Supervisor: Frank Slisser Student: L.M. Dignadice Balster Student no.: 12000965 Master Thesis EPMS Digital Track Amsterdam Business School THE INFLUENCE OF COLLABORATION SOFTWARE ON INTRINSIC MOTIVATION AND TEAM PERFORMANCE

(1)

THE INFLUENCE OF COLLABORATION SOFTWARE ON INTRINSIC MOTIVATION AND TEAM PERFORMANCE

Master Thesis EPMS Digital Track Amsterdam Business School

Student: L.M. Dignadice Balster Student no.: 12000965

Supervisor: Frank Slisser

(2)

Acknowledgments

I would like to thank my thesis supervisor Frank Slisser for his advice and guidance throughout my thesis writing process, it has been a valuable learning experience.

I would also like to thank my employer IBFD for the support and making it possible for me to study and work at the same time. I’m especially thankful to colleagues and friends who have encouraged me along the way.

I am grateful for my family for always being there and for believing in me, to my sister Jeanne and to my parents Jershon and Evelyn, they are my rock.

Most of all I am grateful to my son, Rens, who is my source of inspiration.

Statement of Originality

I, L.M. Dignadice Balster, 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, but not for the contents.

(3)

Abstract

This research examines the effect of collaboration software on internal motivation and team performance of employees. Collaboration software is analyzed according to its frequency usage through a mediation model and according to its type, which are basic and advance, through a moderation model. Due to the prevalence of basic collaboration software and the rapid increase of popularity of its advanced counterpart, this study is particularly interesting for organizations that would like to assess the implications of using or acquiring collaboration software. A survey is conducted among 125 respondents of various employments backgrounds and industries. Results confirmed 4 out of 6 hypotheses. As mediated by internal motivation, collaboration software usage frequency influences team performance. The type of collaboration software being used, on the other hand, does not implicate any difference as to how its usage affects internal motivation and to how it affects team performance. This concludes that the use of any collaboration software, when encouraged to be used on a frequent basis, has a positive impact on team performance.

Key Words:

Collaboration software, collaboration, intrinsic motivation, team performance

(4)

1. INTRODUCTION

In the business setting, employers and employees alike strive to maintain an acceptable level of professionalism and work quality on a daily basis. There are several factors to take into consideration when working with each other to keep the business running and maintain a dynamic working environment.

Performance in organizations is a progressive continuum that is widely studied in its different facets. It is a multidimensional construct reflecting the totality of behaviors or individual actions needed to accomplish business objectives over a standard period of time (Motowidlo, 2003). Job performance is defined as “scalable actions, behavior and outcomes that employees engage in or bring about that are linked with and contribute to organizational goals” (Viswesvaran and Ones, 2000).

Each employee is expected to be able to perform his or her task to the best of abilities, knowledge and expertise. Certain targets have to be met and goals to be achieved. Motivation has a large impact on the level of performance done by employees. How motivated an employee is can depend on two factors: intrinsic or extrinsic motivation. The term intrinsic motivation refers to doing an activity for its inherent satisfaction while extrinsic motivation refers to performing an activity in order to attain some separable outcome (Deci and Ryan, 2000). Intrinsically motivated behaviors are pursued because they are personally rewarding while extrinsically motivated behaviors are done in order to earn a reward or avoid punishment (Bernazzani, 2017).

Being motivated means having the compelling reason or encouragement to act. The motivations attributable to individuals can differ significantly and can be differentiated not only by the degree of motivation required but also its underlying reasons (Battistella and Nonino, 2012). There is a social aspect to both intrinsic and extrinsic motivation, which refers

(7)

to the collective sphere of an employee who joins a community of collaborative innovation (Battistella and Nonino, 2012).

The relationship between motivation and performance has been supported in previous research, connoting that the power of reward and punishment as extrinsic motivators are deemed to be limited. In the organization, extrinsic motivations such as deadlines and threats of being fired (known as the stick) and salary increases and bonuses (known as the carrot) prove to be less successful tools to motivate employees. Furthermore, people are more creative when they feel motivated primarily by interest, satisfaction and work challenge.

(Amabile, 1998 in Battistella and Nonino, 2012). Also, intrinsically motivated individuals tend to elicit higher levels of task persistence (Cerasoli and Ford, 2014). For this reason, it is the intention of this thesis not to focus on extrinsic motivation and instead, explore how intrinsic motivation and its very nature affects job performance.

Looking at the literature for students, being intrinsically motivated shows greater desire toward learning behaviors such as attending classes and staying in school (Hardre and Reeve, 2003; Robbins, et al., 2004). Furthermore, in a case study conducted by Thompson and Ku (2010), they investigated the relationship between the degree of online collaboration and quality of team performance among students. They confirmed that a collaborative team participated more in group discussions, discussed new ideas and solved problems, and ultimately yielded better learning results. In order to maximize the benefits of online collaboration, they recommended providing teams with online collaboration guidelines.

Emerging technologies such as Web 2.0 tools would be an area of interest according to them, but their design and implementation would have to be evaluated (Thompson and Ku, 2010).

It is worthy to recognize the social connectedness in the workplace as a major contributor in the success of an organization. There is a value-added dimension when collaboration is promoted. Firstly, tasks are assigned to the team instead of individuals that

(8)

makes it a unified project. Secondly, working together brings about a feeling of belonging to something larger than oneself, while getting one’s own part in the success of a project.

Collaborative teams are motivated because full control is given to them, instead of a top-down structure where it can be a matter of following orders (Landau, 2018). Collaborative relationships also help companies achieve “win-win” results and outperform power-based relationships. In a world of increasing complexity, the ability to cooperate efficiently with suppliers, customers and other strategic partners provides a significant path to competitive advantage, innovation and growth (Frydlinger, et al, 2013).

Collaboration software is categorized into two types, mainly: basic which is commonly used by most businesses such as email, instant messaging apps and tele- and video conferencing. The main focus is to communicate, but its usage tends to be unstructured. The next type: advanced, addresses a more coordinating focus and arguably has a more holistic and integrative approach to conduct tasks. These are e-calendars, enterprise resource planning (ERP) software as well as scheduling and project management systems (CTI, 2019). This type of software requires teams to work both interactively and independently (Chen 2004). The increase of advanced collaborative software offered in the market caters to a variety of features that could be customized according to what an organization requires.

