The effect of an organizationâ s CSR initiatives on employee outcomes.

(1)

The effect of an organization’s CSR initiatives on employee outcomes

Author: Latoya van de Kerkhof

Student number: 10365559

Date and version of submission: 31 January 2014 – Final version

Qualification: MSc. in Business Studies – Marketing Track Institution: University of Amsterdam (UvA)

First Supervisor: Drs. Ing. A.C.J. Meulemans Second Supervisor: Prof. Dr. J. Tettero

(2)

(3)

Abstract

The past couple of decades the interest and concern of Corporate Social Responsibility (CSR) dramatically increased in society, industries, and with academia. Organizations seek ways to use their CSR activities to generate more favorable attitudes from customers and prospects. Extensive research is done about what the effect is of CSR activities on companies, but little research has focused on the effect on employee outcomes relative to the research that is done about what CSR does towards customers and other external stakeholders. This study investigates the effect of an organization’s CSR initiatives on employee outcomes. The employee outcomes specified in this study are: job satisfaction, job performance, and affective commitment. This study also looks how this effect varies within the stakeholder group in terms of generational differences, industry sensitivity, and communication channel discrepancy. The analysis is based on a sample of 120 employees. The results show that an organization’s CSR activities positively influence the 3 employee outcomes. This has some important implications. No support was found for generational differences, industry sensitivity, and communication channel discrepancy. However, communication channel discrepancy certainly does play a role in terms of an organization’s CSR activities and job performance. These implications are elaborated on in the discussion section. This study goes beyond mere investigating whether an organization’s CSR activities in general have an impact on employee outcomes. Next to the main research question and its belonging hypotheses, in a small secondary research is examined whether the three CSR typologies each have a distinct impact on the employee outcomes. This may be a stepping stone for other researchers to build upon these first notions in terms of this topic.

Key words: corporate social responsibility, job satisfaction, job performance, affective commitment, generation, industry sensitivity, communication discrepancy, CSR typology

(6)

1. Introduction

The past couple of decades the interest and concern of Corporate Social Responsibility (CSR) dramatically increased in society, industries, and with academia. We live in an era in which the climate is changing, where business is having an impact on the environment, resources are getting depleted if we do not handle them more carefully, and nuclear reactors are having a tremendous impact on the environment.

There have been a lot of attempts and a lot of debate to form a correct definition of CSR. Bhattacharya & Sen (2004) broadly defined CSR as ‘’a company’s status and activities with respect to its perceived societal or, at least, stakeholder obligations (beyond legal obligations)’’. So, business must satisfy multiple stakeholders (Freeman, 1984).

Organizations seek ways to use their CSR activities to generate more favorable attitudes from customers and prospects. Extensive research is done about what the effect is of CSR activities on companies. The main focus of previous research was towards external stakeholders (Yoon et al., 2006; Du et al., 2010; Hildebrand et al., 2011; Lii et al., 2013). Nevertheless, a very important stakeholder group is the organization’s employees, because they are the communicators and the representatives of an organization’s identity and message. Or as Joyner & Payne (2002) put it: ‘’values, ethics, and CSR are intertwined in the culture and management of a firm’’ (Joyner & Payne, 2002). Employees can influence the (success of the) firm (Lee et al., 2012). When an organization has a reputation of unsatisfied employees and high turn-over rates this will not do the organization any good (Peterson, 2004).

This implies that the credibility of an organization’s CSR activities would be endangered in case of such reputation. Due to the idea that CSR, among others, is intertwined in the culture and management of a firm, it is even more interesting to look at it the other way around. Because of the fact that employees are the communicators and representatives of an organization’s identity and

(7)

message, it is fascinating to know what the effect is of and organization’s CSR activities on employee outcomes (performance, commitment, and satisfaction).

However, little research has focused on the effect on employee outcomes compared to the research that is done about what CSR does towards customers and other external stakeholders. Due to this, little is known about the effects of CSR initiatives on employee outcomes in the field of organizational behavior. Three important concepts of organizational behavior are job performance, organizational commitment, and job satisfaction, taken together as ‘employee outcomes’, which will be explained in the upcoming sections.

Moreover, this study does not only look at the effect an organization’s CSR has on employees, but also looks to how this effect varies within the stakeholder group in terms of generational differences. Findings of a study of Gursoy et al. (2008) indicate significant generational differences in world views, attitudes towards authority and work perspectives. However, other empirical studies challenged this claim that generational differences actually exist. So more research on generational differences at work appears to be needed (Kowske et al., 2010).

1.1 Contribution

The aim of this study is to make several contributions to the existing knowledge concerning the relationship between CSR and organizational behavior. First, this is one of the few studies that pulls apart the CSR initiatives of an organization and looks at the effect these different initiatives have on employee outcomes. In so doing, this study adds to the knowledge of the impact of CSR on internal stakeholders. As mentioned before, previous research mainly focused on what CSR does towards customers and other stakeholders, however, few studies have investigated the effect of CSR initiatives on employee outcomes while employees ensure organizations’ long-term survival and performance (Turker, 2009).

Second, this study contributes to the existing literature, because Roeck et al. (2013) state that research that does explore the impact of CSR on employees’ attitudes is limited because the

(8)

underlying psychological mechanisms that drive and ensure employees’ favorable responses to CSR initiatives still remain unclear. This study aims to address these issues to gain a better understanding of the positive influences of CSR on employees.

Third, this study contributes to the literature on generational differences at work, because this is one of the few that examines generational differences in work attitudes linked to the CSR initiatives of an organization.

Moreover, by pulling apart the CSR construct, this study creates a start or a stepping stone for other researchers who are interested in CSR and its effects on stakeholders. Also, this study looks to potential discrepancies in terms of internal CSR communication, which provides this study with essential implications.

The subsequent section will theoretically provide information about the CSR typology, employee outcomes, generational differences, and industrial and communicational influences. These subtopics will help in addressing the main research question. To answer these questions, an evaluation and integration of the existing literature is given, followed by a conceptual framework and the belonging hypotheses. After this, the research method and results will be presented. This research concludes with a discussion, implications of the findings, limitations and foundations for further research.

2. Literature Review

2.1 CSR Typology

There are many different ways an organization can execute their CSR initiatives. There is such an overload, that organizations sometimes do not know which ‘CSR-road’ to follow, which results in that these efforts are not as productive as they could be. According to Porter & Cramer (2006), companies are pressured to think of CSR in generic ways instead of in the way most appropriate to

(9)

each firm’s strategy. ‘’Strategy is always about making choices, and success in CSR is no different, it is about choosing which social issue to focus on’’ (Porter & Kramer, 2006).

