Moderating effect of individual resources on innovative behaviour in the work environment

(1)

Moderating effect of

individual resources on

innovative behaviour in the

work environment

Final version dissertation

Word count: 13.925

Daniella Weber, S2550075/B4067464

DD International Business Management and Marketing

(2)

Abstract

This study aims to contribute knowledge in the field of creativity and innovation literature. The focus of this research is in the area of innovative work behaviour and the interplay between individual and contextual resources. In particular, this study is concerned with the moderating effect of individual resources on the relationship between the work environment and innovative work behaviour. Such a study is important in order to understand these relations to take full advantage of internal innovation resources. The research approach adopted in this dissertation includes a survey approach conducted with an online questionnaire. This questionnaire is distributed among 105 employees from 11 companies located in the Netherlands. The findings from this study provide empirical evidence that a supportive work environment has a positive influence on innovative behaviour. In particular, management practices, organizational motivation to innovate and the psychological climate for creativity and innovation are increasing innovative behaviour. In addition, this study expected that individual resources (knowledge, thinking style, intellectual ability, personality and intrinsic motivation) would influence the relationship between the work environment and innovative work behaviour. However, such an interaction effect was not empirically confirmed. The main conclusion drawn from this study is that managers could stimulate all employees to be creative and to come up with new ideas, by creating a supportive work environment. On the long term this could be turned into a competitive advantage by using all internal sources available, instead of only investing in innovation at the R&D department.

(3)

Acknowledgements

I want any reader of this dissertation to know that I could not write this dissertation without the help of my supervisors; dr. K. van Veen and dr. H. Bahemia. Their insightful feedback provided guidance and the opportunity to improve my thesis as much as possible. Furthermore, I would like to thank all participating companies and their employees for taking the time and effort to help me finish this dissertation by filling in the questionnaire. Without them I could not have done any analysis.

(4)

1. Introduction

“We live in a society where technology is a very important force in business, in our daily lives. And all technology starts as a spark in someone’s brain. An idea of something that

didn’t exist before, that once they have invented it – brought it into existence – could change everything. And that activity is generally one that’s not very well supported.”

– Nathan Myhrvold, CEO, Intellectual Ventures (2010)

Background

Innovation is derived from the Latin verb in novus or innovare, which literally means “into new”. Thus, in its simplest form innovation means doing something different or new. (Costello and Prohaska, 2013). Innovation is defined as: “The multi-stage process whereby ideas are transformed into new or improved products, services or processes, in order to advance, compete and differentiate themselves successfully in their marketplace” (Baragheh et al., 2009. In the last decade several theories and models of innovation are developed, such as user-driven innovation (von Hippel, 1988, 2006), continuous innovation (Boer and Gertsen, 2003), open innovation (Chesbrough, 2003; Chesbrough et al., 2006; Lindegaard, 2010) and employee-driven innovation (EDI) (Høyrup, 2010). EDI is based on the concepts of workplace innovation and innovative work behaviour (IWB) (Montani et al., 2014). West and Farr (1990) defined workplace innovation as: “...The initial introduction and application within a role, group or organization of ideas, processes, products or procedures, new to the relevant unit of adoption, designed to significantly benefit the individual, the group, the organization or wider organization”. This definition is widely adopted in the field of employee innovation, and is seen as IWB on individual level (Anderson et al., 2004). Innovative behaviour is influenced by personal factors, like motivation (Amabile, 1997), and organizational factors such as psychological climate for creativity and innovation (Ren and Zhang, 2015).

(7)

engagement, improvements in change orientation, and more general interest in work improvements. Taken these advantages together, EDI can be a source of competitive advantage for an organization and vital for organizational success (Egan, 2005; Heller et al., 1998; Hoyrup, 2010; Kelley, 2010). It is therefore critical to know how to use employees as innovation capital for businesses.

Even though this importance is known in the literature, employees are often overlooked as source of innovation by businesses (Birkinshaw and Duke, 2013). In addition, it remains difficult for organizations to realize this hidden potential in the most supportive way (Amundsen et al., 2014; Birkinshaw and Duke, 2013; Kesting and Ulhoi, 2010). The Investment Theory of Creativity developed by Sternberg provides an answer on how to realize the creativity potential of individuals. This theory argues that a company has to invest in six resources (supportive work environment, knowledge, intrinsic motivation, thinking style, intellectual ability and personality) in order to create individual creativity and innovation (Sternberg et al., 1997). However, the theory does not fully describe what a supportive work environment exists of, but it rather describes that the work environment must be supportive of creativity and innovation otherwise the hidden potential of employees will not be displayed. However, the Componential Theory of Amabile (1997) addresses the work environment more deeply by defining three environmental components; management practices, available resources and organizational motivation to innovate. However, psychological climate for creativity and innovation is also an enabler of innovative work behaviour (Amabile, 2012; Kheng and Mahmood, 2013). The relationship between individual and contextual factors is not straightforward but rather a complex system of interactions between individual and contextual factors, and within these factors. This view is known as the Interactionist Perspective by Woodman et al (1993). A combination of those three theoretical lenses will be used in order to develop a conceptual model.

(8)

culture and climate (Isaksen et al., 2010; Jung et al., 2008; Patterson et al., 2005), external environment (e.g. Damanpour and Schneider, 2006; Damanpour, 2010; Wu et al., 2005), innovation diffusion (Ferly et al., 2005) and lastly corporate entrepreneurship as innovation (e.g. Kaya, 2006; Z. Zhang & Jia, 2010).

Furthermore, there are also person-level antecedents of innovation. Patterson (2002) defined intelligence, knowledge, personality, and motivation as individual factors. These factors can also be found in theories on creativity, e.g. the Componential Theory of Amabile and the Investment Theory of Sternberg. Several studies have examined the relationship between contextual factors, individual factors and innovative work behaviour (E.g. Hammond et al., 2011). Amabile (1997) argues that the social environment does influence the level and frequency of creative behaviour performed by an employee. This view is shared by Shalley et al (2009) who examined the relationship between contextual factors and creative behaviour focusing on the effect of intrinsic motivation. In their study they found that the effect of contextual factors on behaviour is a function of personal characteristics. However, research on the moderating effects of personal characteristics on the relationship between contextual factors and IWB is scarce. Only Bysted (2013) aimed to study this relationship, and found that job satisfaction and personal mental involvement positively moderates this relation.

