Web's Largest listing of crowdsourcing and crowdfunding events

Events

Advertise

Web's Largest Directory of Sites

2,342 crowdsourcing and crowdfunding sites

Understanding Crowdsourcing (Effects of Motivation and Rewards on Participation and Performance in Voluntary Online Activities)

document Distributed Knowledge

IRMA BORST

Understanding Crowdsourcing
Effects of Motivation and Rewards on Participation and Performance in Voluntary Online Activities

Understanding Crowdsourcing
Effects of motivation and rewards on participation and performance in voluntary online activities

Understanding Crowdsourcing
Effects of motivation and rewards on participation and performance in voluntary online activities

Over uitbesteden aan de massa
Effecten van motivatie en beloningen op deelname en prestaties in vrijwillige online activiteiten

Proefschrift

ter verkrijging van de graad van doctor aan de Erasmus Universiteit Rotterdam op gezag van de rector magnificus Prof.dr. H.G. Schmidt en volgens besluit van het College voor Promoties. De openbare verdediging zal plaatsvinden op donderdag 23 december 2010 om 15.30 uur door

Wilhelmina Angelina Maria Borst
geboren te Utrecht.

Promotiecommissie

Promotoren:

Prof.dr.ir. J.C.M. van den Ende Prof.dr.ir. H.W.G.M. van Heck Prof.dr. G.H. van Bruggen Prof.dr. N. van Yperen Prof.dr. C.L. Tucci

Overige leden:

Erasmus Research Institute of Management – ERIM Rotterdam School of Management (RSM) Erasmus School of Economics (ESE) Erasmus University Rotterdam Internet: http://www.erim.eur.nl ERIM Electronic Series Portal: http://hdl.handle.net/1765/1 ERIM PhD Series in Research in Management, 221 ERIM reference number: EPS-2010-221-LIS ISBN 978-90-5892-262-5 © 2010, Irma Borst Design: B&T Ontwerp en advies www.b-en-t.nl Layout & pictures: Michael van Roosmalen, Appello BV Print: Haveka www.haveka.nl All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the author.

FOREWORD

V

Foreword

It was not obvious that I would start a PhD project; it was not my long-cherished wish. When I started my working life as a nurse, I did not dream to become a doctor somewhere in the distant future. But after my move from the health sector to management consultancy and involvement in a number of research projects, I slowly grew into the idea of engaging in scientific research. Through the projects Ecolead and B@home, I discovered that research is more interesting and challenging than I suspected. Together with Jan van den Ende, I submitted a research proposal to NWO (Dutch Scientific Association). We were very happy that financing for our project was granted.

Jan has not only played a crucial role in the initiation of my PhD project, but also in the entire execution. As my promotor, Jan was very much involved in my day-to-day research activities. I could always knock on his door for advice, which I certainly did. In particular, the mathematical knowledge that Jan shared in the interpretation of the Negative Binomial Regression was invaluable to me. When looking back, I see clearly that Jan’s critical and additional, new questions – which I was not happy with at the time they were raised – led to the significantly improved results of my research. Jan, thank you very much for being a very supportive and stimulating promoter!

Two other persons, Eric van Heck and my Logica colleague Geleyn Meijer were closely involved in my thesis project. Thank you, Eric and Geleyn for providing regular feedback on the intermediate results and providing advice on how to move on to the next activities, all with the goal of spurring on my research project. A further word of thanks goes to the other professors that are in my committee: Gerrit van Bruggen, Petra de Weerd-Nederhof, Nico van Yperen, Chris Tucci and Harry Barkema; thank you for taking up the task of committee member.

Although not a member of my committee, I really would like to thank Michael Jensen who acted as a friendly reviewer before Jan and I submitted our article to a journal. His

VI

UNDERSTANDING CROWDSOURCING

suggestions for new analysis appeared to be extremely valuable and I hope that we will succeed in publishing our article in a top journal.

Besides the support of academic colleagues, I also received a lot of support from nonacademic people. First of all, I would like to thank the crowd of respondents. Over 1900 respondents revealed their motivations for their online behavior. I calculated that this crowd jointly spent 282 hours on filling in the websurveys, therewith contributing more than 7 weeks of work to my research. I would also like to thank Wilbert de Vries (Deputy General Editor, Tweakers.net), Jesse Burkunk (Product Manager, NUfoto.nl) and Femke Rotteveel (Coordinator Green Challenge, Dutch Postcode Lottery) who provided me with essential information on the participation and performance of respondents. They facilitated the data gathering from multiple sources which is a major strength in my research design. I would also like to thank the expert panelists that plaid a crucial role in the NUfoto.nl and the Green Challenge studies. Renata Bauer, Bart de Rijk and Dirk Schiemanck: thank you for spending your free time on the assessment of more than 750 newsphotos. Marjolijn Bloemmen, Jeffrey Prins and Femke Rotteveel: thank you for assessing the business plans, each 5 to 10 pages long, of the Green Challenge respondents. Without the work of these two expert juries, I would not have been able to reach the conclusions presented in chapters 6 and 7.

Next I am convinced that the help of Jordan Srour and Michael van Roosmalen substantially improved the quality and readability of my thesis. Jordan, thank you for the language check. And Michael, my personal graphical advisor, thank you for the nice graphics and the lay-out of this thesis. It is a pity that our experiment to use graphical tools in the interpretation of empirical data, did not work out as we expected. I was convinced that you could sell your solution to SPSS.

Financial compensation is not only an important topic in my studies, but it is also enabled my PhD project. NWO financed two years of research as part of the Network of Networks Program. Novay and RSM provided additional funding so that I could extend my research with another nine months. And finally Logica allowed me to engage in this research project.

FOREWORD

VII

I would like to mention that Logica’s support fit extremely well with Logica’s penultimate slogan ‘releasing your potential’. I doubt whether I would have started without financial support. Therefore the contributions of NWO, Novay, RSM and Logica were crucial for my thesis.

In this thesis, I cannot leave my social networks unmentioned. I very much enjoyed being part of RSM’s Department 6. I especially appreciated the lunches with my colleagues Dirk Deichmann, José Larco, Mahmut Ozdemir, Yugang Yu, Erik van Raaij, Henk de Vries, Nima Zaerpour, Amir Gharehgozli, Melek Akin, René de Koster, Daan Stam, Nishant Mishra, Costas Lioukas, Serge Rijsdijk, Koen Dittrich and Tony Hak, in which we had nice discussions on cultural differences and in particular on the Dutch habits. I noticed the hard work of my university colleagues; I was definitely not the sole person working on Saturdays on the 10th floor of the T-building. Department 6 was a very stimulating environment for me.

