Data Visualization of Quantitative Data

A data visualization has been created using Tableau Public. Data from the survey findings was used to create this data visualization which is a summary of the low contributors, their demographics and their initial motivations. This information can be extracted from the data visualization by scrolling over each individual circle.

Viz Pic

 

The data viz is titled Meitheal Duchas.ie Low Contributors and is available at this link: 

https://public.tableau.com/profile/publish/MeithealDuchas_ieLowContributors/DataViz#!/publish-confirm

References

Alam, S.L. and Campbell, J. (2012): Crowdsourcing Motivations in a not-for-profit GLAM context: The Australian Newspapers Digitisation Program, 23rd Australasian Conference on Information Systems Crowdsourcing motivations in GLAM, 3-5 Dec 2012, Geelong.

Bhattacherjee, A. (2012): Social Science Research: Principles, Methods and Practices, Textbooks Collection, Book 3, University of South Florida, Scholar Commons.

Bonney, R., Ballard, H., Jordan, R., McCallie, E., Phillips, T., Shirk, J., and Wilderman, C. (2009): Public Participation in Scientific Research: Defining the Field and Assessing Its Potential for Informal Science Education. A CAISE Inquiry Group Report. Washington, D.C.L: Centre for Advancement of Informal Science Education (CAISE).

Buecheler T., Sieg J.H, Fu¨chslin R.M., & Pfeifer R. (2010) Crowdsourcing, open innovation and collective intelligence in the scientific method: a research agenda and operational framework. In: Fellerman H et al. (eds) Artificial life XII. Proceedings of the twelfth international conference on the synthesis and simulation of living systems, Odense, Denmark, 19–23 August 2010, pp. 679–686.

Causer, T. and Wallace, V. (2012): Building a Volunteer Community: Results and Findings from Transcribe Bentham, Digital Humanities Quarterly, Vol. 6.

http://www.digitalhumanities.org/dhq/vol/6/2/000125/000125.html

Carletti, L., McAuley. D., Price, D., Giannachi, G., Benford, S. (2013): Digital Humanities and Crowdsourcing: An Exploration. In Museums and the Web 2013, The annual conference of museums and the web, held on April 17 – 20, 2013, Portland, Oregon, USA.

Chanal V., Caron-Fasan M.L. (2008): How to invent a new business model based on crowdsourcing: the crowdspirit case. In: EURAM, Lubjana, Slovenia.

Chandler, D. and Kapelner, A. (2013): Breaking monotony with meaning: motivation in crowdsourcing markets, Journal of Economic Behaviour & Organization, Vol. 90 No. 1, pp.123-133.

Clary, E.G., Snyder, M., Ridge, R.D., Copeland, J., Stukas, A., Haugen, J. and Miene, P. (1998):

Understanding and Assessing the Motivations of Volunteers: A Functional Approach, Journal of Personality and Social Psychology. 1998, Vol. 74, No. 6, 1516-1530.

Crowston, K. and Fagnot, I. (2008): The motivational arc of massive virtual collaboration, In

Proceedings of the IFIP WG 9.5 Working Conference on Virtuality and Society: Massive Virtual Communities, Lüneberg, Germany, 1–2 July 2008.

CSO.2015. Information Society Statistics – Household.

http://www.cso.ie/en/releasesandpublications/er/isshh/informationsocietystatistics-households2015/ [27/08/2016]

CSO.2016. Population Estimates by Age Group and Sex.

http://www.cso.ie/multiquicktables/quickTables.aspx?id=pea01 [14/08/2016]

 

CSO.2012. Press Release Census 2011 Profile 9 What We Know.

http://www.cso.ie/en/newsandevents/pressreleases/2012pressreleases/pressreleasecensus2011profile9whatweknow/ [14/08/2016]

Dawson, R. and Bynghall, S. (2010): Getting Results from Crowds – the definitive guide to using crowdsourcing to grow your business, California: Advanced Human Technologies.

Deci, E. L., and Ryan, R. M. (1985) Intrinsic motivation and self-determination in human behaviour, Springer.

Dillon, L. Y. (2002): International Partners, Local Volunteers and Lots of Data: The 1881 Canadian Census Project, History and Computing, Vol. 12 (2) pp.163 – 176.

Duchas.ie.2016. Digitization of the National Folklore Collection

http://www.duchas.ie/pdf/16.02.24_bileog_duchas.ie.pdf [25/07/2016]

Dunn, S. and Hedges, M. (2014): How the Crowd Can Surprise Us: Humanities Crowdsourcing and the Creation of Knowledge, In: Ridge, Mia ed. Crowdsourcing Our Cultural Heritage. Digital Research in the Arts and Humanities. Farnham: Ashgate.

Estellas-Arolas, E. & Gonzalez-ladron-de Guevara, F. (2012): Towards an integrated crowdsourcing definition, Journal of Information Science, Vol.38 (2) pp.189 – 200.

Eveleigh, A., Jennett, C., Blandford, A., Brohan, P. and Cox, A.L. (2014): Designing for Dabblers and Deterring Drop-Outs in Citizen Science, CHI 2014 Proceedings of the SIGCHI Conference on Human Factors in Computing Systems pp. 2985 – 2994.
 

Eveleigh, A., Jennett, C., Lynn, S. and Cox, A.L. (2013): “I want to be a Captain! I want to be a Captain!”: Gamification in the Old Weather Citizen Science Project, Proceedings of Gamification 2013 Stratford, Ontario, Canada, October 2-4, 2013 pp.79 – 82..

 

 

Gregg, D. G. (2010): Designing for collective intelligence, Communications of the ACM, 53(4), 134–138.

Holley, R. (2009): Many Hands Make Light Work: Public Collaborative OCR Text Correction in Australian Historic Newspapers, Canberra: National Library of Australia.

http://www.nla.gov.au/ndp/project_details/documents/ANDP_ManyHands.pdf

Holley, R. (2010) Crowdsourcing: how and why should libraries do it?, D-Lib Magazine, 16 (3/4), www.dlib.org/dlib/march10/holley/03holley.html .

Hossain, M. (2012): Users’ motivation to participate in online crowdsourcing platforms, International Conference on Innovation Management and Technology Research (ICIMTR) IEEE, pp.310 – 315.

Hossain, M. and Kauranen, I. (2015): Crowdsourcing: A comprehensive literature review, Strategic Outsourcing: An International Journal, Vol. 8, No. 1 pp.2 – 22.

Howe, J. (2006): The rise of crowdsourcing. Wired. 14(6). http://www.wired.com/2006/06/crowds/

Howe J. (2008): Crowdsourcing: how the power of the crowd is driving the future of business. UK: Business Books.

Huberman B.A., Romero D.M., & Wu F. (2009): Crowdsourcing, attention and productivity. Journal of Information Science 35(6): 758–765.

Jeppesen, L.B. and Lakhani, K.R. (2010): Marginality and problem-solving effectiveness in broadcast search, Organization Science, Vol. 21 No. 5, pp. 1016-1033.

Kleeman F., Voss GG. & Rieder K. (2008): Un(der)paid innovators: the commercial utilization of consumer work through crowdsourcing. Science, Technology and Innovation Studies 4(1): 5–26.

Leimeister, J. M. (2010): Collective Intelligence. Business & Information Systems Engineering, April, pp. 245–248.

Leon, S.M. (2014): Build, Analyse and Generalise: Community Transcription of the Papers of the War Department, In: Ridge, Mia ed. Crowdsourcing Our Cultural Heritage. Digital Research in the Arts and Humanities. Farnham: Ashgate.

Litwin, M.S. (1995): How to Measure Survey Reliability and Validity, SAGE: London.

Mazzola D., & Distefano A. (2010): Crowdsourcing and the participation process for problem solving: the case of BP. In: VII conference of the Italian chapter of AIS information technology and innovation trend in organization, itAIS 2007, Naples, Italy.

Moyle, M., Tonra, J. and Wallace, V. (2011): Manuscript Transcription by Crowdsourcing: Transcribe Bentham, Liber Quarterly 20 (3/4), March 2011 P347–356.

Nov, O., Arazy, O. and Anderson, D. (2011): Dusting for science: motivation and participation of digital citizen science volunteers, Proceedings of iConference 2011, February 8-11, Seattle, pp. 68 – 74.

Oomen J., and Aroyo, L. (2011): Crowdsourcing in the cultural heritage domain: opportunities and challenges. Proceedings of the 5th International Conference on Communities and Technologies. 19 June – 2 July. Brisbane, Australia.

Oomen, J., Brinkerink,, M., Heijmans, L., van Exel, T. (2010): Emerging Practices in the Cultural Heritage Domain – Social Tagging of Audiovisual Heritage. In: Proceedings of the WebSci10: Extending the Frontiers of Society On-Line, April 26-27th, 2010, Raleigh, NC: US.

Owens, T. (2013): Digital Cultural Heritage and the Crowd. Curator: The Museum Journal. Vol 56 (1).

Owens, T. (2014): Making Crowdsourcing Compatible with the Missions and Values of Cultural Heritage Organizations, In Ridge, Mia ed. Chapter 12 Crowdsourcing Our Cultural Heritage. Digital Research in the Arts and Humanities. Farnham: Ashgate.

Porta, M., House B., Buckley L., Blitz A. (2008): Value 2.0: eight new rules for creating and capturing value from innovative technologies. Strategy & Leadership 36(4): 10–18.

Raddick, M.J., Bracey, G., Gay, P., Lintott, C.J., Cardamone, C., Murray, P., Schawinski, K., Szalay, A.S., and Vandengerb, J. (2013): Galaxy Zoo: Motivations of Citizen Scientists, Astronomy Education Review Vol. 12 (1).

Ridge, Mia (2014). Introduction. In: Ridge, Mia ed. Crowdsourcing Our Cultural Heritage. Digital Research in the Arts and Humanities. Farnham: Ashgate, pp. 1–13.

http://www.ashgate.com/pdf/SamplePages/Crowdsourcing-our-Cultural-Heritage-Intro.pdf [12/12/2015]

Robson, C. (2002): Real World Research, 2nd Edition, Wiley.

Ross, C. (2012): Social media for digital humanities and community engagement, In Digital Humanities in Practice, Edited by Warwick, C, Terras, M. and Nyhan, J. London: Facet Publishing.

Rotman, D., Preece, J., Hammock, J., Procita, K., Hansen, D., Parr, C., Lewis, D. and Jacobs, D. (2012): Dynamic Changes in Motivation in Collaborative Citizen-Science Projects, Session: Civic and Community Engagement 2012, February 11 – 15, Seattle, Washington.

Somekh, B. and Lewin, C. (2011): Theory and Methods in Social Research, 2nd Edition, London: Sage.

Surowiecki, J. (2004): The wisdom of crowds: Why the many are smarter than the few and how collective wisdom shapes business, economies, societies, and nations, Random House: New York.