Nowadays every collaboration process is virtual to a certain degree (Griffith et al. 2003;

Martins et al., 2004). In the business setting, taking advantage of new technologies provides an opportunity to accomplish tasks at a faster rate. According to Fichter (2005) online collaboration tools help teams collectively author, edit, and review materials in a group workspace which speeds up the creation and review of project materials. It increases efficiency and productivity in the team and foster better decision-making. It is however important to define the business requirements in choosing a collaborative tool in order to find a good and effective match between organizational needs and the value it can give to employees’ tasks and targets.

(9)

Businesses operate in an ever fast-changing scenario with work demands increasing in complexity. Adapting to new ways of working and being flexible to utilize the latest available technology and software solutions are increasingly being adapted in order to produce the maximum desired effects such as high productivity and improved efficiency (Hillsberg, 2017).

1.1 Research Problem

Whereas various studies have touched upon the relational complexities of workplace motivation and performance, there is little research done on how the use of collaboration software can play a major role.

Furthermore, although it is increasingly recognized that the use of collaboration software improves efficiency and workflows, the question whether or not the frequency of usage and the type being used have an effect to the motivation of employees are not readily available and widely explored in the academic and business literature. It is therefore the aim of this research to answer the following question:

1.2 Research Objectives and Relevance

The focus of this research is to explore the impact of collaboration software to an employee’s intrinsic motivation and team performance. Focus is given to job performance in the team level as collaboration entails working with others. In addition, substantive information will be provided in the academic literature, which currently lacks studies on the effects of the use of collaboration software in the work setting.

Results obtained will determine the causal relationship of the variables and will How does the usage frequency and type of collaboration software influence the intrinsic motivation and team performance of employees?

(10)

collaboration software can influence the intrinsic motivation of employees working within a team.

1.3 Research Methods

In order to determine the effect of collaboration software on the motivation of employees to their work, an online survey will be administered to collect the data. The information obtained will then be subjected to statistical treatment. This statistical approach signifies a quantitative research design that builds upon existing theories with the intention to establish, confirm or validate relationships and to develop generalizations that contribute to theory (Williams, 2007).

1.4 Thesis Structure

This thesis is structured as follows. Chapter two describes the existing literature on motivation, job performance, collaboration and collaboration software. The chapter concludes with a conceptual model and hypotheses from the literature. Chapter three substantiates the research in terms of the type of research, data collection, and analysis. Chapter four presents the results of the research. Chapter five discusses the results further in detail. Finally, chapter six contains the conclusion and discussion sections. The limitations of the research and recommendations for future research are also presented in chapter six.

(11)

2. THEORETICAL BACKGROUND

In order to explore how motivation affects performance through collaboration software, we will delve into each factor, their very nature and studies previously conducted, and how they interrelate to each other. The role of collaboration, specifically through online means, will also be tackled and how it affects motivation and job performance.

2.1 Aspects of Job Performance

Job performance is defined as the aggregated value to the organization that pertains to discrete behavioral episodes an employee performs over a standard interval of time. It is multidimensional and evaluative (Motowidlo et al, 1997). Job performance should be distinguished from job productivity, which could be easily interchanged but have different concepts. Job productivity is defined as input divided by output. Thus, it is a narrower concept than job performance (Koopmans et.al, 2011).

This thesis will delve into job performance as done by an individual in a team and how it relates to motivation. A team consists of at least two individuals who have been assigned to specific tasks that work together towards a common goal, the completion of which requires dependency among each other (Salas et al. 1992 in DeShon et al. 2004). There are several researches that distinguish between two types of tracks when performing in teams (Baker et al, 1998; Dyer, 1984; Fleishman and Zaccaro, 1993; McIntyre and Salas 1195 in Rasker 2002).

The below framework shows the perspective that team performance is a result of task work and team work. Task work refers to activities and behaviors related to the tasks performed by individual team members which is independent of other members. Team work refers to those that serve to strengthen the quality of functional cooperation of team members. Because tasks have to be performed in a team, members perform in accordance to specific knowledge, skills and attitudes (Rasker, 2002).

(12)

Figure 1: Framework for team performance factors

It is important not only to focus on the extent of goal accomplishment but also on the state of the team and its members (Hackman, 1987 and Tannenbaum et al, 1992 in Rasker, 2002). Teams have to perform subsequent tasks and it is therefore important to maintain the motivation and ability to perform those tasks (Rasker, 2002).

Ancona and Caldwell defined team performance as the extent to which a team is able to meet its output goals, member expectations and cost-time objectives (Ancona and Caldwell, 1992). Since individual behavior is influenced by the presence of others (Crano, 2000), it is no wonder that individuals work harder and faster when others are present, which is also known as social facilitation. Social Orientation Theory of social facilitation has suggested that people differ in their orientation towards social situations. The theory states that those who have positive orientation are more likely to display social facilitation effects while those with negative orientation will lead to impairment in performance (Aiello and Douthitt. 2001).

The theory posits that social facilitation is a product of an individual’s motivation to

(13)

maintain positive self-image in the presence of others. This motivation leads to people behaving in ways that form good impressions and therefore results in social facilitation in evaluative situations (Aiello and Douthitt. 2001). It is therefore worthy to note that motivation plays a major role in how employees perform their job. Based on the classical Self- Determination Theory (SDT) which will be discussed further below, intrinsic motivation describes the innate propensity to pursue interesting tasks that challenge one’s skills and foster growth (Deci and Ryan, 1985) and should have a positive link to performance.

Task performance can be defined as the proficiency or competency on how a job or task is done. It includes both job quantity and quality as well as job knowledge. It is important to note that what constitutes a core job tasks can differ from job to job. Contextual performance focuses on the individual behaviors that support the organizational, social, and psychological environment in which the core job must function. It can be an extra-role or interpersonal relations. The ability and extent to which an employee adapts to changes in the work system or roles is called adaptive performance. It includes creatively solving problems, learning new tasks, dealing with uncertain work situations as well as adapting to other employees, work cultures, technology or physical surroundings. Counterproductive work behavior is defined as those that harm the well-being of the business which includes absenteeism, tardiness, off-task behavior, theft and substance abuse (Koopmans et al, 2011).