My networks of (ex)colleagues, friends and family showed considerable interest in the progress of my PhD project. I can now say: it’s almost done. The defense is the only task left. My collegial network mailed me a variety of questions that I can use in the preparation of my defense. I believe that I have two excellent paranimfen who will support me during the defense. Thank you, Elfi and Lonneke for your willingness to take up this task. But more importantly: thank you for your friendship. The two of you show that it is not true that good friendships can only be forged at young ages; it appears to be possible after your 30th or even 40th year.

I finally come to my strongest ties. My parents, Arnold and Thea Borst, taught me the attitude that ‘you can win if you want, if you want it you can win’. They now can be proud of their daughter. I am convinced that Joop and Wies Balk share these feelings for their daughter-in-law. Dear Annemijn and Sarah, both of you would like to become writers. I think that it is an excellent idea, because it gives so much satisfaction to have a self-written book. Dearest Marcel, together, we live our ‘have-it-all life’ with our two daughters, our jobs and our life at home. Thank you for sharing this altogether with me.

VIII

UNDERSTANDING CROWDSOURCING

TABLE OF CONTENT

IX

Table of Content

Chapter 1.

Introduction

1

1.1. 1.2.

Research challenge Online voluntary contributions 1.2.1. Size and nature 1.2.2. Benefits for the organizing firm 1.2.3. Types of organizations 1.2.4. Online voluntary resources versus employees Research questions and design Scientific relevance Managerial relevance Conclusion Reader guide

1 2 2 3 4 7 8 10 11 12 12
15

1.3. 1.4. 1.5. 1.6. 1.7.
Chapter 2.

Literature Review: Motivation Theories

2.1. 2.2.

Introduction Motivation in online and open source literature 2.2.1. History and status 2.2.2. Identified motives 2.2.3. Quantification of motivation 2.2.4. Motivation theories for online behavior Motivation theories in cognitive psychology literature 2.3.1. Cognitive Evaluation Theory 2.3.2. Self Determination Theory 2.3.3. General Interest Theory 2.3.4. Illustrative studies 2.3.5. The reward-performance controversy Conclusion

15 15 15 16 17 19 21 21 26 28 29 31 33

2.3.

2.4.

X

UNDERSTANDING CROWDSOURCING

Chapter 3.

Theoretical Framework

35

3.1. 3.2.

Our approach in resolving the controversy Development of hypotheses 3.2.1. Effects of intrinsic motivation 3.2.2. Effects of extrinsic motivation 3.2.3. Interplay of intrinsic and extrinsic motivation Conclusion

35 37 38 39 40 41
43

3.3.
Chapter 4.

Methodology

4.1. 4.2.

Introduction Measurement of motivation 4.2.1. Approaches to measure motivation 4.2.2. Motivation measurement tool 4.2.3. Other issues regarding questionnaire development 4.2.4. Websurvey procedure Collection of participation and performance data 4.3.1. Data sources 4.3.2. Expert panels Variables 4.4.1. Independent variables: motivations 4.4.2. Control variables 4.4.3. Dependent variables: participation and performance measures Statistical methods 4.5.1. Confirmatory factor analysis 4.5.2. Regression analyses Conclusion

43 44 44 44 45 46 46 46 47 48 48 48 49 50 50 52 53

4.3.

4.4.

4.5.

4.6.

TABLE OF CONTENT

XI

Chapter 5.

Case 1: Tweakers.net

55

5.1. 5.2. 5.3.

Introduction to Tweakers.net 5.1.1. Financial and reputation rewards Data collection Tweakers.net Measurement of variables 5.3.1. Decision to contribute 5.3.2. Quantity of contributions 5.3.3. Usefulness of contributions 5.3.4. Novelty of contributions 5.3.5. Motives 5.3.6. Control variables 5.3.7. Validity Analysis methods Results 5.5.1. Effects of intrinsic motivation 5.5.2. Effects of extrinsic motivation 5.5.3. Interplay between extrinsic and intrinsic motivation Conclusions 5.6.1. Summary of main findings 5.6.2. Theoretical implications 5.6.3. Managerial implications 5.6.4. Limitations

55 56 57 58 58 58 58 59 59 59 59 60 62 66 66 67 70 70 72 73 78

5.4. 5.5.

5.6.

XII

UNDERSTANDING CROWDSOURCING

Chapter 6.

Case 2: NUfoto.nl

81

6.1.

Introduction to NUfoto.nl 6.1.1. Financial rewards 6.1.2. Reputation rewards 6.1.3. Reward criteria 6.1.4. Hypotheses testing Data collection NUfoto.nl Measurement of variables 6.3.1. Decision to contribute 6.3.2. Quantity of contributions 6.3.3. Usefulness of contributions 6.3.4. Novelty of contributions 6.3.5. Motives 6.3.6. Control variables 6.3.7. Validity Analysis methods Results 6.5.1. 6.5.2. 6.5.3. 6.5.4. Effects of control variables Effects of intrinsic motivations Effects of extrinsic motivations Interplay between intrinsic and extrinsic motivations

81 83 83 83 84 84 86 86 86 86 86 87 87 88 89 90 94 94 95 96 98 98 100 102 103

6.2. 6.3.

6.4. 6.5.

6.6.

Conclusions 6.6.1. Summary of main findings 6.6.2. Theoretical implications 6.6.3. Managerial implications 6.6.4. Limitations and directions for future research

TABLE OF CONTENT

XIII

Chapter 7.

Case 3: Green Challenge

105

7.1. 7.2.

Introduction Green Challenge contest design 7.2.1. Objective of the Green Challenge contest 7.2.2. Contest procedure 7.2.3. Rewards 7.2.4. Reward criteria 7.2.5. Research challenge Case specific methodology Green Challenge 7.3.1. Contributions versus participants 7.3.2. Websurvey 7.3.3. Expert jury Measurement of variables 7.4.1. Usefulness of contributions 7.4.2. Sustainability of contributions 7.4.3. Novelty of contributions 7.4.4. Motives 7.4.5. Control variables 7.4.6. Validity Regression analysis models Results quantitative analysis Additional analysis Conclusions 7.8.1. Summary of main findings 7.8.2. Theoretical implications 7.8.3. Managerial implications 7.8.4. Limitations

105 106 106 108 108 109 109 110 110 110 111 112 112 113 113 114 115 115 116 117 123 126 126 128 129 130

7.3.

7.4.

7.5. 7.6. 7.7. 7.8.

XIV

UNDERSTANDING CROWDSOURCING

Chapter 8.

Conclusion and Future Research

131

8.1. 8.2. 8.3. 8.4. 8.5. 8.6.

Introduction Summary of research questions Findings and conclusion Green Challenge study Scientific contributions Managerial impact 8.6.1. Recommendations for crowdsourcing design 8.6.2. Towards a crowdsourcing classification Limitations and directions for future research

131 131 137 138 144 144 146 148

Findings and conclusions Tweakers.net and NUfoto.nl studies 132

8.7.