Terras, M. (2012): Digitization and digital resources in the humanities, In Digital Humanities in Practice, Edited by Warwick, C, Terras, M. and Nyhan, J. London: Facet Publishing.

Vroom, V. H. (1964): Work and Motivation, New York: Wiley.

Wiggins, A. and Crowston, K. (2011): From Conservation to Crowdsourcing: A Typology of Citizen Science, In 44th Hawaii International Conference on System Sciences (HICSS), 4 – 7 January 2011.

Whitla P. (2009):  Crowdsourcing and its application in marketing. Contemporary Management Research 5(1): 15–28.

Yang J., Adamic L.A., Ackerman M.S. (2008): Crowdsourcing and knowledge sharing: strategic user behaviour on taskcn. In: Proceedings of the 9th ACM conference on electronic commerce. New York: ACM, 2008, pp. 246–255.

Yin, R.K. (2009): Case Study Research, Design and Methods, 4th Edition, London: Sage.

Zastrow, J. (2014): Crowdsourcing Cultural Heritage: ‘Citizen Archivists’ for the Future, The Digital Archivist, Vol. 34 (8) October.

Zhao, Y. and Zhu, Q. (2014): Evaluation on crowdsourcing research: Current status and future direction, Information Systems Front, Vol. 16 pp. 417 – 434.

 

Website References

 

Open Plaques.2016. About.

http://openplaques.org/about [accessed 29/05/2016]

 

 

 

 

 

 

 

Conclusion

The research survey conducted on Meitheal Dúchas.ie volunteers sought to answer the research question; “What motivates individuals to participate in GLAM crowdsourcing projects?”.

The research findings demonstrate that analysis of the survey data has addressed this question. From the survey findings analysis, we see similarities which give support to findings from previous research on GLAM crowdsourcing volunteers; the highest percentage of volunteers are over forty, the majority of volunteers are educated to degree level or higher, low contributors make up the largest group volunteers, the most dominant motivating factor for participation is linked to the projects subject area.

Similarities with other crowdsourcing project research also became apparent from the research findings in terms of the types of motivations of volunteers; egoism based motivations for participation ranked highest, such as personal interest in the project or topic of interest, learning new knowledge, followed by community based motivations such as desire to contribute for the greater good. Task and enjoyment based motivations ranked lower, but were still evident in the form of volunteers being motivated to participate for fun and enjoying the task of transcribing.

Differences between previous research and this study’s findings emerged as volunteer’s viewing gamification in a negative light, and extrinsic motivators such as recognition and feedback ranking low as motivators to participate in the project (however, this must be viewed in the context of very few extrinsic motivators existing for the projects volunteers. Finally, one area identified as standing apart from other research was the finding that the subject matter of a crowdsourcing project may influence the percentages of gender participation.

The research findings will benefit the Meitheal Dúchas.ie project by identify the primary motivations of the low contributor groups and through the suggested marketing and communication strategies to encourage further participation from low contributor’s. In addition, other GLAM institutions may use this information in the design of their crowdsourcing projects to target marketing and communication strategies that appeal to the intrinsic motivations of potential volunteer’s.

By discovering the unique motivations of GLAM crowdsourcing project volunteer’s, we can practically use this information to successfully initiate crowdsourcing projects and to sustainably develop and maintain crowdsourcing communities.

Further Research: This study focused on exploring the negative and positive motivations of transcribers. In the context of negative motivations, or indeed de-motivators, further research would be useful to explore the barriers to participation. This issue is discussed in Eveleigh’s (2013) research where barriers to participation include lack of time and concern that one’s contribution’s may not be of good enough quality.

 

 

10.0 Main Findings Discussion: Transcribers Motivations

Transcribers initial interest in transcribing for Meitheal Dúchas.ie stemmed from a variety of egoism-based motivation categories which included personal, community and enjoyment based motivations.

10.1 Egoism Based Motivations

 Topic of Interest: Volunteers’ most frequent motivation for participating in the project was an interest in the topic, which was either an interest in Irish folklore in general, an interest in the ‘Old Irish’ language to be found in some of the texts for transcription, or indeed a combination of both.

“I find the old way of writing Irish really beautiful (and kind of wish we’d bring it back) so I enjoy interacting with that…” 

“I have a keen interest in Gaeilge, history and folklore”.

“I am merely interested in the subject matter and wish to have it transcribed”.

Personal Interest: Almost sixty percent of volunteers’ motivations to transcribe originated from their personal interest. A large proportion of participants in the Meitheal Dúchas.ie project were people interested in their own family connection to the material for transcription, or through a geographical connection.

Personally, curiosity about life in the area (the field where I built a house 70 years later happened to up in one of the stories!) and how my neighbours (I knew some of the schoolchildren, some are still living) lived in the late 1930s”.

 “Seeing my father’s and mother’s essays online was hugely motivational for me and I wanted others to share the joy of discovering their family connections online through this project”.

 “My grandfather was a school master – so I get an enormous kick out of transcribing his words.  I can almost hear his voice in my head as I type. My late father was involved in collection and was amazed to discover his stories years later. My late mother-in-law also contributed”.

Learning New Knowledge: Volunteers felt motivated to participate in the project as they recognized that they could learn new things from the collection as they transcribed the pages, such as local folklore

(“…it was fascinating to learn more about local folklore and legend”, “Increases my knowledge of Irish folklore…”), life in Ireland in the 1930’s (“It helps me learn about others’ experiences”), to gain knowledge to further personal study (” …to learn more to supplement my genealogy studies….”) and to learn the Irish language (“helps to improve my knowledge of Irish language”).

Competition: Similar to Dunn and Hedges (2014) findings with regard to Transcribe Bentham volunteers, competition with other participants was not a motivator for volunteers of Meitheal Dúchas.ie. Similarly, a large proportion of participants in the Low Contributor groups (88%) did not view recognition of the achievement of transcribers on the projects website as a motivator.

This finding demonstrates that competition with other transcribers is not important to low contributors and that they need to be encouraged to return and contribute again through other strategies, such as marketing messages which appeal to their altruistic motivations to help and contribute to research.

10.2 Community Based Motivations

 Many transcribers viewed transcribing for the project as an altruistic patriotic task that would help make a contribution to society, preserve and disseminate the folklore and would benefit future generations.

folklore is the largely unwritten, uncredited history of regular people, and it is important to me as a matter of social and historical justice to do what I can to help those voices be heard”.

These are community based Intrinsic motivations, that are altruistic in nature, such as the desire to share one’s skills for the greater good. Other community based motivations also exist that relate to the social psychological function of volunteerism, such as engaging with others, yet this motivation ranked low among respondents. Requests for community based discussion forums were only mentioned by one respondent (“I would love to have a forum where stories (or problems transcribing) could be discussed”).

 10.3 Task Based and Enjoyment Based Motivations

Simplicity: Transcription was viewed as an easy and simple task by over one third of the low contributors, who cited this intrinsic task based motivation as an initial motivating factor to participate. No training or complex IT skills are required as the transcription user face is very simple, containing a small blank page for typing beside the digital version of the original document, a clearly visible Save button below the page and a link to a Transcription Guide is provided.

Fun: As an enjoyment-based motivation, fun ranked low (26%) among survey respondents which is consistent with the negative views of gamification for GLAM crowdsourcing projects as discussed in the literature review. This finding implies that gamification is not the way to attract contributors to initially be motivated to participate in GLAM crowdsourcing projects or indeed encourage participants to return.

10.4 Extrinsic Motivations

Attribution: Acknowledgements in the Meitheal Duchas.ie project took the form of private attribution which was done by transcribers being able to view their transcription history via their account handles. However, to outside viewers, and indeed other transcribers, the acknowledgement was given yet the volunteers identity (. i.e. account handle) was not revealed. Below each page transcribed by a volunteer is a message which reads; Transcribed by a member of our volunteer transcription project.

Although, acknowledgement of volunteers’ contributions on a website ranking table were viewed as a motivator by a small number of respondents (20%) it may be a worthwhile experiment for the project to enable a volunteers account handle to be viewed at the end of any pages they have transcribed and review if this change makes any difference to transcription numbers.

Recognition:  Of the respondents surveyed, 25% agreed that recognition of the achievement of transcribers on the website homepage was a motivator. Recognition can be a way for contributors to obtain positive feedback for their efforts, enable themselves and others to see their progress and instill an amount of competitiveness for interested parties (Alam and Campbell, 2012).

Reviewing the findings on extrinsic motivations, it is obvious that they ranked low among contributors to Meitheal Dúchas.ie, yet perhaps this could be due to the fact that these type of motivators are not available to contributors on the website. In this instance it may be possible that the data has made visible effects not otherwise noticed by the project staff.

Research from the other GLAM crowdsourcing projects reviewed demonstrate that while these motivations are not of paramount importance to all volunteers they do motivate a certain number of contributors who have a strong tendency for extrinsic motivation.

A successful example used by ANDP included a ‘hall of fame’ ranking table which listed any contributor who corrected more than a certain amount of lines in month and included an overall ranking for each contributor in their user profile. This approach showed the ranking tables to be more about the ‘big picture of contribution’ rather than specifically about competition. Most importantly, contributors who were mostly highly intrinsically motivated did not see the reward criteria as a restriction on their autonomy (Alam and Campbell, 2012).

 In-Direct Feedback: a progress toolbar on the website homepage is the only form of indirect feedback available to Meitheal Dúchas.ie contributors. An interesting finding from the survey showed how the low contributors differed from the high contributors when it came to finding feedback from the project organizers as a motivator; most of the high contributors agreed, yet most of the low contributors disagreed.

Considering this, perhaps more feedback mechanisms on Meitheal Dúchas.ie such as regular email acknowledgement of a transcribers outstanding work and postings on the website blog by the project team recognizing contributor’s efforts could encourage the low contributors to contribute a little more, thereby potentially creating more high level contributors.

 

Psychological Functions of Volunteerism

 

Constructs

 

Motivation Type

 

Motivation Category

And Rating for Meitheal Dúchas.ie Volunteers

 

Values

 

 

Altruism

 

Sharing One’s Skills for the Greater Good.

Benefit Society

Benefit the Project

 

Intrinsic

 

Community Based (high)

 

Understanding

 

 

New knowledge learning

Personal Interest

 

Fun

 

Enjoy Transcribing

 

Intrinsic

 

Egoism Based (high)

 

 

Enjoyment Based (low)

 

Task Based (low)

 

Social

 

 

Engaging with others

 

Intrinsic

 

Community Based (very low)

 

Career

 

 

Career related benefits

 

Intrinsic

 

Egoism Based (very low)

 

Protective

 

 

Protect the ego

 

Intrinsic

 

Egoism Based (very low)

 

Enhancement

 

 

Enhance self-esteem, personal growth

 

Extrinsic

 

Social Based (low)

 

10.5 Limitations: As the Low Contributor Groups made up the bulk of respondents, it was only therefore natural that the low contributor groups would appear as higher percentages in the survey responses. Further studies on the motivations of low contributor groups are required to consolidate the generalizability of these findings.

