Loading...

Drivers of knowledge base adoption. Analysis of Czech corporate environment

by Zuzana Rakovska (Author) David Voňka (Author)

Master's Thesis 2015 88 Pages

Business economics - Miscellaneous

Excerpt

Table of Contents

Chapter 1
Introduction

Chapter 2
Literature Review and Theoretical Background
2.1 Critical Success Factor approach
2.2 Technology Acceptance Model
2.3 Gamification

Chapter 3
Design and Elements of Knowledge Bases

Chapter 4
Hypotheses
4.1 Hypothesis #1
4.2 Hypothesis #2
4.3 Hypothesis #3

Chapter 5
Data
5.1 Knowledge Management Systems
5.2 Capturing the Data
5.3 Data for the First Hypothesis
5.4 Data for the Second Hypothesis
5.5 Data for the Third Hypothesis

Chapter 6
Methodology
6.1 Poisson Regression and Negative Binomial Model
6.2 Zero-Inflated Negative Binomial Model
6.3 Random Effects Negative Binomial Model
6.4 Testing

Chapter 7
Results
7.1 Results for the First Hypothesis
7.2 Results for the Second Hypothesis
7.3 Results for the Third Hypothesis

Chapter 8
Conclusion
Bibliography
Appendices

Abstract

This paper analyses the process of knowledge-base adoption in the enterprise en­vironment. Using data from two knowledge-management systems operated by the company, Semanta, s.r.o. we studied the day-to-day interactions of employees us­ing the system and identified the important drivers of system adoption. We began by studying the effect of co-workers’ collaborative activities on knowledge creation within the system. It was found that they had a positive and significant impact upon overall knowledge creation and thus on adoption. Secondly, we explored how the newly defined concept of gamification could help determine and encourage an increase in knowledge creation. The use of gamification tools, such as the ”Hall of Fame” page, turned out to have significant influence in the adoption process. Thirdly, we examined how users continually seek knowledge within the system and how asking for missing information and being supplied with answers has an impact on adoption rates. It was shown that the quicker the responses and the more experts dealing with requests the greater the impact on knowledge base adoption. Finally, we showed that the size and character of the company deploying the knowledge management system does not influence the adoption drivers. This paper represents an effort to fill the literature gap surrounding effective knowledge-base adoption in an intra-company environment. Moreover, as far as we know, it represents the first attempt to esti­mate the relationship between gamification concepts and knowledge-base adoption not only in the Czech Republic but also worldwide.

illustration not visible in this excerpt

Chapter 1 Introduction

The world as of today depends highly on exchange of information, its process­ing and utilization. Knowledge represents new intangible asset that companies accumulate and use to achieve their business goals. Effective knowledge man­agement is capable of inducing cost reduction as well as creating competitive advantage in the market. However, the extraction of such benefits does not depend on installation of knowledge management system. Its cornerstone is knowledge-base adoption by firms’ culture.

Every employee, not only directors and managers, possesses a certain knowl­edge that is unique for company. For example, one might know how to provide best services to customer, another one is experienced in product design and there might be a project manager who knows how to lead a project to be profitable. All these workers represent company’s intellectual capital that is essential in creating competitive products and services. However, if such ac­quired knowledge remains only in their minds, company might simply lose part of its know-how when the employee leaves. To prevent this, many firms are de­ploying knowledge management systems (KMS). It is a widely spread solution that captures workers’ unique insights and stores them into knowledge base. As a result, such collected experience is transformed into corporate one which can not be simply removed because firm is now able to control it. Moreover, all knowledge is stored in one place that is available to every worker.

Knowledge management system can be considered as a ’’modern production technology” whose output - knowledge, exhibits increasing returns to scale. Firstly, KMS makes its content available to experts who are able to extract any previously used and shared solution, and adapt it to a current problem. It provides expertise to less experienced personnel and also avoid delays when expertize is needed (Smith 1985). Hence, each such task-execution is facilitated and workers’ costs are reduced. For example, a newly hired consultant saves her working hours when using stored knowledge of senior consultants about customers’ needs, etc. (Ofek & Sarvary 2001). Secondly, accumulated knowl­edge creates space for learning from experience and lead to better solutions. Knowledge management system simply keeps record of decisions and actions that are consistent and available over time. This leads to higher-quality pro­cesses and services that create competitive advantage and superior performance in the relevant market (Gjurovikj 2000).

