Ethical decision making in organizations
The relationship between age and ethical decision making
Master thesis
Msc. in Business Administration – Strategy track
Amsterdam Business School
Name Supervisor Date
Lars Heger Dr. ir. J.W. Stoelhorst 08-19-2016
Statement of originality.
This document is written by Student Lars Heger who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Purpose – The relationship between age and ethical decision making is extensively studied in scientific literature and has produced mixed findings. Past research focused mainly on direct influences, either individual or organizational but limited attention has been given to crucial moderating and mediating influences. The aim of this study is to fill this gap by introducing organizational tenure as a moderator. Furthermore, the moderating role of ethical culture is examined.
Design/methodology/approach – This thesis surveyed 97 respondents and tested the hypotheses with a questionnaire and a vignette study.
Findings – Age and ethical culture are positively related to ethical judgment. No evidence was found for the moderating effect of organizational tenure. Furthermore, no evidence was found for a three-way interaction effect between age, organizational tenure and ethical culture.
Research limitations/implications - This research is based on data by respondents approached within the researcher‟s social network. The results show a biased dataset. Therefore, the
generalizability of the findings is affected. Nonetheless, researchers should use a more
sophisticated approach to investigate the relationship between age and ethical decision making by introducing new moderators and mediators.
Practical implications - Organizations with the aim of improving their ethical environment should take preemptive measures before hiring new employees. The role of managers and a code of ethics should be considered.
Table of contents
1. Literature review ... 6
1.1 Ethical decision making in the literature ... 6
1.2 Age and ethical decision making ... 10
1.3 Organizational tenure ... 13
1.4 Ethical culture ... 15
1.5 Literature gap and research question ... 16
2. Conceptual model and hypotheses ... 18
3.1 Conceptual model ... 18
3.1 Hypotheses ... 19
3.1.1 Age and ethical decision making ... 19
3.1.2 The moderating effect of organizational tenure ... 20
3. Research design ... 24 4.1 Research method ... 24 4.2 Sample ... 25 4.3 Operationalization of variables... 25 4.3.1 Ethical culture ... 26 4.3.2 Ethical judgment ... 27 4.3.3 Control variables ... 28 4. Results ... 30 5.1 Reliability analysis ... 30 5.2 Descriptive statistics ... 31 5.3 Normality analysis ... 34 5.4 Correlations ... 35
5.5.1 The main effect: The relationship between age and ethical judgment ... 40
5.5.2 The moderating effect of organizational tenure ... 41
5.5.3 The effect of ethical culture on ethical judgment ... 44
5. Discussion ... 47
5.1 Interpretation of the results ... 47
5.2 Implications for theory and practice ... 50
6.3 Limitations and recommendations for further research ... 51
6. Conclusion ... 53
Introduction
“Respect, integrity, and professionalism are the core values of our bank.” said Rabobank
chairman Piet Moerland in a statement where he announced his resignation. He resigned as a result of the Libor affair, which eventually led to a settlement of €774 million euro with U.S, U.K. and Dutch authorities in 2013. Despite the alleged core values of the bank, Rabobank became involved in a criminal conviction for fraud. More recently, the four largest banks in the Netherlands (Rabobank, ABN Amro, SNS and ING) had to compensate SMEs for selling interest swaps without pointing out to the risks of falling interest rates. Total loss for compensation is expected to be between € 1.7 and € 3 bln (Bökkerink, 2016). Within Dutch borders there are multiple examples of fraud and misbehavior at multinational companies (e.g. KPMG, Ballast Nedam, SBM Offshore). It is, however, not a Dutch problem; white-collar crime truly is a world phenomenon. The infamous Enron scandal and the more recent emission scandal involving German car builder Volkswagen are well-known examples of this global fraud. How is it possible that these renowned companies, to a greater or lesser extent literally depending on trust and reliability, engage in questionable behavior? This makes clear that fraud still is a hot topic in international business.
Ethical decision making in business has received a substantial amount of attention in scientific research focusing on the conditions, factors and influences on the ethical decision making process. However, it is definitely not the case that the well has run dry. While past
research focused mainly on direct influences, either individual or organizational, limited attention has been given to crucial moderating and mediating influences (Lehnert, Park, & Singh, 2015).
The aim of this thesis is to fill this particular gap. More specifically, in this research the relationship between age and ethical decision making is discussed. Although age is positively associated with moral development and reasoning (Elm & Nichols, 1993; Trevino, Weaver, & Reynolds, 2006), research on the topic of ethical behavior within organizations has yielded mixed results. Findings on the relationship between age and ethical behavior within organizations show both negative and positive relationships. This might indicate that organizations can have a negative influence on ethical behavior, despite an employees‟ age. It is therefore important to research the mechanisms behind this phenomenon that detrimentally affects ethical decision making. In this study, the moderating effect of organizational tenure is suggested and tested. Additionally, the moderated moderation effect (i.e. a three-way interaction effect) of ethical culture is tested.
1. Literature review
This chapter summarizes the different literature used in this study. It starts with explaining ethical decision making, provides the basis of this research and explains Rest‟s (1986) four-step model. Secondly, the current view of the relationship between age and ethical behavior is discussed. Thirdly, the concept of organizational tenure is explained in the context of ethical behavior. Fourthly, the concept of ethical culture is introduced. This chapter ends with the literature gap and research question.
1.1 Ethical decision making in the literature
An ethical decision is defined as “a decision that is both legal and morally acceptable to the larger community” (Jones, 1991, p. 367). Trevino (1986) proposed an interactionist model to explain the interaction of individual and situational components to explain the ethical decision
making process. Trevino‟s interactionist model of ethical decision making in organizations provides the fundament of this study and is schematically presented in figure 1. An individual reacts to an ethical decision by using its moral compass. However, cognition of what‟s right or wrong is not enough to explain an individual‟s ethical decision-making behavior. Supplementary individual and external factors interact with the cognitive component to determine the outcome of an ethical dilemma. The individual factors ego strength, field dependence and locus of control influence an individual‟s ethical decision making. External or situational factors arise from the job context, organizational culture and characteristics of the work. Examples of these factors are the normative structure of the organization, obedience to authority, role taking, responsibility for consequences and other pressures. Interestingly, characteristics of the work itself and the
organizational culture may impact the moral development of the individual, which is illustrated in figure 1. This raises question about what happens inside the organization that influences an individual‟s cognitive moral development and provides the basis of this research.
Figure 1. Trevino‟s (1986) interactionist model of ethical decision making in organizations.