This implies that, for CSR to be a source of opportunity and competitive advantage, it is important that each organization decides for itself which CSR initiative works best for them. Furthermore, for a CSR initiative to be sincere and successful, the internal stakeholders (employees) must carry out these activities and believe in them. However, not every CSR activity has the same impact on the organization’s employees. Different activities may yield different outcomes.

It therefore seems interesting to not only look at CSR initiatives as a whole, but also pull apart the concept of CSR, and look at which of the distinct activities has (more of) an impact on the employee outcomes. This classification of CSR activities is not completely new. Carroll (1991), perhaps the most well-known, presents a four-stage pyramid classification, including economic, legal, ethical and philanthropic responsibilities. However, Halme and Laurila (2009) argue that previous literature has neglected to address the varying outcomes that result from each type of CSR activity. A possible reason for this is ‘’that the typologies most often encountered in the corporate responsibility literature do not easily lend themselves to empirically observable linkages with business outcomes’’ (Halme & Laurila, 2008), let alone employee outcomes.

Therefore, this study will not only focus on CSR in general, but also on the categorization of Kourula and Halme (2008). They apply a new, more, pragmatic categorization of 3 CR types: philanthropy, CR integration and CR innovation. Philanthropy focuses on ‘’charitable actions and using corporate resources for doing good’’ (Kourula & Halme, 2008). This includes donations, sponsorships, and employee voluntarism. These are extra activities and not part of the core business.

On the contrary, organizations concentrating on CR integration combine responsibility aspects into their core business operations. This means ensuring ‘’high product quality and investments to R&D (responsibility to customers), paying just wages (responsibility to employees), paying in time to suppliers, supplier training programs, no child labor in the supply chain

(10)

(responsibility to suppliers), and applying environmentally benign practices and policies (responsibility toward the local community)’’ (Kourula & Halme, 2008).

The third type of CSR is CR innovation, where CR is seen as a source of business innovations. An organization ‘’takes an environmental or social problem as a source of business innovation and seek to develop new products or services, which provide a solution to the problem (win-win situation)’’ (Kourula & Halme, 2008).

By the use of this CSR typology in this study is aimed to make a start for further research, for both theoretical and practical contributions.

2.2 Employee outcomes

As mentioned in the introduction section, relatively little is known about the effects of CSR initiatives on employee outcomes. There is some prior literature that shows that CSR initiatives do influence employee attitudes and behavior (Sen et al., 2006; Brammer et al., 2007; Bhattacharya et al., 2008; Turker, 2009; Gond et al., 2010; Mueller et al., 2012; Lee et al., 2013). The three employee outcomes discussed in this paper (job satisfaction, job performance, and organizational commitment) are much examined topics (Allen & Meyer, 1990; Tett & Meyer, 1993; Podsakoff et al., 2000; Judge et al., 2001; Meyer et al., 2002). Notable, however, previous research has focused more on the three outcomes in relation to one another (Locke, 1970; Shore & Martin, 1989; Tett & Meyer, 1993; Judge et al., 2001; Sutanto, 2004), which brought about inconclusive results.

For instance, commitment to organizations is proven to be positively related to job satisfaction (Mowday et al., 1982; Bateman & Strasser, 1984). However the relationship between organizational commitment and job performance is more vague (Becker et al., 1996). Mathieu & Zajac (1990) conclude that ‘’commitment has relatively little direct influence on performance in most instances’’. Mowday et al. (1982) support this conclusion. Conversely, Sutanto (2004) concludes that the statement ‘’commitment is largely unrelated to performance’ can be dropped based on new

(11)

‘organizational commitment’, which can cause confusion and bias. Also, the results of several models of the job satisfaction-job performance relationship show inconsistencies (Judge et al., 2001). This implies that it is interesting and efficient to pull these employee outcomes apart, so that there is a clear vision on which part of an employee’s behavior is affected.

2.2.1 Job Satisfaction

The definition of job satisfaction used in this paper stems back from professor Edwin A. Locke (1969), the most published organizational psychologist in the history of the field. According to the Association for Psychological Science, ‘’his pioneering research has advanced and enriched our understanding of work motivation and job satisfaction’’ (Association for Psychological Science, 2005). According to Locke (1969), job satisfaction is the ‘’pleasurable emotional state resulting from the appraisal of one's job as achieving or facilitating the achievement of one's job values’’ (Locke, 1969). In turn, job dissatisfaction is the ‘’unpleasurable emotional state resulting from the appraisal of one's job as frustrating or blocking the attainment of one's job values or as entailing disvalues’’.

In other words, job satisfaction (dissatisfaction) describes a positive (negative) feeling about a job, resulting from an evaluation of its characteristics (Robbins et al., 2001). This implies that when an employee sees a CSR activity as positive, this would also positively influence job satisfaction. De Roeck et al. (2013) follow this statement. Their results suggest that employees appear to use CSR initiatives to assess their organization’s character and identify with it. CSR initiatives can be used to create strong relationships with employees and improve attitudes at work (De Roeck et al., 2013).

According to Robbins et al. (2001) organizations have the responsibility to provide employees with jobs that are challenging and intrinsically rewarding. The basic of CSR is that companies integrate social and environmental concerns in their business operations on a voluntary basis (Dahlsrud, 2008). For employees this can be seen as doing something good for society, and thus makes him/her feel good because s/he works for this organization (rewarding). In line with this, Koh & El'fred’s (2001) findings suggest that employees experience greater job satisfaction when they

(12)

consider the organization where they work for as ethical. Valentine & Fleischman (2008) define organizational ethics as ‘’a company’s adoption of desired ethical standards and business practices’’. Moreover, they claim that a natural extension of organizational ethics is an organization’s involvement in CSR (Valentine & Fleischman, 2008). So, this implies that an organization that is ethical by engaging in CSR creates greater job satisfaction with its employees.

In according with these findings, it is argued that job satisfaction can be positively influenced by CSR initiatives, which leads to the following hypothesis:

Hypothesis 1: An organization’s CSR initiatives will have a positive impact on job satisfaction.

However, not all CSR initiatives will elicit (the same kind of) positive feelings. As mentioned in the introduction section, different activities may yield different outcomes. It is, therefore, interesting to look at which CSR initiative elicits the most positive feelings in comparison with the other two. This will be elaborated on in the results section.

2.2.2 Job performance

Based on prior literature, Podsakoff et al. (2000) formulate a working definition of job performance and can be defined as ‘’employee behavior, which consist out of the attitude, skills and knowledge of an individual, cooperating with others, protecting the organization, volunteering constructive ideas, self-training, and maintaining a favorable attitude toward the company’’ (Podsakoff et al., 2000). In short, job performance is about how well an employee performs his or her job. It seems apparent that employees’ attitudes would be affected by aspects of a firm’s performance. According to Carmeli et al. (2007), perceived social responsibility is also a form of organizational performance. This study finds that perceived social responsibility and development had an effect on organizational identification (Carmeli et al., 2007). According to Peterson (2004), when employees proudly identify themselves with an esteemed organization, in this case a socially

(13)

responsible organization, this will positively influence their work attitudes, and thus job performance.