In theory, knowledge management system is a powerful tool in achieving strategic objectives. However, like every production technology, also KMS needs inputs for proper functionality. On the one hand, experts are irreplace­able intake in knowledge-base production (Davenport et al. 1989). If they are not willing to share their unique experience to others via knowledge-base channel, benefits from economies of scale are not in place and company is los­ing competitiveness (Wong & Aspinwall 2005; Ritchie et al. 2011). On the other hand, workers that are not sufficiently motivated to seek and use already created knowledge are less effective and increase firm’s costs. Hence, incor­poration of knowledge management into company’s processes is not the final success factor. Employees must be willing to hand over their knowledge and use the corporate information in order to induce cost-efficiency. In other words, knowledge base must be adopted among them.

Although knowledge management has been widely discussed in the last decade, there are only few studies capturing the process of knowledge-base adoption within a firm culture. The respective literature gap results mainly from the subject’s novelty and from the lack of empirical data in the intra­company and also in the inter-company level. Knowledge base acceptance is a cornerstone for successful KMS which ensures long term sustainability of its benefits (Huang & Lai 2014; Suresh 2013; Yeoh & Koronios 2010, and oth­ers). However, users’/employees’ adoption is not self-acting. It needs stimulus through which workers are motivated to create and seek for contents of knowl­edge base. This paper hence, represents effort to fill the literature gap on knowledge-base adoption and provides a comprehensive explanation and esti­mation of drivers affecting knowledge-base adoption.

We center our study on analysis of knowledge bases designed by company Semanta, s.r.o., which develops and deploys knowledge management systems (KMSs) for enterprise clients all over the world. Its KMSs are available through internet and are in form of web application similar to Wikipedia. It is repre­sented by set of pages organized into trees with hierarchies of classes and sub­classes referencing to each other. These pages differ from usual web pages in a sense that every user has access to them and is able to produce their content by editing them, adding new information, creating sub-pages, etc., just like in Wikipedia. In addition, Semanta’s KMSs employ additional tool for content creation: inserting comments to already generated pages. Knowledge base is thus a result of collaborative, non-proprietary production process, based on sharing resources and outputs among individuals (Aaltonen & Seiler 2014).

Semanta stores information on every user’s action performed in its system. These captured actions are organized in tables in which every row represents detailed information on who did what, when and where (exact page) it hap­pened, etc. We were thus able to extract data capturing history of system­users activities and collaboration with other users, or observations on certain actions performed. We have already discussed two parts of knowledge-base adoption that are considered in this paper: continuous knowledge (con­tent) creation and continuous knowledge-seeking. Knowledge creation arises when employees generate new pages or when they edit them or com­ment them. Knowledge seeking means using knowledge base, which done by visiting its content (pages) by system users. To assess continuous actions, we are studying counts of such events (knowledge creation and knowledge seeking) performed by studied employees in one-week long periods. On the one hand, adoption is induced by factors affecting amount of pages created, edited or commented by a user in consecutive weeks. On the other hand, it is induced by drivers affecting amount of system visits by a user in a week.

Firstly, we analyze activity of knowledge creator (a user who creates con­tent) and how it is induced by activity of other users within knowledge-base space. Nature of KMSs studied in this paper, allows all users to add small pieces of information, while relying on subsequent editors or commenters to develop the content further (a knowledge creator might create a page with only raw data and reviewers then add more information by subsequent edits of the page, for example). We consider such collaboration to be strong motivational tool for knowledge creator leading her to generate another content. Hence, the first studied factor is collaborative activity of other co-workers, including commenting, viewing or editing the created knowledge. Secondly, in order to identify other drivers related to content creation, we exploit new concept called gamification. This construct employs those elements from games that engage ’’players” to stay in game (like points, badges, leader-boards, etc.) and apply them in other non-game contexts (Leeson 2013). Semanta is directly incorpo­rating leader-board-based gamified tool in its KMSs, named Hall of Fame. It is in a form of page showing users who were the most active in a previous week and achieved first five positions in different categories. Such tracked category is the Commenter, for example. Selecting this category, Hall of Fame page shows the first five contributors who inserted the highest number of comments into knowledge base during the previous week. We studied a number of activities connected to Semanta’s gamified Hall of Fame page and how these activities influence knowledge creators in content development. The main motivational mechanisms we study in such manner are: viewing placements reached in Hall of Fame leader-board and the incentive resulting from reaching/not reaching the actual placements. Finally, we study drivers of continuous knowledge seek­ing as the second important part of knowledge-base adoption. We assume that an employee continuously search for precious information in knowledge base when she was satisfied with previous experience in seeking any of it. We em­ploy feature that is a part of Semanta’s KMSs and which allows workers to ask system’s experts to deliver missing knowledge in the base. This is done by using an Ask button placed in pages within knowledge base. Here the analyzed drivers are: the speed with which system experts (users of KMS) answer the request, that is set by other employees, and the variety and amount of these answers.