Trevino‟s model is built upon Lawrence Kohlberg‟s (1969) influential model of cognitive development. Kohlberg found that a person‟s morality changes as the individual matures. He identified six stages of moral development. Each stage explains how individuals think about ethical dilemmas and how they determine what is right or wrong in any given situation (Trevino L. K., 1986). In the first two stages, individual‟s primary concerns are immediate consequences, particularly rewards and punishments and self-interest. At stages three and four, an individual‟s primary concerns are the expectations of the larger society or some segment like a group of
colleagues. According to Kohlberg, most adults in society are in stage three or four. Less than 20% of the adults reach level 5 or 6. At level 5 and 6. an individual has reached a higher consciousness of the variety of values that people hold and is aware that rules are relative to groups. These individuals can persevere in their own moral judgment regardless of the majority‟s opinion. Following Kohlberg‟s theory, approximately 80% of the people are susceptible to the opinion of the majority and will easily conform to group norms. From Trevino‟s model, it may be inferred that younger professionals, who are commonly at a lower stage of moral development, would be more likely to be influenced by external factors to help resolve ethical dilemmas (Peterson, Rhoads, & Vaught, 2001).
According to Rest‟s (1986) four-step model for ethical decision making, a moral concern must first be recognized and judged. After the first two steps, a moral agent establishes his or hers ethical intentions before any actions take place. Moral awareness (1) is the ability to recognize and interpret a moral issue. Moral judgment (2) is the ability to decide which course of action is morally correct. Moral intent (3) is the ability to prioritize moral actions over immoral actions. Moral behavior (4) is the application of the moral intent to the situation. Rest (1986) argues that each step in the process is different and that success in one step of the process does not imply success in any other phase in the process (Jones, 1991). For example, the fact that an individual recognizes a moral concern (first step), does not automatically mean that the individual will act ethical. Rest‟s four-step model is one of the most prevalent models in ethical decision making research (Craft, 2013; Lehnert, Park, & Singh, 2015).
The empirical literature on ethical decision making can be broadly divided into two categories. The first category explores variables related to individual factors. The second
organization. In the literature on ethical decision making, individual factors have received the most attention by far (Richardson & Ford, 1994). This category relates to factors uniquely associated with the individual, such as gender, religion, age and nationality (Craft, 2013). Of the individual components, age is the factor that produces one of the most mixed results regarding ethical decision making in organizations (Lehnert, Park, & Singh, 2015).
1.2 Age and ethical decision making
Even though the relationship between age and ethical decision making is an extensively studied subject, it remains relevant. In fact, it may even become more relevant considering the rapidly aging (working) population in many countries. In the Netherlands in 2016. the proportion of people aged 65 and older is 3.6% in relation to the working population. This number is
expected to increase to 43.1% in 2030 and 51.1% in 2040 (Centraal Bureau voor de Statistiek, 2016). As of today, research findings on the relationship between age and ethical behavior are at odds. Even though there is a rather substantial amount of research on this topic, there is no consensus. In their review, Lehnert, Park and Singh (2015) identified 49 findings on this topic, 23 of them reported a significant relationship while the remaining 26 did not. Of the researches that reported a significant result, 16 found a positive relationship between age and ethical decision making, while seven of them found a negative relationship. Hence, the landscape of the research results can be encapsulated in the following figure:
Figure 1. Distribution of research findings according to Lehnert, Park and Singh (2015).
The current research on the relationship between age and ethical decision making resulted in a wide variety of outcomes. For instance, a study among graduates of a business school and entrepreneurs concluded that ethical beliefs were higher for business professionals over thirty years of age. This study also found a gender difference. Males of thirty years and younger are more susceptible to external factors compared to their female counterparts. This effect diminishes over time which suggests that females reach higher levels of moral development at a younger age than males (Peterson, Rhoads, & Vaught, 2001) This result was acknowledged in a study by Krambia‐Kapardis and Zopiatis (2008) among Cypriot business-related graduates. Furthermore, the results of a survey completed by students enrolled in business courses at the University of Southern Mississippi suggest that age is a determining positive factor in making ethical decisions. Age groups of 40+ years are the most ethical, followed by groups of 31-40 years, 22-30 years and under 21 years respectively (Ruegger & King, 1992). A reasonably similar research and outcome is obtained in a study regarding the differences in ethical evaluations between business faculty and students at a Southern University in the U.S. (Stevens, Harris, & Williamson, 1993). Krambia-Kapardis and Zopiatis‟ (2008) research findings revealed that individuals currently working or pursuing a business-related graduate degree in Cyprus over the age of 30 were more ethical than those under 3. In a study among accounting students at two universities in Hong
Research findings regarding the relationship between age
and ethical behavior.
Significant (47%) Positive relationship (33%) Negative relationship (14%) Insignificant (53%)
Kong, age was also found to have a positive correlation with ethical sensitivity. Older accounting students were more likely to identify ethical matters in the professional scenario (Chan & Leung, 2006).
Contrariwise, there is literature that acknowledges a negative relationship between age and ethical decision making. Browning and Zabriskie (1983) find evidence in the industrial buying industry that particularly older, more experienced buyers engage in questionable buying
practices. The older the manager, the more likely it is that the manager believes it is acceptable to accept favors from vendors who are currently not a supplier. Comparable evidence is to be found in the accounting profession. An unsettling negative relationship is found for age and MRA (moral reasoning abilities) (Eynon, Hills, & Stevens, 1997).
There is a considerable amount of studies that did not find a significant relationship. Age was found to be a poor predictor of ethical behavior for U.S. internal auditors, according to Larkin (2001). This conclusion is supported by Shafer, Morris, and Ketchand (2001), who did not find a significant correlation between the age of auditors and their ethical judgments or behavioral intentions. Similarly, in a study among 97 Israeli managers, age did not explain more than 5% of the variance in managers‟ behavior (Izraeli, 1988).
These contradictory findings lay bare that the effect of age on ethical decision making remains vague. Even though age is normally positively associated with moral reasoning (Elm & Nichols, 1993), which is confirmed by the majority of the studies which do find a significant relationship (Lehnert, Park, & Singh, 2015), there is a fairly large amount of studies stating otherwise. There seems to be a more complex relationship between age and ethical decision making than discussed in earlier studies (O‟Fallon & Butterfield, 2005). It is likely that other
factors interact with age to produce these mixed results. According to Lehnert, Park and Singh (2015), researchers should not just look at the boundary conditions where age is concerned, but also at the theoretical implications of age and ethical decision making. The latter is basically underlined by Trevino‟s (1986) interactionist model of ethical decision making. Other factors should be taken into account that may act as moderators of the relationship between age and ethical behavior.
1.3 Organizational tenure
Organizational tenure can be defined as the length of employment in an organization. Even though it is positively related to job tenure, industry tenure, group tenure and hierarchical level, it is important to recognize the distinction between these terms (Ng & Feldman, 2010). The number of years in the current function (job tenure) does not have a significant influence on the ethical perception of an employee. The work experience in the company, however, does have a
significant negative correlation with ethical decision making (Roozen, De Pelsmacker, & Bostyn, 2001).