Also, meta-analytic evidence shows a relationship between perceived justice and job performance (Colquitt et al., 2001), and according to Rupp et al. (2006), there is ‘’no evidence to suggest that similar outcomes would not result from CSR perceptions’’ (Rupp et al., 2006). Corporate Social Responsibility can also be perceived as a dimension of justice and ethics (Valentine & Fleischman, 2008), which leads to the following hypothesis.

Hypothesis 2: An organization’s CSR initiatives will have a positive impact on job performance.

Again, in the results section will be looked at the (possible) different effects the distinct CSR initiatives have on job performance.

2.2.3 Organizational commitment

According to Allen & Meyer (1996), organizational commitment can be generally defined as ‘’a psychological link between the employee and his or her organization that makes it less likely that the employee will voluntary leave the organization’’. Organizational commitment can be separated into 3 separate dimensions (Allen & Meyer, 1996; Meyer et al., 2002; Robbins et al., 2001). Affective commitment is an ‘’emotional attachment to the organization and a belief in its values’’. Continuance commitment is the ‘’perceived economic value of remaining with an organization compared to leaving it’’. And, finally, normative commitment is an ‘’obligation to remain with the organization for moral or ethical reasons’’ (Allen & Meyer, 1996; Robbins et al., 2001). Although it would be interesting to compare different types of organizational commitment in the context of CSR initiatives, this study is limited to affective commitment. The main reasons are the relative scarcity of studies involving the other forms (Tett & Meyer, 1993), and weak results for continuance commitment in previous research, which implies that continuance commitment isn’t a strong commitment at all (Robbins et al., 2001).

(14)

Studies based on Social Identity Theory claim that employees will be proud to identify with organizations that have a favorable reputation (Peterson, 2004). According to Social Identity Theory people define themselves partly in terms of salient group memberships. Identification is the perception of oneness with or belongingness to a group, involving direct or vicarious experience of its successes and failures. Identification is associated with groups that are distinctive, and prestigious (Ashforth & Mael, 1989). As previously mentioned, according to Peterson (2004), when employees proudly identify themselves with an esteemed organization, this will positively influence their work attitudes. In line with this reasoning, if an employee identifies with and is proud to be a member of an organization, s/he is emotionally attached, which in turn is affective commitment.

Furthermore, previous studies show that socially responsible corporations enjoy certain benefits, like positive employee responses (S. Brammer et al., 2007). This implies that an employee will identify with a socially responsible organization, and therefore is emotionally attached, and thus will improve affective commitment. This leads to the following hypothesis.

Hypothesis 3: An organization’s CSR initiatives will have a positive impact on affective commitment.

Here also, in the results section will be looked at the (possible) different effects the distinct CSR initiatives have on affective commitment.

2.3 Generational influences on the effect of CSR

Due to the Internet and other technological developments we have become more aware of social and environmental problems (Turker, 2009). The idea exists that organizations should ‘give something back’ to the environment. We are taking much more from nature than we give back. When reading a newspaper or watching television, ‘all’ we see is that governments, organizations, and consumers/individuals are engaging in socially responsible activities. Being socially responsible is ‘hot’. It is all about sustainability, the environment, recycling, and charity.

(15)

Young adults are raised with the idea that the environment is deteriorating and that resources are getting depleted (Scott, 2000). Scott builds on the contention of Moscovici (1988) that ‘’generations may have distinctive social representations and that what is unfamiliar and unperceived in one generation becomes familiar and obvious in the next’’ (Moscovici, 1988). Also, she finds that environmental concerns are far more common among younger than older age-groups due to the technological changes, which caused that these issues are more mentioned in the media. It is impossible for them to ignore the environmental issues these days (Scott, 2000).

So, this implies that young adults have been and are faced more with environmental and social problems than ‘older’ generations. The younger generation seems more globally concerned, wants to do work that has meaning, and is more environmentally conscious (Lancaster & Stillman, 2002).

Society, governments, activists, and the media are holding companies accountable for the social consequences of their actions, and therefore, a lot of organizations have made CSR a priority (Porter & Kramer, 2006). This means, among others, changes in production, technology, and strategy. However, Robbins & Judge (2001) state that older workers are perceived as lacking flexibility and as being resistant to new technology. This would have a serious impact, for an organization, on the ability to successfully implement a CSR-strategy.

A generational cohort is a group of individuals that fall into the same range of age, and experienced the same historical events within the same time period (Ryder, 1965). The differences between these age groups are becoming a social problem (Bengtson, 1970). According to Coomes (2004), each generation has its own history, developments, and popular culture. And, each on its own, seem to play an important role in shaping the values, beliefs, attitudes, and worldview of individuals and groups. Because of this ‘generation gap’, these different ages, values, beliefs, attitudes, and worldviews also seem to exist within one organization. However, results of previous

(16)

research have been both complementary and contradictory in terms of generational differences and relationships in personality and work attitudes (Kowske et al., 2010).

2.3.1 Generation taxonomy

Before proceeding, first the generational taxonomy used in this study and derived from Howe & Strauss (2009) is explained. They used historical data to define U.S. generations. 3 age categories will be used in this study. The first one is the age group with people between 20 and 34 years old, the so-called Generation Y or Millennials. The next two generations that will be used in this research are the Generation X generation (age category 35-49) and the Baby Boomers (between the age of 50 and 64). They represent a large portion of the current workforce (Kowske et al., 2010). It is possible that some age categories deviate a little from the standards, but for this study is also been taken into account the retirement age, the age of the largest part of the workforce, and to make sure the age categories are equally divided.

The Millennial generation is the newest to enter the workforce (Kowske et al., 2010). Howe and Strauss (2009) describe them as people who do not ‘operate’ merely for themselves, but also for the future of society and the world. They believe in collective action, with optimism for the future (Gursoy et al., 2008). Results from Gursoy et al. (2008) show that Boomers are very resistant to change because they are comfortable with the way they have always done things. Gen X-ers seem to be more self-centered, they work to live their own lives (Gursoy et al., 2008). This implies that they care less about others and society than the Millennials.

All this, together with the finding of Turker (2009) that the relationship between CSR and organizational commitment increases when ‘’the beliefs of an employee to the importance of CSR increases’’ (Turker, 2009), leads to the notion that that the young generation believe highly in the importance of businesses being socially responsible, more than the ‘older’ generations. So, in line with this information, the following hypotheses say:

(17)

Hypothesis 4a: The effect of CSR initiatives on job satisfaction will be stronger for the young generation.

Hypothesis 4b: The effect of CSR initiatives on job performance will be stronger for the young generation.