10.6 Hypothesis

 Hypothesis 1 (H1): that intrinsically motivated Meitheal Dúchas.ie participants are more likely to engage with the project in depth. Therefore, intrinsic motivation is positively correlated with greater levels of contribution.

H1 was not supported by the findings; the majority of low contributors were intrinsically motivated. Extrinsic motivations were established in a small number of volunteers.

Hypothesis 2 (H2): extrinsic motivators such as recognition are negatively correlated with greater levels of contribution.

H1 is supported by the survey findings; the highest motivators were found to be intrinsic, such as personal interest in the subject matter and learning new knowledge, thereby confirming that volunteers are less likely to be motivated to participate due to extrinsic motivations such as reward.

 

9.0 Analysis, Discussion, Implications and Recommendations

Explanation and interpretation of the survey questionnaire data depends on incorporating into the research, information from other variables and then analysing patterns of correlations and seeking out where relationships are strong, weak or non-existent. From this exercise the story within the data is revealed (Robson, 2002). The variables from which this study relies upon to analyse patterns of correlations and relationships are data collected from a self-completion survey questionnaire of Meitheal Dúchas.ie members, previous studies on the motivations of contributors to crowdsourced digital humanities projects, theoretical frameworks on the motivations of crowdsourcing contributors and a theoretical framework developed for this research by the author. 

9.1 Demographics: Gender, Age and Education

Of the other crowdsourcing motivation studies reviewed for this paper, three conducted demographic studies of their contributors; Galaxy Zoo (Raddick et al, 2013), Old Weather (Eveleigh et al, 2014) and Transcribe Bentham (Causer and Wallace, 2012).

Analysis of the survey results regarding the demographics of participants in the Meitheal Dúchas.ie project revealed that there are a larger group of females than males contributing to the project; 61% Female to 39% Male. This was comparably different to Galaxy Zoo contributors, who reported as 82% Male to 18% Female and Old Weather contributors who reported 53.8% Male to 48.3% Female (Raddick et al, 2013; Eveleigh et al 2014). However, Meitheal Dúchas.ie contributors gender statistics were very similar to Transcribe Bentham’s contributors who were 66.6% Female (Causer and Wallace, 2012).

Survey analysis of the Meitheal Dúchas.ie project revealed that the highest group of female contributors were in the 56 – 65 age bracket, and the highest group of males were in the 36 – 45 age bracket. Similar to Raddick’s (2013) study of Galaxy Zoo contributors, a larger percentage of respondents in the older age groups (66+) were male. Old Weather’s largest age brackets were 46 – 59 (26.8%) and 60 – 79 (32.1%). Galaxy Zoo’s largest age brackets were males aged 55 – 59 and 50 – 54 and females aged 25 – 29 and 45 – 49 / 50 – 54. Meitheal Dúchas.ie contributors were similar to Old Weather and Galaxy Zoo as the largest age brackets were 46 – 55 (22.08%) and 56 – 65 (22.08%), revealing that the highest percentage of contributors in all the studies were over forty.

Meitheal Dúchas.ie contributors were revealed as highly educated with 75% of males and 76% of females citing having primary degrees or higher. This is very similar to Galaxy Zoo contributors, who reported in Raddick’s (2013) study as 70% of respondents having at least a bachelor’s degree. Transcribe Bentham’s survey of transcribers revealed that 97% are educated to primary degree or higher. It is safe to say that crowdsourcing in a GLAM context is attracting a highly educated crowd.

Implications: The marketing strategy for crowdsourcing in a GLAM context needs to attract and maintain the demographic group identified (highly educated, over forty, male and female).

Crowdsourcing projects need to consider if their subject matter may be of more interest to males or females and market to these groups accordingly. For example, its appears from the projects reviewed that the subjects of ships logs (Old Weather) and astronomy (Galaxy Zoo) were more appealing to males and that the subject of personal historical papers / letters (Transcribe Bentham and Meitheal Dúchas.ie) were more appealing to females. Further marketing research to establish this hypothesis could be conducted.

Recommendations: Market GLAM crowdsourcing projects to University Alumni by advertising via University Alumni e-zines and e-mail notifications.

9.2 Contribution Levels and Benefits

Corresponding with other studies on crowdsourcing (Holley, 2010, Oomen and Aroyo, 2011) who note that around 10% of participants are high contributors, the Meitheal Dúchas.ie project’s survey showed that 13% were high contributors and that the majority of transcribers were in the low contributor categories (87%).

Implications: GLAM crowdsourcing projects need to encourage the low contributors to return more frequently.

Recommendations: Send crowdsourcing project contributors an e-mail thanking them for their initial contribution, which is then followed 7 days later by another e-mail inviting them to transcribe again. Follow up reminder e-mails could be sent out on a phased basis (for example, monthly) to all low contributors.

These notifications and the website homepage need to clearly demonstrate the benefit to others and / or the benefit to society of a participant’s contribution to the project, as the respondents in the Meitheal Dúchas.ie survey revealed they contribute for the benefit of others and their own interest. This correlates with Dunn and Hedges (2012) noting that most studies conclude that the majority of crowd-sourcing contributors have both personal and extrinsic motivations; that they do it both for themselves and for others.  In terms of motivations this shows that contributors are motivated by their altruistic tendencies, which GLAM crowdsourcing projects need to tap into through marketing campaigns / strategies in order to attract contributors. 

9.3 Social Network Use

Social network use was low (17%) among survey respondents for the purpose of discussing transcribing for the project. Of those who regularly used social networks (3%) the majority were in the low contribution groups (80%).

Implications and Recommendations: Social networking is the second most common activity (66%) (after e-mail at 84%) of individuals who accessed the internet in Ireland in 2015 (CSO, 2015). The findings of the Meitheal Dúchas.ie survey present an opportunity for the project to utilize social networking as a tool to attract and encourage low contributor transcribers to return by discussing and / or referring to topics or articles such as ‘how to transcribe’ or ‘good transcription practices’ which can be embedded on the project website and for example, shared via Twitter links. Transcribers can be invited to add their social network contact details when registering as a user of the website. However, it must also be acknowledged that Meitheal Dúchas.ie contains a small project staff, and that recommendations such as this are not always possible to implement as the human resources may not be available.

9.4 Initial Motivations

The top five initial motivating factors to participate in the project cited by respondents were subject matter (i.e. Interest in Folklore / Irish Language), the opportunity to contribute to research, feeling they had the necessary skills to contribute to the project, the opportunity to help the project and a personal or family connection to the material.

The top motivating factor to participate in the project was cited by respondents as subject matter, which correlates with Dunn and Hedges (2014) study which noted that in most cases, a single dominant motivating factor exists, which is nearly always linked with the project or activity’s subject area. This finding also correlates with the crowdsourcing projects reviewed for this study as Galaxy Zoo, ANDP, Old Weather and Papers of the War Department contributors all cited interest in the subject matter as a top motivator. Galaxy Zoo contributors (39.8%) cited opportunity to contribute to research as the top motivator; Meitheal Dúchas.ie contributors revealed that the opportunity to contribute to research is also a top motivator. Contributing for fun ranked low at 26% overall, declining even lower to 10% for the 1 – 4 Hours contributor group. However, compared to Galaxy Zoo contributors (2.8%) it is higher.

The most striking finding from the survey responses related to the difference between the High Contributor groups and Low Contributor groups in relation to finding transcribing text a simple and enjoyable task. 80% of high contributors cited cited enjoying text transcription as an initial motivator for participating in the project, yet only 39% of the Less than 1 Hours and 35% of the 1 – 4 Hours groups agreed. In addition, some respondents in the qualitative data analyzed cited not knowing how to transcribe or were waiting for an invite from the project to start.

Does this mean lower contribution groups see transcription as difficult? Could user interfaces make it more user friendly by providing step by step instructions on the website homepage? These already exist on the Meitheal Dúchas.ie website, yet they need to be made larger in text than the surrounding text, so they are clearly marked out, alongside a larger Transcribe Now button which easily stands out, as per the example from the Old Weather homepage below.

Old W

In contrast the Meitheal Dúchas.ie hompage main area contains a Register or Log In button and the Transcribe area needs to be scrolled down to, and it is not a button on its own, making it difficult to see.

DU

Website homepages for GLAM crowdsourcing projects need to be designed with encouraging participation in mind, minimizing barriers and clearly defining tasks. A Contribute button and Transcribe Now button need to be the prominent features on the homepage. This simple interface design reduces cognitive overhead, simplifies tasks and increases participant enjoyment (Ridge, 2013).

In addition, these findings support other research by Eveleigh et al (2013) that place value on designing interfaces that tempt low contributors to complete ‘just another page’ as the majority of contributors in GLAM crowdsourcing projects are in the low contributor groups. To this end, Eveleigh et al (2013) propose several design recommendations, such as breaking the work into components which can be completed without a significant commitment of time and effort, and providing feedback on the quality and value of these contributions. These practical changes, according to Eveleigh et al (2013), should make it easier for low contributors to contribute a little more and feel that their contribution is valuable and valued.

9.5 Motivational Framework Aspects

Referring back to the motivational framework these findings show us that the top five motivators, relate to intrinsic motivations such as egoism based motivations (interest in subject matter) and community based motivations such as the altruistic motivations to help the project.

 Understanding

54% of participants felt they had the necessary skills to contribute to the project. This is an intrinsic type of motivation which corresponds with the Understanding psychological function of volunteerism.  Practical implications of this finding are that crowdsourcing projects in their marketing strategy need to appeal to this psychological function of potential volunteers by clearly communicating how volunteers’ skills such as, knowledge of the Irish language or ability to type, are needed to make the project successful.

Values

Over half of respondents (52%) cited that the opportunity to help the project was a motivator, coupled with 63% who cited that the opportunity to contribute to research was a motivator.

In the Low Contributor groups, 52% of respondents cited the opportunity to help the project was a motivator and 61% cited the opportunity to contribute to research was a motivator. This was relatively similar for the High Contributors groups, 58% who cited helping was a motivator and 73% cited the opportunity to contribute to research was a motivator.

These are intrinsic and altruistic types of motivation, which correspond with the Values psychological function of volunteerism, where individuals are motivated to volunteer out of a desire for an opportunity to express their altruistic concerns. From a practical standpoint, crowdsourcing projects can use this information to appeal to the altruistic nature of potential volunteers. One suggestion is marketing messages in the media, via social media or on the website homepage which ask ‘Would you like to help our project’? or ‘Would you like to contribute to research?’