Bergh (2001) applied the Resource-based view (RBV) of the firm to organizational tenure and suggested that top executives offer the most valuable resource. These executives usually are those with longer rather than shorter tenure. Longer tenured employees have information of the firm like traditions, culture and relationships with buyers and suppliers. The RBV way of viewing the firm describes firms as bundles of tangible and intangible resources. In order to achieve a sustained competitive advantage, these resources must be valuable, rare, inimitable and non-subsitutable (Barney, 1991). Longer tenured employees have gained valuable firm-specific information of the firm throughout their careers. Their knowledge is, in RBV-terms, valuable, rare, inimitable and non-substitutable and therefore a source of sustained competitive advantage.
Therefore, the presence of tenured employees can be considered crucial for a firm‟s performance. In fact, in poorly performing firms, new senior leadership often comes from external hiring rather than from inside the organization. Therefore, firms should pay more attention to high potentials within the organization (Rost, 2008).
Employees with longer organizational tenure are important for a firm‟s performance but it could come with a price. Research in organizational psychology suggested that through socialization, which develops continuously along successive levels of organizational tenure, employees change their attitude towards the organization (Kraemer & Gouthier, 2014). In their study, Obloj and Sengul (2012) show that over time, as employees gain experience under an incentive scheme, they will learn to game the system which is defined as adverse learning. Through organizational learning, employees learn how to, and will appropriate a higher share of the value created (Obloj & Sengul, 2012). More tenured employees are thus associated with gaming incentive systems: activities that increase their pay but hurt the objectives of their
employer (Larkin I. , 2014). Ng and Feldman (2010) found that organizational tenure is positively related with job and citizenship performance. Surprisingly, in the same research organizational tenure is also associated with unethical behavior (e.g. aggressive behavior and nonsickness absence). The authors suggest that long-tenured employees distinguish themselves from shorter-tenured employees and consider assertive behavior as a legitimate right that comes with seniority within the firm. They conform less to group norms because they have more power and freedom in their perspective (Ng & Feldman, 2010). New employees, however, are more concerned with proving themselves as contributing members of the organization and are unsure about the organization‟s expectations as well as their ability to meet those expectations. As their
& Gouthier, 2014). New employees tend to react more positively towards common organizational practices. New employees are characterized by a high motivation and the willingness to make a good impression. However, this behavior tends to change with three months to two years (Huang, Shi, Zhang, & Cheung, 2006). After the first period of time employees lose their enthusiasm. Apparantly, employees change their attitude and behavior along succesive lines of organizational tenure. Of course, merely organizational tenure doesn‟t explain any compromise on ethical norms by longer tenured employees. The environment in which they act could explain their change in behavior. This is in line with Trevino‟s (1986) interactionist model of ethical decision making, the basis of this research, which states that organizational culture may influence an individual‟s moral development. Part of the organizational culture is the organization‟s ethical culture.
1.4 Ethical culture
According to Bowen (2004), for an organization to be ethical, it must have an organizational culture that values ethical decision making. Organizations are systems of shared beliefs and orientations (i.e. organizational culture) that unite the members of the firm and guide their
conduct (Blau & Scott, 1978). Organizational culture unites the members of the organization and gets every member on the same line. Barney (1986) indicated that organizational culture can even contribute to a sustainable competitive advantage. Considering the prior, organizational culture can be considered the backbone of a firm. Organizational culture can have a substantial influence on the emphasis that is placed on ethical decision making, both for the organization as a whole and the individual (Bowen, 2004). Ethical culture can be seen as a subset of organizational culture that can moderate the relationship between an individual‟s moral reasoning and its behavior (Brown & Treviño, 2006). Ethical culture can be either formal (e.g. incentive systems, codes and policies etc.) or informal (e.g. peer behavior) systems that support ethical or unethical
behavior. In every formal organization, informal organizations arise. As employees work together, they form their own ideas, practices, values, norms and social relations. However, this informal culture is originated from the formal organization (Blau & Scott, 1978).
In general, research regarding organizational culture and ethical decision making strongly supports that the culture of the organization contributes to managing organizational ethics (Loe, Ferrell, & Mansfield, 2000). The collective conclusions of many empirical studies suggest that the employees‟ perception of the ethical climate in their organization strongly influences the likelihood of unethical behavior (Shafer & Simmons, 2011). In other words, a unethical culture within the organization results in unethical behavior. In a study among practicing tax
practitioners in China, it was found that organizational cultures characterized by strong ethical norms and incentives for ethical behavior reduced the vulnerability of the organization to unethical behavior. Conversely, in organizations where top managers are unethical and reward unethical behavior, the likelihood that employees engage in unethical behavior is significantly higher (Shafer & Simmons, 2011). Armstrong, Williams, and Barrett (2004) found that a „banal wrong-doing environment‟ impacts the overall health and morality of an organization.
Adding ethical culture as an explanatory factor regarding ethical behavior could provide a more in-depth insight in understanding employee‟s behavior. This may explain differences in ethical behavior between short and long tenured employees specifically.
1.5 Literature gap and research question
Analyzing the existing research on the relationship between age and ethical decision making, it becomes clear that a more ticated approach is needed to achieve consensus in the ongoing debate about the relationship between age and ethical decision making. Even though age is typically associated with moral reasoning (Elm & Nichols, 1993), the results of research on this
topic are inconsistent. Organizational tenure as a moderating variable for age could provide new insights in the present literature about ethical decision making. From the cited articles in
paragraph 2.3 it can be concluded that organizational tenure can influence an employees‟ moral compass. To fully understand the potential moderating influence of organizational tenure, it is important to identify the underlying mechanisms regarding organizational tenure and the influence of the environment (organizational ethical culture) in which the employee operates. Trevino‟s (1986) interactional model of ethical decision making is the starting point of this research. It shows that individuals while they mature develop their cognitive moral abilities. Besides aging, moral development is influenced by both individual moderators and situational moderators. Ethical culture is in this research referred to as a situational moderator. Adding organizational tenure and ethical culture as explanatory variables provide a new perspective to the debate, a debate which is becoming even more relevant nowadays considering the greying working population. The recognition of organizational tenure as an influential factor in ethical decision making could help managers to adapt their management style accordingly. More attention to incumbent employees could prevent unethical behavior.
This leads to the following research questions:
Is age related to ethical decision making?
Does organizational tenure moderate the relationship between age and ethical decision making?
2. Conceptual model and hypotheses
This chapter contains the theoretical framework of this study. As discussed in the previous two chapters, the relationship between age and ethical decision making will be explored. Additionally, the moderating effect of organizational tenure and ethical culture on this
relationship will be examined. It is expected that organizational tenure negatively moderates the relationship between age and ethical decision making. This negative moderation is expected to be moderated by ethical culture. An unethical culture will strengthen the negative moderation. An ethical culture is expected to diminish the negative relationship, or even make it positive. Figure 2 schematically represents the corresponding basis of this research in a conceptual model.