These factors, however considered as separate dimensions, all inter-relate with each other.

The distinction may get blurred to some extent depending on the nature of the job and the fact that how a certain factor is considered by one company can differ on the viewpoint of another.

According to a theory of performance by Campbell (1990) there are three distinct determinants of job performance: declarative knowledge, procedural knowledge and skill, and motivation. Declarative knowledge is knowing what to do – it pertains to knowledge of principles, facts and ideas. This can be similar to task performance as explained on the

(14)

framework above. Procedural knowledge and skill are knowing how to do the job – it includes cognitive, perceptual and interpersonal skills. This factor can also be related to contextual performance as previously mentioned. Motivation as a third predictor of performance refers to the direction, intensity and persistence of volitional behaviors (Campbell, 1990). As it is also a main variable of this study, motivation will be discussed next in further detail.

2.2 The Nature of Motivation

According to R. Ryan and E. Deci, motivation concerns energy, direction, persistence and equifinality - all aspects of activation and intention. Motivation can be classified as either intrinsic or extrinsic. Intrinsic motivation is the inherent tendency to seek out novelty and challenges, to extend and exercise one's capacities, to explore, and to learn. Extrinsic motivation refers to the performance of an activity in order to attain some separable outcome, not focusing on the inherent satisfaction of doing the activity itself. According to them, people can be motivated because they value an activity or because there is strong external coercion (Ryan and Deci, 2000). Furthermore, their focus was to specify the conditions that tend to support people’s natural activity versus elicit or exploit their vulnerability. They formulated the Self-Determination Theory (SDT) which is an approach to human motivation and personality. It uses traditional empirical methods while employing an organismic metatheory, highlighting the importance of evolved inner resources for personality development and behavioral self-regulation (Ryan, Kuhl, & Deci, 1997).

Basing on the SDT model, intrinsic motivation occurs when a task or activity is performed for one’s own enjoyment or seen as an opportunity to explore, learn and actualize potentials on a personal level rather than a desire for an external reward. Extrinsic motivation, on the other hand, occurs when one is motivated because of inner desires but due to external rewards. Tasks can also be performed out of avoiding punishment even though the tasks themselves are not rewarding or enjoyable (Asha and Warrier, 2017).

(15)

Figure 2: Different types of motivation according to Self-Determination Theory (SDT).

2.3 Key determinants of Intrinsic Motivation

Following Ryan and Deci’s Self-Determination Theory (SDT), they have concluded that intrinsic motivation arises from three psychological needs: self-determination, competence and relatedness (Deci, 1985). Further studies conducted by Reeve present an additional factor:

excitement which he derived from earlier studies done by Izard in 1977 (Reeve, 1988).

2.3.1 Self-determination

Self-determination is defined as the attitudes and abilities that lead individuals to define goals for themselves and to take the initiative in achieving these goals (Wehmeyer, 1999). It prefers the choices made by oneself rather than external rewards and pressures (Reeve, 1988).

In a meta-analysis conducted by Patall (2008) on the effects of choice on intrinsic motivation, the results indicated that providing choice enhanced intrinsic motivation, effort,

(16)

task performance, and perceived competence. When individuals have autonomy through choice their motivation is enhanced, this depending also on the self-regulatory costs (Patall, 2008).

2.3.2 Competence

Competence is the need to be effective in exercising ones’ capacities and skills with the aim to master optimal challenges (Vallerand and Reid, 1984).

Perceived competence represents the extent to which an individual believes he performs an activity well (Bandura, 1982). It is presumed to affect intrinsic motivation following feedback, either during or at the conclusion of performing a task (Bandura, 1982;

Harackiewicz, 1989). Competence valuation is presumed to be operative prior, in that it is hypothesized to influence how important it is to do well at the beginning of a task which in turn influences intrinsic motivation (Elliot et al., 2016).

2.3.3 Relatedness

Relatedness or affiliation is the need to establish close emotional bonds and attachments from person to person (Grolnick, Ryan and Deci, 1991).

According to Hill (1987) social motivation have four underlying factors as suggested by previous theories in social psychology and personality, they are: positive affect or stimulation related to a sense of closeness to others, attention or praise from others, social comparison, and emotional support through social contact. He has developed the Interpersonal Orientation Scale to support the multidimensional construct (Hill, 1987).

2.3.4 Excitement

Excitement is “what it feels to be interested” (Izard, 1977) and “being engaged in activities that are motivating to their own right” (Watanabe and Kanazawa, 2009). They investigated whether openness to experience, excitement to do new things, would affect intrinsic motivation. Their test of personality-based view of intrinsic motivation yielded results

(17)

that indicated excitement to be a significant predictor of intrinsic motivation with the effects of task-related motivators.

In another experiment conducted by Reeve, Cole and Olson (1986) they added excitement as a fundamental intrinsic reward that is operative in the intrinsic motivational processes. In a majority of cases, they found out that participants reported excitement as the feeling most associated with favorite activities, rather than competence or autonomy. They used the traditional intrinsic motivation paradigm of Deci (1971) and found a significant positive association with the other indices of intrinsic motivation.

Figure 3: Sources of intrinsic motivation

(18)

A more current article explores intrinsic motivation on a similar approach. According to Morikawa, there are four sources of intrinsic motivation as shown on Figure 3. They are:

meaningfulness, choice, competence and progress. An employees’ creativity and quality of work is enhanced by these sources of intrinsic motivation. When employees have intrinsic motivation, they tend to look for better ways to perform tasks and concentrate more on the overall quality of their work (Morikawa, 2017).

2.4 Collaboration as Precursor to Business Success

In an article by Hearn (2018) he mentioned that according to former Yahoo president and CEO Marissa Mayer: “To become the absolute best place to work, communication and collaboration will be important, so we need to be working side-by-side.” If a modern business would like to compete and succeed, there will be a need to find ways to encourage collaboration at all costs (Hearn, 2018).