We work with different data in each analyzed hypothesis therefore, our results are estimated using three different methodologies. In the first two hy­potheses, we are dealing with panel data of users across weeks in which de­pendent variables are weekly amounts of content (knowledge) generated by a knowledge creator. Independent variables are build similarly, and represent counts of given activities interacting with the content build by knowledge cre­ator. Dependent variables used in both hypothesis are suffering from overdis­persion. Moreover, in the first hypothesis, the excess zero problem is detected. As a result, we transformed the first panel into cross-sectional data by using dummy variables to estimate fixed effects and employed zero-inflated negative binomial model (ZINB). The second panel was estimated using random effects negative binomial model (RE NegBin). In the third hypothesis, we concentrate only on users that asked experts for a missing knowledge (using the Ask but­ton). We study effects on number of visits performed by these employees in a week after they obtained answers. Hence, we are not dealing with panel of users across weeks which were applicable in the first two hypotheses. Instead, the data are structured into cross-section of questions that were once asked by an employee whose activity (further knowledge seeking) is than subject of our analysis. Since overdispersion is present also in this case, we use standard negative binomial model for event counts again to control it.

Our framework is innovative in the way that we will assess intra-company interactions between workers as main factors, while the literature concentrates mainly on studying those that are arising from inter-company relations. This enables us to study direct influences on KMS acceptance on a firm level. Fur­ther, as far as we know, this paper represents the first attempt not only in the Czech Republic but also worldwide, to estimate relation between newly defined concept of gamification and knowledge-base adoption. And finally, employing two knowledge management systems, that differ in size of deploying firm, allows us to study the importance of knowledge-base size in intra-company environ­ment.

Using the data and methodology described above, we found overall pos­itive and significant effects of co-workers’ collaborative activities on further knowledge creation. Moreover, usage of gamified tools within knowledge bases, turned to be another important driver for the content generation. Study of factors affecting knowledge seeking proved that quick responses and number of experts dealing with requests boost knowledge-base adoption. And finally, we showed that the size and character of company deploying knowledge manage­ment system do not generate different results in our setting (results for knowl­edge management systems of both, small and big company, revealed almost the same drivers for knowledge base adoption).

The paper is structured as follows: Chapter 2 summarizes possible ap­proaches in analyzing adoption process and introduce the gamification concept in relation to knowledge management systems. Chapter 3 offers description of knowledge bases designed by Semanta and their elements. In Chapter 4 we discuss studied hypotheses in detail. Chapter 5 characterizes extraction of data, provides its description and defines variables used. Chapter 6 specifies methodology that we work with and Chapter 7 reports results of our empirical research. Chapter 8 summarizes our findings and offers suggestions for further study.

Chapter 2 Literature Review and Theoretical Background

The importance of knowledge management (KM) adoption in corporate envi­ronment was emphasized in several studies. Although, this area is very recent, a number of approaches have been developed to examine the forces that im­pact effective knowledge management implementation. These concepts differ mainly in understanding of knowledge management system (KMS) but also in interpretation of the adoption process. They can be specified as follows:

1. Critical Success Factor (CSF) approach - studies and ranks criti­cal factors that affect successful adoption of knowledge management and suggests the construction of a hierarchy according to importance.
2. Approach that utilizes Technology Acceptance Model (TAM) -
regards knowledge management system as innovation and examines the behavioural intentions of users to accept this innovation.
3. Approach utilizing the concept of Gamification - leverages from the structure of game elements and explains their effect on knowledge management adoption.