Furthermore, hypotheses are developed based on the literature review in chapter two. Each arrow in the conceptual model represents a hypothesis.
3.1 Conceptual model
3.1 Hypotheses
In this chapter, the hypotheses are formulated based on prior literature.
3.1.1 Age and ethical decision making
A lot of research is done about the influence of age on ethical decision making. As shown in the previous section, this resulted in a wide variety of outcomes. The researches were performed under different circumstances, settings and roles. Most of the studies looked at the moral
reasoning in business-related contexts (accountants/industrial buyers). To measure the influence of the number of years in a business-context, it is important to know the influence of age on ethical decision making in generic terms. A 20-year longitudinal research of moral development performed by Colby, Kohlberg, Gibbs, Lieberman, Fischer, and Saltzstein (1983), clearly indicates that people gradually develop their moral reasoning as they get older. In other words, older people are considered to be more ethical than younger people. Important for this study is what happens to their moral reasoning in a working environment. Even though aged people are associated with a more developed moral reasoning, they might alter their behavior in another context. Weeks et al (1999) say that people below their 30‟s, focus on finding a job in which they can be succesful and make progress. People in this age group are very much concerned about peer acceptance. Their longing for peer acceptance could lead to ethically questionable conduct and shortcuts while pursuing this goal. Weeks, Moore, McKinney, and Longenecker (1999) also mention their susceptibility to the organizational environment. An organization‟s emphasis on certain performance benchmarks may tempt the employee in this age group to compromise on their ethical standards. In this age group, the superior/subordinate relationship plays a significant role. If the superior pushes the employee to achieve their standards of performance this may tempt the employee to engage in questionable actions. This effect is even stronger when the
organization‟s management doesn‟t represent high ethical standards (Weeks, Moore, McKinney, & Longenecker, 1999).
Weeks et al (1999) recognize a second age group between the age of 30 and 45.
Employees in this age group typically commit themselves to a particular organization and aims for progress within the organization. Progress is often associated with promotion which makes this a competitive age group. Even though people in this age group are less susceptible to external influences, their intrinsic motivation for achievement may cause them to comprimise on their ethical standards. On the other hand, employees in this age group typically gradually increased their salary and have made themselves financially independent. Financial incentives will influence this age group less than the younger age group.
The final age group, 45 until retirement, lacks interest in competition but instead looks for security. Their prior achievements in their career and organization increased their self-assurance and helped them to establish a certain job security. They no longer feel the need to prove
themselves and may consequently be less tempted to compromise on ethical values.
Considering the prior it is expected that older people will present higher ethical standards than younger people. Therefore, ceteris paribus, I expect a positive relationship between age and ethical decision making which is the first hypothesis in this study.
H1: Age is positively related to ethical decision making.
3.1.2 The moderating effect of organizational tenure
In the study by Roozen, De Pelsmacker, and Bostyn (2001) it is found that the most
career of the employee‟. Slocum and Cron (1985) state that among researchers, there is agreement that career stage impact attitutude toward the job. Career stage researchers have proposed, based on the theory initially developed by Super (1957), that under regular circumstances people will experience four career stages: establishment, advancement, maintenance and decline (Gould & Hawkins, 1978). Decline happens after the career so that stage is out of the scope for this study. The establishment stage occurs in the first two years within the organization. In this phase, employees‟ primary concern is to establish oneself which requires developing the right competences for the new role and being accepted by the colleagues. Also, new hires typically enter the organization at a low hierarchical level and pay rates. There is evidence that pay is an important need at lower pay levels and that this need is descending at each subsequent management level (Gould & Hawkins, 1978). Buchanan (1974) stated that the
primary concern of employees in the first year is safety; getting established with and accepted by the organization. It seems that in the beginning of the career stage, employees most important concerns are dominated by being accepted by their colleagues and organization, pay, and developing the right competences.
In the next stage, the advancement stage, employees face opportunities of promotion and higher salary. This stage occurs in the time frame of two to ten years within the organization. Employees have commonly established themselves in their work roles after the establishment stage (Gregersen, 1993). Individuals tend to turn their focus on achievement and promotion. Pay itself, is of lesser importance (Gould & Hawkins, 1978). It appears that employees‟ focus of interest changes from the organization to the individual.
The advancement stage is followed by the maintenance stage which occurs after ten years of organizational tenure. These employees have typically proved themselves in a number of
business contexts and have gained responsibilities. This stage is characterized by reduced
competitiveness and self-indulgence (Gould & Hawkins, 1978). In this stage, employees enjoy a higher job security and the need to prove oneself declines. It seems that employees in this stage feel more confident in their role and enjoy a high job security. Therefore, they might become self-indulgent and as a consequence less concerned with the organization‟s goals (Gould & Hawkins, 1978).
It became clear in the prior that employees enter the organization at a low pay rate. Their pay rate typically increases along increasing career stages. According to Roozen, De Pelsmacker, and Bostyn (2001), people with a higher income have a significantly lower ethical attitude than people with lower income. Also, in the advancement stage and the maintenance stage, employees gain responsibilities. The position in the company (in other words, the amount of people who have to report to the employee) has a negative correlation with ethical attitude (Roozen, De Pelsmacker, & Bostyn, 2001). These results indicate that the organizational tenure (career stage) of employees negatively influences their ethical attitude. Therefore, the following set of
hypotheses is introduced:
H2: Organizational tenure is negatively related to ethical decision making.
H3: The relationship between age and ethical decision making is negatively moderated by organizational tenure.
3.1.3 The conditional effect of ethical culture
The collective conclusions of many empirical studies suggest that the employees‟ perception of the ethical climate in their organization strongly influences the likelihood of unethical behavior (Shafer & Simmons, 2011). Also, in organizations where top managers are unethical and reward
unethical behavior, the likelihood that employees engage in unethical behavior is significantly higher (Shafer & Simmons, 2011). It is sensible to expext that ethical culture can positively or negatively influence an employees‟ moral compass. According to Key (1999), ethical culture describes how an individual is likely to respond to an ethical dilemma. Hence, ethical culture can be used to predict and steer an inidividual‟s behavior. An organization‟s ethical culture can influence an individual‟s cognitive moral reasoning and predicts how people act when faced with ethical dilemmas.