Bibliography Annex A Annex B Annex C Annex D Annex E Summary Samenvatting About the Author Defining motivation and rewards Websurvey Tweakers.net Websurvey NUfoto.nl Websurvey Green Challenge Criteria expert jury Green Challenge

151 169 171 175 179 183 185 187 189

LIST OF TABLES

XV

List of Tables

Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 11 Table 12 Table 13 Table 14 Table 15 Table 16 Table 17 Table 18 Table 19 Table 20 Table 21 Table 22 Table 23 Table 24

Effects of motivation on quantity of contribution in online literature ............ 18 Comparison of effects of tangible rewards in high interest tasks (Cameron, 2001) .............................................................................................................. 32 Net effects resulting from hypotheses 1 to 6 .................................................. 42 Data source per participation and performance measure ................................ 47 Descriptive statistics and correlations – Tweakers.net ................................... 62 Results hurdle model – Tweakers.net ............................................................. 64 Results linear and logistic regression – Tweakers.net .................................... 65 Comparison of quantity and usefulness for groups with different motivations profiles............................................................................................................ 78 Descriptive statistics and correlations – NUfoto.nl ........................................ 90 Results hurdle model – NUfoto.nl .................................................................. 92 Results negative binomial and linear regression – NUfoto.nl ........................ 93 Region of origin participants Green Challenge 2008 ................................... 107 Descriptive statistics and correlations – Green Challenge............................ 117 Results linear regression usefulness – Green Challenge .............................. 119 Results linear regression sustainability – Green Challenge .......................... 120 Results linear regression novelty – Green Challenge ................................... 121 Linear regression of control variables and performance measures ............... 126 Testing of hypotheses 1, 2 and 3 .................................................................. 133 Testing of hypotheses 4a, 4b and 5 ............................................................... 134 Testing of hypothesis 6 ................................................................................. 135 Summary of direct and indirect effects of intrinsic and extrinsic motivation on participation and performance in absence and presence of rewards ............. 136 Comparison of behavior of non-rewarded and rewarded volunteers ............ 140 Motivation orientation optimal performers per crowdsourcing type ............ 147 Typology of reward contingencies (Ryan et al, 1985; Deci et al, 1999; Cameron 2001) ............................................................................................. 170

XVI

UNDERSTANDING CROWDSOURCING

Table of Figures

Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 Figure 18 Figure 19 Figure 20

Relation between crowd souring – online and open source communities ........ 7 Impact of contextual factors on intrinsic motivation and behavior according to CET ................................................................................................................ 22 Application of Cognitive Evaluation Theory ................................................. 24 Application of Cognitive Evaluation Theory ................................................. 25 The Self-Determination Continuum (Ryan and Deco, 2000) ......................... 26 Adjusted theoretical model ............................................................................. 37 Theoretical model and hypotheses ................................................................. 41 First and higher order effects of desire for compensation and challenge on quantity – Tweakers.net ............................................................................. 68 First and higher order effects of desire for compensation and challenge on usefulness – Tweakers.net ......................................................................... 69 Expected effects of intrinsic and extrinsic motivation on quantity in absence of rewards ....................................................................................................... 74 Expected effects of intrinsic and extrinsic motivation on quantity in presence of rewards ....................................................................................................... 75 Expected effects of intrinsic and extrinsic motivation on usefulness in absence of rewards ....................................................................................................... 76 Expected effects of intrinsic and extrinsic motivation on usefulness in presence of rewards ........................................................................................ 77 First and higher order effects of desire for recognition and pleasure on quantity –NU.nl ......................................................................................... 97 First and higher order effects of desire for recognition and pleasure on usefulness – NU.nl ......................................................................................... 98 Direct effects of extrinsic motives in absence or presence of rewards considering relevance of reward criteria ...................................................... 101 Interaction effects of extrinsic motives in absence or presence of rewards, considering clearness of reward criteria ....................................................... 102 First and higher order effects of desire for compensation and pleasure on usefulness – Green Challenge ...................................................................... 123 Direct and interaction effects of challenge and desire for compensation on quantity of contributions – Tweakers.net ...................................................... 141 Direct effects of challenge and desire for compensation on quantity of contributions – Tweakers.net ........................................................................ 142

CHAPTER 1. INTRODUCTION

1

Chapter 1.

Introduction

1.1.

Research challenge

Companies increasingly outsource activities to volunteers approached via an open call on the internet. In general rewards are absent or small. When rewards are present, they take the form of recognition on the website or monetary prizes for the best contributions. Well known examples of firms outsourcing business activities to internet communities are YouTube (production of user generated content) and Lego Factory (design of new products). Although the benefits of outsourcing to online volunteers are obvious, outsourcing organisations become dependent on online community members for delivering the desired number of contributions and adequate level of usefulness and novelty. Firms, however, run the risk of receiving high numbers of low quality contributions. Firms using online volunteers often wonder how the different motivations of its community members relate to their participation and performance in terms of quantity, usefulness and novelty of their contributions, and how might rewards affect these relations. To date, the literature on open source and online communities has not investigated the effects of motivation on participation and multiple performance aspects. The effects of rewards in online communities are also not addressed. Although psychologists have done extensive research on the motivation of volunteers in the offline world, these researchers did not reach consensus on the effects of motivation and rewards. Therefore the research challenge is to study the effects of intrinsic and extrinsic motivation of online volunteers and how rewards influence this relation. In this thesis empirical studies on the effects of motivation and rewards on participation and performance are described. We studied three online initiatives: a discussion forum on IT news Tweakers.net, user generated news photographs on NU.nl and submissions to the Green Challenge innovation contest for sustainable products and services. These three cases differ significantly in the provision of rewards. While Tweakers.net did not provide any financial rewards, NU.nl paid small prizes for exceptional contributions and finally the Green Challenge paid a substantial amount to the winner of the contest. In all cases reputation rewards were provided.

2

UNDERSTANDING CROWDSOURCING

In the three studies the motivation of online community members was measured through a websurvey – almost 1,900 respondents completed this survey – while data on participation and performance measures was gathered at the firm organizing the voluntary contributions. We present the results and conclusions of these studies and lay the path for future research.

1.2.
1.2.1.