Hypothesis 4c: The effect of CSR initiatives on affective commitment will be stronger for the young generation.

2.4 Industrial influences on the effect of CSR

Organizations are influenced by many internal and external factors, and so are the employees. The industry in which an organization operates is one of those factors. Prior work of Bhattacharya et al. (2008) claim that stakeholders’ responses to CSR are generally favorable, yet highly dependent upon the line of industry. Brammer & Millington (2004) seem to confirm this because the results of their study provide evidence of a relationship between type of industry and CSR salience. Firms in socially and environmentally sensitive industries were found to have higher levels of relative and absolute expenditure (Brammer & Millington, 2004). This seems to suggest that stakeholders in such sensitive industries are more aware of the damage organizations cause in terms of the environment, and consequently hold these organizations accountable for these issues. So, there seems to be a relationship between these stakeholder pressures and corporate charitable contributions (Brammer & Millington, 2004). Moreover, Brammer and Pavelin (2006) find that the reputational effect of social performance varies across sectors. For the same reason as abovementioned, because of the presence of industry-specific stakeholder pressures for improved social responsiveness from primary stakeholders such as employees, customers, communities, and legal and regulatory bodies (Brammer & Pavelin, 2006; Chiu & Sharfman, 2011).

When all this is reflected to the current study, CSR salience seem to depend on the industry in which the organization operates, because the line of industry influences stakeholders’ awareness

(18)

and pressure. And thus, this seems to have an impact on whether CSR initiatives have an effect on employee outcomes. This leads to the following hypotheses.

Hypothesis 5a: The effect of CSR initiatives on job satisfaction will be stronger for higher sensitive industries.

Hypothesis 5b: The effect of CSR initiatives on job performance will be stronger for higher sensitive industries.

Hypothesis 5C: The effect of CSR initiatives on affective commitment will be stronger for higher sensitive industries.

2.4.1 High vs. low sensitive industries

In this study, two types of industries will be taken into account: high-sensitive and low-sensitive industries. Firms which operate in industries with potentially significant environmental and social consequences are seen has high-sensitive (Brammer & Millington, 2004). Construction, industry, and (non-food/food) retail are sectors with high potential in environmental and social consequences, and therefore fall under high-sensitive industries. In this study, the lower-sensitive industries are: healthcare, education, F&B, and services.

2.5 Communicational influences on the effect of CSR

As human beings from a very young age, we learn to communicate. The significant basic of a theory of communication comes from Shannon and Weaver (1949), and, very simplified, is about sending and receiving a message (Shannon & Weaver, 1949).

The sender must choose whether and how to communicate its information and the receiver must choose how to interpret the information. Naturally, the source by which the message is communicated is very important since it can influence the receiver’s interpretation (Shannon & Weaver, 1949; Pornpitakpan, 2004; Connelly et al., 2011).

(19)

However, in communication should not only be taken into account the message source, but also the interaction with channel variables. In terms of organizations, business returns to CSR are contingent on stakeholders’ (e.g. employees) awareness of a company’s CSR activities (Du et al., 2010). Stakeholders seem to do react positively to an organization’s CSR activities, but a large portion of the employees are not aware of the company’s CSR activities (Sen et al. 2006), even though this is of such high importance. Namely, as Dawkins (2005) put it, internal communication can have a powerful impact on employee advocacy of the company.

So, it is essential that employees are aware of a company’s CSR activities, and the only way they receive this information is by receiving this information through the right communication channel. The communication channel they prefer the most. Companies must bring employees closer to their CSR activities, and this is done through communication; companies need to inform employees about their CSR programs in a concrete, coherent and consistent manner (Bhattacharya et al., 2008). To correctly communicate the right message through the right channel to employees is not only important because it can influence the employees itself. Also, employees have the power and reach to become the (CSR) communicators (Dawkins, 2005). Employees have a wide reach among other stakeholder groups through their social ties, and are often considered credible source (Du et al., 2010). This emphasizes again the importance of choosing the right channel to reach the right employee group.

There are a variety of communication channels by which information about a company’s CSR activities can be distributed. Not only concerning external stakeholders, but also concerning internal stakeholders (i.e. employees). Especially in the past decades, where highly quality technological developments took place, the number of different communication channels have increased. With this increase, there is a possibility that discrepancies exist between the way employees receive organizational information in terms of CSR and the way they most prefer to receive CSR information. If these discrepancies do exist, this would mean that employee awareness of an organization’s CSR

(20)

activities is low, which could influence employee outcomes. All this leads to the following hypotheses.

Hypothesis 6a: The effect of CSR initiatives on job satisfaction will be stronger when there is no communication channel discrepancy.

Hypothesis 6b: The effect of CSR initiatives on job performance will be stronger when there is no communication channel discrepancy.

Hypothesis 6c: The effect of CSR initiatives on affective commitment will be stronger when there is no communication channel discrepancy.

2.6 Additional analysis

This study goes beyond mere investigating whether an organization’s CSR activities in general have an impact on employee outcomes. Next to the main research question and its belonging hypotheses, in a small secondary research is examined whether the three CSR typologies each have a distinct impact on the employee outcomes. This may be a stepping stone for other researchers to build upon these first notions in terms of this topic. The belonging hypothesis is as follows.

Hypothesis 7: There will be differences among the 3 CSR typologies in the effect they

have on employee outcomes.

2.7 Conceptual Model

The previous literature review, with theoretical framework and hypotheses, leads to a conceptual model presented in figure 1. The conceptual framework gives a clear overview, a summary, of the variables and hypotheses. The reason for not showing the last hypothesis (H7) in the conceptual model is, because this is a secondary research, still in its infancy. It is a way to open up new possible fields of research.

(21)

Figure 1. Conceptual Model

3. Research Method

This section will describe the research design, procedure, participants, and the measures. Into detail will be explained how the independent, dependent, and moderator variables are operationalized.

3.1 Research design

The hypotheses are tested by means of a quantitative research method with a deductive approach. Quantitative research methods ‘’study social phenomena by primarily aiming to develop a mathematical or statistical model between measurable constructs based on numerical data from, for example, surveys’’ (Huysmans & De Bruyn, 2013). Consequently, a standardized questionnaire is conducted and sent out online. This, by means of direct mailing, using the snowball method, and by publishing the survey on Social Media Platforms.

Deduction is a research approach where a theoretical proposition is tested by a certain research strategy. The key characteristics of deduction are, first, to explain causal relationships between variables. Other characteristics are: defining (from the general theory that exists) and operationalizing research questions, the collection and analysis of data to answer the research

(22)

question, and at last, the use of a clearly structured methodology to facilitate replication. This is important to achieve reliability (Saunders & Lewis, 2012).