2.1 Critical Success Factor approach

The goal of this framework is to determine drivers that systematically pre­dict knowledge-base acceptance among single users or firms. Such extraction of important factors, that impacts the effective functionality and adoption of knowledge management, have been studied in the number of areas and from different perspectives (inter-company or intra-company level, etc.). The Crit­ical Success Factor concept was employed in Wong & Aspinwall (2005) who studied small and medium-sized enterprises. Authors used data from postal surveys in analysis of the hierarchy of eleven factors affecting the adoption. These factors were extracted using review of studies rooted in what ’’early adopters”, i.e. large companies, were doing to take advantage of their knowl­edge. In the next step, the respondents of postal questionnaire were asked to rank the factors according to importance. The unit of analysis used here was the organization, thus, single form approach rather than multi-form one was followed (postal survey was answered once by company as a whole and not by every manager in a firm). The first place in the final ranking of critical success factors was encroached by management leadership and support. The manage­ment, thus, promotes co-operation and knowledge sharing across company and also provides support to initiate and sustain effort of employees to create con­tent. The second place belonged to culture of the company. This means that knowledge-oriented cultural foundation determined by trust, collaboration and openness is more important than deployment of KMS. Moreover, result sug­gests that management and firm’s culture, that create company’s environment and that determines the willingness of employees to participate in knowledge accumulation (Leeson 2013), is an important critical success factor.

Suresh (2013) investigated factors affecting adoption of knowledge manage­ment system in various Indian industries. The methodology used resembles the one employed in the previous study of Wong & Aspinwall (2005) and differs in subject matter, which is in this case, middle and top level managers in a firm in­stead of a single organization (multi-form approach). The results are ranked ac­cording to the quality of success and in detail describe all the elements engaged in a knowledge management system acceptance process. Recognition of knowl­edge and organization culture were placed in the top of the hierarchy and are considered to be certainly more important predictors of adoption than deploy­ment of KMS technology. Suresh (2013) identified components of these factors for better understanding of how they drive knowledge acceptance within a com­pany, and such components were submitted into the questionnaire. The above mentioned recognition of knowledge thus, includes for example, recognition of employee’s contribution towards knowledge management (firm should attract and retain talented people who are able to deliver good knowledge), or knowl­edge sharing that firm induce by making content of knowledge base available. The other factor, organizational culture, is determined by knowledge-intensive environment, collaboration, emphasize on knowledge sharing and trust. The final ranking of factors divided into components provides a deeper analysis of drivers for knowledge management adoption, practice and innovation.

Yeoh & Koronios (2010) employed critical success factor approach to study successful implementation of business intelligence systems. He argues that crit­ical success factors applicable to other types of information systems may not necessarily apply to a contemporary business intelligence system.1 In contrary to previous studies, he thus, utilized different method for critical success fac­tors, and success measures extraction was applied on five different organizations (cases). According to his findings, system use is one of the three measures that determine successful implementation of business intelligence systems (in addi­tion to system quality and information quality). Moreover, he indicates that organizational and process-related factors are more influential than technical factors.

To analyze drivers that influence knowledge accumulation in knowledge base, it is vital to look at the behavior of knowledge creators, users that system­atically interact within installed system. So far, authors employed companies (or executive officers per each company) as unit of analysis (Yeoh & Koronios 2010; Suresh 2013; Wong & Aspinwall 2005). On the one hand, this approach provides an insight from unit that controls all the processes and thus, under­stands the application of knowledge management system. On the other hand, knowledge base is a collaborative product conducted by workers who are willing to share their precious knowledge. Hence, analysis within a firm instead of anal­ysis between firms is needed. Abril (2007) applied this approach and studied adoption of KMS through behavioral model aimed on workers in a single corpo­ration. Using the shadowing and action research he identified following drivers of behavioral change towards KM adoption: personalized value, executive spon­sorship, enabling support organization or incremental perceived success. By personalized value the author means that if managers who are responsible for hiring employees are perceived about value of knowledge-base adoption then employees’ cultural change towards the adoption would be induced. Executive sponsorship also induce adoption, this time, by means of inclusion of knowledge[1] management objectives at the leadership team. Enabling-support-organization driver represents creation of collaborative environment for teams. And finally, incremental perceived success assures that knowledge management system has to be perceived to successfully affect such behavioral change. This study pro­vides complex outlook to the day-in-a-life storyboards of employees and explain their motivational aspects to participate in KMS. However, users’ interaction via knowledge base is not captured, and study lacks this deeper insight into firm’s processes.