Online voluntary contributions
Size and nature

Every day, millions of people make all kinds of voluntary online contributions. YouTube receives hundreds of thousands of videos daily. Every minute, 24 hours of video are uploaded to YouTube 1 . Thanks to its voluntary contributions, the Dutch version of Wikipedia counted 300,000 lemmas six years after its start (May 2007)2 and was able to extend its encyclopaedia to over 600,000 lemmas in June 20103. Not only are the massive numbers of contributions impressive, but also the variety of contributions. People searching for the crashed airplane of billionaire Steve Fossets in the Nevada desert with Google Earth4 is, for example, a complete different activity than the design of new toys with software from Lego Factory5. Cook (2008) provided a taxonomy for user contributions. He distinguished between active and passive user contributions. Examples of passive user contributions are the searches generated by a massive number of people, which form the basis of Google’s search engine algorithm, or persons’ buying behavior which determine the product recommendations at Amazon.com. In case of passive user contributions, people do not provide their contribution intentionally. They may even be unaware of the value that their behavior has for the firm aggregating these contributions. Active user contributions consist of those contributions that users provide intentionally, such as multimedia content (text, pictures, audio or video), software code and ratings. In this study we will focus on active user contributions, since rewards and motivation are only relevant in situations where people provide their contributions intentionally.

1 2 3 4 5

http://www.youtube.com/t/fact_sheet http://www.fan.tv/digitaal/toontext.asp?id=5027 http://wikipedia.josemanuelperez.es/nl?lang=nl http://www.gearthblog.com/blog/archives/2007/09/help_find_steve_fosset_with_google.html http://factory.lego.com/

CHAPTER 1. INTRODUCTION

3

A different categorization can be derived when following a firm’s perspective. A firm can use online volunteers for operational and for research and development (R&D) activities. Online volunteers provide, for example, an operational contribution when serving as ‘citizen journalists’ for an online news site (e.g. Newsvine 6 ), when tagging objects or content (e.g. identifying and documenting new astronomical stars for GalaxyZoo 7 ) and when providing user generated videos (YouTube8). Contributions to the R&D activities of a firm consist of generating ideas for improved or new products (sport equipment such as basketball shoes (Füller, 2006), t-shirts and shoes9) or solving expert or scientific R&D problems (e.g. InnoCentive10, Amazon Mechanical Turk11). In order to detect differences in motivation between operational and R&D activities, we include both activities in our study. 1.2.2. Benefits for the organizing firm

The use of online voluntary resources provides some clear benefits. A key benefit is cost savings (Howe, 2006a) since online volunteers are not rewarded in the same manner as employees. They provide their contributions frequently without being paid for it (e.g. Wu et al, 2007; Lampel and Bhalla, 2007). When financial compensation is offered, they are generally linked with contributions that represent value for the organizing company. When for example a solver provides a solution to an unsolved problem at InnoCentive, the solver gets a money prize for it varying from USD 1,000 to USD 1 million. No prize is paid when community members did not solve the problem. This makes the posting of a problem at the InnoCentive website a relatively cheap and non-risky activity. However, the story of cheap or even free resources should be tempered. The experiences of companies organizing online innovation contests show that the evaluation process can be very time and cost consuming (Jouret, 2009). Other benefits refer to the improvement of product quality and customer intimacy and to the acceleration of development activities or large routine tasks. Quality improvement can be achieved when large numbers of users are pre-testing new products or when groups of experts are involved in forecasting (Bonabeau, 2009). Higher customer intimacy is reached through more intensive communication with online customers and increased engagement

http://www.newsvine.com http://www.galaxyzoo.org/ 8 http://www.youtube.com 9 http://www.threadless.com/ and http://www.dreamheels.com/ 10 http://www.innocentive.com/ 11 https://www.mturk.com/
7

6

4

UNDERSTANDING CROWDSOURCING

with the product and firm when customers are contributing (Bonabeau, 2009). Acceleration of time-to-market can be realized when using external expert knowledge which is not available within the company. Proctor and Gamble wanted to boost the sales of Pringle potato chips by having them printed with trivia questions. An Italian professor had an inkjet technology available, which could be quickly adapted for the intended use of P&G12. The execution of routine activities can also be accelerated through the use of large groups of volunteers. Within 4 weeks, members translated the entire content of Facebook into Spanish and within one year translation into 100 languages and dialects was achieved (Van Den Ende et al, 2009). 1.2.3. Types of organizations

New types of organizations have come into existence to organize voluntary online activities. Well known types of organizations are online communities, open source software development communities, and crowdsourcing. In this section we first provide a general description of these organization types followed by a section on similarities and differences. Online communities O’Mahony and Ferraro (2007) explain that a community is a social group with a shared basis of authority. Such a social group consists of people sharing common interests and needs. The specific characteristic of online communities is that members primarily interact via online communication media instead of face-to-face contacts (O’Mahony and Ferraro, 2007; Moon and Sproull, 2008). Some authors highlight that online communities are guided by protocols and norms (Porter, 2004; Wise, Hamman and Thorson, 2006; Preece and Maloney, 2005). Frequently these protocols and norms are implemented in a formal structure that ranges from professional editors to teams of voluntary moderators (Poor, 2005; Preece, 2000). The primary function of moderators is to clarify which contributions are relevant – for example by keeping a conversation on topic – and to prevent harmful contributions. It is believed that moderation becomes more crucial when the size of an online community grows (Lampe and Resnick, 2004).

12

http://marketing.boomja.com/index.php?ITEM=115065

CHAPTER 1. INTRODUCTION

5

Online communities use different methods to select their members (Plant, 2004; Wenger & Snyder, 2000). Although online communities usually have open-membership whereby anyone who has access to a computer and an Internet connection can become a member and participate, online communities can also use a closed-membership policy (Ciffolilli, 2003). Closed membership means that only people meeting a predetermined list of criteria are admitted. The purpose of closed-membership is to increase control over its members, making management, identification of common interests, and meeting easier (Dubé, Bourhis and Jacob, 2006). An example of a closed-membership community – or the so called gated community – is the community set up by Suncorp which is only open to carefully selected customers and non-customers; a group representing the core of the Suncorp target market13. An open-membership policy is more in line with the idea of using worldwide expertise and resources. Two main types of online communities can be distinguished based on their focus: knowledge sharing communities and production communities. Communities of practice are focused on knowledge sharing (Wenger, McDermott and Snyder, 2002), while collective models of innovation can be classified as a production community (Von Hippel and Von Krogh, 2003). O’Mahony and Ferraro (2007) also acknowledge this difference in focus. In our study we consider both knowledge sharing and production communities since we would like to improve our understanding of similarities and differences between these two types of communities. Open source communities Originally the term ‘open source’ was exclusively used to describe groups of voluntary software developers at many different locations and organizations, sharing software code to develop and refine programs (Raymond, 1999; Lerner and Tirole, 2002). In recent years the meaning of the term ‘open source’ has broadened to groups engaged in other activities than software development. Although the activity differs, these groups of volunteers still follow one or more key characteristics of the open source software development community. Below we discuss two main characteristics. The first characteristic concerns deviating licensing regimes. Key to open source software is a distinct class of software licenses certified by the Open Source Initiative (OSI)14. Such licenses must guarantee openness of the code and the rights to use, modify and distribute

13 14

http://bankingreview.blogspot.com/2009/07/online-communities-innovation.html http://www.opensource.org/