A Stanford research conducted by Gregory Walton and Priyanka Carr (2014) also proved that working together boosts motivation. The results of their study showed that, even if employees work alone, the perception of being involved in a collaborative process substantially increased their intrinsic motivation. These employees persevered on tasks 64 percent longer than those who had no sense of collaboration at all (Parker, 2014).

Moreover, according to Economist Intelligence Unit’s white paper, they have forecasted the future belongs to those who collaborate. In the “Foresight 2020” study conducted in 2006 composing of 394 executives from various industries, it is predicted that markets will become more global, functions will automize across geographies and competition will intensify. The need for agility in a fast-paced and ever-changing environment will therefore compel companies to increase collaboration in order to succeed and continually grow.

Companies see collaboration in achieving the following goals: improving efficiency and

(19)

productivity, problem-solving, knowledge-sharing and competitive differentiation (Rubin, 2017).

Figure 4: The benefits of collaboration

Collaboration is also defined as “a process through which parties who see different aspects of a problem can constructively explore their differences and search for solutions that go beyond employees’ own limited vision of what is possible” (Gray, 1989). Morton et. al.

defined collaboration further: “Collaboration consists of the following elements: a common purpose, separate professional contributions, and a process of cooperative joint thinking and communication” (Morton et al, 2009).

The research study made by Fan et. al. formulated a theory to explain the suitability of virtual collaboration which they call “Collaboration Virtualization Theory” or CVT. According to CVT, the collaboration process is neither purely virtual nor physical but a hybrid of the two.

It incorporates three categories of constructs: task, technology and team characteristics. Under

(20)

task the urgency (time-based), complexity (difficulty) and sensitivity (confidentiality) of data and processes are vital components to be taken into consideration. For technology, IT plays an important role. This also includes the functionality capacity and information accessibility provided by information technology (Sengupta and Zhao, 1998). As to team characteristics, the nature of groups will contribute to the effectiveness of virtual collaboration, in that familiarity among team members will allow them to know the available expertise and reduce knowledge barriers (Aubert and Kelsey, 2003 in Fan et. al., 2012).

Figure 5: Conceptual Model of Collaboration Virtualization Theory (CVT)

2.5 Two Types of Collaboration Software

The classification of collaborative software can be distinguished in two types: Firstly, Communication Technology which includes the basic communication technology used by most businesses: email, instant messaging, chat forums, digital voicemail applications, voice- over-internet protocol (VoIP) calls, and tele- and video conferencing. The second type is the Coordination Technology which makes use of a more advanced collaborative software in order to integrate both teamwork and taskwork functions. These tools include e-calendars, employee

(21)

time trackers or scheduling systems, project management systems, enterprise resource planning software and employee portals (CTI, 2019).

Due to digital transformation in the business landscape a huge shift in how teams work has emerged. Information is shared more and more online therefore companies feel the need to organize efficiently while maintaining transparency and productivity among teams. Global Market Insights predict that the use of collaboration software will continue to grow and exceed USD 8.5 billion by 2024 (Litsa, 2017).

Figure 6: Collaboration platforms top communication channels 2018 and onwards

2.5.1 Basic collaboration software

E-mail is the simplest form of collaboration software that promotes communication from the sender to one or more recipients (Lopes et. al., 2016). Although it may not necessarily provide immediate feedback and recipients find it challenging to understand the context of information being sent, it is easily accessible for employees to correspond through email. Other

(22)

basic collaboration tools include audio and videoconference (Skype, ZOHO Meeting, GoToMeeting), file sharing applications (Dropbox, WeTransfer), Shared Presentations (Slideshare, Prezi), CMS (Joomla, Drupal), Blogs (Wordpress, Blogger) and Shared Documents (Google Drive, Microsoft Web Apps). Companies usually have a variety of these tools available to their employees according to their business needs and in order to promote team collaboration (Lopes et. al., 2016).

2.5.2 Advanced collaboration software

Collaboration software has been increasingly used from communication to project management (Bika, 2019). The advantages of using advanced collaboration software mainly highlight its project tracking capabilities – both reading and reviewing, on working within the team. These include the option to work remotely for as long as there is internet, document sharing and retrieval, as well as interactive options with users and even stakeholders. When a good online collaboration tool is used, it becomes easy and quick to generate detailed reports that include all of the activities associated with a certain project, giving team members more time to work on result-generating activities (Warren, 2019). Additional benefits of collaboration software include quicker completion of end goals, enhanced project management efforts, strengthening of team relationships and improving the organization’s workflows through open communication (Gilbert, 2019).

Examples of widely used advanced collaborative software in the market include:

Asana, Proofhub, Microsoft Teams, Slack, Trello, WebEx, Wrike and the list goes on. New tools are developed each year while existing ones are constantly improving their features and functionality. Users choose according to ease of use, multitasking features, privacy options, and integrative compatibility with other applications.

(23)

2.6 Effects of Usage Frequency

Collaboration is measured as a function of three variables according to Asdemir (2012), these variables are: frequency of interactions, content of information exchange, and openness to share information during collaborative interactions (Asdemir et al, 2012). A collaboration- based approach consists of several simultaneous work flows where teams coordinate frequently in order to decide which information gaps must be filled, therefore eliminating the wait time and reduces cycle times significantly (Holman et al, 2003).

The increased usage of collaboration software has also been explored by Komarov and co-researchers (2014) with the aim to help social software developers and providers improve and increase software adoption rate. They cited the results from the Australian Bureau of Statistics that more than 41% of companies who are open to an innovative collaboration at work have increased in their productivity as compared to previous years. Furthermore, their analysis showed that social collaboration software plays a vital role in e-business (Komarov et al, 2014).

In a study conducted by Asdemir and academic associates (2012), they have concluded that the frequency and intensity of collaboration leads to greater product design quality, lower design cycle time, and reduced product development cost (Asdemir et al, 2012). By considering the effect of usage frequency on collaboration software, it is the intention of this research to determine whether usage frequency impacts the intrinsic motivation and performance of employees.

In line with the literature mentioned above, this research proceeds with connecting the concepts in order to address and test the hypotheses mentioned hereafter.

(24)

2.7 Conceptual Model and Hypotheses

The literature above has shown the positive impact of collaboration in the workplace and it can be implored that the use of collaboration software has increased performance and productivity in organizations. This research aims to find how intrinsic motivation can be affected and whether or not the type and frequency of its usage contributes to an increase in intrinsic motivation of an employees’ performance within the team.

