Beyond trust and reliability

While previous CSCW research on data sharing and reuse has focused on how researchers assess the trust and reliability of the data of others, we know little about scientists' data use practices after that decision has been taken. This qualitative study of post-doctoral researchers' use of preexisting datasets investigates the practices of cancer-epidemiology post-docs working to understand their "Small Data" datasets. We report the ongoing and iterative nature of information seeking inherent in using unfamiliar data and the time-consuming and highly-collaborative process post-docs used to understand aspects of the dataset important to their scientific questions. Understanding data use practices can help inform the design of both Small Data projects and large cyberinfrastructure projects where multi-source data are collected and combined.