6

UNDERSTANDING CROWDSOURCING

the source code. In recent years, similar licenses were developed for products other than software code (e.g. creative commons licenses for art and content15). Despite these new licenses, in many online communities the intellectual rights are still automatically transferred to the organizer when volunteers accept the General Terms and Conditions before uploading their contribution. The second main characteristic of open source communities is that open source developers collaborate with a common set of software tools and internet-enabled communication. Although examples of tools enabling people to co-produce or co-design (e.g. Lego Factory, c,mm,n16) exist, development of web-based collaboration tools is relatively premature. The majority of collaboration platforms only provide the opportunity to sign up, upload one’s contribution and comment and vote on others’ contributions. It is expected that in the coming years substantial efforts will be put into the development of tools which really enable collaborative design or production. Crowd sourcing Jeff Howe (2006a) coined the term crowdsourcing when he described a new web-based business model that harnesses the creative solutions of a distributed network of individuals. Crucial to crowdsourcing is the use of an open call format and a large network of potential laborers (Howe, 2006b). Howe further clarifies that ‘it's only crowdsourcing once a company takes that design, fabricates it in mass quantity and sells it’ (Howe, 2006c). It is less clear what ‘mass fabrication’ means in the current information society: whether it is limited to the production of physical goods or can also be extended to making information available to a large audience without any reproduction costs via the internet. We concentrate on the first part of Howe’s definition of crowdsourcing focused on the outsourcing of business activities to the internet crowd via an open call. Similarities and differences It should be clear that online communities and crowdsourcing are not identical organizational forms. Online communities are groups in which members experience social connection. This social connection is not by definition a requirement of crowdsourcing; crowdsourcing can also be successful outside an online community. When requesting user generated ideas or designs via an open call, these are not necessarily shared with other participants. So group mechanisms applicable to community activities are not always

15 16

http://creativecommons.org/ http://www.cmmn.org

CHAPTER 1. INTRODUCTION

7

relevant to crowdsourcing initiatives relying on individual – and not collaborative – contributions. After concluding that not all crowdsourcing happens in online communities, we argue that not all online communities deliver crowdsourcing. For example social networks facilitating communication between its members do not outsource activities to its members and can therefore not be classified as crowdsourcing initiatives. We consider all open source software development initiatives as examples of crowdsourcing since software development is in general a business activity. Finally we consider open source communities as a specific form of online communities; a conclusion that we share with O’Mahony and Ferrero (2007). In Figure 1, the relations between crowdsourcing and online and open source communities are shown. In this figure our research cases are plotted. It can be concluded that all our cases can be classified as crowdsourcing initiatives, but not all of them are online communities: participants of the Green Challenge did not form a social group that interacts via online media; they remain individual participants throughout the contest procedure. We did not include an open source case since we expect that the motivation of open source software developers is somewhat different compared to the other crowdsourcing initiatives.

Figure 1

Relation between crowd souring – online and open source communities

1.2.4.

Online voluntary resources versus employees

Online communities differ from the normal work situation in organizations with respect to the motivation of members and coordination mechanisms. In communities, expectations are not specified in contractual obligations; therefore key to participation is the contributor’s self selection to assist with a task (Lakhani and Panetta, 2007). In general strong incentive schemes intended to influence self-selection are missing. In most online communities rewards are absent or small, and have the form of reputation systems on the

8

UNDERSTANDING CROWDSOURCING

website or small monetary prizes for the best contributions. As a consequence members’ intrinsic motivations are considered to be important stimuli of behavior (Hertel, Niedner and Herrmann, 2003; Von Krogh and Von Hippel, 2006; Dahlander and Magnusson, 2005; Shah, 2006). In addition, reputation works out differently in an online community compared to an organization, since online communities provide a far more distributed production or knowledge sharing system (Kollock, 1999; Lakhani and Panetta, 2007; Dahlander, Frederiksen and Rullani, 2008; Brabham, 2008), in which members are often anonymous or only known by their nickname (Mesch and Talmud, 2006). It is yet unclear how in this situation extrinsic motivations and rewards affect behavior that is strongly driven by intrinsic motivation.

1.3.

Research questions and design

The main objective of this thesis is to increase our understanding of how participation and performance of online volunteers can be stimulated towards the levels that the organizer desires. This requires an extension of our knowledge on the effects of motivation and rewards on behavior. Therefore the central question in this thesis is: how do motivation and rewards affect participation and performance of volunteers in online communities? We expect that different motives also have different effects. Therefore we distinguish between intrinsic and extrinsic motivation. Intrinsic motivation implies that people perform an activity because they find it interesting and derive spontaneous satisfaction from the activity itself (Gagné and Deci, 2005; Calder and Staw, 1975). Extrinsic motivation implies that people perform an activity for the sake of receiving compensation or other rewards (Frey and Oberholzer-Gee, 1997; Deci, 1971). We foresee that the effects of extrinsic motivation are conditional on the presence or absence of rewards. Although in many articles extrinsic motivation and rewards are used interchangeably, we would like to stress that these terms are not synonyms. We consider rewards and extrinsic motivation as related but not identical concepts. Motivation is a psychological feature that arouses a person to action, while rewards are the goal objectives that reinforce behavior (Porter, 1970). Thus, motivation is an internal condition while rewards are provided by external parties.

CHAPTER 1. INTRODUCTION

9

Besides the effects of the presence or absence of rewards on extrinsic motivation, we expect an interplay of intrinsic and extrinsic motivations resulting in both enhancing and undermining effects on behavior. These interaction effects are mainly expected when rewards are absent. We test which combinations of motivation levels result in positive and negative effects. Finally, we consider multiple measures of behavior: the decision to contribute, the quantity, the usefulness and novelty of contributions. The decision to contribute is a participation measure and indicates whether the person is an active contributor or a non-contributor. Quantity, usefulness and novelty are performance criteria. Quantity is defined as an output measure, namely the number of contributions that a contributor provides in a certain time period. Usefulness is defined as the value that a contribution has for other visitors of the site or for the organizer of the crowdsourcing activity. Finally novelty means the newness of one’s contribution. The research questions underlying the central question are: • How do intrinsic motivations of online volunteers affect the decision to contribute and the quantity, usefulness and novelty of contributions? • When rewards are provided, how do extrinsic motivations of online volunteers affect the decision to contribute and the quantity, usefulness and novelty of contributions? • When no rewards are provided, how do extrinsic motivations of online volunteers affect the decision to contribute, the quantity, usefulness and novelty of contributions? • When no rewards are provided, which combinations of intrinsic and extrinsic motivation levels result in enhancing or undermining effects on the decision to contribute and on the quantity, usefulness and novelty of contributions? We started our research with a literature review in which we identified existing theoretical models on motivation, rewards and behavior. In this review we concluded that existing motivation theories are not fully able to explain the results of empirical field studies and therefore we develop an adjusted theoretical model that we subsequently tested in our three studies. The full set of hypotheses is tested in the Tweakers.net study. The NUfoto.nl study is a replication of the Tweakers.net study in which we investigated the relevance of criteria for receiving the rewards in more detail. Finally, the Green Challenge study focuses on the effects of extreme financial rewards since the winner of the Green Challenge can earn an amount of €0.5 million.