Figure 8: Proposed conceptual model

Figure 8 illustrates the proposed research model for this study. Five hypotheses are deduced from the literature review revolving around four key variables. These four key variables are identified based on the above conceptual model.

H1 Collaboration Software (CS) usage frequency influences team performance. The higher the CS usage frequency the better employees perform their job in the team.

H2 The relationship between CS usage frequency and team performance is mediated by intrinsic motivation.

CS Usage Frequency

Intrinsic Motivation

Team Performance CS Type

*Basic -

*Advanced +

H1 H3

H2

H2a

H4

H2b

(25)

H2a CS usage frequency influences intrinsic motivation. The higher the CS usage frequency the higher the employees' intrinsic motivation.

H2b Intrinsic motivation influences team performance. The more intrinsically motivated employees are the better they perform their job in the team.

H3: The relationship between CS usage frequency and intrinsic motivation is moderated by CS type. The intrinsic motivation of using CS frequently will be affected by the type of CS used.

H4: The relationship between CS usage frequency and team performance is moderated by CS type. The team performance of employees using CS frequently will be affected by the type of CS used.

Independent variables: Collaboration software usage frequency is an independent variable defined as the amount of time an employee uses this tool in their workplace.

Collaboration software type is another independent variable and is categorized as basic or advanced. Basic category is for employees that are offered the basic communication package by their employers such as email, instant messaging and video conferencing. The second category is advanced for employees using more advanced and sophisticated collaboration software.

Mediator: Motivation is classified as intrinsic or extrinsic. However, this study will focus on the intrinsic nature. Literature shows that the more intrinsically motivated an employee is, the better chances of performing tasks well individually and as a team. Intrinsic motivation is seen as a mediator showing the relationship between collaboration software usage frequency and team performance. Employees who use collaboration software on a more frequent basis are intrinsically motivated, and as a result perform their job well.

Moderator: The type of collaboration software, either basic or advanced, will be seen as a moderator between usage frequency and intrinsic motivation, and between usage

(26)

frequency and performance. Those who use the more advanced collaboration software tend to perform better compared to the ones using only the basic counterpart.

Dependent variable: Team performance is seen as the dependent variable of this research. The focus is given to team performance as the use of collaboration software involves working with others (Gajda, 2004).

(27)

3. METHODOLOGY

This chapter describes the methodologies that are applied for this study. The research design will be explored and the collection and analysis of data are explained. Furthermore, the operationalization of constructs will be discussed as to how variables are defined, collected and measured.

3.1 Research Design and Sample

The main purpose of this study is to identify the influence of the mentioned independent variables on the dependent variable, whether the type and usage frequency of collaboration software influences intrinsic motivation and team performance.

This research makes use of quantitative method to conduct causal research. This method falls under the deductive approach wherein questions are based on specific relationships and existing theories, which are then tested using quantitative data (Thornhill et al, 2009).

An online survey is created in Qualtrics software, a website which not only creates surveys but also automatically collects all data from the respondents. The use of Qualtrics will also make it easier to transfer data to SPSS and thereafter proceed with analysis and interpretation. Moreover, advantages of choosing this type of questionnaire includes: quick deployment to respondents, user-friendly online environment, likelihood of high response rate, easy to forward through a link, and reduced costs (Ervans and Mathur, 2005).

The respondents are contacted through email and LinkedIn, with chain referrals being encouraged and done by colleagues and acquaintances in order to reach the desired sample.

The sample will be derived from companies with employees that have varying degrees of collaboration software usage. Each respondent will be asked, after a brief explanation of terminologies, as to how often and what type or types of collaboration software are being

(28)

3.2 Data Collection

A total of 125 respondents (N=125) comprising of men and women between 25 – 65 years old are asked to fill-in the survey. This sample size is similar to a study made by I. Lopes et. al. on collaboration tools (Lopes et al, 2015).

The online survey includes a brief introduction followed by an explanation of the research study. The type of software collaboration will be defined, which will be helpful for the respondents to understand which category they belong to. Anonymous manner of collecting data is also mentioned thereby adhering to confidentiality of respondent’s details.

A link will be created in Qualtrics which will be forwarded by email to respondents.

Respondents will be requested to forward the survey to their professional and personal network. In order to reach a sufficient number of participants, two non-probability sampling techniques are used, namely: convenience and snowball sampling (Emerson, 2015). Through these sampling techniques, this research aims to attain a minimum of 120 respondents. A timeframe of two weeks is given for responding and collection of data.

3.3 Data Analysis

All data obtained from Qualtrics is exported to SPSS software. Errors will be checked through a frequency and normal distribution test. After this is found to be satisfactory, a bivariate correlation analysis is performed using the four variables: collaboration software (CS) usage frequency (independent variable), collaboration software type (independent variable), intrinsic motivation (independent variable), and team performance (dependent variable). The first regression will test the relationship between CS usage frequency and intrinsic motivation, the second on the relationship between CS usage frequency and team performance, the third on intrinsic motivation and team performance, the fourth on CS type and intrinsic motivation, and the fifth on CS type and team performance. SPSS Macro Process

(29)

is used to analyze the moderation and mediation effects. In the event of missing data, these will be substituted with the mean value of the concerning variable.

3.4 Operationalization of Constructs

The four constructs in this research are intrinsic motivation, team performance, collaboration software frequency and collaboration software type. For the intrinsic motivation and team performance scales, items are measured on a 7-Point Likert scale from 1 (strongly disagree) to 7 (strongly agree). For the collaboration software type and usage frequency, items are measured on a 5-Point Likert scale from 1 (Never) to 5 (Always).

3.4.1 Intrinsic Motivation

This research will focus on the four key determinants mentioned in Figure 3 in order to measure the intrinsic motivation of the respondents. Questions related to competence, self- determination, relatedness and excitement derived from past scientific researches (Deci and Ryan, 1991; Reeve and Olson, 1986) will comprise the survey questions.

