National Network of Libraries of Medicine
English Arabic Chinese (Simplified) French Hindi Japanese Korean Persian Portuguese Russian Spanish

Understanding Rates of Attrition in Biomedical Data Challenges: A Study of Failure

Biomedical data challenges are interdisciplinary live or virtual competitive events that require participants to analyze big biomedical data and develop prototype solutions to real-world problems that accelerate innovation of medical applications, improve healthcare technology design, and help streamline healthcare business models (Wang, et al., 2018). Participation in biomedical data challenges helps current and future health professionals identify, consider, and develop solutions to challenges associated with how big biomedical data influences patient well-being, and data challenges are particularly useful for illustrating issues associated with the underuse of uniform data standards and with navigating siloed data. However, data challenges are often criticized for promoting a culture unwelcoming to women and underrepresented minorities by fostering an overly competitive environment; for using a format that potential participants might find unappealing--i.e. staying up for days on end; and for lacking in diversity, which leads to feelings of otherness (Decker, et al., 2015). In 2016, Purdue University Libraries launched the Biomedical Big Data Hacking for Civic Health Awareness Project to help alleviate these issues by offering inclusive biomedical data challenges that leverage current interest in data competitions and extend that appeal to members of groups commonly underrepresented in data challenges. The project has succeeded in its goal. Across its 2.5 years of offering data challenges, 28% of total participants have been either health professionals or students majoring in a health sciences discipline, 38% of attendees have been women, and 46% have been members of a racial minority group. However, the project has seen a high rate of attrition amongst the number of participants who register for the challenges and the number of participants who submit final projects. Similar rates of attrition have been observed amongst participants in other data challenges, for example, in business case competitions where participants are tasked with analyzing health-related data to develop a marketing plan. As the Biomedical Big Data Hacking for Civic Health Awareness Project continues, we seek to understand the success factors that motivate students to complete data challenges and to understand what factors lead to failure, in hopes of improving attrition rates.
Identify any specific population(s) this project will serve: 
African Americans or Black
Alaska Natives
American Indian
Latino or Hispanic
Native Hawaiians
Pacific Islanders
Identify roles of participants this project will serve: 
Data resource or tool developer
Data scientist
Health care provider
Library or information professional
Public health professional
Student, college & post-grad
Project Lead:
Bethany McGowan
Funding Source: 
Project Funding:
Federal Fiscal Year: 
Funding Amount: 
Funding Period:
May 1, 2019 to Apr 30, 2020
Project Status:
In Progress