10

UNDERSTANDING CROWDSOURCING

1.4.

Scientific relevance

Due to major changes in society and technology, such as increased diversity in the workforce and the use of information technology, work environments have changed dramatically. This change in work environments results in an urgent need for an adjustment of existing work motivation theories (Steers et al, 2004). We endorse that the open nature of outsourcing activities to online volunteers indeed increases the diversity of the workforce since no function requirements are specified and neither the skills nor experience of contributors are checked. We also note that crowdsourcing has changed the manner and location of work activities resulting in highly divergent needs and demands which require new theory development. The literature on online communities has investigated the different motives of members of online communities (Lerner and Tirole, 2002; Hertel, Niedner and Herrmann, 2003; Von Krogh and Von Hippel, 2003; Dahlander and Magnusson, 2005; Shah, 2006; Jeppesen and Molin, 2003; Wasko and Faraj, 2000; Füller et al, 2007; Baldwin et al, 2006; Harhoff et al, 2003), but has left the relationship of motives to performance and the role of rewards in this relationship largely unclear. The few authors who have actually measured behavior (Wasko and Faraj, 2005; Jeppesen and Frederiksen, 2006; Lampel and Bhalla, 2007) did not find unambiguous effects of motivation since both positive and negative effects of extrinsic motives are found (Nov, 2007; Shah, 2006; Wasko and Faraj, 2005; Füller, 2006; Roberts et al, 2006). The fact that they did not take the presence or absence of rewards into account may explain the lack of consistent results. Moreover, this literature usually defines performance as just the quantity of behavior, for instance the number of hours spent on contributing to the online community or the number of contributions. This study contributes by investigating effects of intrinsic and extrinsic motivation on behavior in the presence and absence of rewards and by including multiple performance criteria, particularly the decision to contribute, and the quantity, usability and novelty of contributions. Amongst others we show that intrinsic motivation contributes to quantity and novelty of contributions, and that in the presence of rewards extrinsic motivation contributes to the usefulness of contributions. Motivation literature by psychologists also does not provide a clear picture on how motivations and rewards affect voluntary activities or so-called free choice behavior. The end of last century saw a debate (e.g. Deci et al, 1999; Cameron and Pierce, 1994; Eisenberger and Cameron, 1996) which is still not solved. One school of scholars, advancing the Self Determination Theory, argues that rewards diminish autonomy (e.g. Deci et al, 1999; Ryan and Deci, 2000; Bear et al, 2003) and thus have negative effects on

CHAPTER 1. INTRODUCTION

11

intrinsic motivation and free choice behavior. The other school, advancing the General Interest Theory, emphasize that rewards have a signaling function regarding the importance of the task, and consequently these authors claim positive effects on intrinsic motivation and behavior (Eisenberger et al, 1998; Eisenberger et al, 1999a). Both schools support their claims with empirical evidence. Our study provides a middle ground by showing that the effects of rewards depend on a person’s motivation levels. In the absence of rewards, members with a combination of high intrinsic and low extrinsic motives had improved performance and members with both high intrinsic and high extrinsic motives had diminished performance. These results indicate that the effects of rewards depend on specific combinations of intrinsic and extrinsic motivations and provide a possible solution to the debate between psychologists researching motivation for voluntary behavior. We provide a contribution to the literature by performing a more fine-grained analysis of the relation between motives, rewards and voluntary behavior. Whereas the psychology literature has focused on the effects of rewards on voluntary behavior, we also measure the self-reported levels of intrinsic and extrinsic motivation. And whereas most of the online and open source literature just studies the motives of the contributors to online communication, irrespective of the provision of rewards (e.g. Nov, 2007; Füller, 2006; Shah 2006; Hars and Ou, 2002), we explicitly take the presence or absence of rewards into account. Moreover, both the psychology and the open source literature usually define performance as just the quantity of effort spent on the activity (e.g. Nov, 2007) or quantity of output, for example the number of creative ideas or code generated (e.g. Roberts et al, 2006). We are far more specific by analyzing the effects of motives on the decision to contribute, and on the quantity, usefulness and novelty of contributions. Finally we analyze both direct and interaction effects of intrinsic and extrinsic motives on these performance measures.

1.5.

Managerial relevance

Firms using voluntary contributions aim to stimulate the participation and performance of those volunteers (Antoniadis and Le Grand, 2007; Harper et al, 2007). These firms are experimenting with rewards following classic motivation theories arguing that employees can be motivated and actively managed with rewards and that without these employees will work less effectively (e.g. Vroom 1964, 2005). From interviews with executives of firms using online volunteers, it became clear that they had not investigated which motivations are relevant for the contributions that volunteers provide (Borst and Van Den

12

UNDERSTANDING CROWDSOURCING

Ende, 2007). This lack of knowledge can result in inappropriately designed reward systems with sometimes fatal consequences such as bankruptcy (Borst and Van Den Ende, 2008). Considering that firms applying this type of outsourcing, are growing into large scale and profitable companies (e.g. in 2008, iStockPhoto had a turnover of USD 130 million with a profit margin of 50%17), the relevance of knowledge that improves the effectiveness of these firms also increases.

1.6.

Conclusion

This thesis describes the empirical testing of the effects of motivation and rewards on participation and performance of online volunteers. The selection of this topic has its origin in managerial practice: practitioners do not have solid knowledge on which motives drive the behavior of online volunteers and the effectiveness of rewards in active management of online volunteers. The topic also appears to be a research challenge since online and open source researchers have hardly investigated the effects of motivation on the behavior of online volunteers and have not explored the effects of reward systems. Psychology researchers on motivation for voluntary behaviour have been debating the effects of motivation and rewards for more than a decade. We will provide a contribution to both literature streams by performing a more fine-grained analysis of the effects of intrinsic and extrinsic motivation and rewards on participation and performance.

1.7.

Reader guide

In Chapter 2, we start with a review of motivational literature. We review papers on motivation in online and open source communities and psychology literature on motivation for voluntary behavior. This literature review forms the basis for our theoretical model that we tested with three empirical studies. The development of our theoretical model and underlying hypothesis is described in Chapter 3. In Chapter 4 the general research design that we followed in our three studies is described. Case specific methodologies are not described in this chapter, but are included in the chapter of each specific study. The three studies are extensively described in Chapters 5 to 7. Chapter 5 is dedicated to the study of Tweakers.net, Chapter 6 to NUfoto.nl and Chapter 7 to the Green Challenge 2008. Per

17

http://www.ethanzuckerman.com/blog/2009/03/17/jeff-howe-on-crowdsourcing/#comments

CHAPTER 1. INTRODUCTION

13

study, the research context, study specific methodologies, results and conclusions are reported. In the last chapter, Chapter 8, we present a summary of our conclusions.