However, the mentioned factors are often studied in isolation or with only a few related factors at the same time (Arad et al., 1997; Axtell et al., 2000; Hammond et al., 2011; Shalley et al., 2009; Smith et al., 2012). This have led to a lack of understanding how and why individual differences and contextual factors are affecting individual innovation, how the factors behave in combination, and which factor is deemed most important through the eyes of individual employees (Axtell et al., 2000; Pratoom and Savatsomboon, 2010; Shalley and Gilson, 2004; De Spiegelaere, 2015). To really understand how the work environment influences creativity and innovation it is important to consider moderators such as individual factors (Runco, 2007; Shalley et al., 2014). Specifically, in both climate and innovation studies there is still a lack of studies on multilevel effects, e.g. measuring the psychological climate as well as the organizational climate (Anderson et al., 2004; Kuenzi and Schminke, 2009).

Value of the research and its focus

(9)

factors. While, at the other hand, studies on individual level of analysis are more concerned with personal characteristics. Following the Interactionist Perspective these factors are part of a complex system of interactions. Therefore, research on the moderating effects of factors and multilevel analysis is needed to gain understanding in how and why personal and contextual factors affect individual innovation and how these factors behave in combination. This study seeks to answer research calls to study interaction effects of factors in individual innovation, multi-level analysis and an extension of Amabile’s work environment by adding psychological climate for creativity and innovation. This is therefore an area worthy of study and one that contributes knowledge in the field of EDI, climate studies and behavioural studies.

From a practical point of view, this study presents the answer in which personal and environment resources to invest as a company to increase creative and innovative behaviour of an employee. For businesses it is also interesting to know how the work environment influences innovative behaviour and how individual differences can affect this relationship. When a business manager knows where to invest he can turn his employees in a competitive advantage.

A major focus of this dissertation will concentrate on the relationship between the work environment and innovative work behaviour, and the moderating effects of personal characteristics. To gain a deeper understanding of these relationships and their effects, two main activities will be tackled: a review of relevant literature to ascertain current research findings, and empirical data collection using surveys. The literature review will examine the relationship between factors of the work environment and IWB. In addition, it will examine how personal characteristics potentially can influence this relationship. Empirical data collection will examine whether the developed hypotheses can be accepted.

Overall research aim and research objectives

(10)

“How does the work environment influence innovative work behaviour, and how does individual resources influence this relationship?”

However, in order to understand these relationships, it is felt necessary to gain insights in the moderating effect of individual resources, and in the direct relationship between the contextual factors and IWB. Besides, it is necessary to understand the difference between creativity and innovation, and its connection to behaviour. In turn, two main research methods will be exploited to facilitate this research: an in-depth and critical review of the literature and the collection and appropriate analysis of empirical data. Chapter 3 entitled Research Methodology contains a detailed description of the research strategy and data collection techniques used to obtain the empirical data.

The objectives of the research are to:

1. Identify the difference between creativity and innovation in the workplace.

2. Evaluate critically which contextual factors are enhancing creativity and innovation in a work environment.

3. Clarify how personal characteristics, in terms of knowledge, intellectual ability, thinking styles, motivation and personality, are influencing the relationship between the work environment and innovative work behaviour.

4. Study the work environment using multi-level analysis, i.e. individual level and organizational level.

5. Formulate recommendations on further research in the field of employee-driven innovation.

Structure of the dissertation

(11)

(12)

2. Literature review

Chapter introduction

This chapter identifies the difference between creativity and innovation, evaluate the critical factors for innovative work behaviour, and clarify the moderating effect of personal characteristics on the relationship between contextual factors and IWB. This literature review focuses on objectives 1, 2 and 3 as set-out in sub-section 1.3 of the introduction. This will lead to a significant contribution in this dissertation by critically reviewing existing literature in the field of EDI. At the end of this chapter a critical understanding of key issues is revealed, and a clear focus and justification for empirical research is emerged.

Distinguishing creativity and innovation

(13)

role, group of organization, in order to benefit role performance, the group or the organization (Janssen, 2000; West and Farr, 1990). In short, it is known as behaviour that stimulates the introduction and development of innovations at the workplace (De Spiegelaere et al., 2014). As mentioned earlier, innovation can be distinguished from creativity since it encompasses different phases. The workplace innovation process can be divided in three phases; idea generation, idea promotion and idea implementation. In the innovation process an employee generate creative and innovative ideas, seek support for these ideas from co-workers and supervisors, and implement or realize the ideas in the workplace (De Spiegelaere et al., 2014; Ng and Feldman, 2010; Scott and Bruce, 1994). A visual explanation of IWB is shown in figure 2.1.

Figure 2.1 Visual explanation of IWB

Contextual and individual factors

2.3.1 Theoretical lens

(14)

creativity to organizational innovation where creativity is the first step towards innovation. This is in line with the definition of IWB given earlier. In addition, the Componential Theory provides a foundation for other theories and empirical investigations (e.g. McAdam and McClelland, 2002; Amabile et al., 2004; Hirst at al., 2009; Wang and Tsai, 2014). Contrary, a main shortcoming is that the focus of the theory is on factors within an organization. It does not include outside forces. In addition, the theory does not include the psychological climate for creativity and innovation, which also has shown to have an influence on creativity and innovative behaviour (Amabile, 2012).

Sternberg and Lubart (1992) conceptualize two approaches within the creativity research field: centred approaches and context-centred approaches. The person-centred approach looks at the individual differences, and how this affect creativity. The context-centred approach takes environmental factors in account. The Investment Theory of Creativity by Sternberg et al (1997) does combine these two research approaches. This theory argues that six resources are necessary to increase creativity within an organization. The six resources can be divided in five individual resources and one contextual resource. The five individual resources are knowledge, intellectual ability, thinking style, motivation and personality. The contextual resource is the work environment. These resources are brought together using an investment perspective – ‘buy low and sell high’. This means that people invest in ideas that are new, slightly out of favour or unknown. To earn a creative return on this investment the six resources are needed. If a company only invests in one resource, the probability of creativity to take place is low. Two studies by Sternberg and Lubart (1995) and one study by Zhang and Sternberg (2011) have been conducted to directly test the Investment Theory of Creativity. A limitation of these studies is that all of them did not take the contextual resource in account specified in the Investment Theory.