Online survey research is very useful because it is possible to collect a large amount of data due to its increased reach (Saunders & Lewis, 2012). Moreover, costs in terms of time and money are lower compared to conventional surveying methods (Schmidt, 1997). Also, the ease of data entry and analysis because of the possibility to import data into SPSS is a major advantage (Evans & Mathur, 2005). Some disadvantages are privacy issues, limited computer literacy of respondents, and spam filters (Evans & Mathur, 2005; Saunders & Lewis, 2012). The survey was developed in Qualtrics.

3.2 Procedure and participants

The survey was distributed online, in the third week of November 2013, through a direct mail towards the participants including a link that lead to the survey. The respondents were asked kindly to forward the survey (snowball effect). Also, the survey was put online on Facebook and LinkedIn. The questionnaire was conducted in such a way that it was made impossible to go back or to proceed with the questions if a question was left open. This to decrease the chance of missing data.

The participants were previously informed by means of an introduction letter, including the assurance that the data would be handled uttermost discretely. This introduction letter told the participants that the questionnaire was part of a general employee study. Nothing was mentioned about the questionnaire being a way to measure the effect of different CSR activities on employees to avoid any bias. Questions that needed more detailed information, started with an explanatory paragraph in advance. The survey contained questions about demographics (control variables), and predetermined questions to measure the variables. The participants are Dutch, so both the introduction letter and the survey itself were presented in Dutch.

In total, 120 respondents completely filled out the questionnaire. Only individuals who fitted the description of ‘working people between the age of 20 to 65’ were able to fill out the survey.

(23)

Table 1. Demographic statistics of the respondents

Before sending out the survey, a small pilot test was done with 5 people to polish the survey before sending it to the ‘large crowd’. This to avoid uncertainties and errors. This is very important because ‘’questionnaire pretesting identifies questions that respondents have difficulty understanding or interpret differently than the researcher intended’’ (Krosnick, 1999).

3.3 Description of measures

3.3.1 CSR activities

The purpose of this study is to identify the effect of CSR activities on employee outcomes. To measure the effect of CSR activities, items from the modified Perceived Role of Ethics and Social Responsibility (PRESOR) scale (Singhapakdi et al., 1996; Etheredge, 1999) and the four-dimensional approach of Turker (2009) are utilized. Apart from Turker (2009), measures of CSR are rare in prior

(24)

research (De Roeck et al., 2013). The scale has four factorial subscales comprised of 17 items and measures CSR to: social and non-social stakeholders, employees, customers, and government (Turker, 2009). To operationalize these different types, Turker’s four-dimensional scale was slightly modified in this study. Each of the measures of CSR is based on employee perceptions of social performance rather than the objective measurement of CSR policy or commitment (Brammer et al., 2007). This is because employee outcomes partly reflect employee beliefs. Table 2 presents this scale. The responses are obtained on a 5-point Likert-type scale from strongly disagree to strongly agree.

Table 2. CSR scale items

Number Items

1 Being socially responsible is the most important thing our firm can do 6 Social responsibility of our firm is essential to its long-term profitability

7 The overall effectiveness of our business can be determined to a great extent by the degree to which it is socially responsible

10 Social responsibility is critical to the survival of our business enterprise 12 Business has a social responsibility beyond making profit

1 I find it important that our company engages in extra activities concerning society and environment that are not part of the core business

7 I find it important that our company encourages its employees to participate to the voluntary activities

- I find it important that our company combines responsibility aspects into their core

business operations

6 I find it important that our company contributes to the campaigns and projects that promote the well-being of the environment and society

8 I find it important that our company policies encourage employees to develop their

skills and careers

13 I find it important that our company protects consumer rights beyond the legal

requirements

- I find it important that CSR is seen as a source of business innovations

2 I find it important that our company develops new products or services to solve social and environmental problems

(25)

To find out whether CSR is really seen as important, even in terms of money, a ‘control-question’ was asked concerning the year-end bonus in comparison to CSR activities of an organization. Respondents were able to answer with ‘yes’ or ‘no’. This section in the questionnaire ends with a ranking-question to find out which of the 3 types of CSR activities was seen as most important. As mentioned in the introduction section, the categorization of Kourula and Halme (2008) is used: philanthropy, CR integration, and CR innovation. The data that results from this question is used to assess the differences between the 3 CSR typologies in the effect they have on employee outcomes.

3.3.2 Job Satisfaction

Job satisfaction is a very important variable in organizational studies. Job satisfaction and its different facets are seen as factors affecting the feelings and attitudes, and in turn, the behavior of employees (Smith, 1969). One of its measures, the ‘Job Descriptive Index’ is used very frequently (Kinicki et al., 2002). Therefore, also in this study.

The JDI is developed by Smith et al. (1969), and it measures satisfaction with 5 areas of a job. So, the index contains of 5 scales: the supervision, the co-workers, type of work, the pay, and the opportunities for promotion (Smith et al., 1969). The construct validity of the JDI is described and supported by Kinicky et al. (2002).

All items are short words or phrases, and respondents can respond with Y, N, and ? (yes, no, and ?), depending whether each word or phrase applies with respect to the particular facet of his job in question (Smith et al., 1969). For a complete description of the development of the JDI check Smith et al. (1969).

It is a different way of scaling response categories, but format research has shown that reliability, stability, and validity of the 5 subscales were not significantly different across yes-no-? scaling and Likert-type scaling (Johnson et al., 1982; Kinicki et al., 2002). Even though the results of Johnson et al. (1982) support continued use of the JDI in its original format, for the sake of this study,

(26)

to be consistent throughout the questionnaire, the responses are obtained on a 5-point Likert-type scale from strongly agree to strongly disagree (instead of the yes-no-? scale) (Johnson et al., 1982).

However, previous results indicate that a single-item facet measure was significantly correlated with each of the JDI facets (Nagy, 2002). This implies that it is not needed to have 72 items (distributed among the 5 subscales), as what the JDI contains. Moreover, single-item measures may be easier, less expensive, and may take less time to complete (Nagy, 2002). Therefore, a single-item approach, derived from the JDI, was used. Table 3 presents the items. Again, the responses are obtained on a 5-point Likert-type scale from strongly disagree to strongly agree.

Table 3: Job satisfaction scale items Items

- I am satisfied with my supervisor - I am satisfied with my co-workers

- I am satisfied with the type of work I have to do - I am satisfied with how much I am getting paid

- I am satisfied with the opportunities I get in terms of promotion - Overall, I am satisfied with my job

Source: Smith et al. (1969)

3.3.3 Job Performance

To measure job performance, the items of Williams & Anderson’s (1991) scale were used. In this performance measure, 3 classes of performance were included: performance of organizational citizenship behaviors that have a specific individual as the target (OCBI), those that focus on primarily benefiting the organization (OCBO), and employee performance of in-role behavior (IRB) (Williams & Anderson, 1991). Initially, the scale consists of 21 items (7 items per class), however, in the context of this study, per class, 3 items will be included. Table 4 shows the items that were presented to the respondents. The responses are obtained on a 5-point Likert-type scale from strongly disagree to strongly agree.