2.2 Technology Acceptance Model

The Technology Acceptance Model was introduced by Davis (1992) to explain why a user want to use technological innovation. These individuals’ intentions are determined by two beliefs: perceived usefulness, defined as the extent to which a worker believes that the use of a particular system would increase his job performance, and the second, perceived ease of use, defined as the extent to which a user believes that using such a system will be free of effort. In this sense, knowledge management system is considered to be an innovation in a company and these studies examine the factors that lead workers to ac­cept this innovation. Huang & Lai (2014) utilized the technology acceptance model approach to study the effects of three factors on attitude towards knowl­edge management adoption: perceived usefulness, complexity of the system and the subjective norm, defined as perceived pressure or expectations of the community that affect the decision to engage or not to engage in a certain behavior. Author found the positive relationship between perceived usefulness and technology acceptance, and also between the subjective norms and behav­ioral intentions to accept the technology. In case of complexity, its relationship to users’ attitude to accept knowledge management was proved to be negative. Ritchie et al. (2011) employed the technology acceptance model and analyzed influences of perceived usefulness and perceived ease of use on the behavioral intentions. He states that user acceptance of knowledge management system depends not only on a technology acceptance, but also on the organizational and cultural influences. Technology acceptance model was also utilized by Kuo & Lee (2011) who studied effects of perceived usefulness and perceived ease of use on users’ behavior. Additionally, he determined compatibility factor to be important in a sense that if the usage of knowledge management system is com­patible with the work practices of the users, it also enhances their intention to use the KMS. Employing structural model and principal component analysis, he confirmed the positive relationships between behavioral intentions to accept knowledge management system and all three factors.

Hou (2014) investigated determinants leading users to accept business intel­ligence systems using technology acceptance model. Additionally, he employed its extensions that considers attitudes, subjective norms and perceived behav­ioral control. According to his findings, the important influences on users’ be­havioral intention to use business intelligence systems are employees’ attitude cultivation and subjective norms. Thus, both peer opinions and managers’ appreciation of successful use of business intelligence platform, may motivate users to use this platform.

Taking into account the same implications as in the technology acceptance model (suggesting that the process of knowledge management system imple­mentation can be considered as the process of innovation), interesting results can be found in paper written by Gopalakrishnan & Bierly (2001). In this study author examines impact of three innovation types based on dimensions of knowledge on innovation adoption.[2] Results suggest that the more tacit (un­able to codify or articulate) and complex knowledge associated with innovation, the higher level of innovation adoption is reached.

The goal of studies utilizing technology acceptance model and critical suc­cess factor approach is to determine factors that should be emphasized in order to enhance adoption of knowledge management system. The findings of such studies serve as systematic guidance for companies according to which they might direct their management. For example, Suresh (2013) highlighted shar­ing knowledge as one of the important components of critical success factors. Following technology acceptance approach, Hou (2014) identified employees’ attitude cultivation as a driver of behavioral intention to use knowledge man­agement system. Although, these results define functionality of knowledge-base platforms, they lack deeper explanation of how such factors can be used to mo­tivate users to create content. In other words, knowledge-base creation is in hands of knowers (Davenport et al. 1989) thus, analysis of drivers that affect users’ motivation and behavior towards collaboration should be emphasized before general firm-level factors are considered. The following section thus, of­fers the new concept, called Gamification, that might be capable of influencing productive behaviors of users (Leeson 2013).

2.3 Gamification

In an innovative paper Leeson (2013) argues that the culture is a lynchpin that will determine the workers collaboration and system adoption and that the valuable tool to encourage this process is a new concept, called gamification.

2.3.1 Gamification Concept

Hamari (2013) defines gamification as a process in which services are enhanced with motivational stimulus in order to invoke gameful experience and further behavioral outcomes. Deterding & Dixon (2011) provides simpler approach and define gamification as ”the use of game design elements in non-game con­texts”. It is a new term for relatively old method. One example might be education and its gamified approaches from Scrabble used to teach spelling to duoLingo - application for learning languages.[3] We can also find its main characteristics in strategies that are used to maintain customers interaction like Customers Relationship Management including loyalty systems, etc. (Bal- lance 2013). In practice, only one part of games is incorporated in non-game contexts - scoring.[4] Users of gamified systems are thus motivated to use such system more by obtaining points, badges or reaching leader-boards and higher levels. Since new technology era, principally era of smart-phones and tablets, gamification is strongly connected to social interaction (likes or dislikes from other users/players, etc.). Hence, gamified experience brings not only feeling of self pride (by reaching leader-board or more points in some activity) but also satisfies the need for socializing (Moise 2013). Gamification then seems like reasonable approach for motivating knowers/employees to deliver and seek further content.