(15)

extension of Amabile’s componential theory (Sternberg, 1999; Amabile and Pillemer, 2011). Third, the Componential Theory of Creativity and Innovation specifies three components of the work environment that are important for fostering creativity and innovation. The Investment Theory only states that the environment must be supportive, and risk spreading. Therefore, the Componential Theory can be better used in assessing the work environment.

In sum, both theories might be helpful in understanding which factors play a role in enhancing IWB. Since the Investment Theory is an extension of the Componential Theory, this will be the leading theoretical lens in this study. All six resources will be part of the conceptual model, because only by investing in all resources creativity and innovation is more likely to take place. The Componential Theory will be used to derive to which factors must be included in assessing the work environment. A perspective that relates to but also expands the componential and investment theory is the interactionist perspective on creativity from Woodman et al (1993). The perspective argues that behaviour is a complex interaction of person and situation. To really understand how the work environment influences creativity and innovation it is important to consider moderators such as individual factors (Runco, 2007).

Figure 2.2 Conceptual model

2.3.2 The work environment and IWB

(16)

environment can obtain stimulants as well as obstacles to creativity and innovation. For example, political problems within the organization can block creativity and innovation, while some other factors such as work autonomy can stimulate creativity and innovation (Amabile, 2012; Shalley et al., 2000). More specifically, the work environment can provide opportunities for acquiring expertise and developing new ideas (Mumford, 2000). Following the Investment Theory, the right environment displays individual creativity and IWB in the organization. A supportive work environment has therefore a direct postive influence on creativity and IWB. This positive direct relationship is confirmed by multiple other studies (Amabile, 2012; Hammond et al., 2011; Shalley et al., 2000; Shalley and Gilson, 2004).

Hypothesis 1a: A supportive work environment has a positive relationship with IWB.

In order to display the hidden capability of an employee, the environment must be supportive and rewarding of creative ideas (Zhang and Sternberg, 2011). Amabile (1997) defined three major components of the work environment that enhance creativity and innovation. These components are organizational motivation to innovate, resources and management practices. These components exist out of several factors as can be seen in table 2.1.

Work environment component Related factors Level of analysis

Organizational motivation to innovate

Challenging work and tasks Free and open communication Trust

Supervisory encouragement and support

(17)

Autonomy

Resources Sufficient resources (money, people, space) Time pressure

Organizational level

Management practice Organizational encouragement Risk orientation

Rewards and recognition

Organizational level

Psychological climate for creativity and innovation

Factors from meta-analysis Hunter (2007): Mission clarity

Positive interpersonal exchange Intellectual stimulation

Top management support Flexibility and risk-taking Product emphasis Participation

Organizational integration

Individual level

Table 2.1 Factors in the work environment

(18)

opportunity to schedule their own work, make job-relevant decisions, and determine the most appropriate method for own work (Meyer et al., 2010). Giving employees autonomy results in learning, development of new skills, felt responsibility, and a broader ownership of problems and ideas (De Jong and Kemp, 2003; Hammond, 2011; Krause, 2004; Meyer et al., 2010), which in turn leads to new suggestions and implementations of ideas (Parker et al., 1997). Contrary, it is expected that too much autonomy can erode decision structures and authority (Smith et al., 2012).

All factors, i.e. challenging job, free and open communication, trust, autonomy and positive supervisor relations, concerned within organizational motivation to innovate showed to be positively related to individual creativity and innovation in previous studies. Only autonomy showed that it could also have some negative effects when there is too much autonomy. Overall, the discussion suggests that there is a positive relation between this component of the work environment and IWB.

Hypothesis 1b: Organizational motivation to innovate has a positive relationship to IWB.

Resources facilitate, encourage and implement creative ideas (Hunter et al., 2007). Supervisors and management distribute available resources like materials, people, time and space (Amabile, 1997; Janssen, 2005; Wang et al., 2015). Time pressure can have both positive and negative effects on IWB (Wu et al., 2014). At one hand, time pressure can impair innovation since it exhausts employees and since there is no time left to search for alternatives and explore different perspectives (Amabile and Gryskiewicz, 1987; De Spiegelaere et al., 2015; Shalley and Gilson, 2004). At the other, time pressure can provide motivation to seek new ways of efficient working processes. It also can result in high levels of activation and proactive behaviour (Hammond et al., 2011; Ohly and Frits, 2010; Sonnentag, 2003). Likewise, the other resources can have a double effect on innovation. A lack of material resources, space and people can foster creativity because it may stretch employees to perform their tasks in a different way. However, resources are needed to encourage creativity and innovation (Wu et al., 2014). The study of Scott and Bruce (1997) is the only study that did not found a positive relation between resources and IWB. Following the majority of research outcomes in the field, this study hypothesize that sufficient resources positively influences IWB.

(19)

Hypothesis 1c: Available resources have a positive relationship to IWB.

Management practice concerns management at the level of departments and individual projects (Amabile, 1997). Creative requirements and expectations for creative behaviour at work are related to creative work involvement (Carmeli and Schaubroeck, 2007; Hammond et al., 2011; Unsworth et al., 2005). Also setting (project) goals lead to increased attention and clear targets (Shalley and Gilson, 2004). This results in more aspired employees to be innovative and more proactive (Parker et al., 2006). In addition, a manager or supervisor can provide employees with helpful and valuable feedback that enables them to develop, learn, and make improvements on the job (Zhou, 2003). Feedback is, therefore, an opportunity for employees to learn, which in turn enables an employee to try new things and to come up with new ideas (Zhou and Li, 2013).

Hypothesis 1d: Management practices are positively related to IWB.

In addition to these components, the psychological climate for creativity and innovation also influences IWB (Amabile, 2012; Hunter et al., 2007; Kheng and Mahmood, 2013; Shalley et al., 2000). The work climate can be assessed at different levels of analysis; the individual level, the group level and the organizational level (Schneider, 2000). The individual perception of the work environment is known as the psychological climate, and this climate is the individuals’ perception of, or experiences in, their immediate work environment (Hunter et al., 2007; Kuenzi and Schminke, 2009; Montani et al., 2014).

