSR elements, such as fair practices regarding recruitment, career, and talent development, meeting employee needs, and industry compatible remuneration (De Roeck et al., 2014; Hofman & Newman, 2013). In this way CSR activities could help retain and attract talented candidates and increase commitment and identification with the organisation (Roudaki & Arslan, 2017). In deciding upon CSR initiatives or strategies, managers should show that they regard employee perceptions as important and as a driving force behind their CSR initiatives (Moon, et al., 2014). In developing internal communications regarding CSR, managers must incorporate the viewpoints of employees as stakeholders and monitor how the employees perceive the organisation in terms of CSR. Top management must model socially responsible behaviour to create a trusting, fair work environment.

The nomological network of variables that was found to be relevant to the relationship between Perceived CSR and Affective Commitment demonstrates the complexity or difficulty of understanding human behaviour. Gond, et al. (2012) claim that positive outcomes such as Organisational Trust, Job Satisfaction, Organisational Citizenship Behaviours, and reducing Workplace Deviance are to be expected as well. Studies on CSR have also linked it to other outcomes, such as investor relations, customer perceptions, and company image or reputation (Pomering & Johnson, 2009; Sankar & Bhattacharya, 2001). Thus, increased communication of CSR policies may not only result in a more committed workforce but yield further positive external results.

Limitations and recommendations

The study is subject to several limitations. Data were obtained from a single organisation in a particular industry (FMCG) in South Africa. The convenience sampling method added to this limitation, as the sample may not be representative of the entire South African workforce and thus generalisation of the results might be problematic. The sample size was satisfactory, but a larger sample, with participants from more organisations and different industries, would have provided more credible results.

Corporate Social Responsibility is measured based on employee perceptions, but it is possible that some individuals may not have had sufficient information about the organisation’s involvement in CSR (Turker, 2009). The ‘do not know’ response allowed participants not to feel pressured by selecting a viewpoint if they did not have knowledge with respect to a specific question. It is also possible that actual CSR policies implemented by the organisation are not effectively communicated by the organisation.

The lower AVE values for Perceived CSR: Economical Citizenship, could be viewed as a limitation or a constraining factor in the model, as a large proportion of variance is left unexplained by the construct. One may thus have to review the scale and items in future research. Some employees may further have struggled to understand the questions, such as questions pertaining to Collectivism or Perceived CSR, where the statistics indicated that there were problems with some measurement items. It is acknowledged that forms of self-report bias may have had an impact on the results of the study. As noted by Donaldson and Grant-Vallone (2002), accurate measurement of organisational behaviour is important when conducting organisational research and most of the research in the field is conducted utilising different forms of self-report questionnaires.

In terms of future research, cultural influences should be further investigated, as several studies have established that cultural differences can have an impact on the results (Hofman & Newman, 2013; Mueller et al., 2012; Thanetsunthorn, 2015) and the inclusion of cultural differences in future research has been suggested in other research articles (Rego et al., 2007; Thanetsunthorn, 2015).

It is further possible that other variables and mechanisms not in the current model, may play a role in the relationship between Perceived CSR and Affective Commitment. In addition, Perceived CSR may be linked to other outcomes or attitudes, such as increased Organisational Trust, or Job Satisfaction (Farooq et al., 2014; Gond, et al., 2009).

Conclusion

From the above, it is evident that Organisational Social Identity is a key variable in the relationship between Perceived CSR and Affective Commitment, although in a different way than initially hypothesised. The findings suggest that neither Perceived Organisational Justice nor the Relational Contract had a positive effect on Affective Commitment, but they influence Affective Commitment through increased Organisational Social Identity. The results of the study demonstrate the complex nomological network of variables in the relationship between Affective Commitment and Perceived Corporate Social Responsibility.

Organisational Social Identity remains a key variable as all other variables in the proposed model had a positive impact on Organisational Social Identity. Positive perceptions regarding CSR will thus indirectly impact Organisational Social Identity, even though the direct relationship was not found to be statistically significant in the present study.

The study contributes to the extant body of knowledge by shedding light on the relationships between the variables impacting on the relationship between Perceived CSR and Affective Commitment within the South African work environment at an FMCG company. Employees’ perceptions of CSR directly and indirectly influence Affective Commitment through their relationships with other variables.

Acknowledgements

The authors thank the participating organisation and the actual participants. Without their cooperation this study would not have been possible.

The author(s) declare that they have no financial or personal relationship(s) that may have inappropriately influenced them in writing this article.

The article is based on the unpublished master’s thesis of the first author under the supervision of the second author and their contribution to the final article is reflected by the sequence in the naming of the authors.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The data that support the findings of this study are available on request from the corresponding author, J.M.

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.

References