(27)

Table 4: Job performance scale items

Number Items

1 I adequately complete assigned duties

2 I fulfill responsibilities specified in the job description 3 I perform tasks that are expected of me

9 I help others who have heavy workloads

10 I assist my supervisor with his/her work (when not asked) 14 I pass along information to co-workers

15 I attend work above the norm

20 I conserve and protect organizational property 21 I adhere to informal rules devised to maintain order Source: Williams & Anderson (1991)

3.3.4 Affective commitment

As previously explained in the literature review, this research is limited to measuring affective commitment because of relative scarcity of studies involving the other forms and weak results. Previous results from Allen & Meyer (1996) indicate that there appears to be considerable evidence regarding the construct validity of the three measures (normative, affective, continuance) (Allen & Meyer, 1996). So, they are appropriate measures of organizational commitment. Therefore, the affective commitment component is going to be used. Allen & Meyer (1990) selected 8 items in the Affective Commitment Scale (ACS). Reliability of this scale was ensured (Allen & Meyer, 1990). Table 5 presents this scale. The responses are obtained on a 5-point Likert-type scale from strongly disagree to strongly agree.

Table 5: Affective commitment scale items

Number Items

1 I would be very happy to spend the rest of my career with this organization 2 I enjoy discussing my organization with people outside it

3 I really feel as if this organization's problems are my own

4 I think that I could easily become as attached to another organization as I am to this

one (R)

5 I do not feel like 'part of the family' at my organization (R) 6 I do not feel 'emotionally attached' to this organization (R) 7 This organization has a great deal of personal meaning for me 8 I do not feel a strong sense of belonging to my organization (R) Source: Allen & Meyer (1990)

(28)

3.3.5 Age (generation)

In this study, I am interested in the moderating effect of a certain generation on the CSR-employee outcome relationship. Therefore, age will not be asked for on a continuous basis, but on a categorical one (20-34, 35-49, 50-64). The categorical variable Age is recoded into a dichotomous variable. The number 0 is assigned to the age category 20-34, and the number 1 is assigned to age categories 35-49 and 50-64. This is done because in this study it is argued that they younger generation is influenced most by an organization’s CSR activities.

3.3.6 Industry sensitivity

As mentioned before, the sectors organizations operate in can be divided into high sensitive and low sensitive industries. To work with a categorical variable Industry, the variable is recoded into a dichotomous variable. The number 0 is assigned to low(er)-sensitive industries, and the number 1 is assigned to high(er)-sensitive industries.

3.3.7 Communication channel discrepancy

To find out whether there is a discrepancy between the way employees receive information in terms of CSR and the way they actually prefer receiving this kind of information (and whether this has an effect on employee outcomes), the respondents had to answer these questions by selecting all possible communication channels they received information from and preferred receiving information from. With both questions they could give multiple responses. A new variable was computed in SPSS, the so-called Discrepancy variable. When there was an overlap in communication channels the number 0 was appointed. If there was no overlap at all, then apparently there is a discrepancy, and the number 1 was appointed.

3.3.8 Demographics and control variables

There are several variables that can influence the outcome of the study (Turker, 2009; Mueller et al., 2012). Therefore, the questionnaire started with a demographic section which included control questions about gender, age (moderator), educational background, income, tenure, and the sector (moderator) in which the respondents were currently working in.

(29)

3.4 Strategy of Analysis

This section will describe all the preliminary work that has been done with the existing data before testing the hypotheses.

3.4.1 Preparing the data and data entry

The online survey tool ‘Qualtrics’ was used to develop the questionnaire. Online survey research is very useful because it is possible to collect a large amount of data due to its increased reach (Saunders & Lewis, 2012). Moreover, costs in terms of time and money are lower compared to conventional surveying methods (Schmidt, 1997). Also, the ease of data entry and analysis because of the possibility to import data into SPSS is a major advantage (Evans & Mathur, 2005). By means of a step-by-step tutorial brought by Qualtrics, it was very easy to import the data into SPSS.

3.4.2 Data screening and cleaning

The imported data from Qualtrics first needed to be explored and screened for any possible discrepancies and/or errors. Due to the fact that the data was imported from Qualtrics, I have scanned the data in the ‘Variable View’-tab whether ‘type’ (e.g. numeric or string), ‘label’, and ‘measure’ (nominal or ordinal) was implemented correctly. If not, I corrected it.

When scanning the data in the ‘Data View’-tab I found out that there was missing data with some variables. To easily indicate missing values in the data I assigned a specific value to these missing data; 999. I also specified this in the ‘Variable View’-tab in the ‘missing’ column, otherwise SPSS would use this as a legitimate value in any statistical analyses (Pallant, 2010).

3.4.3 Preliminary analyses

The questionnaire existed of predetermined and premeasured questions. Some of these questions can be labeled as counter-indicative, because they are asked in the opposite way other questions are asked. To prevent the risk of getting strange results, it is important that SPSS ‘knows’ these kinds of questions are used. Therefore, some values are recoded.

(30)

4. Results

This section will provide the results of the study. Multiple hierarchical analyses and a multivariate analysis of variance (MANOVA) were applied to test the hypotheses.

4.1 Descriptive statistics, correlation matrix, and reliability

Table 6 presents descriptive statistics and correlations coefficient of the dependent, independent, and moderating variables. Before computing the correlation matrix, the mean scores of the independent variable and dependent variables are computed. Pearson correlation coefficient is used for the continuous variables and the dichotomous variables age, industry, communication channel discrepancy, and gender. Spearman rank order correlations is used for the data that does not meet the criteria for Pearson’s correlation; tenure, income, and education. These are all ordinal level or ranked data.

(31)

If you look at the first three hypothesized correlations in table 6, they all are different from zero and have the expected sign. The variables CSR and job satisfaction have a significant but weak positive correlation (r=0.19). Correlations between CSR and job performance also are significant but are moderately correlated (r=0.32). The CSR and affective commitment variables also significantly correlate, may it be weak (r=0.23). In terms of the moderators, the variable Age only shows a weak significant positive correlation with affective commitment (r=0.21). This could be explained by the notion that older workers have a higher tenure, and therefore, are more committed. This seem to be confirmed by the correlation between age and tenure (r=0.49) and between tenure and affective commitment (r=0.34). This follows the results of Meyer et al. (2002). They already found that age and tenure are correlated. The moderating variable Industry shows moderate correlation with job satisfaction (r=0.28) and with affective commitment (r=0.19). The moderating variable Communication Discrepancy shows a negative significant correlation with job performance (r=-0.20). This seems to imply that the more communication discrepancy exists within an organization, the less people can perform their jobs properly. Also, communication discrepancy seems to be negatively correlated with CSR (r=-0.19). This is an interesting notion because it may suggest that the higher the communication discrepancy within an organization, the lower the importance of CSR activities of organizations by employees.