(20)

involved in various functions and increase proactive problem solving and commitment. This in turn enhances IWB (Axtell, 2000; Campbell, 2000; Martinez-Sánchez and Pérez, 2003, Martinez-Sánchez et al., 2008, Parker et al., 2006). Cooperative teams enable open communication, information sharing and high qualitative relationships characterized by mutual respect and trust (Zeng et al., 2015). This will stimulate creative performance and productivity (Clegg et al., 2002; Munoz-Doyague and Nieto, 2011; Parker et al., 2006; Schniederjans and Schniederjans, 2015)

Also management is able to create an innovative climate by creating a vision that encourages innovation and by revealing a positive attitude towards change (Arad et al., 1997; Thamhain, 2003; McDonald, 2007). This orientation towards innovation can be supported by the quality orientation (Moreno et al., 2011). Contrary, high quality can reduce variation, and therefore, the need to explore new opportunities (Benner and Tushman, 2003; Kyriakopoulos and Moordman, 2004). In addition, employees perform better when they are involved in setting future organizational objectives and decision-making, because it fosters commitment and proactive behaviour (Eduardo, 2015; Rasulzada and Dackert, 2009). Psychological climate theory would hence suggest that, when perceiving a company’s norms that encourages and favour creativity and innovation-oriented endeavours, employees will be more likely to attribute value and meaning to change and innovation (Montani et al., 2014). As results, they will be more willing to invest their efforts in performing creative and innovative behaviour.

Hypothesis 1e: The presence of a psychological climate for creativity and innovation has a

positive relation to IWB.

The moderating effect of individual resources

According the Investment Theory, creative people are willing and able to invest in new and unknown ideas (Sternberg and Lubart, 1995). Individual differences in thinking style, intellectual ability, knowledge, personality and internal motivation impacts the creativity performed by an individual (Zhang and Sternberg, 2011).

(21)

2007). This personal factor is interrelated with an individuals’ thinking style. A person needs to do things in their own way (legislative style), a person needs to see the holistic situation in order to identify new ideas (global style), and a person needs to think in a new, out-of-the box way (Zhang and Sternberg, 2011). An individual can only do this when he is intelligence enough. A creative relevant skill is divergent thinking (Amabile, 1988; Runco, 2007). The third creative resource, knowledge, is necessary for creative behaviour. New and creative ideas derive from past knowledge and experience of the individual. However, when a person is too knowledgeable he can be absorbed in seeing a situation in only one way (Runcorn, 2007; Zhang and Sternberg, 2011). In addition, Patterson (2002) argues that knowledge is necessary bit not sufficient for innovation to occur. Fourth, personality influences creativity. A creative person is willing to overcome obstacles, to take reasonable risk, and to learn (Sternberg and Lubart, 1995). Personality is studied much in the literature and showed to have an effect on creativity outcomes (e.g. Silvia et al., 2009; Thompson et al., 2009). At last but not least, an individual must be intrinsically motivated to perform creative behaviour. The important premise of this resource is that people love what they do, and are not fully focuses on potential rewards they get from it (Amabile, 1997; Zhang and Sternberg, 2011).

(22)

work environment is present (Shalley et al., 2004). It is therefore expected that persons with creative characteristics strengthen the relationship between the work environment and IWB.

(23)

3. Research methodology

Chapter introduction

This chapter clarifies how empirical research takes place in this dissertation. The empirical research is needed to study the work environment using multi-level analysis; individual-level and organizational-level, and to explore the moderating effect of individual characteristics. Empirical research focuses on objectives 4 and 5 as set-out in sub-section 1.3 of the introduction.

Chapter 2 identified a gap in existing literature. The study and analysis of empirical data is an important contribution of this dissertation. This chapter provides the details of the research strategy adopted, together with the means of collecting for analysis, including company and sample selection, and the statistical approach to be adopted. In addition, this chapter addresses ethical issues, potential limitations and problems occurring with the chosen research strategy and its implementation.

Overall research strategy

3.2.1 Research philosophy and approach

In business studies there are two main research philosophies: pragmatism, positivism, realism and interpretivism (Saunders et al., 2012). The philosophical view adopted in this study is the positivist philosophy to the development of knowledge. The ontology, epistemology, axiology and methodology most often used are explained in table 3.1. This philosophy fits this research best since existing theories are used to develop hypotheses. Following this view, a highly structured methodology is used in order to facilitate replication. This methodology will be described in the remaining part of this chapter.

Ontology Epistemology Axiology Methodology

Objective, external and independent of social actors

Observable phenomena provide credible data and facts

Researcher maintains objective stance and is independent of data.

Highly structured Large samples Measurements Quantitative

(24)

This research identified a gap in the existing literature to generate testable hypotheses. In addition, a conceptual framework is developed in the literature review. This means that the research approach adopted in this study is abduction. In the abductive approach existing theories are modified and tested with data (Saunders et al., 2012). This study does that by adding psychological climate for innovation to the work environment, and by testing the moderating effect of individual factors instead of the direct effect on IWB.

3.2.2 Research design

This study follows a quantitative research design since this is associated with positivism. Adopting a quantitative approach enables to measure variables numerically and to analyse the data using statistical techniques (Saunders et al., 2012). This study aims to examine relationships between variables, which indicates that the research is explanatory in nature.

There are multiple research strategies available for collecting the data such as a case study, archival research, ethnography, action research, experiments and a survey. The first four named strategies are not appropriate since they include in-depth and qualitative research designs (Biggam, 2011). An experiment is concerned with whether there is a link between multiple variables. However, this strategy is conducted in laboratories rather than in the field. Another research strategy is the survey research. This research strategy allows to examine particular relationship between variables, and to produce a model of these relationships (Saunders et al., 2012). Since this study adopted a positivist approach with a quantitative research design, a survey research is most appropriate. The advantages of the survey approach are that data is collected in an economical way, allowing easy comparison. In addition, it is possible to use a sample to generate findings over the whole population. However, the collection technique is relatively time consuming in the designing phase and a limited amount of questions can be generated (Biggam, 2011; Saunders et al., 2012).