14

UNDERSTANDING CROWDSOURCING

CHAPTER 2. LITERATURE REVIEW: MOTIVATION THEORIES

15

Chapter 2.

Literature Review: Motivation Theories

2.1.

Introduction

In this chapter, we first provide a description of motivation theories that explain behavior in online or open source communities. Although a substantial number of scientific articles address this topic, motivation theories for voluntary online behavior are barely developed. In order to form a proper theoretical foundation for our studies, we also consider motivational theories developed by cognitive psychologists explaining voluntary behavior in offline situations. In particular we describe the Cognitive Evaluation Theory, the Self Determination Theory and the General Interest Theory.

2.2.
2.2.1.

Motivation in online and open source literature
History and status

Since the turn of the century an impressive body of academic research on online and open source communities has emerged. Von Krogh and Von Hippel (2006) suggest a framework for organising the existing research papers. They distinguish three research areas: the organisation process, competitive dynamics and the motivations of contributors. The first research area focuses on the governance and other organisation issues such as leadership. The second group of researchers explores the impact of voluntary contributions to online and open source communities on competition with traditional firms. Finally the topic of motivation appears as a separate research area. It appears that a large number of studies deal with the question “what makes individuals voluntarily participate in online and open source communities?” Most researchers performed explorative research, identifying motives for participation. A limited number of researchers also performed quantitative research by actually measuring motivation levels. Finally, theory development embraced empirical results (Von Krogh and Von Hippel, 2006). These three types of motivation studies are discussed below.

16

UNDERSTANDING CROWDSOURCING

2.2.2.

Identified motives

A major number of explorative motivational studies are performed in open source communities (e.g. Lakhani and Wolf, 2005; Shah, 2006; Hars and Ou, 2002) although some other online communities are studied as well; for example communities engaged in computer game development (Jeppesen and Molin, 2003), knowledge sharing (Hars and Ou, 2002) and the design of basketball shoes (Füller et al, 2007). When clustering these motives into intrinsic and extrinsic motivations – note that researchers did not, in general, cluster the motives themselves – these exploratory studies appear to have similar findings. As expected for voluntary activities, intrinsic motives, such as fun and learning play, an important role, but these were not the sole reason for participation since extrinsic motives also appear to be relevant. A mix of intrinsic and extrinsic motives determines the participation of online volunteers. The main extrinsic motive found to be relevant for voluntary online contributions is peer recognition (Hars and Ou, 2002; Jeppesen and Molin, 2003: Wasko and Faraj, 2000; Lerner and Tirole, 2002, Lakhani and Wolf, 2005; Shah, 2006). Peer recognition includes the signaling of competencies by colleague experts. In addition to the motives of fun, learning and recognition, research on open source communities identified a limited number of very specific motives: the motive ‘desire to satisfy own needs’ (Franke and Von Hippel, 2003; Lakhani and Von Hippel, 2003; West and Gallagher, 2006) and pro-social feelings, such as altruism. The desire to satisfy own needs can be classified as extrinsic motivation while the pro-social feelings are part of one’s intrinsic motivations. The desire to satisfy own needs refers to the situation in which a software developer solves their own problem and subsequently reveals his or her solution to the community. In our view the desire to satisfy own needs explain why the person develops the software code, but not the subsequent step to reveal it to the community. Motives such as recognition and reciprocity expectancy (‘tit for tat’) are more likely to explain the uploading and publishing of self-produced code. We argue that this motive, desire to satisfy own needs, is more applicable in communities were design and production of goods and materials are separated, in other words where a consumer can not take care of the production him/herself. When you for example want to buy a personalized toy, you can send your own design to the producer to satisfy your need. Again other motives, such as recognition for your design skills and the desire to receive a revenue share of the sales from your design, become relevant when placing your design in a catalogue of the producer so that other customers can also order this toy (e.g. Lego Factory).

CHAPTER 2. LITERATURE REVIEW: MOTIVATION THEORIES

17

Pro-social motivations or altruism can be described as the desire to increase the welfare of other people or the obligation to do something for another at the costs of oneself (Ozinga, 1999). Pro-social motivations are classified as intrinsic motivation since these feelings of obligation arise in the person him/herself. Pro-social motivations are found to be relevant in communities such as Wikipedia (Nov, 2007) or a community of legal experts providing free advice (Franke and Shah, 2003). Again, this type of motivation is not relevant for all online voluntary contributions since some contributions do not increase the welfare of visitors. For example a person uploading entertainment videos on YouTube is not expected to experience altruistic feelings. Feelings of recognition are expected to be far more relevant for uploading video clips. Therefore we conclude that the motives ‘desire to satisfy own needs’ and pro-social motivations are not always applicable in online contexts. 2.2.3. Quantification of motivation

While the majority of explorative studies rely on indirect methods for analysing motivation, for example through an analysis of weblogs and posts or through interviews of product development managers organizing the community activities (Jeppesen and Molin, 2003; Füller et al, 2007), most quantitative studies measure motivation at the source, namely through websurveys among the contributors. Filling in the websurvey require respondents to score their own motivation on a Likert scale. These studies show that the average level of intrinsic motivation of online volunteers is substantially higher than their extrinsic motivation (Nov, 2007; Wasko and Faraj, 2005) sometimes even twice as high (Füller, 2006). It should be noted that these studies do not analyze the motivation levels in relation to the presence or absence of rewards. Even more interesting are those studies that explore how motives produce a mix of outcomes. This exploration requires the measurement of not only motivation, but also performance. We observe that the performance measure, quantity, is the most frequently investigated. The following table shows the findings of studies on the relation between motivation and the performance measure, quantity. Effects found appear to be inconsistent.