2.3.2 Gamification in Knowledge-Base Adoption

In theory, gamification can be divided into three parts: 1) the implemented mo­tivational stimulus, 2) the resulting psychological outcomes, and 3) the further behavioral outcomes (Hamari & Sarsa 2014). Leeson (2013) suggests that the correct combination of game mechanics and behavioral economics may lead to long run increase in users’ intention to accept KM and share their knowledge.

However, he emphasizes that the most important issue is to direct employ­ees to realize the inherent benefits of collaboration via boosting their intrinsic motivation instead of the extrinsic one.[5] Thus, even the introduction of such gamified tools as badges or leaderboards in knowledge-base platform can lead only to short run change in users’ behavior. This statement is also supported by Nicholson (2012). In his paper, he claims that rewards can reduce internal motivations as firm which temporarily implements external payoff system will be after quitting such a program worse off than before implementation. Users will be simply less likely to return to the behavior without the external reward.

The working idea utilized in the paper by Leeson (2013) is approach pro­posed by Pink (2009). He argues that human motivation is largely intrinsic and he identifies three powerful ways to induce this kind of motivation: au­tonomy, mastery and purpose. Autonomy allows users to set their own goals and to control their activity. The more free is a knower to decide how to col­laborate (write a comment, thank for created page, etc.) the more he will be engaged in sharing a knowledge. Mastery is about obtaining a good skill in something which yields own inherent benefits. And finally, purpose ensures the social connection to the larger entity via the channel of making a broader impact. Collaboration within a knowledge management system therefore, leads to a higher purpose - further creation of a collective wealth of information and experience.

Although, the concept of gamification is new and is still evolving, some studies have analyzed effects of implementing the gamified tools into knowl­edge management system on further content creation. Farzan (2008, 2008a) in his work utilized the system of points in networking website for employees and studied the impact on their collaboration. In his framework, he divided users into two groups. The experimental group which was rewarded by points if en­gaged in the knowledge creation and the control group that was not rewarded and did not know about point system. The framework has several elements, but the most important is the idea that the user from control group (without rewards) is in the long run motivated by the higher activity of other group members and also by previous activity of experimental group. In particular, results of both studies showed that the point system does increase the knowl­edge creation of the experimental group and that their higher activity serves as an intrinsic trigger for control group participation in the next period. More­over, Farzan et al. (2008) states that gamified tools installed in the knowledge base stimulate the discussion among users and that workers’ action depends on what others do. Farzan & DiMicco (2008) in their study utilized data in form of a log into the database, where every action of all users is recorded, so as an independent observer can analyze which activity within a studied sys­tem contributes to content creation. The main idea is that, this method of data accumulation provides detailed insight into creation of each component of knowledge base and hence, allows studying the incentives’ characteristics.

In the next study, Farzan & Brusilovsky (2011) employed new incentive scheme installed in community-based course recommended system that pro­vides personalized access to information about courses and which turns user participation into a self-beneficial activity. Users are here provided with in­centive scheme that motivates them to collaborate and rank courses. In par­ticular, the users evaluate the relevance of each taken course to each of their self-selected career goals. When subsequent users are choosing courses they can decide according to the degree of relevance toward their goals and hence, contributors’ activity is beneficial to the community as a whole when users engage for the activity itself. Students are then supposed to be motivated by the tool that shows their progress towards their self-selected goal. This was followed by subsequent analysis in which the positive relationship between working mechanism and users’ collaboration was found. Nevertheless, author emphasizes the problem of self-deception that can cause the higher rating by students who want to attain a higher visible progress. The study thus, hints the deep consideration of the incentive mechanisms used, as effect of extrinsic motivation on intrinsic one can raise the possible drawbacks. Author further argues that in both large and small communities, the most important issue is to motivate the largest percentage of users possible to contribute. While small knowledge-management-systems’ survival depend on contribution of ma­jority of users, larger communities (like Wikipedia) with large amount of users is able to survive with small percentage of contributors. However, even such big knowledge base can suffer from participation inequality bias problem when small percentage of users represents the views of larger population.[6]

Finally, the connection between possible implications of gamified knowledge base and behavioral economics was offered by Hamari (2011). He suggests that concepts utilized in behavioral economics can be used to explain the effects of game patterns installed directly in knowledge management system. The main concept used in this study is loss aversion in connection with prospect theory, according which losses loom larger than corresponding gains (Kahne- man & Tversky 1979). This framework of decision making under risk that systematically violate the predictions of expected utility theory has been found in decision making in different areas like consumption-savings decisions, labor supply or insurance (Barberis 2013). Hamari (2011) further suggests sunk-cost fallacy theory (Arkes & Blumer 1985) to explain potential intentions leading users to participate in further content generation. Along the lines of this the­ory, people are far more willing to invest to the activity that they have already invested in. Therefore, supposing risk aversion of users and an assumption that a proper incentive scheme is in place, users participate because they have already participated before. Adoption of knowledge base therefore, depends on users’ previous activity.