Data collection

3.3.1 Data collection technique: Online questionnaire

(25)

questionnaires. Respondents completed the questionnaire and then sent it to the researcher. Questionnaires are often used for explanatory research and enable to examine and explain relationship between variables by collecting precise data to test hypotheses. The main attributes and advantages of this instrument is that the size of the sample can increase due to easy distribution, an average response rate between 30% and 50% is likely, and the length can be between 6-8 pages with closed questions. However, it takes around two until six weeks to receive completed questionnaires (Saunders et al., 2012). However, since not all employees on the work floor did have an email, the same questionnaire was also distributed by hand to make sure their views are not missing in this research. The questionnaire contained close-ended questions to encourage more specific and precise answers. This is needed for the analysis using SPSS to test a theory and the formulated hypotheses (Biggam, 2011; Saunders et al., 2012).

In addition, the questionnaire was available in Dutch and English, since the questionnaire is distributed in the Netherlands. The questionnaire was directly translated by three people whose native language is Dutch. Appendix A contains a copy of the questionnaire used in this study.

3.3.2 Population and sample

(26)

Company Branch International

focus Nr. of employees asked Nr. of participants Return rate

Company A Electrical equipment

Yes 14 12 85.7%

Company B Financial Business services No 12 8 66.7% Company C Wholesaler No 15 7 46.7% Company D Electrical equipment Yes 25 9 36%

Company E Financial Business services

Yes 10 4 40%

Company F Financial Business services Yes 45 14 31.1% Company G IT Business services No 25 12 48% Company H Governmental sector No 20 10 50% Company I Hospitality industry No 15 7 46.7%

Company J Healthcare sector No 13 13 100%

Company K Logistics Yes 30 9 30%

Total - - 224 105 46.9%

Table 3.2 Participating companies

Probability sampling techniques are often associated with survey research strategies (Saunders et al., 2012). Systematic random sampling was used to select the employees without management function of the participating companies. This means that the results are more generalizable because the results are randomly taken from employees without management function. This study did not focus on different levels and functions of the employees (Biggam, 2011). This research is conducted among 54 males and 51 females. The total response rate in this study is 85.7% with 15 questionnaires filled in partially.

3.3.3 Variables and their measurements

In designing the questionnaire secondary sources are used in developing the scales. Other researchers developed questions for the same variables with sufficient internal reliabilities. In addition, using existing question formats allowed timesaving in developing a reliable questionnaire and comparability with other studies (Saunders et al., 2012).

(27)

argues that the construct is multidimensional (e.g. Battistelli et al., 2013; De Jong and Den Hartog, 2010; Leung et al., 2011). This means that there is no empirical evidence to make a distinction between different stages of the workplace innovation.

In addition, IWB can be rated in two ways; it can be self-rated or rated by others such as managers and co-workers (De Jong and Den Hartog, 2010). Most studies (60%) in the field have used supervisory ratings (Anderson et al., 2004). However, supervisor’s ratings might be biased due to their overall holistic view of the performance level and capabilities of particular employees (De Jong and Den Hartog, 2011). In developing a measurement of IWB the focus of type of rating has been shifted to self-ratings since 2002 as displayed in table 3.3. This study adopted the nine-item measurement (α = 0.95) of Janssen (2000) developed for a sample of employees, which is also the case in this study. This means that this study will include self-ratings of IWB. Employees have more information about the contextual background of his/her own work activities. In addition, using supervisory ratings may miss genuine innovative activities of the employee, and may only capture those behaviours to impress the supervisor (Janssen, 2000). The response format of this measurement is a 5-point Likert scale ranging from never (1) to always (5). IWB is an ordinal variable, since the construct is divided in categories with a specific order.

Study Items and dimensions Sample Rating type Reliability

Scott and Bruce, 1994

6, one-dimensional R&D management Other ratings α = 0.89

Bruce and West, 1995

5, one-dimensional Employees Self-ratings α = 0.75 and α = 0.80

Spreitzer, 1995 4, one-dimensional Management and employees

Other ratings α = 0.91

Basu and Green, 1997

4, one-dimensional Management Other ratings α = 0.93

Scott and Bruce, 1998

4, one-dimensional R&D management Other ratings α = 0.86 and α = 0.84

Janssen, 2000 9, (three dimensions) Employees Self-ratings α = 0.95

(28)

2001

Krause, 2004 8 (five on creativity and three on implementation)

Middle management Self-ratings α = 0.78 and α = 0.81

Dorenbosch et al., 2005

16 (ten on creativity and six on implementation)

Employees Self-ratings α = 0.92

Reuvers et al., 2008 4, one-dimensional Management and employees

Self-ratings α = 0.86

De Jong and Den Hartog, 2010

10 items (four dimensions)

Management Other ratings α = 0.93

Table 3.3 Previous measurements of IWB

This study contains multiple independent variables. These variables are expected to predict the outcome (Field, 2013). In this study the independent variables reflect the contextual factors of the work environment. There are multiple measurement instruments of the work environment; (1) the Siegel Scale of Support for Innovation (SSSI; Siegel and Kaemmerer, 1978), (2) KEYS (Amabile et al., 1996), (3) the Creative Climate Questionnaire (CCQ; Ekvall, 1996), (4) the Team Climate Inventory (TCI; Anderson and West, 1998), and (5) the Situational Outlook Questionnaire (SOQ; Isaksen et al., 1999). This study adopts the KEYS instrument of Amabile et al (1996), because Amabile’s theory is used in this study. KEYS measure the work environment for creativity and innovation on organizational level. The internal reliability of the instrument is high with a median Cronbach alpha of 0.84 (Mathisen and Einarsen, 2004). Table 3.4 shows the internal reliability of each construct and the amount of items used in the questionnaire. The response format of the items is a 7-point Likert scale ranging from strongly disagree (1) to strongly agree (7).

Construct Dimensions Amount of

items Internal reliability Organizational motivation to innovate Organizational encouragement Organizational impediments Three items Four items α = 0.80

Resources Sufficient resources

Workload pressure

Four items

Three items

α = 0.75

Management practices Challenging work

Work group support

(29)

Three items

Table 3.4 Internal reliabilities and amount of items per work environment construct

In addition, psychological climate for creativity and innovation is added in this study. However, the KEYS instrument does not provide measurements for this construct since it is not included in the Componential Theory. The dimensions of the psychological climate for creativity and innovation are mission clarity, positive interpersonal exchange, intellectual stimulation, top management support, flexibility and risk-taking, product emphasis, participation and organizational integration. There is no measurement instrument including all variables. Therefor items are derived from instruments such as the Organizational Climate Measure (OCM) of Patterson et al (2005), and the Organizational Change Questionnaire (OCQ) of Bouckenooghe et al (2009). The aim was to adopt as much as possible items from the same instrument to reduce the change that items and constructs are overlapping. The response format of the items is a 7-point Likert scale ranging from strongly disagree (1) to strongly agree (7). Table 3.5 contains the background information about the psychological climate for creativity and innovation measurements.