Individuals with longer organizational tenure, have had more exposure to the ethical culture of the organization than shorter tenured employees. Trevino (1986) argued that over time, an organizational culture will influence an individual‟s cognitive moral development. Consider a culture where employees are encouraged to take responsibility for decisions and resolve conflicts at lower organizational levels. In this culture an employees‟ cognitive moral development may be enhanced. On the other hand, in an authoritarian organization where roles are predetermined and decisions are grounded on formal authority, moral development may actually be diminished (Trevino, 1986). From Trevino‟s model, it may be inferred that younger professionals, who are commonly at a lower stage of moral development, would be more likely to be influenced by external factors to help resolve ethical dilemmas Therefore, it is expected that the moderating effect of organizational tenure depends on the organization itself. In an unethical environment, organizational tenure will negatively moderate the relationship between age and ethical behavior. In an ethical environment, organizational tenure will positively moderate the relationship
between age and ethical behavior.
H5: The moderating effect of organizational tenure on ethical decision making is moderated by ethical culture, such that higher levels of ethical cutlure will positively affect organizational tenure, which in turn positively affects the relationship between age and ethical judgment.
3. Research design
In this chapter the design of the research and a description of the sample are presented. First, the chosen research method is explained. The second part consists of a description of the studied sample. Lastly, the operationalization of the variables is explained.
4.1 Research method
To study the relationship between age and ethical decision making a cross-sectional survey was used. Surveys allow a researcher to collect data through methods such as mail questionnaires and telephone interviews and to perform quantitative analysis on the data. The time period in which this thesis was written constrains the possibilities for the researcher. Therefore, a cross-sectional survey was chosen even though a longitudinal study would provide stronger evidences for the hypotheses. The survey was developed with Qualtrics, an online survey tool. A survey enables the researcher to measure behavioral intentions as opposed to behavioral actions. Therefore, the dependent variable, ethical decision making, is operationalized by measuring ethical judgment. Ethical intent or behavior cannot be measured through a questionnaire. The surveys will be conducted through an online questionnaire. An advantage of an online
questionnaire is that it allows the researcher to gather a substantial amount of data in a short period of time. Also, online data can easily be transferred to SPSS. An advantage of a questionnaire over a personal interview is that privacy is as good as guaranteed. That‟s
ethical dilemmas and answer sensitive questions about their job and coworkers. The data are expected to be more reliable if the respondents feel comfortable in telling the truth. Lastly, to perform statistical analysis, SPSS was used.
4.2 Sample
The aim for this study is to measure the moderating effect of organizational tenure on the relationship between age and ethical decision making. Therefore, the targeted population is the employed working population. The data was collected within the own network, mostly by using social media, LinkedIn and Facebook, and email. Even though convenience sampling is prone to biased selection, it‟s the most convenient way to get as much respondents as possible in a short period of time. A total of 97 people completed the survey of which 51.5% was male. The majority of the sample, 44%, had a master degree at a university. Furthermore, 43% had a background in business or economics. A complete overview of the respondents is presented in table 1 and table 1.1.
4.3 Operationalization of variables
The survey is designed to measure the variables organizational tenure, ethical culture, ethical judgment, age and other personal characteristics. Due to the sensitive nature of this questionnaire, steps were taken to assure reliable responses. The survey was accompanied by a cover letter assuring the confidentiality of the respondent‟s information. Additionally, it stated that the subsequent results are for academic purposes only. Apart from some generic personal information, respondents were not asked for personal identifying information.
The measurement scales were adopted from prior related research. This paragraph contains an explanation of the operationalization of ethical culture and ethical judgment and an elaboration on the control variables.
4.3.1 Ethical culture
The moderating effect of ethical culture on the moderator of this research, organizational tenure, is examined. To measure ethical culture, the modified Ethical Culture Questionnaire (ECQ-M) is used. The ECQ-M is developed by Key (1999) and is a modification of the original ECQ developed by Trevino et al (1995). The modification involved reformulating or deleting items that referred to an ethics code. In the original research by Trevino et al (1995), these items were found problematic due to the fact that many organizations don‟t have a formal ethics code. The original questionnaire modified had not been widely used in practice because it was
relatively new, but had undergone empirical testing to test the validity (Key, 1999). Previous researchers using this questionnaire reported a Cronbach‟s Alpha of .91 (Shafer & Simmons, 2011). The questions in the ECQ-M are based on the perception of the respondent of the ethical culture in the organization he or she works for.
In the ECQ-M, respondents are asked to provide information on their perception of the ethical culture of their organizations. The questionnaire contains 18 questions. Respondents were asked to express their personal views regarding the ethical culture of their organization by
circling a number on a 1-7 Likert scale. The possible answers range from (1) „completely false‟ to (4) „not sure‟ to (7) „completely true‟. A high score (5-7) would indicate a more ethical culture whereas a low score (1-3) would indicate a less ethical culture. Five items were reverse-coded and were corrected preceding the data analysis. An example of an item from the ECQ-M:
Please indicate to which extent the statements are true for your organization.
„In this organization, people are not encouraged to take full responsibility for their actions.‟
The complete questionnaire is presented in appendix 1.
4.3.2 Ethical judgment
A vignette approximates a real-life situation by presenting the dilemmas as conrete and detailed as possible. This will prevent the respondent from answering in terms of his own mental picture of the task before him. It enables researchers to measure the variable more accurately than could be achieved by using direct questioning (Alexander & Becker, 1978). Therefore, the quality of data is increased. By adopting vignettes from previously validated studies, the reliability of the data and the consistency with previous studies using this method is increased. The vignette study used to measure ethical judgment in this study is widely used in research regarding ethical decision making (e.g. Conroy & Emerson, 2004; Conroy, Emerson, & Pons, 2010; Vynoslavska, McKinney, Moore, & Longenecker, 2005). This part of the survey consists of 15 hypothetical situations that present ethical dilemmas. Ten items were randomly removed from the original vignette to increase the response rate. Respondents were asked to determine the acceptability of each vignette. The respondents were asked to share their personal views regarding the ethical dilemmas on a 1-7 Likert scale. The possible answers range from (1) „Never acceptable‟ to (4) „sometimes acceptable‟ to (7) „always acceptable‟. Table 1 contains a summary of the statistics of the responses to the vignettes.
To address ethical dilemmas in different functional business areas, the vignette deals with a variety of common business situations (Weeks, Moore, McKinney, & Longenecker, 1999).
Examples of the covered real-life business situations are gender-discrimination, pollution, accounting tricks and bribery.
An example of a vignette used in this study:
Please indicate to which extent you think the following statement is acceptable:
„An engineer discovered what he perceived to be a product design flaw that constituted a safety hazard. His company declined to correct the flaw. The engineer decided to keep quiet, rather than taking his complaint outside the company.‟
4.3.3 Control variables
Control variables can potentially influence the relationship between the dependent and the independent variable. The control variables for this study are gender, education level, working experience, field of study and industry of occupation.
Gender is a frequently studied variable in the ethical decision making literature. Just like age, gender as an independent variable delivers mixed findings. On average though, females were reported to be more ethical than men. However, women‟s behavior tends to be contextually dependent, men decision are more universal than contextual (Craft, 2013). Since ethical context is measured is this study, gender as a control variable could provide interesting findings.
