Social scientists' satisfaction with data reuse

Abstract

Much of the recent research on digital data repositories has focused on assessing either the trustworthiness of the repository or quantifying the frequency of data reuse. Satisfaction with the data reuse experience, however, has not been widely studied. Drawing from the information systems and information science literature, we developed a model to examine the relationship between data quality and data reusers' satisfaction. Based on a survey of 1,480 journal article authors who cited Inter-University Consortium for Political and Social Research (ICPSR) data in published papers from 2008-2012, we found several data quality attributescompleteness, accessibility, ease of operation, and credibility-had significant positive associations with data reusers' satisfaction. There was also a significant positive relationship between documentation quality and data reusers' satisfaction.

Social scientists' data reuse behaviors: Exploring the roles of attitudinal beliefs, attitudes, norms, and data repositories

“…Several recent studies have investigated data reuse practices and behaviors in the social sciences (Daniels, Faniel, Fear, & Yakel, 2012;Faniel, Kriesberg, & Yakel, 2012;Faniel, Kriesberg, & Yakel, 2016;Niu, 2009;Yoon, 2014bYoon, , 2016Yoon, , 2017 as part of a broader drive to understand data practices within the social sciences. While these studies have captured some of the contexts and characteristics of social science data reuse by focusing on specific aspects of data reuse practices, which may overlap with other disciplinary contexts, fewer studies have used theoretical approaches or models to explain social scientists' data reuse behaviors.…”

“…Easy access to data was one of the most influential factors in determining social scientists' satisfaction with data reuse (Faniel, Kriesberg, & Yakel, 2016). Data repositories have a long history in social science, and they are known to support easy access to and reuse of available data through value-adding activities (Daniels, Faniel, Fear, & Yakel, 2012;Yoon, 2014a).…”

“…"Perceived effort", rather than "perceived ease of use," emphasizes the effort expectancy involved in data reuse as compared to the effortlessness or ease of data reuse that a "perceived ease of use" construct would imply (Venkatesh, Morris, Davis, & Davis, 2003). Data reuse literature suggests that when data reusers spend significant time actually using existing data (Rolland & Lee, 2013;Zimmerman, 2008), and less time acquiring and processing data, their level of satisfaction with data reuse improves (Faniel et al, 2016). Thus, it is likely that a greater perceived effort to reuse data would negatively influence social scientists' attitudes toward data reuse and their intention to reuse other scientists' data.…”

“…Compared to prior data reuse studies (Faniel et al, 2016;Yoon, 2014b), this research considered an inclusive range of social science researchers in terms of their data reuse experience (e.g., experienced data reusers, potential data reusers, and even those who never want to reuse data). The survey is also designed to incorporate the wide variety of social science researchers who may have different perceptions (e.g., subjective norm of data reuse) and intentions (e.g.…”

More Data, Less Process? The Applicability of MPLP to Research Data

“…Based on results from a study of users' perceptions of data quality, Wang and Strong (1996) identified four dimensions of data quality: 1) intrinsic, referring to the accuracy and credibility of the data; 2) contextual, or the relevancy of the data to the user's goals; 3) representational, relating to the ability to interpret and use the data; and 4) accessibility, or the ability to obtain the data. A more recent study conducted by Faniel, Kriesberg, and Yakel (2015) to determine the factors that elicit social science researchers' satisfaction with data reuse found that users associate data quality with attributes that align with Wang and Strong's quality dimensions. They include completeness (contextual), accessibility (accessibility), ease of operation (representational), and credibility (intrinsic) of the data.…”