Construct Measurement instrument Reliability Amount of items

Mission clarity OCM α = 0.87 Three items

Positive interpersonal exchange OCQ α = 0.74 Four items Intellectual stimulation Loon et al (2012) α = 0.71 Three items

Top management support OCQ α = 0.82 Four items

Flexibility OCM α = 0.86 Four items

Risk-taking Garcia (2015) α = 0.83 Four items

Product emphasis OCM α = 0.80 Three items

Participation OCM α = 0.87 Four items

Organizational integration OCM α = 0.86 Four items

(30)

At last but not least, individual resources are the moderator variables. Knowledge is measured using items on educational level, age and job tenure. These constructs reflect domain expertise or knowledge (Amabile, 1988; Hammond et al., 2011; Janssen 2000; Wang et al., 2015; Kark and Carmeli, 2009; Shalley and Gilson, 2004; Tierney, 1999).

Often studies are including these variables as controls (e.g. Dul, 2011; Janssen, 2000), but in this study they have a moderating function. Intrinsic motivation is measured using the items of Tierney et al (1999). The items had an internal reliability of 0.77. Thinking style is measured with the Thinking Style Inventory of Sternberg (1999). Three styles are measured; legislative (α = 0.65), global (α =0.59) and liberal (α = 0.80). Personality can be measured using extensive personality instruments. However, to keep the length of the questionnaire as short as possible a 10-item version of the Big Five personality characteristics is used. The extended version gives a higher reliability, but it is impossible to use 44 items on measuring one construct in the current context due to time limit and the length of the questionnaire (Rammstedt and John, 2007). The Cronbach alpha of the short personality measure is 0.44. Lastly, intellectual ability will be measured using self-efficacy items from Tierney and Farmer (2002). The internal reliability of their items are α = 0.83 and α = 0.87 for manufacturing and operations respectively.

Consistent with previous research, this study will include gender as control variable. Previous studies argue that females are rated lower than males in displaying creativity and innovative behaviour (e.g. Dul, 2011; Janssen, 2000; Wang et al., 2015; Rasulzada and Dackert, 2009).

3.3.4 Pilot study

The questionnaire is pilot tested to test the internal reliability of the test before sending the online questionnaire out. It also enables to see whether instructions, questions and scale items are clear (Pallant, 2010; Saunders et al., 2012). The pilot study is conducted among twenty employees without management function. As can be concluded from table 3.6 there is an internal consistency among all items. However, some variables showed a lower alpha than expected. Pallant (2010) states when there are short scales (less than ten items) it is sometimes more appropriate to look at the mean of inter-item correlation with the optimal range of .2 to .4. All dependent and independent variables showed an acceptable or good reliability, and are included in the main study.

Variable Internal consistency in original study

Internal consistency in pilot study

(31)

Innovative work behaviour α = 0.95 α = 0.72 0.225 Acceptable Resources α = 0.75 α = 0.74 0.293 Acceptable Management practices α = 0.77 α = 0.85 0.346 Good Organizational motivation α = 0.80 α = 0.65 0.214 Acceptable

Mission clarity α = 0.87 α = 0.98 0.962 Good

Intellectual stimulation α = 0.71 α = 0.88 0.719 Good Top management support α = 0.82 α = 0.64 0.319 Acceptable

Risk taking α = 0.83 α = 0.70 0.373 Acceptable

Organizational integration

α = 0.86 α = 0.66 0.363 Acceptable

Flexibility α = 0.86 α = 0.57 0.289 Acceptable

Product emphasis α = 0.80 α = 0.72 0.493 Acceptable

Positive interpersonal exchange

α = 0.74 α = 0.64 0.319 Acceptable

Participation α = 0.87 α = 0.84 0.592 Good

Creative ability α = 0.83 α = 0.75 0.519 Acceptable

Table 3.6 Results from pilot test

3.3.5 Validity and reliability

A valid research is one where appropriate choices are made regarding the research strategy, data collection and data analysis techniques (Biggam, 2011; Field, 2013). This research methodology took the research aims in account in developing the research strategy. The appropriateness of the choices made are discussed in the previous paragraphs. However, there is chosen for self-report measures in this study, which can affect the content validity (Field, 2013). However, to ensure a better validity this study used existing measurements of the variables. This study design also ensures ecological validity because the research is not influencing what happens and the researcher does not bias the measures.

(32)

participating companies and the questions asked are added in the appendices to increase the reliability by being as transparent as possible (Biggam, 2011). At last but not least, to increase the reliability of the used scales, a pilot study is conducted on internal consistency as mentioned earlier.

Framework for data analysis

Since this is a quantitative study, statistical analyses are involved. SPSS will be used to analyse and describe the data, collected with the online questionnaire. SPSS is often used to present quantitative data in managerial form and enables statistical analyses (Biggam, 2011; Field, 2013). Before analysing the results, incomplete responses of which more than 5 percent of the questions were unanswered were deleted, which resulted in a total of 95 included responses.

First of all, a codebook is developed with information about the variables and their coding instructions and scales of measurements. After data entry, the total scale scores are calculated after negatively worded items are reversed. Appendix B contains the codebook and questionnaire. Before conducting any statistical tests to test the hypotheses, descriptive statistics and factor analysis are performed on the data. These statistics will be used to test the assumptions and choose the right test. At last but not least, the items of the scales are added up to receive an overall scale score. This research is interested in the strength of the relationship between variables. The hypotheses will be tested with a multiple regression analyses in the main effect model and hierarchical regression analysis in the moderation model, which allows to study not only the value of the dependent variable based on one or more independent variables (Verma, 2013), but also the interaction effect of a third variable (Field, 2013).

Figure 3.1 illustrates graphically the approach that is adopted to analyse data.