Considering the cognitive nature of moral development, among age, moral development has been most strongly associated with education level (Trevino, Weaver, & Reynolds, 2006). Education level also produced mixed findings in previous literature. For example purchasing managers with more education viewed gifts and favors to be more unethical than less educated managers.
Table 1. Summary statistics of responses to vignettes
Vignette Brief description Mean response SD n
A B C D E F G H I J K L M N O
Pad expense account Exceed legal lim. pollution Recommend bad stock Underreport income for tax Bribe to foreign official Hire employee for secret Collusion to reduce comp. Gifts to purchasing agents Insider stock trading
Promotion of friend over other Safety design flaw cover-up Accounting trick to conceal Hire male employee over female Deceptive advertising
Free copy of software
2.4 1.7 1.9 1.9 2.2 3.2 2.7 3 1.9 2.6 2.3 2.7 2.0 2.6 3.0 1.4 1.0 1.1 1.2 1.3 1.7 1.4 1.4 1.2 1.6 1.1 1.5 1.3 1.7 1.6
Other studies, on the contrary, found no significant relationship at all (Richardson & Ford, 1994). However, research focusing this particular subject generally indicates a positive
relationship between education and ethical decision making (O‟Fallon & Butterfield, 2005).
Among gender and education, work experience is one of the most expansively examined variables. Research provided mixed findings regarding work experience which suggests that the
role of work experience in ethical decision making needs more research (Loe, Ferrell, & Mansfield, 2000). Also, work experience is obviously related to age.
Researching regarding the effect of type of education on ethical decision making often focuses on the difference between business majors and non-business majors. Some studies found that non-business majors were more ethical than business majors (O‟Fallon & Butterfield, 2005; Richardson & Ford, 1994). The variable is included because of the potential effect on ethical judgment.
Industry of occupation, in the literature commonly referred to as industry type, is not a commonly studied subject (Craft, 2013). However, the organizational culture of a company can inlfuence moral development (Trevino L. K., 1986). Assuming that each industry has its own culture, the type of industry could affect ethical decision making and is therefore added to the list of control variables.
4. Results
In this chapter the results of the analysis of the dataset are discussed. The data is analyzed with SPSS. The first part addresses the reliability of the dataset. The second part contains the descriptive statistics of the sample group. The third paragraph in this chapter addresses normality. In the fourth paragraph, the correlation matrix is presented followed by the regression analysis in the fifth paragraph.
5.1 Reliability analysis
To test the reliability of the questionnaire Cronbach‟s alpha was measured. Cronbach‟s alpha is used as an estimate for the internal consistency of the variables. Cronbach‟s alpha is considered
to be a measure of scale reliability. An alpha equal or above .7 is considered to be an accepted score. There are two variables for which Cronbach‟s alpha was measured:
- ethical judgment; - ethical culture.
The Cronbach‟s alpha for the variable ethical culture is .877. Ethical judgment‟s Cronbach‟s alpha is .881. Therefore, the reliability of the scales may be assumed.
5.2 Descriptive statistics
The data is collected with the online data collection software Qualtrics. Within the fourteen days of data collecting, 131 respondents started the survey and 97 finished it. The data didn‟t have to be corrected for missing values because respondents couldn‟t proceed in the survey without answering each question. Table 2 and table 2.1 contain a description of the sample. 50 of a total of 97 respondents were male and 47 were female. Their age (M=32.5; SD=12.4) ranges from 21 to 65. Their work experience (M= 1.7; SD=12.4) ranges from 1 to 47 years and their
organizational tenure (M=6.1; SD=8.5) ranges from 1 to 40 years. The majority (56.7%) of the respondents completed their study at a university.
In table 2.1 the field of study and the industry of occupation is given. Business and/or economics is the dominant study among the respondents, 35.1 % of the respondents, followed by social studies (21.6%). The biggest part of the group declared to work in the services (profit and non-profit) industry. The government (15.6%) is the second largest employer in the sample.
Table 2. Demographic of the respondents n Percentage Gender Male Female 50 47 51.5 % 48.5 % Age Mean SD 32.4 12.4 Education level HAVO VWO MBO HBO University Ba University Ma 4 4 9 25 14 41 4.1 % 4.1 % 9.3 % 25.8 % 14.4 % 42.3 % Working experience Mean SD 1.7 12.4 Organizational tenure Mean SD 6.1 8.5
Table 2.1 Demographic of the respondents n Percentage Field of study Business or economics Law Social studies Engineering Health care Humanities Other 34 6 21 11 3 7 22 35.1 % 6.2 % 21.6 % 11.3 % 3.1 % 7.2 % 15.5 % Industry of occupation Construction Education Finance/insurance/real estate Government Healthcare High tech Manufacturing
Services (non-profit & profit) Wholesale/Retail trade Other 6 3 8 15 4 2 0 26 10 22 6.3 % 3.1% 8.3 % 15.6 % 4.2 % 4.2 % 0 % 27.1 % 1.4 % 22.9 %
5.3 Normality analysis
In order to determine the normality of the variables, a normality test needs to be conducted. The independent variable age had a mean of 32.5 and a 5% trimmed mean of 31.3. Removing the top five percent scores of the top and bottom would change the mean by 1.1. According to the descriptive statistics in SPSS, the variable stress has a skewness of 1.616 and a kurtosis of 1.344. A skewness greater or smaller than zero means that the distribution deviates from a normal distribution. Kurtosis measures the peakedness of the distribution. A perfectly normal distributed model will have a kurtosis of zero. The positive skewness is an indication that the distribution‟s tail on the right side is longer than the left side. Hence, the majority of the values lie to the left of the mean (32.5). The positive kurtosis is an indication for a more peak-topped distribution than a normal distribution.
The moderator tenure had a mean of 6.1 and a 5% trimmed mean of 4.7. Removing the top five percent scores of the top and bottom would change the mean by 1.4. The variable organizational tenure has a skewness of 2.457 and a kurtosis of 5.654. The positive skewness indicates that the majority of the values lie to the left of the mean (6.1). As with age, the positive kurtosis indicates a more peak-topped distribution than a normal distribution. Both values are quite high which indicates that the data is non-normally distributed. Therefore, tenure has been recoded into the following values:
Range Value 0-2 years 2-10 years 10+ years 1 2 3
Recoding produced an acceptable skewness of .395 and a kurtosis of -.929. Normality of this distribution may be assumed.
Ethical culture had a mean of 4.93 and a 5% trimmed mean of 4.98. Removing the top five percent scores of the top and bottom would change the mean by .05 which means that the outliers don‟t have a strong influence on the mean. Ethical culture has a skewness of -.980 and a kurtosis of 1.216. The negative skewness indicates that the majority of the values lie to the right of the mean (4.93). The positive kurtosis indicates a more peak-topped distribution than a normal distribution.