Galaxy Zoo, Meitheal Dúchas.ie, Papers of the War Department and ANDP all state the on their homepages that the help of volunteers is required, however the Transcribe Bentham homepage is the only website that refers to volunteers contributing to research.

The qualitative data retrieved from the responses to Question 5 provided further evidence that interest in the subject matter, an egoism based motivation, is a major initial motivator to participate.

Data analysis revealed that word ‘Time’ appeared frequently in the context of respondents citing they wished they had more time to contribute. This provides an opportunity for the project to use this information in a way that appeals to transcribers who feel they may not have enough time to contribute by sending out notifications via social media, email and on the website of short pages available for transcription.

Another popular topic within the corpus of text was the value of the ‘old Irish’ dialect contained within the national folklore collection pages. This finding is useful to the project as a means to attract transcribers, through creating marketing messages that invite transcribers to explore ‘old Irish’ dialect. In addition, social media messages could be created which contain screen shots of pages for transcription which contain ‘Old Irish’ or even Irish specific to a particular geographic region.

A re-occurring theme which appeared in the qualitative data for Question 5 and throughout the entire survey responses is how the Meitheal Dúchas.ie project had a family or personal connection for the respondent and how this was an initial motivating factor to participate. This resonates with subject matter being of high importance as a motivating factor to volunteer for a GLAM crowdsourcing project

9.6 Motivations to Volunteer Online

Question 8 in the survey provided respondents with a set of choices as to why they volunteered online.

The choices available included some ego-based motivation aspects of the motivations of participants, such as increasing one’s self-esteem. There was a need to explore to what degree are volunteers’ motivations egoism based, as all volunteers cannot be purely altruistically motivated. Essentially, the purpose of this question was to establish what type of other prominent motivational aspects, besides altruism, were at play and were they significant enough for GLAM crowdsourcing projects to acknowledge them.

The responses to Question 8 revealed that 71% of contributors like to volunteer on-line for a cause that is important to them, which is an altruistic motivation. Looking at the motivations that are egoism based and fall into the protective psychological function of volunteerism, we see that the survey responses for these categories are very low; Increases self-esteem (8%), makes me feel important (4.96%), makes me feel needed (7.8%), helps me forget my troubles (7%), helps me feel less lonely (3.5%), helps relieve some of the guilt I feel over being more fortunate than others (2.13%).

Essentially this tell us that egoism based motivations of this context are not prominent, and that GLAM crowdsourcing projects should focus on appealing to the egoism based motivations related to volunteers’ interest in the subject matters and the altruistic nature of potential volunteers.

34 respondents provided text for qualitative analysis. The most striking finding from analysis of this data was the amount of references to how volunteering online for the Meitheal Dúchas.ie project made contributors feel a greater connection to their ‘roots’ and cultural heritage. This aspect had not appeared in other responses.

This is an interesting finding which could be used by the Meitheal Dúchas.ie to attract volunteers by creating media messages via social media and messages on the website homepage which refer to connecting to one’s roots by engaging with the national folklore contained therein.

 

 

 

 

 

 

Data Visualizations of Qualitative Data

Qualitative Data

Qualitative Data was obtained from the survey responses to Question 8 and Question 5 in the ‘Other – please specify’ box. Computer-aided content analysis was used for coding the responses, with the purpose of yielding substantially interesting and theoretically useful generalizations. Traditional coding methods were also used to gain other insights into the qualitative data.

Computer-aided Analysis

Voyant-Tools.org, a web-based suite of analysis and exploration tools for digital text, was used to analyse the frequency of the words in the text of the respondent’s answers.

Question 8 asked respondents to select a statement in relation to volunteering on-line which applied to them. 34 respondents selected the ‘Other’ box, providing a corpus of text for qualitative analysis. Using Voyant-Tools for analysis, the most frequent words in the corpus were revealed as Project, Volunteering, Feel and Interested. With regard to the word Project, the responses were in a positive light. Respondents had volunteered on-line due to the Meitheal Dúchas.ie project being ‘marvellous’, ‘worthwhile’ and ‘ahead of its time’.

A word cloud of the corpus can be found here 

Volunteering frequently appeared in the corpus of text at the same time as the word Feel;

“Volunteering on a project like this makes me feel like part of something bigger.”

“As I live abroad volunteering for this project also makes me feel a little more connected to my roots”.

“Volunteering online makes me feel like I’m making a contribution to society”.

“Volunteering online allows me, in a way to ‘connect’ practically with my forebears”.

Finally, the last frequent word in the corpus was ‘Interested’, which appeared in the text solely in relation to interest in the subject matter, this being the reason for volunteering online.

Question 10 concerned respondent’s initial motivation for participating in the project and also contained a ‘Other- please specify’ box where respondents could write text. Using Voyant-tools.org for text analysis of the corpus of text the most frequent word was ‘Time’, within the context of respondents citing they wished they had more time to contribute.

 “Initially, I had more free time at my disposal for doing the transcribing but more recently had less time. I do plan, however, to pick up again and give more time to the transcribing as soon as possible”.

Transcribing, Irish, Project and Folklore followed Time as the next most popular words in the corpus.

A Word Cloud of the corpus can be found here.

Transcribing appeared in relation to Time, as previously discussed, yet the word also appeared within the context of how transcribing can allow one to feel more connected to and get to know the material within (i.e. the stories) to a greater level.

Irish appeared in much of the corpus in relation to the value of the Irish language contained therein, specifically, ‘old Irish’, such as the different dialects and styles of native speakers.

Project appeared within the text in primarily in relation to how the Meitheal Dúchas.ie project had a family or personal connection for the respondent. This is a re-occurring theme throughout the entire survey responses.

“Seeing my father’s and mother’s essays online was hugely motivational for me.”

 “I’ve been largely influenced [to participate] by the fact that my late mother was involved in collecting and writing much of the material in the 1930s and this project was very dear to her heart”.

Folklore appeared in the corpus in relation to how the subject matter of folklore was of interest to respondents, i.e. interest in the subject matter being an initial motivator to participate.

 

 

8.0 Research Survey Findings

8.1 Demographics: Demographic variables offer a picture of the Meitheal Dúchas.ie population, which provides insights into who is participating in the crowdsourcing project. The survey for this study asked about respondent’s gender, age, employment and level of education. In total, 154 respondents answered the survey. Statistical analysis was conducted using Survey Monkey and Microsoft Excel.

Gender and Age

Gender and age are key characteristics to determine the audience. Table 1 shows the gender distribution of responses.

Table 1: Respondents by Gender

Table 1 Resp. b y Gender

Of the 154 respondents, 61% were female and 39% were male. Irish population estimate figures for 2015 from the Central statistics Office (CSO, 2016) show that there is a relative equality between male and female genders in the population.  Therefore, the survey results demonstrate that there is a larger group of females participating in the Meitheal Dúchas.ie project relative to the population distribution of males and females.

Table 2: Age Groups of Contributors

Table 2

Table 2 displays the Age Groups of Contributors based on the survey responses. The Table below shows the female respondents by age group. The highest age category of females contributing to the project were in the 46 – 55 and 56 – 65 age category.

Survey Results in PowerPoint Presentation2

The largest group of males contributing to the project were in the 36 – 45 age category, followed by 46 – 55 and 66+.

Population estimates by age group show the greatest age categories lie in the 25 – 44 years’ group followed closely by the 45 – 64 years’ age group. Therefore, the female population of contributors to the project correlate with this and are a representative sample of the population. Male contributors to the project, for the most part correlated with the population estimates by having the largest age categories ranging from 36 – 45, 46 – 55 and 56 – 65. However, the male contributors also contained a large proportion of respondents in the 66+ age group.12% of the male population are over 65 years old (CSO, 2016). Adding the 66+ age group contributors plus 1 respondent aged 78 shows that 20% of survey respondents were 66+, demonstrating the project has attracted a relatively high number of male participants in the 66+ age group. In comparison only 10% of female respondents were in the 66+ age category.

8.1.1 Education

Education is an important audience measure for any project. The Meitheal Dúchas.ie contributors spanned a range of educational levels, however the majority held primary degrees or higher; with 75% of males and 76% of females citing holding primary degrees of higher. The Irish 2011 Census revealed 26 % of the population hold primary degrees or higher (CSO, 2012).

Table 3: Education Levels of Contributors

Contributors pci

Therefore, the survey results demonstrate that there is a larger group of participants in the Meitheal Dúchas.ie project with primary degrees or higher relative to the population demonstrating the ‘crowd’ for the Meitheal Duchas.ie project is highly educated.

Table 4: Employment Status of Contributors

Employment status

8.1.2 Employment Status

The survey results showed that the largest employment status among respondents was Full Time Employment, followed by Retired and Part Time Employment.

Focusing on the Low Contributor groups, i.e. Less than 1 Hour and 1 to 4 Hours, the survey results show that 37% of this group were in full-time employment, 22% in part-time employment and 22% retired.

8.2 Contribution Levels

In this study, the low contributors (the Less than 1 Hour and 1 to 4 Hours Groups) significantly outnumbered the high contributors. This provided a good sample of low contributors for analysis, as this group was intended to be the focus of the study. Of the 150 respondents who answered this question (4 skipped), 131 (87%) were classed as Low Contributors (Less than 1 Hour and 1 – 4 Hours) and 19 (13%) were classed as High Contributors (4 – 10 Hours, 10 – 15 Hours and 15 – 20 Hours).

Table 5: Contribution Levels of Contributors

Contibution L

Benefit of Others or Own Interest

Table 6 displays the overall responses for Question 7 which asked respondents if they contributed to the project for the benefit of others or for their own interest. From the graph below it appears that the majority of respondents contributed for the benefit of others.

Table 6: Question 7 Responses

Responses

However, on closer examination of the data it became apparent that a lot of respondents selected both options. Therefore 80 out of the 151 respondents (53%) who answered the question cited that they contributed to the project for their own interest and for the benefit of others.

Only 13 respondents cited that they contributed to the project solely for their own interest. Further analysis of responses indicated that 72 % of highly active contributors (4 hours +) have both personal and extrinsic motivations; they contributed for themselves and the benefit of others. Lower contributors (1 – 4 Hour group) were relatively similar in this regard, with 63% of respondents contributing for themselves and the benefit of others. The lowest contributor group (less than 1 hour) were more divided with 45% of respondents contributing for themselves and the benefit of others and 38% showed more altruistic tendencies by revealing their contribution was for the benefit of others alone.

 8.3 Initial Motivating Factors for Participation

In many crowdsourcing studies it is possible to identify a single, dominant motivating factor, which is almost always concerned directly with the project or activity’s subject area (Dunn and Hedges, 2014).

The dominant initial motivating factor in this study for all contribution groups was being motivated by the subject matter, for this project the subject matter being folklore and the Irish language. This highest score was in the 1 to 4 Hour Group with 83% of respondents citing that subject matter was an initial motivating factor to participate in the project. This was followed by the Less than 1 Hour and 4 to 10 Hours Groups both at 77%, and the 10 to 15 Hours Group at 60%.