Figure 3.1 Data analysis approach

(33)

Ethical issues, limitations and potential problems

3.5.1 Limitations and potential problems

Using an online questionnaire contains some limitations. First of all, there is no physical control on how and by whom the questionnaire is filled in. To achieve the right sample, this questionnaire was only send to employees without management function. Furthermore, the introduction of the questionnaire contained the characteristics of the required respondents for this study. However, this survey contained a small sample size due to time limitations, which affects the generalizability of the results. Another disadvantage of the online questionnaire is that no interview intervention was available to explain some questions.

In addition, there are some limitations coupled to the chosen measurement of variables in this study. First of all, the items of Janssen (2000) adopted in this research are measuring IWB based on self-ratings. This implies that interpretation of data must be done cautious since they may suffer systematic biases related to the respondent context (Prieto and Perez-Santana, 2014). However, using self-ratings also have positive influences on the measurement scale as discussed earlier in this chapter. Another limitation in the measurements chosen is the measurement of intellectual ability. Studies on intellectual ability are extensive in length and time, and complex in nature. Due to practical reasons such as time and length limitations this measurement scale is not included in this study. Instead, this study uses the self-efficacy scale. Self-efficacy is the perceived ability to produce creative outcomes, and not the actual ability to produce creative outcomes. Multiple studies used this measurement in relation to creativity and innovation (e.g. Carmeli and Schaubroeck, 2007). To see how intellectual ability of an individual influence the proposed relationships, extensive intellectual tests should be performed.

(34)

The results of this study do not provide insights in the effects of IWB on the work environment or individual resources.

3.5.2 Ethical issues

(35)

4. Survey findings: Description, Analysis and Discussion

Chapter introduction

This chapter reveals the results of the survey described in Chapter 3. The research concentrates on employees without any management function. The survey is approached in a structured way. First, a simple description of the results is provided using a correlation matrix. The gathering of empirical data for this study is based on a questionnaire distributed by email and by hand. This allows analysis of precise data to accept or reject the set hypotheses based on the literature review. Summaries of the main findings and the associated statistical procedures are given to describe the results. This analysis of data is followed by a discussion on what is found. Finally, an integrative analysis of the empirical data against the literature review findings is provided.

Descriptives and factor analysis

This study uses exploratory factor analysis in order to determine if a series of dimensions or factors exist in the data. This will be done following the steps of factor analysis as described by Malhotra (2006).

4.2.1 Problem formulation

(36)

4.2.2 Correlation matrix

Means, standard deviations, correlations, and reliabilities among all variables are presented in table 4.1. The relationship between the main variables was investigated using Pearson correlation coefficient. The correlation matrix reveals correlation among the variables indicating latent underlying variables. For example, correlation between the dimensions of IWB are .799 between idea generation and promotion, .683 between idea generation and idea realization, .708 between idea promotion and idea realization.

Table 4.1 Correlation matrix

(37)

Table 4.2 Kaiser-Meyer-Olkin measure and significance

4.2.3 Number of factors

The second step of the analysis is deciding on the number of factors to extract. EFA revealed the presence of four factors with initial eigenvalues exceeding 1, explaining 43%, 14%, 9.8% and 9.7% of the variance respectively, shown in table 4.3. However, only using this method can lead to arbitrary decisions since it can reject factors with an Eigenvalue of 0.99 (Ledesma and Pedro, 2007). Therefore, a scree test is used in addition. An inspection of the Scree Plot in figure 4.1 reveals a clear break (inflection point) after the fifth factor implying that four factors must be retained. This is in line with the Eigenvalue criterion. Retaining four factors results in 64.789% cumulative variance meeting the threshold of 60% (Malhotra, 2006). A four-factor solution is, therefore, the most appropriate outcome.

(38)

Figure 4.1 Scree Plot

4.2.4 Rotate and interpret factors

To aid the interpretation of four factors, oblique rotation with promax procedure was performed. This method allows for correlations between factors. The Investment Theory stated that the six resources are correlated to each other, which is reason to believe that the dimensions will be correlated with one another. The Pattern Matrix containing loadings of all values and the Structure Matrix are included in Appendix C. A version of the Pattern Matrix showing coefficients larger than 0.4 can be found in table 4.4.

The rotated solution revealed the presence of simple structure, with all factors showing a number of strong loadings and all variables loading substantially on only one component. In conclusion, this table is free from any cross-loadings. In addition, all loadings are greater than 0.4. This indicates reliable items that should be included in any further analysis.

(39)

factor show that individual resources are multidimensional. The Investment Theory argues that there are five resources, but the factor analysis reveals only two dimensions, which will be used in further analysis.

Table 4.4 Pattern Matrix with loading >.4

4.2.5 Reliability of factor analysis

Finally, to determine the reliability of the used factor analysis a quick look is given into the reliability using Cronbach’s alpha. The results are displayed in table 4.5. The dependent variable, the work environment, and individual resources are highly reliable. Knowledge has a lower Cronbach’s alpha, which is not reliable. However, following the Investment Theory this variable remains included. Since the other items have high reliabilities, summated scales are developed for further analysis.

Factor Cronbach’s Alpha Number of items

1. Innovative work behaviour 0.896 9

2. Work environment 0.938 33

3. Individual resources 0.849 30

4. Knowledge 0.529 3

Table 4.5 Reliability analysis on factorized variables

4.2.6 Adjusted correlation matrix

(40)

reveals that all defined control variables (gender, branch, company and international focus) are significant correlated to IWB. However, IWB is not significant correlated with knowledge. These results preliminary confirm hypothesis 1a; there is a positive correlation between the work environment and innovative behaviour. However, the matrix does not reveal the direction of the relationship. This will be done using multiple regression analysis in the next section of this chapter.

Furthermore, work environment and individual resources are also significantly (r=.452, p<0.01) related to each other. This could imply multicollinearity, but since the coefficient is below 0.5 this is not a problem in further analysis. A final note on the correlation matrix is that a high correlation (r=.898, p=.000) between branch and company exists. Since regression analyses is extremely sensitive to multicollinearity (Field, 2013), one of the variables will be dropped in further analysis. Since branch is better correlated (r=-.321, p=.002) to IWB, this variable will be included as control variable.

Table 4.6 Adjusted correlation matrix

Main effect model

(41)

and the related hypotheses. Analyses include controlling for gender, international focus and branch.