The dependent variable ethical judgment had a mean of 2.4 and a 5% trimmed mean of 2.37. This indicates that the outlying scores don‟t have a strong influence on the mean. Ethical culture has a skewness of .530 and a kurtosis of -.315. The positive skewness indicates that the majority of the values lie to the left of the mean (2.4). The negative kurtosis is an indication for a more flat-topped distribution than a normal distribution.
5.4 Correlations
In this paragraph, the correlation between all variables used in this study is checked. The analysis is performed with the Pearson correlation coefficients. Prior to the analysis, several variables were recoded into dummy variables. Age and organizational tenure were recoded into three levels. The control variables gender, education level, field of study and industry of
occupation were recoded into two levels. For education level, VMBO, HAVO, VWO and MBO are considered to be lower education. HBO and university are considered to be higher education. Table 3 contains an overview of the recoded variables.
Table 3. Overview of recoded variables.
Variable Range Value
Gender Male
Female
0 1
Education level Lower
Higher
0 1
Field of study Other
Economics or business
0 1 Industry of occupation Other
Finance
0 1
The Pearson correlation coefficients are displayed in in the correlation matrix in table 4. In the correlation matrix, the correlations between the variables are presented. An asterisk (*) indicates a significance at the .05 level (2-tailed). Two asterisks (**) indicates a significance at the .01 level (1-tailed).
In the first column in the correlation matrix, the correlations for the dependent variable ethical judgment with the other variables are presented. As expected, ethical judgment shows a positive correlation with the independent variable age (r = .219; .01 < p < .05). This suggests that maturing has a positive effect on ethical judgment and possibly support for the first
hypothesis. Furthermore, ethical culture positively correlates with the independent variable (r = .380; p < .01) which indicates that ethical culture has a positive influence on an individual‟s ethical decision making. A positive correlation with ethical judgment was also found for
This result, however, is insignificant. Ethical judgment has a significant negative correlation with field of study (r = -.255; .01 < p < .05). This indicates that a background in economics or business negatively influences an individual‟s ethical decision making. Ethical judgment positively
correlates with gender, which would mean that women are more ethical than men. This result however is just insignificant (r = .198; p = .052). Furthermore, the results indicate that higher educated people (r = -.150; p = .142) and employees in the finance industry (r = -.147; p = .151) are less ethical even though there is not enough evidence to support these claims.
The second column shows a strong positive relationship between age and organizational tenure (r = .739; p < .01).
Column three presents the relationship between organizational tenure and several variables. Organizational tenure is negatively influenced by education (r = .462; p < .01). Apparently, higher educated people switch jobs/organizations more often than lower educated people. To conclude, in the fourth column the relationship between ethical culture and different variables is presented. Ethical culture shows a positive relationship (r = .218; .01 < p < .05) with education. This indicates that higher educated people work for or choose organizations with a higher ethical culture compared to lower educated people.
The strong relationship between age and work experience (r = .969; p < .01) and between age and organizational tenure (r = .739; p < .01) unsurprisingly, stand out in the correlation matrix. This could be an indication for multicollinearity. To test for multicollinearity between the two variables, the tolerance levels and Variance Inflation Factors (VIF) were checked in SPSS. The tolerance level for age and work experience is .316 and the VIF is 3.16. The VIF value is quite high compared to the other values. Therefore, considering the strong relationship and high
VIF value, work experience and age are expected to measure the same. Therefore, work experience is excluded from further analysis.
Table 4. Mean, standard definition and correlations. Mean SD 1 2 3 4 5 6 7 8 1. Ethical judgment 32.46 12.45 1 2. Age 5.69 .87 .219 * 1 3. Organizational tenure 1.63 .73 .145 .739** 1 4. Ethical culture 4.94 .78 .380 ** -.081 -.094 1 5. Gender Female = 1 .48 .50 .198 -.166 -.187 .035 1 6. Education High = 1 .82 .38 -.150 -.500** -.462 ** .218 * .067 1 7. Field of study Eco/business = 1 .35 .48 -.255* -.366** -.340 * -.111 -.064 .282 ** 1 8. Industry of occupation Finance = 1 .08 .28 -.147 -.129 -.157 -.018 .009 .138 .408 ** 1
5.5 Regression analysis
In this chapter the hypothesis are tested by conducting multiple regression analyses in SPSS. First, the main effect from this study, the relationship between age and ethical judgment, will be tested. In the second paragraph, the moderating effect of organizational tenure is
presented.
5.5.1 The main effect: The relationship between age and ethical judgment
First, a hierarchical multiple regression was performed to examine the ability of the variable age to predict levels of ethical judgment, after controlling for the control variables gender, education, field of study, industry of occupation and work experience. This means that for this hypothesis, two models are created. The first model contains the control variables and in the second model, the independent variable is included. For the first model of hierarchical multiple regression, the control variables were entered in SPSS (table 5). This resulted in the following statistics: F (4, 92) = 2.52; p < .05; R² = .099. Therefore, the model was statistically significant and explains 9.9% of the variance in ethical judgment. Only the variable field of study (ß = -.222; p < .05) significantly contributes to the explanation of the variance in ethical
judgment. For the second model, age was added. This model was statistically significant with F (5, 91) = 4.761; p < .01; R² = .164. This model explains 2.7% of the variance in ethical judgment, which means an increase of 1.9 percentage point after controlling for gender, education, field of study, industry of occupation and work experience. Furthermore, this result is significant (R² change = .109; F (1, 91) = 12.476; p < .01). Gender remains the only significant variable in the model (ß = .180; p < .05). Because of the significant outcome of the model, support is found for the first hypothesis. I.e., there is a significant positive relationship between age and ethical judgment. Table 5 and 5.1 contain the regression data for this analysis.
Table 5. Regression analysis. Dependent variable is ethical judgment.
Model 1 Model 2
Control variables Control variables + IV ß Sig. ß Sig. (Constant) .000 0 Gender .117 .244 .180 .063 Education -.117 .264 .052 .637 Field of study -.222 .048 * -.111 .319 Industry type -.026 .810 -.043 .674 Age .403 .001 **
* = significant at the .05 level (2-tailed) ** = significant at the .01 level (1-tailed)
Table 5.1. Regression analysis.
Model 1 Model 2 R² .099 .207 F 2.518 4.761 Sig. for F .047 .001 ** R² change .099 .109 F change 2.518 12.476
Sig. for R² change .047 * .001 **
* = significant at the .05 level (2-tailed) ** = significant at the .01 level (1-tailed)
5.5.2 The moderating effect of organizational tenure
For the second hypothesis, a negative relationship between organizational tenure and ethical judgment is expected. In the third hypothesis, organizational tenure is expected to
negatively moderate the relationship between age and ethical judgment. To test for a moderating effect, the interaction variable age * organizational tenure was created in SPSS. Four models of hierarchical multiple regression will be examined. Model one and two are the same as in the previous paragraph. In model three and four the variable organizational tenure and the interaction variable are added respectively. Table 6 and 7 contain the analysis.