The study revealed only one high contributor (15 – 20 Hours) who also identified subject matter as an initial motivating factor, along with opportunity to contribute to research, opportunity to learn new skills and opportunity to help. This respondent did not have a personal or family connection to the material.

Table 7: Initial Motivating Factors to Participate

Particpate

The top five initial motivating factors to participate in the project cited by respondents were subject matter (i.e. Interest in Folklore / Irish Language), the opportunity to contribute to research, feeling they had the necessary skills to contribute to the project, the opportunity to help the project and a personal or family connection to the material*.

*The Meitheal Duchas.ie project is unique as a public humanities crowdsourcing project in two regards; Firstly, due to the large corpus of material collected by the National Folklore Commission in the 1930’s. A vast amount of schools all over Ireland were involved in the study, making it easy for personal or family connections to the material to be established, either by school, location, teacher or public. This creates opportunities for contributors to be motivated to participate in order to seek folklore written by family members such as parents or grandparents.

 The highest overall group with a personal / family connection to the material were the 1 to 4 Hours and 4 to 10 Hours Group’s, at 50% and 53%, showing that over half of the total of respondents in these groups were initially motivated to contribute to the project because of a personal or family connection.

Similar data emerged from the Less than 1 Hours and 10 to 15 Hours Groups, with 35% of respondents in the first group and 40 % of respondents in the 10 to 15 Hours group being motivated to participate in the project due to a family or personal connection.

Overall, 26% of the total of respondents cited fun as an initial motivator, of which 80% of respondents were from the low contributor groups. Fun ranked lowest as an initial motivator among the 1 to 4 Hour group of respondents at 10%, and at 23% for the Less than Hour and 4 to 10 Hour groups.

In total, 39% of respondents found transcribing text a simple and enjoyable task. In the lower contribution groups this figure remained almost static, with the Less than 1 Hour Group at 39% and the 1 to 4 Hours Group at 35%. In contrast, 80% of the higher contributors cited enjoying text transcription as an initial motivator for participating in the project.

8.4 Motivations to Volunteer On-line

Question 8 asked contributors of the Meitheal Dúchas.ie project if they felt any of the statements listed regarding on-line volunteering applied to them. Of the 141 respondents who answered this question (13 skipped) 71% felt that they liked to volunteer online for a cause that is important to them.

Table 8: Results of Question 8 

 Volunteer

Q8

The second highest choice for respondents at almost 30%, was that volunteering online allowed them to explore their strengths. In the Discussion Section of this study, these findings will be compared to the six psychological functions of volunteerism, which can provide insight into how volunteers can be attracted to participate in crowdsourcing projects by appealing to the most likely psychological function of their potential volunteers.

 8.5 Social Network Use

In Question 9 of the survey, respondents were asked if they ever use social media sites such as Twitter or Facebook to discuss transcribing for the Meitheal Dúchas.ie project with others.

Table 9: Social Media Usage

T9

Of the 153 respondents who answered the question (1 skipped), 78% stated that they never use social media for the purpose of discussing transcribing for the project and 14% stated that they sometimes do.

4 out of the 5 respondents who stated that they regularly use social networks to discuss transcribing were in the Low Contribution level groups (Less Than 1 Hour and 1 to 4 Hour).

 8.6 Motivating Factors for Participation

In Question 10 survey respondents were asked to rate on a Likert Scale from 1 (not at all motivating) to 7 (very motivating) reasons that would motivate them to participate in a crowdsourcing project. 149 answered respondents answered this question (5 skipped).

In order for data analysis, the response categories in Question were assigned numerical values; 1 = Strongly Disagree, 2 = Moderately Disagree, 3 = Disagree Slightly, 4 = Neutral, 5 = Slightly Agree, 6 = Moderately Agree, 7 = Strongly Agree.

In the Feedback from the Project Organizers category 29% of respondents strongly disagreed. When combined with the Moderately Disagree and Slightly Disagree groups, it tells us that 53% of respondents do not find feedback from the project organizers as motivating.

By combining all three groups who agree, the results show that 33% of respondents find feedback from project organizers motivating. The top transcriber (15 – 20 Hours) strongly agreed that feedback from the project organizers was a motivator. The 10 – 15 Hours Group of transcribers did not find feedback a motivator, as all within this group selected 1, 2 or 3. In the 4 – 10 Hours Group of transcribers, by combining the three Agree categories we see that 62% of this group find feedback motivating.

In the lower contributor groups (Less Than 1 Hour and 1 to 4 Hours) when all Disagree categories are combined we see that 54% do not find feedback motivating. In this group, 17% were Neutral and by combining all three Agree categories the results show that 29% of lower contributors find feedback motivating. Strongly Agree was selected alone by only 13% of the lower contributor groups.

Social contact with others was not viewed as a motivator to participate with 41% of respondents selecting this category as ‘not at all motivating’ and only 7% selecting Strongly Agree.

Developing skills for future career development was not deemed a motivator for participants to contribute to crowdsourcing projects. When the Disagree and Agree categories are combined we see that in total 64% disagree and 26% agree.

Desire to share one’s own skills for the greater good was viewed as a motivator for participation by 68% of respondents overall when all Agree categories are combined (39.58% Strongly Agreed, 14.58 Moderately and 13.89% Slightly Agreed), and by 86% of respondents in the Low Contribution groups.

Competition with other transcribers ranked very low as a motivating factor to participate, with 73% of respondents selecting Strongly Disagree. By combining results from all three Disagree categories we see that 91% of respondents feel that competition with other transcribers is not a motivator. When Slightly Agree and Strongly Agree figures are combined, 7 out of the 6 respondents were from the Low Contributor Groups, and one respondent from the 4 – 10 Hours Group.

Recognition of the achievement of transcribers on the projects website was not viewed as a motivator for participation. Almost 67% of respondents disagreed with this statement, of which 88% were from the Low Contributors group.  Of the 5% who Strongly Agreed, the top contributor (15 – 20 hours) was included.

Similarly, acknowledgment of volunteers’ contributions on the websites ranking tables were not viewed as a motivator for participation. 46% strongly disagreed and when all three disagree categories are combined we see that 70% of respondents disagree with the statement.

Generating a feeling of achievement by contributing to the Meitheal Dúchas.ie project in one’s own spare time was considered as a motivating factor by 68% of respondents overall (all 3 Agree categories combined) and by 85% of Low Contributors.

The opportunity to learn more about the Meitheal Dúchas.ie project was viewed as a motivating factor to participate by 65% of respondents overall (20% Slightly Agree, 20% Moderately Agree and 25% Strongly Agree) and by 84% of respondents in the Low Contribution groups.

The responses for the addition of new content as a motivator to participate in the project were mixed. There was only a small difference in percentages in those who Strongly Disagreed (18%) and those who Strongly Agreed (24%).

However, overall, the Agree category outweighed the Disagree category at 54% versus 34%. The top transcriber (15 – 20 Hours) strongly agreed that addition of new content was a motivator.

 

7.0 Research Design

Survey research was selected as the research design method for this dissertation, which is regarded as a research strategy for conducting social research, rather than just a specific tactic or method (Robson, 2002). Bryman (1989: 104) provides a definition;

‘Survey research entails the collection of data on a number of units and usually at a single juncture in time, with a view to collecting systematically a body of quantifiable data in respect to a number of variables which are then examined to discern patterns of association’.

However, Robson (2002) cites Bryman noting that the ‘single juncture in time’ need not be taken too literately, as the realities that dictate survey data collection mean that surveys are often taken over a number of days or weeks and treated as if the collection was simultaneous.

Survey research is a positivist method of data collecting aimed at theory or hypotheses testing and takes a deductive approach to research, i.e. the researcher starts with a theory / hypotheses and tests theoretical postulates using the empirical data collected (Bhattacherjee, 2012). As survey research is a fixed design method, it is suited to hypothesis testing (Robson, 2002) and was therefore used within this study as the researcher was in a position to make predictions before the data was gathered and analysed.

7.1 Hypotheses and Assumptions

In the context of the motivation framework outlined in Table ??, a number of assumptions can be made about the motivations of contributors to the Meitheal Dúchas.ie crowdsourcing project;

Hypothesis 1 (H1): that intrinsically motivated Meitheal Dúchas.ie participants are more likely to engage with the project in depth. Therefore, intrinsic motivation is positively correlated with greater levels of contribution.

Hypothesis 2 (H2): that extrinsic motivators such as recognition are negatively correlated with greater levels of contribution.

For hypothesis 1, we draw upon data from two of the survey questions – Q6 which asked ‘How much time per week do you spend contributing to the Meitheal Dúchas.ie project?’ and Q7 which asked “Do you see your contribution to the project as being for your own interest, or for the benefit of others?’

For hypothesis 2 we draw upon data from survey question – Q10 which asked participants to rate a scale from 1(not at all motivating) to 7 (very motivating), the reasons that motivated them to participate in the project.

At this juncture, the researcher acknowledges a challenge to this study, set out by Eveleigh et al (2014), that comparing the results of empirical studies of motivation in citizen science is complicated because of the diversity of contributions and because typically the participants are motivated by more than one factor simultaneously.

7.2 Unit of Analysis

The survey research for this study involved the collection of data on a unit of analysis, i.e. the individuals who are contributors to the Meitheal Dúchas.ie project, at a single juncture in time, which consisted of a period of 8 days, from June 8th to 15th 2016 inclusive.

7.3 Survey Research Advantages and Limitations

Survey research has several strengths compared to other research methods. Surveys are an excellent vehicle for measuring a wide variety of unobservable data, such as attitudes, beliefs, behaviours and factual information such as employment status (Bhattacherjee, 2012), all of which were required for this study.

Advantages associated with questionnaire surveys: The survey method provided the researcher with a simple and straight forward approach to the study of attitudes, values, beliefs and motives, which was necessary given that the purpose of the survey was to reveal the motivations of the contributors of a crowdsourced digital humanities project, essentially ‘people characteristics’. Other advantages associated with the survey method include how it provided the researcher with an easy way of retrieving information on a large group of people (over 600 registered users). The self-completion on-line survey was an efficient method for providing large amounts of data at a low cost in a short period of time and involved the researcher using the first level ‘pro package’ of a survey software tool, Survey Monkey. This allowed for analysis of up to 1000 respondents and an export facility for the data obtained within the survey in a variety of formats such as PDF, PowerPoint, Microsoft Excel and most importantly CSV file format which would allow the data to be exported to data analysis software such as Tableau.

