Red flags in data: Learning from failed data reuse experiences

Abstract

This study examined the data reusers' failed or unsuccessful experience to understand what constituted reusers' failure. Learning from failed experiences is necessary to understand why the failure occurred and to prevent the failure or convert the failure to success. This study offers an alternative view on data reuse practices and provides insights for facilitating data reuse processes by eliminating core components of failure. From the interviews with 23 quantitative social science data reusers who had failed data reuse experiences, the study findings suggest: (a) ease of reuse, particularly the issue of access and interoperability, is the important initial condition for a successful data reuse experience; (b) understanding data through documentation may be less of an issue, at least for experienced researchers to make their data reuse unsuccessful, although the process can still be challenging; and (c) the major component of failed experience is the lack of support in reusing data, which emphasizes the need to develop a support system for data reusers.

Role of communication in data reuse

“…This finding suggests that effective communications can be the key to support successful data reuse. Existing research has argued that the lack of support in data reuse is one of the critical components of failed reuse (Yoon, 2016a). Communications with peers, data professionals, and relevant research communities are ways to provide support for problem solving, which can even turn unsuccessful data reuse experiences to successful ones.…”

“…Because data reusers build their work on other researchers' findings, the process of data reuse involves these interactions and communications with other relevant parties. Most commonly, previous research has found that data reusers contact data producers (the original investigators of the project in which the data for reuse were created) with questions (Niu, 2009;Yoon, 2016aYoon, , 2017. Sometimes mentors have been involved in locating and understanding data (Kriesberg, Frank, Faniel, & Yakel, 2013;Rolland & Lee, 2013).…”

“…Several studies have pointed out the difficulty of locating data for reuse, and understanding data is often a time-consuming process for data reusers, which sometimes lead to a failure in data reuse (e.g., Yoon, 2016aYoon, , 2017. Birnholtz and Bietz (2003) suggest that data are not simple carriers of meanings, and "converting raw data into scientific and social meaning is an active, context dependent process" (p. 341).…”

Empowering communities with data: Role of data intermediaries for communities' data utilization

“…While data intermediary organizations shared some common challenges relating to data reuse, such as difficulties accessing secondary data, difficulties aggregating data due to interoperability issues, and difficulties working with sensitive data (Lucero et al, 2015;Yoon, 2016), one notable challenge that staff from all intermediaries identified was a lack of data representing local areas or neighborhoods. Although all staff utilized publicly available data for community data work, they agreed that "the public data are fairly general, and usually the scope is fairly large (S41)."…”

“…Although only two intermediary organizations in this study identified data democratization as one of their missions, all intermediaries contributed to democratizing data in practice. Easy access to data when needed at any given moment is an important first step towards data democratization, as previous research has suggested that access to data is the single most important component of successful data reuse (Yoon, 2016). Most intermediaries tried to improve communities' access to data.…”