Table 6 also shows the reliability of the scales of the variables of CSR, job satisfaction, job performance, and affective commitment. Before starting with testing the hypotheses, the quality of the variables scales is tested. This was done by means of a reliability analysis. Ideally, Cronbach’s Alpha coefficient of a scale should be above 0,7 (Pallant, 2010). As can be seen from the matrix, the scales used in this study are internally consistent.

(32)

4.2 Hypothesis testing

First, preliminary analyses were conducted to ensure no violation of the assumptions of normality, linearity, multicollinearity and homoscedasticity. Then, the mean of the scales of the independent variable (CSR) and dependent variables (job satisfaction, job performance, affective commitment) was computed. To test the hypothesized relationships, hierarchical multiple regression analysis was applied. The hypotheses hold that an organization’s CSR initiatives will have a positive impact on job satisfaction (H1), job performance (H2), and affective commitment (H3). Hypotheses 4 to 6 say that the effect of an organization’s CSR initiatives on employee outcomes will be moderated by generation (H4), industry sensitivity (H5), and communication discrepancy (H6).

Table 7 present the results of the hypotheses H1, H4a, H5a, and H6a. The regression model itself shows to be significant [F(11, 101)= 2.32, p < 0.05]. In the first step the control variables were entered into the model and together explained 11.3% of the total variance in job satisfaction. Only income emerged as a significant factor. Income is positively correlated with job satisfaction (B=0.16, t(108)= 2.40 , p < 0.05), which means that when income goes up, satisfaction on the job increases.

(33)

In the second step, the independent variable CSR was added to the model. Table 7 shows that the R2 change value of model 2 is 0.036. This means that an organization’s CSR initiatives explain an additional 3.6% of the variance in job satisfaction. This small increase in variance is a statistically significant contribution, as indicated by F change (1, 107)= 4.52, p < 0.05. The results support the first hypothesis that an organization’s CSR initiatives will positively influence job satisfaction (B= 0.23, t(107)= 2.13, p < 0.05).

In the third step, the moderator effects of age, industry sensitivity, and communication discrepancy in hypotheses 4a, 5a, and 6a was tested by calculating interaction terms between CSR and the moderator variables (CSR x Age (dich.), CSR x Industry (dich.), CSR x Comm.Discr. (dich.). Age (B=0.33, t(101)= 1.17, p = 0.25 > 0.05), industry sensitivity (B=-0.16, t(101)= -0.60, p = 0.55 > 0.05), and communication channel discrepancy (B=0.24, t(101)= 0.80, p = 0.42 > 0.05) show no influence on the effect of CSR on job satisfaction. So, these variables are statistically insignificant and therefore H4a, H5a, and H6a were not confirmed by the analysis.

The same hierarchical multiple regression analysis procedure was conducted to test H2, H4b, H5b, and H6b, as presented in Table 8. The regression model itself shows to be significant [F(11,101)= 2.62, p < 0.05]. In the first step, again, the control variables were entered into the model. None of the control variables emerged as a significant factor. This is remarkable, even when the control variables seem to have no effect, the independent variable CSR still is highly significant. Moreover, prior literature indicates that at least gender, and tenure is an adequate control variable when the dependent variable is job performance (Carmeli et al., 2007; Igbaria & Baroudi, 1995; Wright & Cropanzano, 2000). These results seem to contradict this.

(34)

Table 8. The results of hierarchical multiple regression analysis for H2, H4b, H5b, and H6b

In the second step, the independent variable CSR was added to the model. Table 8 shows that the R2 change value of model 2 is 0.104. This means that an organization’s CSR initiatives explain an additional 10.4% of the variance in job performance. This increase in variance is a statistically significant contribution, as indicated by F change (1,107)= 13.37, p <0.001. The results support the second hypothesis that an organization’s CSR initiatives will positively influence job performance (B= 0.33, t(107)= 3.66, p < 0.001).

In the third step, the moderator effects of age, industry sensitivity, and communication discrepancy in hypotheses 4b, 5b, and 6b was tested by calculating interaction terms between CSR and the moderator variables (CSR x Age (dich.), CSR x Industry (dich.), CSR x Comm.Discr. (dich.). Age (B=-0.17, t(101)= -0.74, p = 0.46 > 0.05), industry sensitivity (B=-0.07, t(101)= -0.34, p = 0.74 > 0.05), and communication channel discrepancy (B=-0.19, t(101)= -0.77, p = 0.441 > 0.05) show no influence

(35)

on the effect of CSR on job performance. So, these variables are statistically insignificant and therefore H4b, H5b, and H6b were not confirmed by the analysis.

Again, the same hierarchical multiple regression analysis procedure was conducted to test H3, H4c, H5c, and H6c, as presented in Table 9. The regression model itself shows to be significant [F(11,101)= 2.66, p < 0.05]. In the first step, again, the control variables were entered into the model and together explained 14.7% of the total variance in affective commitment. Only tenure emerged as a significant factor. Tenure is positively correlated with affective commitment (B=0.12, t(108)= 2.56, p < 0.05), which means that when tenure increases, affective commitment increases as well. This is conforming to the literature of Meyer et al. (2002).

Table 9. The results of hierarchical multiple regression analysis for H3, H4c, H5c, and H6c

In the second step, the independent variable CSR was added to the model. Table 9 shows that the R2 change value of model 2 is 0.047. This means that an organization’s CSR initiatives explain an additional 4.7% of the variance in affective commitment. This small increase in variance is a statistically significant contribution, as indicated by F change (1,107)= 6.30, p < 0.05. The results

(36)

support the third hypothesis that an organization’s CSR initiatives will positively influence affective commitment (B= 0.26, t(107)= 2.51, p < 0.05).

In the third step, the moderator effects of age, industry sensitivity, and communication discrepancy in hypotheses 4c, 5c, and 6c was tested by calculating interaction terms between CSR and the moderator variables (CSR x Age (dich.), CSR x Industry (dich.), CSR x Comm.Discr. (dich.). Age (B=0.15, t(101)= 0.59, p = 0.56 > 0.05), industry sensitivity (B=-0.32, t(101)= -1.32, p = 0.19 > 0.05), and communication channel discrepancy (B=-0.33, t(101)= 1.17, p = 0.25 > 0.05) show no influence on the effect of CSR on affective commitment. So, these variables are statistically insignificant and therefore H4c, H5c, and H6c were not confirmed by the analysis.