Chapter 3 Design and Elements of Knowledge Bases

Before we introduce the framework of our study it is essential to describe how knowledge bases designed by Semanta work and what are their elements. As discussed in Chapter 1, knowledge management systems employed in our anal­ysis are available to workers through common web browser. They are appli­cations based on user-generated content similar to Wikipedia, consisting of a huge number of pages organized into trees and hierarchically ordered in classes and subclasses. Users of these systems are able to see how pages are related and can be navigated to other pages using links. The most important characteristic of such systems is that within them pages can be easily created or edited. This allows users to collaborate and continuously create content, compared to the usual web pages that can be only visited without possibility to contribute to them.

Knowledge management systems are not typical open-sources as they are open only to individuals with granted permission - usually employees or other external workers. However, they incorporate many elements used by popular websites (Wikipedia, Facebook, LinkedIn, etc.) used to share individually pro­duced content. There are dozens of such elements and features which change and evolve over time. Therefore, we will concentrate only on those which are most important and most widely used. These elements are: creating pages, editing pages, commenting pages, Thank you buton and Ask button.

- Creating Pages is within analyzed knowledge bases performed by button Add Page that can be found in all system pages from Home Page to the last page in tree-like hierarchy. This means that users are able to create their own content by placing their pages anywhere in a tree, directly specifying its relation to other pages (parent page, child page, etc.). The process of creation is done through automatic form that is displayed after user clicks on Add Page button. The form requires specification of page title and insertion of the content (text, table, picture, figure, attachment, etc.).
- Editing Pages is accessible using Edit button. As in previous case, this button is part of every page in a system, however, some of them might be restricted and can be edited only by some employees (for example, pages containing important information on suppliers can be edited only by administrator). In such situations Edit button is still present but is not active and after clicking on it the edit form is not displayed. Thus, if a user is allowed to edit some page, and she clicks on the button, automatic form appears and is pre-filled by the page content. User can rewrite it, add new passages, insert or delete tables, graphs, attachments, etc. Any such change in original content of the page is considered as edit (even if a user only corrects the grammar).
- Commenting Pages is an element mostly known from social networks. As well as any content inserted in these networks (blogpost, photos, etc.), also pages in knowledge bases can be commented by other users. Users can find pre-inserted box at the bottom of every page, fill it and click on Send button to save it. Comments are then immediately showed on the given page. Comments are usually created by other user than the one that generated the page, while edits are usually performed by page­creator herself.
- Thank you button is again element well known from social networks. Using example of Facebook, this button is similar to Like. The button is placed in the bottom of every page and after user clicks on it, the button changes the color and the information that the page was thanked appears next to it.
- Ask button is present in the bottom of every page next to Thank you button. By clicking on it, a user is provided with a form in which she can specify a question or request and assign it a title. By submitting this form the question is directly sent to a relevant expert and is saved as a new page in a special section of knowledge base designed for requests. This question is then answered by inserting comments to its corresponding page so as it is visible for everyone. Alternatively, the question is answered using ”The Best Answer” box by editing the question-page.[7] Moreover, not only experts but also other co-workers are allowed to join the answering process and insert comments to such pages. This element is very important as it allows users to ask for knowledge when they can not find it or when it is simply missing.

Chapter 4 Hypotheses

The paper estimates the effect of various activities performed by users of knowl­edge management system, on the adoption of this respective system. Following Kuo & Lee (2011), knowledge base is adopted if users continuously share and seek knowledge within it. By sharing knowledge via knowledge management system, users convert their own personal knowledge into corporate one - they are creating knowledge base. By seeking knowledge they are extracting bene­fits of corporate knowledge which leads to facilitation of users’ task execution (Suresh 2013). Thus, the two components of adoption can be stated as:

(i) continuous creation of knowledge base’s content, and

(ii) continuous knowledge seeking.