Table 6 contains the analysis for the direct effect of organizational tenure. In this model, organizational tenure is positively related to ethical judgment (ß = .281; p = .016), and this relationship is significant. Model two is statistically significant and explains 15.4% of the variance in ethical judgment (F (5, 91) = 3.326; p < .01; R² = .154) which means an increase of 15.4 percentage point after adding the variable organizational tenure. This change of R² is significant (R²change = .154; F (1, 91) = .171; p = .016). This means that there is no statistical support for the second hypothesis. A negative relationship between organizational tenure and ethical judgment was hypothesis. The results indicate a significant positive relationship between organizational tenure and ethical judgment.
Table 6.1. Regression analysis. Dependent variable is ethical judgment.
Model 1 Model 2
Control variables Control variables + IV ß Sig. ß Sig. (Constant) .000 0 Gender .117 .244 .167 .096 Education -.117 .264 -.009 .932 Field of study -.222 .048 * -.154 .177 Industry type -.026 .810 -.025 .813 Organizational tenure .281 .016 * * = significant at the .05 level (2-tailed)
Table 6.2. Regression analysis.
Model 1 Model 2 R² .099 .154 F 2.518 3.326 Sig. for F .047 .008 ** R² change .099 .056 F change 2.518 6.008
Sig. for R² change .047 * .016 *
In table 7 the third hypothesis, the moderating effect of organizational tenure, is tested. Model four contains the interaction effect. There is a negative interaction effect (ß = .238; p = .704) as hypothesized. However, this effect is also insignificant. The fourth model is significant and explains 21.0% of the variance in the dependent variable (F (7, 89) = 3.382; p < .01; R² = .210). This means an increase of .1 percentage point after adding the interaction variable. The increase in insignificant (R²change = .001; F (1, 89) = .145; p = .704). Therefore, there is no statistical evidence for the third hypothesis. There is no significant moderating effect of organizational tenure on the relationship between age and ethical judgment.
Table 7.1. Regression analysis. Dependent variable is ethical judgment.
Model 1 Model 2 Model 3 Model 4
Control variables Control variables + IV Control variables + IV + Moderator Control variable + IV + Moderator + Interaction ß Sig. ß Sig. ß Sig. ß Sig.
(Constant) .000 0 0 0 Gender .117 .244 .180 .063 .184 .060 .186 .059 Education -.117 .264 .052 .637 .058 .599 .051 .653 Field of study -.222 .048 * -.111 .319 -.107 .340 -.097 .400 Industry type -.026 .810 -.043 .674 -.041 .689 -.043 .679 Age .403 .001 ** .365 .015 ** .505 .206 Org. tenure .059 .680 .170 .602 Age * Org. tenure -.23 .704 * = significant at the .05 level (2-tailed) ** = significant at the .01 level (1-tailed)
Table 7.2. Regression analysis.
* = significant at the .05 level (2-tailed) ** = significant at the .01 level (1-tailed)
5.5.3 The effect of ethical culture on ethical judgment
To test the fourth hypothesis („ethical culture is postively related to ethical judgment‟), a hierarchical multiple regression is performed. Similar to testing the main effect in the first paragraph, two models are created. The first model contains the control variables and in the second model, the independent variable ethical culture is included. The result of the hierarchical multiple regression is displayed in table 8. The first model is significant (F 4. 92) = 2.52; p < .05; R² = .099) and explains 9.9% of the variance in the dependent variable. The independent variable ethical culture was added for the second model. Model 2 is significant and explains 25.7% of the variance in ethical judgment F (1, 91) = 6.29; p < .01; R² = .257). This means that adding ethical culture significantly increases the explanation of the variance in ethical judgment by 15.8
percentage point (R² change = .158; F (1, 91) = 19.361; p < .01). Therefore, sufficient statistical evidence is found for the fourth hypothesis. There is a positive relationship between ethical culture and ethical judgment.
Model 1 Model 2 Model 3 Model 4
R² .099 .207 .209 .210
F 2.518 4.761 3.960 3.382
Sig. for F .047 .001 ** .001 ** .003 **
R² change .099 .109 .002 .001
F change 2.518 12.476 .171 .145
Table 8. Regression analysis. Dependent variable is ethical judgment.
Model 1 Model 2
Control variables Control variables + IV ß Sig. ß Sig. (Constant) .000 0 Gender .117 .244 .109 .236 Education -.117 .264 -.220 .026 * Field of study -.222 .048 * -.136 .195 Industry type -.026 .810 -.037 .709 Ethical culture .414 .000 **
* = significant at the .05 level (2-tailed) ** = significant at the .01 level (1-tailed) Table 8.1. Regression analysis
Model 1 Model 2 R² .099 .257 F 2.518 6.288 Sig. for F .047 .000 ** R² change .099 .158 F change 2.518 19.361
Sig. for R² change .047 * .000 **
* = significant at the .05 level (2-tailed) ** = significant at the .01 level (1-tailed) 5.5.4 The three-way interaction effect
For the final analysis, the three-way interaction effect of age, organizational tenure and ethical culture on ethical judgment is analyzed. Hypothesis 5 predicts that high values of ethical culture will positively affect organizational tenure, which in turn will positively affect ethical judgment. The analytics were performed with the SPSS plugin Process.
The results of the analysis by process are presented in table 8. The model is significant and explains 34.3% of the variance in ethical judgment (F (7, 89) = 9.1931; p < .001; R² = .343). Furthermore, the results indicate that both age (ß = .0295; p < .01) and ethical culture (ß = .4505;
p < .01) have a significant positive effect on ethical judgment. The R² change is .14 percentage point which is insignificant (R²change = .0014; F (1, 89) = .2071; p = .6501). Hence, no support is found for the fifth hypothesis. Ethical culture doesn‟t moderate the relationship between age and ethical judgment moderated by organizational tenure.
Table 8. Regression analysis performed with Process. Dependent variable is ethical judgment.
Model
IV + Moderators + 2-way interaction effects + 3- way interaction effects
ß Sig.
(Constant) .000
Organizational tenure .0948 .5903
Age .0295 .0088 **
Ethical culture .4505 .0078 **
Age * Org. tenure -.0048 .5891
Age * Ethical culture -.0129 .4976
Org. tenure * Ethical culture .0929 .6475
Age * Org. tenure * Ethical culture .0074 .6501 ** = significant at the .01 level (1-tailed)
Table 8.1. Regression analysis
Model 1 R² .343 F 9.1931 Sig. for F .0000 ** R² change .0014 F change .2071
Sig. for R² change .6501