4.3 Additional analysis

Next to the main research question and its belonging hypotheses, in a small secondary research is examined whether the three CSR typologies each have a distinct impact on the employee outcomes. The results of this study support the notion that the CSR initiatives of an organization have a significantly positive effect on all 3 employee outcomes. This makes it interesting to look further and perform additional analyses with the data. This may be a stepping stone for other researchers to build upon these first notions in terms of this topic. The belonging hypothesis (H7) is that there will be differences among the 3 CSR typologies in the effect they have on employee outcomes. A multivariate analysis of variance (MANOVA) was performed to investigate the difference of CSR typology importance in the 3 employee outcomes. Preliminary assumption testing was conducted to check for normality, linearity, univariate and multivariate outliers, homogeneity of variance-covariance matrices, and multicollinearity, with no serious violations noted. The results show that there was no statistically significant difference among the three CSR groups and the employee outcomes, F (6, 216)= 0.62, p = 0.72; Wilks’ Lambda= 0.97; partial eta squared= 0.02. Because no significant results were obtained, no further investigation will be conducted by means of separating the employee outcomes.

(37)

5. Discussion

This study examines the effect of an organization’s CSR activities on three employee outcomes: job satisfaction, job performance, and affective commitment. The results show that the CSR activities of an organization positively influence all three the employee outcomes.

The finding of this study that an organization’s CSR activities positively influence job satisfaction is consistent with prior literature. For example, De Roeck et al. (2013) reported that CSR initiatives can be used to create stronger relationships with employees and improve attitudes at work. Moreover, Koh & El'fred (2001) suggest that employees experience greater job satisfaction when they consider their organizations as (being) ethical, of which an organization’s involvement in CSR is seen as a natural extension according to Valentine & Fleischman (2008).

Also, prior literature in terms of job performance seems to be consistent with the results of this study. Carmeli et al. (2007) claim that perceived social responsibility affects employee’s attitudes and organizational identification. When employees identify themselves with an esteemed organization, this will positively influence, among others, job performance (Peterson, 2004). Moreover, according to this claim of Peterson (2004), when an employee is a proud member of an organization, s/he is emotionally attached, which in turn is affective commitment. This is in line with the results of this study that CSR initiatives of an organization positively influence affective commitment. Moreover, when employees are involved in and are made enactors of the organization’s CSR activities, attention and commitment most probably will increase, which in turn will do the effectiveness of CSR good (Bhattacharya et al., 2012).

The results show something noteworthy in terms of the job performance variable and the added control variables in the regression model: gender, tenure, income, and education. None of the control variables emerged as a significant factor. This is remarkable, even when the control variables seem to have no effect, the independent variable CSR still is highly significant. Moreover, prior literature indicates that at least gender, and tenure is an adequate control variable when the

(38)

dependent variable is job performance (Igbaria & Baroudi, 1995; Wright & Cropanzano, 2000; Carmeli et al., 2007). This seems to imply that employee’s perceived importance of CSR is so significantly present, while the control factors show no effect, it would even influence their performance on the job.

Another noticeable result from this study is that employees do think CSR is important, however in terms of monetary consequences, the majority of the respondents (67.5%) are not willing to give up their year-end-bonus for the sake of an organization’s socially responsible behavior. So, the perceived importance of CSR of employees only is true to a certain extent. This may provide a funding for future research. For instance, how far will people go in terms of the importance of CSR, what is the turning point, so when will the CSR activities of an organization backfire?

The results did not provide support for generational differences. This was not completely unexpected because Kowske et al. (2010) already stated that there are many inconclusive results from prior research about generational differences. Their principal finding is that generational effect sizes are small, which suggests generations are more similar than different at work. A possible explanation of why employees across generations are similarly satisfied with their jobs is that the basic process of becoming employed and navigating a career has remained largely unchanged. ‘’Regardless of generation, people feel similarly satisfied, perhaps indicating that they continue to select jobs according to what they like to do, and change jobs when they are dissatisfied’’ (Kowske et al., 2010). Another possibility may be that due to the Internet, media, and other technological developments we have become more aware of social and environmental problems (Turker, 2009). Not only the younger generation, but also the older ones. Additional research is needed to understand generational differences at work (Kowske et al., 2010).

Moreover, we are currently dealing with world crises. The lack of support of generational differences in performance and affective commitment may be because employees across

(39)

most concern for employees. Very often, the impact it has on the people who have not lost their jobs is overlooked. (Greenhalgh, 1983). According to Greenhalgh (1983), the greater the job insecurity, the greater are employees’ adverse reactions. Job insecurity would cause insecure employees’ work attitudes, which affect productivity, stability, and adaptability of the workforce (Greenhalgh, 1983). So this suggests that employees’ main concern is, whatever the age, to keep the job, a job in general, not keeping in mind whether this organization is socially involved.

This job insecurity may also explain why the results did not provide support for industrial differences. No significant effect was found of higher sensitive industries in comparison with low sensitive industries on the CSR activities-employee outcome relationship. Again, this may suggest that employees’ main concern is, whatever the industry they work in, to keep the job.

Also, employees may be unaware of the CSR involvement of their organizations. Brammer & Millington (2004) find that firms in socially and environmentally sensitive industries have higher levels of relative and absolute expenditures, and that there seems to be a relationship between stakeholder pressures and corporate charitable contributions. This suggests that stakeholders in such sensitive industries are more aware of the damage organizations cause and, consequently, hold the organizations accountable for these issues. However, when employee awareness is low, they do not know about the CSR initiatives, so cannot form any opinion about it.

Another reason for this effect to be non-significant may be the organization size, which is not part of this study. Perrini et al. (2007) find that size explains the differences in firms’ willingness to implement CSR strategies, since formal CSR approaches still seem to be a privilege of large firms (Perrini et al., 2007). This would mean that SMEs (are) less (able to) involve in CSR and/or not seem to prioritize CSR, and therefore importance and awareness of employees in terms of CSR could be low. Future research could touch upon this topic.

No support was found of the influence of communication channel discrepancy on the CSR-employee outcomes relationship. A reason for this may be that, in this study, the majority of the

The effect of an organizationâ s CSR initiatives on employee outcomes.

Table of Contents

Abstract

1. Introduction

1.1 Contribution

2. Literature Review

2.1 CSR Typology

2.2 Employee outcomes

2.3 Generational influences on the effect of CSR

2.4 Industrial influences on the effect of CSR

2.5 Communicational influences on the effect of CSR

2.6 Additional analysis

2.7 Conceptual Model

3. Research Method

3.1 Research design

3.2 Procedure and participants

3.3 Description of measures

3.4 Strategy of Analysis

4. Results

4.1 Descriptive statistics, correlation matrix, and reliability

4.2 Hypothesis testing

4.3 Additional analysis

5. Discussion