4.1 Hypothesis #1

According to Farzan & DiMicco (2008), user’s content production within knowl­edge management systems is enhanced by activity of other users. As discussed in previous chapters, we assume three different users’ actions leading to content generation: creation of pages, editing of pages, and inserting comments. The nature of KMSs analyzed allows users to visit the created content and collabo­rate on it by commenting or thanking the creator. We assume that such activity of co-workers interacting with the content creator positively affects creator’s in­tentions to generate more content. In simple words, if co-workers are visiting creators pages, or if they are commenting it or giving ’’thanks”, the creator should be motivated to add more pages into knowledge base (as she assume her content to be important to others). Alternatively, depending on nature of comments, creator might be motivated to edit the page in order to correct or elaborate more on ideas, etc. Moreover, broader knowledge base (as for the amount of content), provides more opportunities for collaboration. Thus, we also expect that knowledge-base size positively affects users’ contributions. So, we formulate the first hypothesis as:

Hypothesis #1: Further content creation (creation of pages, edits of pages and comments) depends on collaborative activity of other users - page visits, page comments, thanks for pages as well as on knowledge-base size.

4.2 Hypothesis #2

Users’ collaboration and knowledge sharing are essential determinants of suc­cessful knowledge-base adoption. To study the effects of users’ activity on sharing knowledge (and thus, on knowledge-base adoption) we will also employ the gamification concept (Section 2.3). This construct uses game elements (those that make games engaging and attractive for ’’players”) and apply these components in other contexts (Leeson 2013). It offers an answer on how to promote desirable users’ activity within a system. In other words, gamification is the solution, in which content of knowledge base is created because users are motivated to contribute and collaborate by ”gamified” tools directly in­stalled in knowledge management system. This paper considers such tool to be leader-board-like Hall of Fame placement.

4.2.1 Hall of Fame Page

The Hall of Fame represents a single page in KM system, that serves as an information portal about top 5 positions in workers’ collaboration displayed in several different categories. Activity of all users is here evaluated and any viewer of Hall of Fame page can see chart of people who dominated in given category in a previous week. Tracked categories are:

1. contributor - sequence of maximum of 5 users who created and edited the highest amount of pages in the previous week,
2. commenter - sequence of maximum of 5 users who commented the highest amount of pages in the previous week,
3. consumer - sequence of maximum of 5 users who visited the highest amount of pages in the previous week,

[...]


[1] Business intelligence can be defined as ”a collection of tools and methodologies that transform the raw data that companies collect from their various operations into useable and actionable information” (Kaula 2015). According to Yeoh & Koronios (2010), implementing a business intelligence system is not an activity that includes the purchase of software and hardware but it is a complex adoption requiring appropriate infrastructure and resources over a lengthy period.

[2] The dimensions of knowledge are tacit-explicit, systematic-autonomous, and simple- complex (Gopalakrishnan & Bierly 2001).

[3] www.duolingo.com

[4] Nicholson (2012) suggests ”pointsification” as a label for gamification systems that add nothing more than a scoring system to a non-game activity.

[5] Intrinsic motivation happens when people engage in activities for the activity itself and without any obvious external incentives such as rewards. Extrinsic motivation happens when people engage in activities as a result of an external incentive mechanism such as contingent rewards (Farzan & Brusilovsky 2011).

[6] As of 2008, Wikipedia had 684 milion unique users, while only 75 000 (0.01% ) of them actively contributed (Farzan & Brusilovsky 2011).

[7] To make it clear, suppose we would like to ask the following question: ”What is this paper about?”. We will fill the request form (predetermined by the system), which will allow us to state our question and to specify for example title of a question (let’s make it ”Thesis”) or the field of a question (let’s suppose ’’academic”). This will directly create a single page (our question) in the space dedicated to requests and will automatically notify the expert in ’’academic” field about new item added. The expert (or any other system’s user including the asking person) is able to comment this page or to edit this page (only the expert or the admin) by filling the special prearranged box ”The Best Answer”. The both actions produce the response to the question.

Details

Pages
88
Year
2015
ISBN (eBook)
9783668166295
ISBN (Book)
9783668166301
File size
880 KB
Language
English
Catalog Number
v317285
Institution / College
Charles University in Prague – Institute of Economic Studies
Grade
1,00
Tags
drivers analysis czech

Authors

Previous

Title: Drivers of knowledge base adoption. Analysis of Czech corporate environment