A Geography of Participation in IT-Mediated Crowds

Abstract

In this work we seek to understand how differences in location effect participation outcomes in ITmediated crowds. To do so, we operationalize Crowd Capital Theory with data from a popular international creative crowdsourcing site, to determine whether regional differences exist in crowdsourcing participation outcomes. We present the results of our investigation from data encompassing 1,858,202 observations from 28,214 crowd members on 94 different projects in 2012. Using probit regressions to isolate geographic effects by continental region, we find significant variation across regions in crowdsourcing participation. In doing so, we contribute to the literature by illustrating that geography matters in respect to crowd participation. Further, our work illustrates an initial validation of Crowd Capital Theory as a useful theoretical model to guide empirical inquiry in the fast-growing domain of IT-mediated crowds.

Crowd Science: Measurements, Models, and Methods

“…Researchers and managers operating in different industries and seeking different resources from a crowd will naturally need to develop the operationalization that makes most sense for them. However, in-line with [47,62] among others, we posit that crowd participation/contribution count variables, count variables of the number of problems solved, and count variables of text accumulated, all qualify as meaningful measures of the Crowd Capital generated. Now that our review of some crowd science operationalizations is complete, in the ensuing section we investigate some different research methods for an empirical crowd science.…”

“…From this research beginning, the body of Crowdsourcing research has expanded rapidly, supplying numerous taxonomies, typologies, and findings related to the various processes, attributes, and outcomes of engaging IT-mediated crowds, including some of the following themes; task complexity [25], crowdsourcing models [19], crowdsourcing processes [18], crowd ability [26], solution quality [27], crowdsourced data [28] and data processing [29], enterprise crowdsourcing [30], crowdsourcing as a lens for human-computer interaction [31], organizational crowdsourcing intentions [23,32], value creation [33], crowdsourcing multiple tasks [34], crowdsourcing for behavioral science purposes [35], crowdsourcing and algorithms [36], crowdsourcing for innovation [37], crowdsourcing labor law [38], crowdsourcing workers with disabilities [39], health care crowdsourcing [38], crowdsourcing for policy assessment [40], the geography of crowdsourcing participation [62], and cultural factors in crowdsourcing [41]. The universal characteristics of all Crowdsourcing applications listed above are useful and important, since they allow researchers and practitioners alike to understand the stable, relative differences between the forms of Crowdsourcing, while vividly displaying the inherent trade-offs that organizations face when considering Crowdsourcing initiatives.…”

“…First, IT-structures vary along the Collaborative and Episodic dimensions, as well as by the type of IT carrying said IT-structure, for example, web, mobile, SMS, intermediaries, etc. As others have illustrated [47,62] the use of categorical variables can be useful for these purposes.…”

Next Generation Crowdsourcing for Collective Intelligence

“…These techniques, here termed next generation Crowdsourcing, serve to extend Crowdsourcing efforts beyond the heretofore dominant desktop computing paradigm. Employing new configurations of hardware, software, and people, these techniques represent new forms of organization for IT-mediated crowds 2015b, 2015c, 2015d, Prpić & Shukla 20132016). However, it is not known how these new techniques change the processes and outcomes of IT-mediated crowds for Collective Intelligence purposes?…”