The justification for using a self-completed on-line survey is that this method allowed for respondents’ anonymity. This was necessary to encourage openness and honesty from respondents as several of the survey questions required respondents to provide sensitive personal information about their motives, beliefs and values and by providing such anonymity a greater response rate could be encouraged. Furthermore, by presenting all survey respondents with the same 10 standardised questions, it became possible to obtain a high reliability of response, which lends to the generalizability of the findings.

Limitations associated with questionnaire surveys: Non-response bias -survey research tends to be ill-reputed for its low response rates. A response rate of 15-20% is typical in a mail survey, even after two or three reminders. (Bhattacherjee, 2012).  Low response rates to surveys can be caused by technical issues such as lengthy and poorly designed surveys (Robson 2002); this issue was mitigated by creating a short survey which contained 10 questions taking no more than five minutes to complete. The email which the survey was attached to also informed potential respondents that it would take 5 minutes to complete.

In addition, the survey was reviewed for quality by Meitheal Dúchas.ie project staff who made suggestions for quality improvement which were incorporated into the survey by the researcher prior to the survey being administered. No reminder e-mails were sent to the 600 Meitheal Dúchas.ie members who received the initial e-mail containing the survey. The researcher discussed the possibility of sending an email reminder with project staff who rejected the proposal on the grounds that they prefer not to overburden their members with emails. A response rate of 25.6% was achieved with 154 out of 600 members completing the survey.

7.4 Social Desirability Response Bias: a disadvantage associated with survey questionnaires is the likelihood of social desirability response bias, whereby individuals do not accurately report their beliefs or attitudes and instead respond in a way that shows them in a good light (Robson, 2002). Although there is ‘practically no way of overcoming the social desirability bias’ in a questionnaire survey (Bhattacherjee, 2012: 81) measures taken to mitigate this disadvantage by the researcher included creating the survey as anonymous, with the intention of encouraging respondents to more accurately report their attitudes, beliefs and motivations.

7.5 Other Limitations: Respondents not treating the exercise seriously is a limitation associated with self-completion surveys (Robson, 2002). Given that the survey respondents were a group of 600 individuals who registered with and volunteered to contribute to the Meitheal Dúchas.ie project, the researcher was confident that any individual who chose to complete the survey would do so in a serious and professional manner. A general disadvantage associated with all surveys using respondents is that the data may be affected by the characteristics of the respondents such as their motivation or personality (Robson, 2002). This limitation was not seen as a major disadvantage for this study as the primary focus of the survey was to gather information on respondent’s motivations and personalities.

7.6 Other Research Methods Considered

Case research is an in-depth investigation of a problem in one or more real-life settings (case sites) over an extended period of time (Bhattacherjee, 2012).  Case research, in the form of a case study, was initially selected as the research method for this study as it is regarded as a strong and relevant research method for research questions that require extensive and in-depth description of social phenomenon, (for example, this research study aimed to explore crowdsourcing as a social phenomenon) (Bhattacherjee, 2012, Yin, 2009). In addition, the case study is a relevant research method when ‘What?’ type of questions are used in a study (Yin, 2009), and the research question for this study aimed to explore; What motivates individuals to participate in digital humanities crowdsourcing projects?

Data for case research is collected using a combination of interviews, personal observations, and internal or external documents (Bhattacherjee, 2012). However, due to the recent nature of the Meitheal Dúchas.ie project (in operation since 2013) there was a lack of data to work with. For example, the project had conducted no previous surveys with regard to transcription contributor motivations and no in-house reports or data existed in this regard and no external documents. As the project contributors were on-line volunteers and the researcher was not a Duchas.ie project staff member, the researcher felt there would be too many limitations, such as time-constraints and sourcing willing interviewees from the contributor base, to organise face-to-face interviews for data collection regarding motivation. For these reasons, the alternative research method of survey research was selected.

7.7 Scale Reliability and Validity

Reliability is a statistical measure of how reproducible the survey instruments data are (Litwin, 1995). Alternate-form reliability was used as the reliability method for this research. This involved using different worded questions to obtain the same information about the topic of motivational behaviour of survey respondents, to determine if their motivations were altruistic (for the ‘greater good’ / benefit of others) or egoism based (out of personal interest or desire to learn new knowledge).

For example, Question 7 asked ‘Do you see your contribution to the project as being for your own interest, or for the benefit of others? Question 8, Part 4, asked the same yet worded differently; ‘Do you feel any of the following statements apply to you? – I like to volunteer for a cause that is important to me.

Validity is a measure of a survey instruments accuracy, i.e. how well it measures what it sets out to measure (Litwin, 1995). Content validity, used for this research study, is a subjective measure of how appropriate the items seem to a set of reviewers who have some knowledge on the subject matter (Litwin, 1995).  This involved an organised review of the surveys contents to ensure that it included everything it should and that it did not include anything it shouldn’t have. This process was conducted by staff at Fiontar, DCU, who are proficient in survey administration.

Although this is not a scientific measure of the survey’s accuracy, it is considered as a good foundation for building a methodologically sound assessment of the surveys validity (Litwin, 1995).

7.8 Data Collection

Data collection for this piece of survey research involved the use of a standardized questionnaire (a research instrument consisting of a set of questions intended to capture responses from respondents in a standardized manner) to collect data about people (. i.e. Meitheal Dúchas.ie members) and their preferences, thoughts, and behaviours (.i.e. motivations) in a systematic manner (i.e. use of a fixed quantitative design) (Bhattacherjee, 2012). The ten questions for the survey questionnaire were drawn from frameworks developed by Clary et al (1998), Crowston and Fagnot (2008) and Alam and Campbell (2012).

7.9 Survey Design

 A self-completion survey was chosen for this piece of survey research due to its associated advantages, described by Robson (2002) in the context of humanities research;

  • Surveys work best with standardized questions, where questions will mean the same thing to different respondents.
  • Surveys are carried out for descriptive purposes and provide information about the distribution of a wide range of people characteristics and of relationships between such characteristics.

In addition, the survey method is best suited to studies which have individual people as the unit of analysis (Bhattacherjee, 2012). For this study, the unit of analysis is the individual, i.e individual contributors to a crowdsourced digital humanities project.

The survey was administered over the Internet using interactive forms, i.e. an on-line survey. Respondents received an electronic mail request for participation in the survey from Meitheal Dúchas.ie which contained a link to the Survey Monkey website where the survey could be completed.  The on-line survey method of data collection was chosen by the researcher due to its associated advantages such being inexpensive to administer, results were instantly recorded in an online database, the survey could be easily modified if required and this method was less time consuming than completing a standard mail survey.

The structured survey questions contained a variety of response formats;

  • Dichotomous response – where respondents were asked to select one of two possible choices, for example; male / female.
  • Nominal response – where respondents were presented with more than two unordered options, for example; Which of the following categories best describes your employment status? – In Part-Time Employment / Retired / Unemployed.
  • Ordinal response – where respondents were presented with more than two ordered options, for example; What is the highest level of education you have achieved? – Primary school / Secondary School / Technical Certification.
  • Interval-level response – where respondents were presented with a 7-point Likert scale, for example see survey Question 10 (Appendix 1).

In order to achieve the best response rates possible, questions were sequenced to flow from the least sensitive to the most sensitive and from the factual to the behavioural in order to adhere to the general rules for survey question sequencing as outlined by (Bhattacherjee, 2012). The survey starts with easy non-threatening questions that can be easily recalled, for example, demographics such as gender, age and education level. Following on from here the survey sequence moves onto questions relating to the behavioral, such as questions relating to personal motivations.

Other ‘golden rules’ of survey research identified by Bhattacherjee, (2012) and adhered to for this study included assuring respondents about the confidentiality of their responses, how their data would be used for academic research, how the results would be reported on completion of the study (via the Meitheal Dúchas.ie blog) and thanking respondents for their participation in the study. All of the above measures were included in the e-mail to all 600 registered users of Meitheal Dúchas.ie which contained the link to the on-line survey questionnaire.

The on-line survey questionnaire allowed for close examination of the motivations of transcribers based on a combination of specific and general questions. The questionnaire contained ten structured questions asking respondents to select an answer from a set of choices. In addition, Question 5, Question 7, Question 8 and Question 10, which focused on the motivations of respondents, contained a ‘Other Response’ box to allow for open ended survey questions in order to collect qualitative data that had the potential to produce unforeseen insights that could not be gathered from the structured quantitative data alone (Bhattacherjee, 2012).

7.10 Pre-Testing

 The survey questionnaire was pretested on Meitheal Dúchas.ie project staff who uncovered biases and errors in question wording which were subsequently eliminated prior to the survey being administered to the intended sample.

7.11 Survey Administration

 Duchas.ie, the GLAM institution selected for this research study, executed delivery of the survey questionnaire via email to their 600 registered users on June 8th 2016. Responses were collected for a period of 8 days, with the survey being closed-off by the researcher on June 15th 2016. The decision to close-off the survey at this stage was due to a major decline in response rates and that the Duchas.ie project staff preferred not to send out a reminder e-mail regarding the survey. In total, 154 responses were collected, giving the questionnaire a response rate of 25.6 % by June 15th 2016. As the response rate was 5% over a typical response rate (15 – 20%), the researcher felt this was a reasonable sample to allow for data analysis and decided to close-off the survey.

Respondent Numbers by Date

7.12 Data Analysis

Quantitative Analysis: Quantitative data was obtained from the survey responses and analysed for statistical data using Survey Monkey and Microsoft Excel.

Qualitative Analysis: Qualitative data was obtained from the survey responses to Question 8 and Question 5. Computer-aided content analysis was used for coding the responses, with the purpose of yielding substantially interesting and theoretically useful generalizations. Traditional coding methods were also used to gain other insights into the qualitative data.

Advantages associated with computer-aided content analysis include that the rules for coding text are made explicit and that the computer provides high coder reliability (Somekh and Lewin, 2011).  For this type of data analysis, the word or phrase is the basic unit of analysis.

7.13 Research Ethics

 Anonymity and Confidentiality: To protect subjects’ interests and future well-being their identity was protected in this study, using the dual principles of anonymity and confidentiality. By providing a guarantee of anonymity, the researcher implied to all potential respondents that the researcher and readers of the final research paper could not identify a given response from a specific respondent (Bhattacherjee, 2012).

This research attended to the ethical issue of confidentiality of survey participants and protection of their identities. This issue was addressed by ensuring the survey was anonymous. The Survey Monkey software used to collect the survey questionnaire responses contains a facility to block all access to respondents IP addresses, which the author utilised, therefore making the survey anonymous. Respondents were made aware of the confidential and anonymous nature of the survey on the letter attached to the email.