The Intrinsic Motivation Inventory (IMI) will be used to measure the participants subjective experience related to their intrinsic motivation in the workplace. The instrument consists of varied numbers of items from these subscales, all of which have been shown to be factor coherent and stable across a variety of tasks. In a study by McAuley, Duncan, and Tammen (1987) they examined the validity of the IMI and found strong support for its validity (McAuley et al, 1987). The instrument has often been modified slightly to fit specific activities, and for this reason it can be flexible for questions in this research related to workplace scenarios.

For this scale, the following questions are taken and adjusted from the IMI:

Perceived Competence

- I am satisfied with my performance at work.

- I am very skilled at the work that I do.

(30)

- I think I perform my job better than other colleagues.

- I usually feel competent after working certain tasks for a while.

Self-determination

- I do my work because I wanted to.

- I felt like I had to do some tasks assigned to me. (R) - I believe I have some choice about doing my job.

- I feel that I have a choice on what to do at work.

Relatedness

- I like to interact with colleagues more often.

- I feel like I could really trust my colleagues.

- I don’t feel like I could really trust some of my colleagues. (R)

- It is likely for me to build good collegial relationships with my workmates.

Excitement

- There are tasks at my work that doesn’t challenge me. (R) - I enjoy doing the work that I do.

- I would describe my work as very interesting.

-

- I think my job is quite enjoyable.

In order to score the instrument those with (R) will have to be reversed first. Then, the subscale scores will be averaged across all the items before proceeding with the analysis of the results.

3.4.2 Team Performance

The items from the team performance scale is adapted from Chiocchio’s study on team performance in the context of collaboration (Chiocchio et al, 2012). The instrument is relevant for multilevel theory building, generally in teams and particularly in project teams, which

(31)

supports the very nature of collaboration. Questions derived from the collaborative work questionnaire are:

- My teammates and I share knowledge that promotes work progress.

- My teammates and I understand each other when we talk about the work to be done.

- My teammates and I share resources that help perform tasks.

- My teammates and I make sure our tasks are completed on time.

3.4.3 Collaboration Software Frequency

Respondents are asked how often they use collaboration software in the workplace.

The items from this scale is adapted from a 5-point Likert scale on frequency depicting general timeframes ranging from 1 (never) to 5 (always). It is important to have the frequencies sequentially ordered in order to prevent confusion (Sauro, 2018).

3.4.4 Collaboration Software Type

With two types of collaboration software, being basic and advanced, the items are based on a paired comparison scale (Tsukida and Gupta, 2011). In the introduction section of the survey, both basic and advanced collaboration will be briefly explained in order for respondents to identify which type of software is being currently used.

3.4.5 Demographics

Next to the four mentioned variables, demographic information such as age, gender, job title, length of work experience, educational attainment and field of industry are requested.

(32)

4. RESULTS

Due to the ordinal nature of the variables, an appropriate factor analysis technique must be chosen. A normal factor analysis assumes that the variables are continuous such as interval or ratio type, therefore the ordinal variables from the research results is not suitable for a normal factor analysis. A special version of the factor analysis is chosen that takes into account the ordinal nature of the variables, which is known as the Categorical Component Analysis (CATPCA) with Optimal Scaling.

4.1 Recoding

In order to check if there is consistency in all the answers, analysis is done by applying CATPCA using Cronbach Alpha. The reversed items are excluded and are replaced with the recoded versions. Three items that have to be recoded due to its reverse questions are Q4, Q7 and Q13. Recoding is run via SPSS to transform them into different variables and are renamed to Q4rec, Q7rec and Q13rec.

4.2 Reliability

In order to check if there is consistency in all the answers, analysis is done by applying one similar scale to all variables using Cronbach Alpha. The reversed items are excluded and are replaced with the recoded versions. As exhibited in table 1, all four variables have a Cronbach alpha > .7, which indicates an acceptable high level of internal consistency (Gliem and Gliem, 2003).

Table 1: Cronbach Alpha

Variable Cronbach Alpha % of Variance Total

EigenValue

Items

CS Frequency Usage .843 76.101 2.283 3

CS Type .703 77.122 1.542 2

Intrinsic Motivation .967 67.086 10.734 16

Performance .942 85.218 3.409 4

N = 125

(33)

4.2.1 Collaboration Software (CS) Frequency

After performing CATPCA, the % of variance accounted for is 76.1% which is above the goal of 50% or more implying the result is satisfactory (Saade, 2005). The Cronbach alpha of this solution is .843 which is well above the threshold of .7 (George and Mallery, 2003).

This means there is a good internal consistency. The eigen values is 2.283 by making use of three variables, namely CS usage frequency, CS Basic Type, CS Advanced Type explaining 76.101% of explanatory power.

The component loading shows how much the different variables load onto the theoretical construct and how much similarity there is between the variables. The goal is to achieve a component loading of at least .5 and this is achieved based on the results shown on Appendix 2 for all the variables.

4.2.2 Collaboration Software (CS) Type

The commonality between the usage frequency of CS basic type and CS advanced type is also tested via a CATPCA analysis which takes into account the ordinal nature of both variables. As seen on Table 1, the CATPCA shows a Cronbach Alpha of .703 which indicates an acceptable internal consistency. This implies that respondents who use CS basic type frequently may also use CS advanced type frequently. This can also be shown by the component loading of .878 on both factors which indicates again a high commonality between both types.

4.2.3 Intrinsic Motivation

As seen on Table 1, for intrinsic motivation the eigen values is 10.734 by making use of 16 variables. This explains the 67.086% of explanatory power.

The recoded items Q4, Q11 and Q13 show a low loading that is lower than .5 which

(34)

imply that the respondents did understand the question and are not agreeing these items are indicative of intrinsic motivation. It may also indicate that respondents got confused with the way the questions are asked. Even so, these items will be kept in the analysis in order to align with the validated questions that are used in this research.

4.2.4 Team Performance

For team performance, the eigen values is 3.409 by making use of 4 variables which explains the 85.225% of explanatory power. Furthermore, a Cronbach Alpha of .942 indicates an excellent internal consistency. All factor variables show high component loadings so no further remarks.

It can be reported that Q10 (.986), Q15 (.986) and Q20 (.985) have a very high component loading. Q5 deviated (.703) from the rest with a relative low component loading compared to the rest but not necessarily unacceptable.