Looking first at the correlation matrix in table 4.6, a significant positive correlation (r=.404, p=.000) between the work environment and IWB was found, which is in line with hypothesis 1a. For the remaining hypothesis 1-series, a look is given on the correlation matrix in table 4.1. This matrix reveals a significant positive correlation between organizational motivation (r=.535, p=.000; r=.378, p=.000; r =.296, p=.004), management practices (r=.344, p=.001; r=.369, p=.000; r=.305, p=.003), psychological climate (r=.392; r=.391; r=.394; p=.000 for all coefficients), and all three IWB dimensions. These results are in line with hypotheses 1c, 1d and 1e. Available resources only significantly correlates with one dimension, idea generation, of IWB (r=.217, p=.035). This is partial in line with hypothesis 1b, which predicted that available resources would positively influences IWB. In order to take a closer look at the relationships and to include control variables, regression analyses were performed with IWB (one-dimensional) as dependent variable. Table 4.7 contains all models paired to the hypotheses.

Model 1 only contains the control variables and reveals that gender has significant negative effect (B=-.352, p<.05) on innovative work behaviour; no other significant relationship is observed. This implies that an individual’s gender influences the amount of innovative work behaviour performed. The overall model is significant (p<.01) with a F-value of 5.62. The control variables explain 16.2% of the total variance in the model. This finding shows that creativity and innovation is demographically discriminated based on gender.

(42)

therefore, provides support for the basic premise of the Componential Theory of Creativity and Innovation, and the Investment Theory.

Hypothesis 1b stated that organizational motivation to innovate positively influences IWB. Model 3 is significant (p<0.05) with a F-value of 6.703. The model explains 23.8% (R Square=.238) of the variance explained by organizational motivation. Furthermore, the influence of organizational motivation on IWB is statistically significant at 1% level with an unstandardized regression coefficient of .228. This implies that an increase in organizational motivation to innovate will lead to higher IWB performed, supporting the basic premise of the Componential Theory of Creativity and Innovation. Since the results are significant, it can be concluded that there is enough evidence to accept H1b. Hence, hypothesis 1b is accepted.

Hypothesis 1c predicts that the available resources in an organization positively influences IWB. This relationship is tested in model 4. The overall model is significant (p<.01) with a F-value of 4.279. Moreover, the model shows a R Square of .166, implying that 16.6% of the variance is explained by the model. However, the results show that available resources does not have a statistically significant effect (B=.063, p>0.05) on IWB, and challenge the basic premise of the Componential Theory of Creativity and Innovation. This implies that an increase in available resources do not significantly influences performed innovative work behaviour. Since this result is not significant, it can be concluded that there is not enough significant evidence to accept H1c. Hence, hypothesis 1c is rejected.

Hypothesis 1d stated that management practices that support creativity and innovation positively influences IWB. Model 5 tested this relationship. The overall model is highly significant (p<0.001) with a F-value of 6.773. Moreover, the model explains 24% (R Square=.24) of the overall variance. The influence of management practices on IWB is statistically significant at 1% level. The unstandardized regression coefficient for management practices is .264. This implies that an increase in supportive management practices leads to higher IWB of employees as predicted by the Componential Theory of Creativity and Innovation. Since the results are significant, there is enough statistical evidence to accept H1d. Hence, hypothesis 1d is accepted.

(43)

the work environment he increases his innovative work behaviour. In addition, model 7 shows that psychological climate also has a significant unique contribution in the overall work environment (B=.534, p<.05). This model is highly significant (F-value=5.428, p-value<.001; R Square=.314). Because the results are statistically significant, there is enough evidence to accept H1e. Hence, hypothesis 1e is accepted.

Moderation effect model

For knowledge and individual resources, the resulted factors from EFA, a moderation analysis is conducted to explore what the effect of these individual resources are on the relationship between the work environment and IWB. The second hypothesis is, therefore, tested using hierarchical regression analysis. In order to decrease possible multicollinearity between an interaction term and its corresponding main effects, and to facilitate interpretation, all independent variables are centred (Aiken and West, 1991). The regressions analysis involved three steps. After controlling for the gender, international focus, and branch, the main effects (the work environment, individual resources, and knowledge) were entered in step two. In the third step, the interaction effects of individual resources and knowledge are added. The interaction effect should explain additional variance, above and beyond that explained by the main effects (Battistelli et al., 2013).

(44)

(45)

Table 4.8 Multiple regression analysis to test moderation

Discussion

(46)

“How does the work environment influence innovative work behaviour, and how does individual resources influence this relationship?”

In total, six hypotheses are formulated based on the theoretical framework and the conceptual model. These hypotheses are tested by performing multiple linear regression analyses and a hierarchical regression analysis. It can be concluded that four hypotheses are confirmed by this research. Table 4.9 summarizes the formulated hypotheses, expected outcomes and actual results that are found in this study. In the discussion, a closer look at every relationship will be taken.

Hypotheses Results Notes

H 1a

A supportive work environment has a positive relationship with IWB

Accepted - H

1b Organizational motivation to innovate has a positive relationship to IWB. Accepted - H

1c

Available resources have a positive relationship to IWB. Rejected The strength of the effect of available resources was not found significant. H

1d Management practices have a positive relationship to IWB. Accepted - H

1e

Psychological climate for creativity and innovation has a positive relationship to IWB.

Accepted - H

2

The relationship between the work environment and IWB will be moderated by individual resources.

Rejected No significant interaction effect

Table 4.9 Overview of accepted and rejected hypotheses

First of all, it was expected that a supportive work environment would display innovative work behaviour of employees without a management function. This expectation was based on the Componential Theory and the Investment Theory who assumed that a supportive work environment is necessary for IWB to occur. Both theories argued that when an environment is supportive, employees are encouraged and willing to display creative and innovative behaviour. In line with those theories, this research showed a significant positive effect of the work environment on IWB. More specifically, a supportive work environment resulted in higher innovative work behaviour. According to Shelley et al (2004) and Hammond et al (2011) the work environment and IWB are positively related, because an environment that is ‘safe’ enables individuals to suggest new ideas or to try something new. This research shows that this assumption is correct and confirms H1a.

Moderating effect of individual resources on innovative behaviour in the work environment