Analysis and Reporting: The researcher of this study was aware of the ethical obligations to the scientific community on how data is analyzed and reported in a study, such as the necessity to fully disclose negative findings, that hypotheses should not be designed after the results of the data analysis and that any problems encountered during the research process should be disclosed, so that other researchers may avoid similar problems (Bhattacherjee, 2012).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6.0 Motivational Theoretical Framework

Motivation has been defined differently in a variety of contexts and studies; this study will refer to Humphreys & Revelle’s (1984) definition cited by Raddick et al (2013),

where motivation is considered to be;

a mental construct that a volunteer uses, consciously or unconsciously, to explain their behaviour, arising out of the person’s mental state and properties of the situation they are in.

Motivators are broadly divided into intrinsic (those which stem from the task itself) and extrinsic (the outcomes of an activity) (Eveleigh et al, 2014). The fundamental concerns of motivational inquiry are understanding the processes that move people to action—the processes that initiate, direct, and sustain action (Clary et al, 1998).

Motivations of crowds are a key factor in crowdsourcing. As crowdsourcing is used for a broad range of functions, the motivations of the crowd vary considerably based on the nature of the task. For example, research has shown that in crowdsourcing activities, such as Wikipedia, crowds are intrinsically motivated (Hossain, 2012). Intrinsic motivation is predominant in crowdsourcing forms which include citizen journalism, citizen science and public participation in state and community building (Chandler and Kapelner, 2013). When a crowdsourcing task is complex, extrinsic motivations are seen to be more prevalent than intrinsic motivation, while motivations in intermediary crowdsourcing platforms are seen to be mostly extrinsic (Jeppesen and Lakhani,2010).

Several studies used general motivational studies to formulate motivational frameworks for explaining the motivations of crowdsourcing participants, for example Alam and Campbell (2012), Rotman et al (2012), Kaufeman and Schulze (2011), Leimeister et al (2009), Crowston and Fagnot (2008), Lakhani and Woolfe (2005) and Batson et al (2002). A model of motivations for social participation towards achieving common goals was presented by Batson et al 2002, which identified four types of motivations: egoism, altruism, collectivism and principalism. Egoism occurs when the goal relates to increasing one’s own welfare. Altruism occurs when the goal relates to increasing the welfare of another individual or group of individuals. Collectivism has the goal of increasing the welfare of a specific group that one may belong to and Principalism has the goal of upholding one or more principles dear to one’s heart such as justice or equality (Alam and Campbell, 2012).

 

Kaufmann & Schulze (2011) provided a more detailed model in which their motivational framework drew on classic motivational theories such as Self Determination Theory (Deci & Ryan 1985, 2000), work motivation theory (Hackman & Oldham 1980), education theory (Weis 1995) and established theoretical models such as open source software development model (Lakhani & Wolf 2005). They combined these theories and frameworks to create a worker motivational framework for paid crowdsourcing environments such as Amazon Mechanical Turk. The Worker Motivational Framework model focuses on intrinsic and extrinsic motivations and is divided into five motivation categories; enjoyment based motivation, community based motivation, immediate pay-offs, delayed pay-offs and social motivation. Alam and Campbell (2012) adapted these two frameworks to develop their own framework (see Table ??) to explain the motivations of crowdsourced text correctors in their study on crowdsourcing motivations in a not-for-profit GLAM context.

 

Table 4: Model of Text Correctors Motivation in Crowdsourcing (Alam and Campbell, 2012).

   

Category

 

 

Constructs

 

Source

 

Intrinsic

Motivation

 

Egoism-based motivation

 

Personal interest (e.g. research goals), Trust, Challenge, Learning new knowledge, Competition, Topic of Interest (Australian history), Addiction, Obligation, Supportive Environment

 

 

Batson et al (2002)

Holley (2010)

 

Intrinsic

Motivation

 

Community based motivation (Altruism, Collectivism and Principalism).

 

 

Altruism, Collectivism (e.g. genealogy), Principalism (or action significance by external values)

 

Batson et al 2002

Deci & Ryan 1985 cited in Kaufmann & Schulze 2011

 

Intrinsic

Motivation

 

Enjoyment based motivation (Task based motivation)

 

 

Enjoyable/Fun/pleasure/recreation, Simplicity, Task Autonomy, Pastime

 

Hackman & Oldham 1980, Deci & Ryan 1985 cited in Kaufmann & Schulze 2011

 

Extrinsic

Motivation

 

Social motivation

(non-monetary reward)

 

Acknowledgement, attribution & ownership, Desire for recognition (Ranking table & Hall of fame), Rewards (Australian Day awards), Indirect Feedback, Advocacy

 

 

Hackman & Oldham 1980 cited in Kaufmann & Schulze 2011, Rotman et al 2012

 

The category Community Based Motivation differs from the Kaufmann & Schulze framework as text correctors are more motivated by factors relating to welfare and principles of the group and not motivated by guidance of the platform community. The category enjoyment/task based motivations are similar to Kaufmann & Schulze’s model and include relevant constructs such Fun, Task Autonomy and Pastime. The only extrinsic motivational category relevant for crowdsourced text correction is social motivation which is described as a non-monetary motivation and includes rewards that are non-monetary in nature such as acknowledgement, recognition and reward, indirect feedback and advocacy (Alam and Campbell, 2012)

An aspect neglected by the framework proposed by Alam and Campbell is the motivational behaviours associated with volunteerism. Crowdsourcing in the GLAM sector relies upon on-line volunteers to make contributions. This study therefore proposes a framework which combines Alam and Campbell’s (2012) model with elements of the motivational foundations of volunteerism proposed by Clary et al (1998).

6.1 Psychological Functions of Volunteerism

 Clary et al (1998), drawing on classic functional theories proposed by Katz (1960) and Smith et al (1956), hypothesized 6 psychological functions that are potentially served by volunteerism;

Values – volunteerism provides opportunities for individuals to express values related to altruistic and humanitarian concerns for others.

Understanding – volunteerism provides opportunities to exercise one’s skills, knowledge and abilities and for new learning experiences.

Social – opportunities to engage with others.

Career – opportunity for career related benefits (e.g. preparing for a new career).

Protective – refers to how volunteerism can be used as a mechanism to protect the ego from negative features of the self (e.g. to escape negative feelings or reduce guilt over being more fortunate than others).

Enhancement – the enhancement function involves a motivational process that focuses on the ego’s growth and development and involves positive strivings of the ego (e.g. people use volunteering as a means to enhance or maintain personal growth or self-esteem).

This information can be used in the recruitment of volunteers by appealing to their own psychological function. For example, Clary et als (1998) study created six advertisements that asked readers to become volunteers, with each advertisement corresponding to one of the psychological functions of volunteerism. The hypothesised result is that by appealing to each volunteer’s psychological function, volunteers will be satisfied to the extent that they have engaged in volunteer work that serves their own psychological function, and they will then plan to continue to serve as volunteers.

The defining and characteristic features of volunteerism as voluntary, sustained, and ongoing helpfulness suggest that for this study it may be productive to inquire about the motivations that dispose individuals to seek out volunteer opportunities and the motivations that sustain their involvement in volunteerism over extended periods of time.

The Alam and Campbell (2012) Model of Text Correctors Motivation in Crowdsourcing is merged with the 6 functions served by volunteerism proposed by Clary et al (1998) for this study so we can understand the phenomenon of crowdsourcing motivations in a GLAM context at multiple levels through merging different theoretical perspectives and frameworks.

Using Alam and Campbell’s framework to categorise the 6 functions served by volunteerism proposed by Clary et al (1998), the constructs are categorised as altruism, new knowledge learning, engaging with others, career related benefits, protect the ego, enhance self-esteem, which are then classed by motivation type, which may be intrinsic or extrinsic.

Table 5: The Crowdsourcing Motivations Framework

 

Psychological Functions of Volunteerism

 

Constructs

 

Motivation Type

 

Motivation Category

 

Values

 

 

Altruism

 

Intrinsic

 

Community Based

 

Understanding

 

 

New knowledge learning

 

Fun, pleasure, recreation

 

Intrinsic

 

Egoism Based

 

 

Enjoyment Based

 

Social

 

 

Engaging with others

 

Intrinsic

 

Community Based

 

Career

 

 

Career related benefits

 

Intrinsic

 

Egoism Based

 

Protective

 

 

Protect the ego

 

Intrinsic

 

Egoism Based

 

Enhancement

 

 

Enhance self-esteem, personal growth

 

Extrinsic

 

Social Based

6.2 Implications for Practice

 The implications for practice drawn from this study is that it potentially provides a solution base for addressing issues associated with citizen participation in meeting the needs of society, i.e. how to actively engage, through digital means, the eager GLAM institution volunteer base that exists in society.

 The findings of this study have direct implications for GLAM institutions that are dependent on the services of volunteers; such institutions could use The Crowdsourcing Motivations Framework (see Table 5) to assess the motivations of potential volunteers (i.e. crowdsourced contributors) and in turn utilize this information to strategically promote their organization in a manner that speaks to the predominant motivations of the volunteers they seek to recruit.

A second practical application, recognised by Crowston and Fagnot (2008), focuses on the ongoing nature of volunteerism. The functional approach proposes that a volunteers continued contributions depend on the person-situation fit. Essentially, volunteers who help in roles that complement their own motivations will obtain more satisfaction and more enjoyment from their service and are more likely to continue to help than those whose motivations have not been addressed by their activity.

‘surveys of motives for contribution should be careful to include the level of participation and to separate motives for different levels of contribution’ (Crowston and Fagnot, 2008).

 

 

 

5.0 Previous Research on Crowdsourcing Motivations

Most studies on digital humanities crowdsourcing conclude that the majority of contributors do not have a single motivation; the survey of crowdsourcing participants conducted by Dunn and Hedges (2014) indicated that 79% of highly active contributors have both personal and extrinsic motivations. However, the same study also noted that in most cases a single, dominant motivating factor exists, which is nearly always linked with the project or activity’s subject area. In this context Dunn and Hedges note that;

In general therefore, it may be said that research into crowd-sourcing motivations suggest a clear primary, although certainly not exclusive focus on the subject or activity area, and that this focus can be altruistic, extrinsic or intrinsic (Dunn and Hedges, 2014: 10).

For example, Raddicks et al (2010) study found that for Galaxy Zoo contributors the top motivation (39%) was an interest in astronomy. In addition, most studies on humanities crowdsourcing conclude that it is only a small number of contributors who end up being actively engaged and doing a large percentage of the work (Dunn and Hedges, 2014; Owens, 2013).

 5.1 Australian Newspaper Digitization Programme

 Alam and Campbell’s (2012) study focused on crowdsourcing motivations in a not-for-profit GLAM context and presents findings from a study of the motivational factors affecting participation in the Australian Newspaper Digitization Programme (ANDP) by the National Library of Australia. The study, focused on the crowdsourcing task of text correction, while data for the study was collected in the form of a user survey, forum posts and interviews.

Based on motivational theories, the study found that participants were motivated by a complicated configuration of personal, collective and external factors. Participants were found to be highly intrinsically motivated and valued altruistic and community motivations which were found to play a significant role in their continued involvement in contributions.