4.3 Means, Standard Deviations, Correlations

Based on the results of the Cronbach alpha on Table 1 there is sufficient internal consistency between the survey questions that are creating the different theoretical constructs.

To compute constructs, the average or Mean of all items is taken per theoretical construct.

The variables created with the CATPCA are inspected on their distribution and extreme values, due to regression is sensitive to extreme values (Stevens, 1984). Since the dataset is small, instead of removing outliers, winsorization is chosen in order to replace extreme values for less extreme ones. Winsorization or winsorizing is a process to minimize the influence of outliers in a given data in order to improve its efficiency and robustness (Stephanie, 2016).

The performance variable is winsorized at 2.4% and 77.6%.

The intrinsic motivation variable is winsorized at 7.2% and 99.2%.

The CS Usage Frequency has no outliers so no adjustment is necessary.

(35)

The CS Type is measured on an ordinal scale therefore this variable is not corrected by winsorizing.

4.4 Descriptive Statistics

Team performance which is the dependent variable shows a mean of .108 and standard deviation of .002. Based on a CATPCA analysis which is in essence a special kind of factor analysis, it is the aim to create a variable with a mean of 0 and standard deviation of 1. A mean of .108 implies that the winsorizing of this variable caused a small increase in the mean value of performance. With winsorizing the standard deviation of .002 is considerably reduced. This implies there is a very small variation in this variable which can be due to respondents are choosing similar answers.

Intrinsic motivation shows a mean of .123 and standard deviation of .031. These results show a similar pattern as with the team performance variable implying there are extreme values in the variables which makes winsorizing necessary, except for a few extreme answers it shows that the variation in the data sample is relatively low.

CS frequency shows a mean of .000 and standard deviation of 1.00 showing a classic pattern of a standardized variable as a result of factor analysis (CATPCA).

CS basic type shows a mean of .768. Because it is a dummy variable it can only assume a value of 1 or 0. Therefore a mean of .768 illustrate the percentage of 1’s compared to the 0’s.

This implies that 76.8% of the observations have a basic type of CS.

CS advanced type show a mean of .416. Similar to the basic type, this implies that 41.6% of the observations use the advanced type of CS.

In order to prevent misunderstandings, the basic type doesn’t exclude the advanced type and vice versa, therefore respondents can use the basic type, the advanced type or both simultaneously.

(36)

Table 2: Descriptive Statistics

Variable name N Mean Std

Dev Median Min Max Skew ness

Std Err

Kurtos is

Std Err CS Usage Frequency 12

5

.000 1.00 -.454 -1.57 1.43 .132 .217 -1.41 .430

CS Basic Type 12

5

.768 .424 1.00 0.00 1.00 -1.28 .217 -.354 .430

CS Adv Type 12

5

.416 .494 .000 0.00 1.00 .345 .217 -1.91 .430 Intrinsic Motivation 12

5

.123 .031 .129 0.05 0.20 -.558 .217 .944 .430 Team Performance 12

5

.108 .002 .107 0.11 0.11 .773 .217 -1.23 .430

4.5 Computing Scale Means

A Spearman’s correlation is chosen due to the variables having an ordinal nature, which implies that the variables are not normally distributed (Hauke and Kossowski, 2011). In this regard, a non-parametric correlation technique is chosen. Table 3 shows that there is a positive significant correlation between the CS usage frequency and team performance (r= .244, p=

.006). This suggests that there is an association between the CS usage frequency and team performance which implies that high performing teams seem to use CS frequently or a high usage of collaboration software improves the team performance.

A remarkable finding in Table 3 is that there is a positive but non-significant association between the CS basic type and team performance (r= .078, p= .389). This suggests that the use of CS basic type has no association with team performance. But there is a significant positive correlation between CS basic type and intrinsic motivation (r= .194, p= .031). This suggests that the CS basic type has a stronger association with intrinsic motivation than with team performance.

For the CS advanced type, Table 3 shows there is a positive and significant association between CS advanced type and team performance (r= .181, p= .043). This suggests that the usage of CS advanced type has an association with team performance. There is also a positive

(37)

significant correlation between CS advanced type and intrinsic motivation (r= .194, p= .030).

This also suggests that the use of CS advanced type has an association with intrinsic motivation.

For intrinsic motivation, Table 3 shows there is a positive and significant association between CS usage frequency and intrinsic motivation (r= .241, p= .007). This suggests there is an association between CS usage frequency and intrinsic motivation which implies that employees who are intrinsically motivated seem to use CS frequently or a high usage of collaboration software improves intrinsic motivation.

There is a strong correlation between CS basic type (r= .667, p< .000) and CS advanced type (r= .675, p< .000) with CS usage frequency. This is because the CS usage frequency includes both basic and advanced type.

Table 3: Means, Standard Deviations, Correlations

Variables M SD 1 2 3 4 5

1 CS Usage Frequency .000 1.00 1

2 CS Basic Type .768 .424 .667** 1

3 CS Adv Type .416 .494 .675** .387** 1

4 Intrinsic Motivation .123 .031 .241** .194* .194* 1

5 Team Performance .108 .002 .244** .078 .181* .340** 1

*Correlation is significant at the 0.05 level (2-tailed)

**Correlation is significant at the 0.01 level (2-tailed) N=125

4.6 Testing the Mediation Effect

In order to do an advanced mediation analysis with the benefit of bootstrapping, SPSS Process Macro of Andrew Hayes with a Hubert-White Heteroscedasticity-consistent correction is used in this research.

First, the direct effect of CS usage frequency on performance was positive and

(38)

Second, analysis is done without the moderator using model 4 (Hayes, 2013). Results show that there is a positive significant effect from CS usage frequency on intrinsic motivation (path a) (B= .0083, p= .0024). This leads to the acceptance of Hypothesis 2a. There is a positive significant effect from intrinsic motivation on performance (path b) (B= .0206, p= .0001). This leads to the acceptance of Hypothesis 2b. The direct effect from CS usage frequency on performance is positive and significant (path c’) (B= .0004, p= .0168). The total mediation effect is significant based on the BootLLCI (.0001) and BootULCI (.0003) as reflected on Table 6. This finding results in the acceptance of Hypothesis 2.