18 Table 1

UNDERSTANDING CROWDSOURCING Effects of motivation on quantity of contribution in online literature Quantity definition Self reported future contributions Self-reported hours spent per week Self reported participation level Number of messages posted Intrinsic motives Pleasure Positive Effect Positive Effect Positive Effect No Effect Challenge No Effect Positive Effect Not studied Not studied Not studied Extrinsic motives Desire for compensation Positive Effect Not studied Not studied Not studied Not studied Desire for recognition No Effect Positive Effect Negative Effect Positive Effect Positive Effect Positive Effect

Author(s) Füller (2006) Nov (2007) Shah (2006) Wasko and Faraj (2005) Lampel and Bhalla (2007) Roberts et al (2006)

Number of online Not studied reviews contributed Number of source code contributed and accepted No Effect

Not studied

Not studied

The most remarkable are the contrary effects of the motive ‘desire for recognition’ on the number of contributions; both positive (Nov, 2007; Wasko and Faraj, 2005; Füller, 2006; Roberts et al, 2006) and negative effects are found (Shah, 2006). Furthermore, Shah (2006) and Lakhani and Wolf (2005) find positive effects of the motive pleasure on quantity while results of another study do not confirm these effects (Roberts et al, 2006). One might expect that differences in measurement methods, i.e. subjective self-reported measures versus objective count data (see second column in which the quantity definition is described), explain the contrary findings. You could argue that numbers of self-reported quantities are more optimistic and studies using these quantity measures found more positive effects than actual measures of quantity. This does not appear a valid assumption since Shah (2006) who used a self-reported measure found negative effects of recognition while studies using objective quantity measures report positive effects. The studies using self-reported measures of quantity even show the highest inconsistency in the effects of recognition. Therefore the measurement methodology does not explain the contrary findings. Another possible explanation could be the presence or absence of reputation systems rewarding a contributor for high quantity. The motive, desire for recognition, may work out differently when reputation systems are absent and needs are not satisfied compared to the presence of reputation systems. The researchers do not indicate whether reputation systems were present or absent, which severely hinders the interpretation of the results. In our

CHAPTER 2. LITERATURE REVIEW: MOTIVATION THEORIES

19

studies we analyze the effects of motivation on behavior with both the presence and absence of rewards. To our knowledge, only two studies address performance measures other than the quantity of contribution. Wasko and Faraj (2005) determined the helpfulness of answers to legal questions based on an interpretation of response messages and found that only the motive, desire for reputation, has a positive effect on quality. Jeppesen and Frederiksen (2006) measured the self-reported innovativeness of contributions and concluded that striving for firm recognition increases the innovativeness of contributions. It should be noted that researchers did not include all intrinsic and extrinsic motives in their studies, so we do not gain a complete picture of the effects of intrinsic and extrinsic motives. 2.2.4. Motivation theories for online behavior

Some theorists have tried to explain online behavior with the help of existing organizational models that also include motivation. Von Hippel and Von Krogh (2003) for example tried to apply the private investment model and collective action model to online contexts. The private investment model assumes that people or organizations put time, money and other resources into an innovation because they expect a private return (Demsetz, 1967). It is clear that in an open source project, in which the software developer freely reveals his software code, no private returns are present. The collective action model is typically described as the collective that produces a public or semi-public good. Public goods are defined by their non-excludable and non-rival nature (Von Hippel and Von Krogh, 2003; Wasko et al, 2009). This means that even if a user consumes this public good, it is also open for consumption by other users. Examples of public goods are public roads and parks, national defense, clean environment and a crime-free neighborhood. Von Hippel and Von Krogh (2003) argue that the Public Action model is not applicable to open source projects because members in open source projects have the option to wait for others to contribute and then free-ride on what they have done. It appears that members of open source communities are not stopped by the chance of free-riding. So in the Public Action Model the question still remains as to why people en masse freely reveal their software code or other types of innovations (Lerner and Tirole, 2002). Von Hippel and Von Krogh (2003) introduced the Private-Collective model which is a compound of the Private Investment and Collective Action Model. They suggested that the Private-Collective model would provide the ‘best of both worlds’, by combining the

20

UNDERSTANDING CROWDSOURCING

benefits of the two models. They showed that the application of the Private-Collective model is not limited to open software development, but also applies to the free revelation of product and service designs (Von Hippel and Von Krogh, 2006). They defined free revelation in a broad sense to include the absence of immediate financial compensation for one’s contribution: the voluntary sacrificing of all intellectual property rights of that design and providing all parties equal access it. They consider free revelation a defining characteristic of open innovation (Von Hippel and Von Krogh, 2006). In the Private-Collective model, innovators use their own resources to privately invest in creating product or service innovations. First, the model highlights that, in general, the competitive advantage associated with keeping the code private is relative low. The commercialization process of their developed software is often times very time consuming and very costly (e.g. applying for intellectual property rights), which limits the chance for profits (Von Hippel and Von Krogh, 2003, 2006). In addition Von Hippel and Von Krogh (2006) suggest that some of these innovations are created at low costs, which also strengthen the willingness to freely reveal it. Finally, contributors gain some benefits that are not applicable to the free riders. Contributors retain private benefits, such as learning and enjoyment, and benefits associated with community participation, such as social rewards and feelings of solidarity, altruism, fairness and the like (Von Hippel and Von Krogh, 2003, 2006). These benefits are expected to offset the absence of direct or possible future monetary rewards. The Private-Collective model thus shows that most innovators have difficulties personally commercializing their innovation which positively affects the willingness to freely reveal the innovation. Furthermore, this willingness to share the innovation with others is strengthened through its low investment. Ultimately, this mix of non-monetary motives yields a rationale for freely revealing the innovation. Although the Private-Collective model acknowledges that a mix of motives is relevant for freely revealing one’s contribution, it does not specify t

Summary Researching the behaviors and motivations of 'online volunteers":
• How do intrinsic motivations affect the decision to contribute, the quantity, usefulness and novelty of contributions?
• In a rewards model, how do extrinsic motivations affect the decision to contribute and the quantity, usefulness and novelty of contributions?
• With no rewards, how do extrinsic motivations affect the decision to contribute, the quantity, usefulness and novelty of contributions?
• With no rewards, which combination's of intrinsic and extrinsic motivation levels result in enhancing or undermining effects on the decision to contribute and on the quantity, usefulness and novelty of contributions?

Description A PhD dissertation, "Understanding Crowdsourcing" by Irma Borst sub-titled "Effects of Motivation and Rewards on Participation and Performance in Voluntary Online activities, was published on 23rd December 2010. The research challenge was to study the effects of intrinsic and extrinsic motivation of online volunteers and how rewards influence this relationship.

Tags: Crowdsourcing, extrinsic motivation, financial rewards, intrinsic motivation, motivation profiles, online volunteers, participation, performance, reputation rewards, reward size

Posted by Luis Enrique Morales

Jan 05, 2011 02:32 pm GMT Viewed 5078 times

Author: Irma Borst Source

Contact us if you are owner of this content and want to make changes.

Download Flag This

Events

Understanding Crowdsourcing (Effects of Motivation and Rewards on Participation and Performance in Voluntary Online Activities)

Comments

Crowdsourcing.org

Categories

Content

Crowdsourcing.org, an initiative by massolution®

Events

Related Content

Understanding Crowdsourcing (Effects of Motivation and Rewards on Participation and Performance in Voluntary Online Activities)

Comments

Crowdsourcing.org

Categories

Content

Crowdsourcing.org, an initiative by massolution®