Personal interest in the subject matter, a sense of obligation to contribute out of gratitude to the library for providing the resource, learning new knowledge and gaining insight from the newspaper texts, and a supportive forum environment motivated continued participation. Others viewed text correction for the project as an altruistic patriotic task that would help Australian history and hence this was their motivator. Others reported enjoyment based motivations, such as viewing the task of text correction as fun and an enjoyable was to pass the time.

Extrinsic motivators for the participants were found to be attribution, i.e. naming text correctors on the articles they amended, recognition and rewards, such as public recognition of their achievements on user profile and ranking tables and rewards such as the top 5 contributors being invited to meet project staff. Indirect feedback was also noted as an extrinsic motivator such as impromptu updates on the website about the newspaper correction and its progress, regular email acknowledgement of the text corrector’s work, and public postings on the forum and in NLA newsletters by the project team recognising the efforts of the text correctors.

Overall, the study found that personal and community based motivations were the initial motivators to contribute to the project, while extrinsic motivators such as reward, provided the motivation for long-term participation in the project. Based on the findings of the study, Alam and Campbell (2012), propose a Model of Text Correctors Motivation in Crowdsourcing (see Table 2).

 Table 2: Model of Text Correctors Motivation in Crowdsourcing (Alam and Campbell, 2012).

 

   

Category

 

 

Constructs

 

Intrinsic

Motivation

 

Egoism-based motivation

 

Personal interest (e.g. research goals), Trust, Challenge, Learning new knowledge, Competition, Topic of Interest (Australian history), Addiction, Obligation, Supportive Environment

 

 

Intrinsic

Motivation

 

Community based motivation (Altruism, collectivism and Principalism).

 

 

Altruism, Collectivism (e.g. genealogy), Principalism (or action significance by external values)

 

Intrinsic

Motivation

 

Enjoyment based motivation (Task based motivation)

 

 

Enjoyable/Fun/pleasure/recreation, Simplicity, Task Autonomy, Pastime

 

Extrinsic

Motivation

 

Social motivation

 

(non-monetary reward)

 

Acknowledgement, attribution & ownership, Desire for recognition (Ranking table & Hall of fame), Rewards (Australian Day awards), Indirect Feedback, Advocacy

 

 

5.2 Transcribe Bentham

Similarly, to the findings from the Australian Newspaper Digitization Programme study, a crowd-sourcing scoping study conducted by Dunn and Hedges (2014) of the Transcribe Bentham public humanities crowdsourcing project, reported that most studies on the topic of crowdsourcing motivations conclude that contributors do not have a single motivation. Findings from the survey they conducted, indicated that 79% of respondents had both personal and extrinsic motivations and that they contribute for themselves and for others.

Transcribe Bentham Project Home Page

TB

Based on findings from a literature review on the topic, Dunn and Hedges (2014) posit that in most cases a single, dominant motivating factor exists, which is nearly always concerned directly with the projects subject area. In line with findings from the ANDP project study, Dunne and Hedges (2014) research also identified reward in the form of feedback to participants acknowledging contributions made as a motivator. Competition is seen as a possible motivation for people to participate in crowd-sourcing projects, however very few Transcribe Bentham participants admitted to being motivated by competition with each other.

5.3 Old Weather

In most crowdsourcing projects, a large proportion of participants contribute in small quantities, with a small number of ‘super transcribers’ contributing the bulk of the workload (Eveleigh et al 2014). To investigate how small contributors differed from so called ‘super transcribers’, Eveleigh et al (2014) distributed a survey to members of the Old Weather project, a Zooniverse citizen science project to transcribe the weather observations recorded in historical ships’ log books, and conducted interviews with respondents selected according to their levels of contribution to the project.

OW

The Old Weather project contains a ranking system that recognises the quantity of weather transcriptions made by each volunteer with the aim of motivating sustained and loyal participation through competition where participants can become ‘Captain’ of each ship.

The aim of the survey for this piece of research was to investigate how intrinsic and extrinsic motivations affect the quantity of contributions and the depth of user participation. The study found that participants of Old Weather demonstrated intrinsic motivation factors which included subject interest and curiosity, competence in the transcription task, and an enjoyment from taking part in the project. Both intrinsic task and extrinsic motivations were found to be related to participant’s contribution behaviour in Old Weather, with highly motivated participants contributing more. Intrinsic motivation was linked to wider contributions to the project, such as making more forum contributions and transcribing the optional non-weather information.

This research differed from the previous two studies in that a series of interviews were conducted with selected project participants in order to gain insight specifically into the experiences of low contributors, coined by the researchers as ‘dabblers’.

The study revealed that ‘dabblers’ are less motived compared to ‘super-contributors’ but they are still motivated. This is an important angle to consider given that the majority of participants in crowdsourcing projects display small-scale contribution patterns, and has led to the authors suggesting these findings have value in terms of institutions designing interfaces to tempt contributors to complete another transcription and to lure ‘early drop-outs’ back to participate.

Motivations for crowdsourcing contributors may change over time, for example many Old Weather volunteers are initially interested by the possibility to contribute to climate change research, but become interested in maritime history as they engage with the project’s content. In many projects, the feedback loop, affirming to contributors that their contributions made were correct and valuable, has been established as an important component of the reward for engagement (Dunn and Hedges, 2014) an extrinsic motivator.

 5.4 Galaxy Zoo

In this paper Raddick et al (2013) analyse results from an online survey to measure the motivations of almost 11,000 volunteers participating in Galaxy Zoo, an astronomy citizen science project which invites volunteers to classify the shapes of galaxies seen in images of the Sloan Digital Sky Survey.

Galaxy Zoo Homepage

GZ

A wide range of primary motivations were revealed from the research (see Table???). The most important motivation to Galaxy Zoo participants (almost 40% of respondents) was Contribute, i.e. participants felt excited to contribute to original scientific research. The second most popular primary motivation was Astronomy (almost 13% of respondents), i.e. interest in the subject matter of astronomy. This finding concurs with the other studies discussed demonstrating that interest in the subject matter is a primary motivator for contributors in crowdsourcing humanities projects. The findings similar to the other studies revealed that intrinsic motivators such as Fun, Learning and Helping are primary motivators for participants contributing to the project.

Table 3 Primary Motivations for all respondents (Raddick et al, 2013)

 

 

Motivation Category

 

 

Percentage

Contribute 39.8%
Astronomy 12.4%
Discovery 10.4%
Beauty 8.9%
Vastness 8.3%
Science 6.8%
Zoo 4.1%
Helping 2.8%
Fun 2.8%
Learning 1.6%
Other 1.3%
Teaching 0.7%
Community 0.2%

5.5 Stardust@home

 Nov, Arazy and Anderson (2011) conducted research on the motivations of volunteers of the Startdust@home project, a digital citizen science project in which volunteers classify on-line images from NASA’s Stardust spacecraft.

Stardust@home Homepage

SD

Through a survey administered to volunteers this research revealed similar findings to the research of Raddick et al (2013) in that intrinsic motives such as enjoyment and collective motives such as identification with the goals of the project, were identified as the most prominent motivations.

A study of the motivations of volunteers in citizen projects conducted by Rotman et al (2012) revealed that volunteers presented a range of egoism-related reasons as the initial motivation for participating in citizen science projects, such as familiarity with and personal interest in the subject matter.

Volunteers presented secondary motivational factors, such as recognition and attribution, which affected their ongoing participation in the projects. All volunteers recognized for their individual contributions to the projects identified this as a vital motivational factor. Other motivations, such as collectivism and altruism were identified as effecting long-term engagement in citizen science projects.

5.6 Papers of the War Department

 Leon (2014) discovered similar findings in a study focusing on what initially brought users to the Papers of the War Department, 1784 – 1800 a community transcription project (See Figure ??). Initial motivators for users contributing to the project were intrinsic in nature, such as interest in the topic (U.S. History) and altruistic in nature, such as a sense of civic duty to participate.

Papers of the War Department Homepage

PWD

While these research studies focused on different elements and tasks of digital humanities crowdsourcing projects ranging from transcription to classification, there is a general theme revealed in the primary motivations of participants. The primary motivations for volunteers contributing to the projects are intrinsic motivations such as interest in the subject matter, fun and learning, with collective motivations such as desire to contribute to scientific research, altruistic motivations such as a sense of civic duty to contribute. In addition, extrinsic motivations such as rewards in the form of acknowledgment from project organisers were identified as the secondary motivators which effect long term participation in projects.

5.7 Gamification and Competition

 Previous research suggests that gamification (the use of game design elements in non-game context) can make repetitive tasks, such as transcription, more enjoyable and sustain volunteer engagement. However, concerns exist that gamification may not appeal to all volunteers as some users prefer a ‘more serious’ user interface (Eveleigh et al (2013). Gamification in crowdsourcing projects is seen as a way to potentially motivate participants to contribute and sustain participation. Eveleigh’s et al (2013: 79) study of gamification in the Old Weather project found that;

“the same competitive mechanisms which some volunteers found rewarding and motivating were either ignored by other participants, or contributed to a decision to discontinue participation”.

The study revealed both positive and negative views to gamification. In Old Weather a project a ranking system recognizes the level of contribution made by each volunteer as a way of encouraging participation, with three levels through which a volunteer can progress when they ‘join a ship’; Cadet, Lieutenant and Captain (the top transcriber). There is also a list of leading ‘crew’ for each ship (Eveleigh et al 2013).

Positive views to gamification from volunteers included how the ranking system validated one’s contribution and healthy competition existed so one could become Captain of a ship. Negative views were that some volunteers found gamification de-motivating, for example, high scoring participants were spurred on by vying for the top position, yet low scoring participants were demotivated by a ‘distant competition’ they had no hope of achieving. In addition, other volunteers felt the ranking system undervalued smaller contributions and that being Captain of a ship was stressful. Eveleigh et al (2013: 82) concluded that;

 “the same competitive gamification mechanisms which motivated some leading volunteers were either ignored by more casual participants, or contributed directly to the decision to discontinue participation”.

Further negative views towards gamification within GLAM crowdsourcing were discussed by Dunn and Hedges (2014) who cite Prestnopnik and Crowston (2011) noting that gamification can potentially operate as a disincentive to contributors who have expertise or in-depth interest in the subject area, can be a barrier for users who just want to engage in the task in question, such as transcription, and can trivialise the process of acquiring or processing data.

Transcribe Bentham volunteers, it was noted by Dunn and Hedges (2014) were not motivated by competition with each other. Volunteer’s for Old Weather felt that competition can lead to sacrificing quality over quantity (Eveleigh et al, 2013).

Although a Hall of Fame was included on the ANDP website due to user demand, Alam and Campbell (2012) note that survey results show that inclusion on the ranking table, i.e., competition among contributors, was not considered important in terms of motivation to participate.