Table 6: Bootstrap for mediation effect

4.7 Testing the Moderating Effect

4.7.1 Basic Type

In order to investigate if the basic type has a moderating effect between CS usage frequency and intrinsic motivation (a path) and between CS usage frequency and performance (c path), a second analysis is performed.

Analysis is done with the moderator using model 8 (Hayes, 2013). Results show that there is a positive non-significant effect from CS usage frequency on intrinsic motivation (path a) (B= .0010, p= .9351). There is a positive but non-significant effect from Basic Type on

(39)

intrinsic motivation (B= -0130, p= .3987). In addition, the interaction has a positive but non- significant effect on the intrinsic motivation (B= .0058, p= .6553). This implies that there is no statistically significant moderation effect on the relation between the CS usage frequency and intrinsic motivation. This leads to the rejection of Hypothesis 3. A possible explanation for this is the variation in the intrinsic motivation is relatively low and therefore the detection possibilities are limited.

There is a positive significant effect from intrinsic motivation on performance (path b) (B= .0210, p= .0001). The direct effect from CS usage frequency on performance is positive and non-significant (path c’) (B= .0002, p= .8274). The basic type has a negative non- significant effect on performance (B= -.0002, p= .8712). The interaction effect is positive but not significant (B= .0004, p= .7186). This result shows that there is no moderation on the basic type on the c’ path. This leads to the rejection of Hypothesis 4. The total mediation effect is not significant based on the BootLLCI (-.0004) and BootULCI (.0010).

4.7.2 Advanced Type

In order to investigate if the advanced type has a moderating effect on between CS usage frequency and intrinsic motivation (a path) and between CS usage frequency and performance (c path), a second analysis is performed.

Analysis is done with the moderator using model 8 (Hayes, 2013). Results show that there is a positive non-significant effect from CS usage frequency on intrinsic motivation (path a) (B= .0084, p= .0862). There is a positive but non-significant effect from Advanced Type on intrinsic motivation (B= .0048, p= .4727). In addition, the interaction has a negative but non- significant effect on the intrinsic motivation (B= -.0034, p= .6125). This implies that there is no statistically significant moderation effect on the relation between the CS usage frequency and intrinsic motivation. This leads to the rejection of Hypothesis 4. A possible explanation for

(40)

this is the variation in the intrinsic motivation is relatively low and therefore the detection possibilities are limited.

There is a positive significant effect from intrinsic motivation on performance (path b) (B= .0207, p= .0001). The direct effect from CS usage frequency on performance is positive and non-significant (path c’) (B= .0002, p= .4904). The advanced type has a positive non- significant effect on performance (B= -.0001, p= .8356). The interaction effect is positive but not significant (B= .0004, p= .3150). This result shows that there is no moderation on the advance type on the c’ path. This leads to the rejection of Hypothesis 4. The total mediation effect is not significant based on the BootLLCI (-.0004) and BootULCI (.0002) as reflected on Table 7.

Table 7: Bootstrap for moderation effect

4.8 Summary Hypotheses

To summarize this chapter and thereby the results of the study, Table 8 below is created for a complete overview of all hypotheses and whether they are accepted or rejected.

Hypothesis 1, 2, 2a and 2b are accepted while Hypothesis 3 and 4 are rejected.

(41)

Table 8: Hypotheses Summary

Nr Hypothesis Accepted/Rejected

H1 Collaboration Software (CS) usage frequency influences team performance. The higher the CS usage frequency the better employees perform their job in the team.

Accepted

H2 The relationship between CS usage frequency and team performance is mediated by intrinsic motivation.

Accepted

H2a CS usage frequency influences intrinsic motivation. The higher the CS usage frequency the higher the employees' intrinsic motivation.

Accepted

H2b Intrinsic motivation influences team performance. The more intrinsically motivated employees are the better they perform their job in the team.

Accepted

H3 The relationship between CS usage frequency and intrinsic motivation is moderated by CS type. The intrinsic motivation of using CS frequently will be affected by the type of CS used.

Rejected

H4 The relationship between CS usage frequency and team performance is moderated by CS type. The team performance of employees using CS frequently will be affected by the type of CS used.

Rejected

(42)

Figure 9 shows the strength of the hypothesized relationships. It can be seen that all hypotheses, besides H3 and H4, are supported by the data.

Figure 9: Strength hypothesized relationships Note: *p<.05 | **p<.01 | ***p<.001 CS Usage Frequency

Intrinsic Motivation

Team Performance CS Type

*Basic -

*Advanced +

.0004*

.0058 – -.0034 +

.0004**

.0083**

.0004 – .0004 +

.0206***

(43)

5 DISCUSSION OF RESULTS

This research is conducted in order to obtain a better understanding of how collaboration software can influence the internal motivation and team performance of employees. With respect to the actual hypothesis testing, the results show that most of the hypotheses are significant. To be more specific, hypotheses 1, 2, 2a and 2b could be accepted.

An important finding is that the use of collaboration software has a significant effect on intrinsic motivation and team performance. This confirms findings of earlier research that there is a value-added dimension when collaboration and a unified approach is promoted in the workplace (Landau, 2018).

For hypothesis 1, the more frequent collaboration software is used the better employees perform their job in the team. This supports the notion that the frequency and intensity of collaboration leads higher performance and organizational success (Asdemir et al, 2012). For hypothesis 2, intrinsic motivation plays a major role and is supported by its significant effect on team performance while using collaboration software. Intrinsic motivation occurs when people value an activity, this is in line with the Self-Determination Theory by Ryan and Deci (2000). Hypothesis 2a posits how usage frequency directly affects intrinsic motivation by supporting the premise that the more software collaboration is used, the more intrinsically motivated employees are. This further supports hypothesis 2b that when employees are intrinsically motivated, they are able to perform their job more efficiently in the team. When employees have intrinsic motivation, they tend to look for better ways to perform tasks and concentrate more on the overall quality of their work (Morikawa, 2017).

Hypotheses 3 and 4 are rejected which implies the type of collaboration software has no separate distinction whether one is better than the other, but that both are seen to contribute equally to the overall job process. Respondents who use the basic type of collaboration