1 Increasing Academic Success in Undergraduate Engineering Education using Learning Analytics: A Design-Based Research Project Andrew E. Krumm 1 , R. Joseph Waddington 1 , Steven Lonn 2 , and Stephanie D. Teasley 3 1 University of Michigan School of Education 2 University of Michigan USE Lab 3 University of Michigan School of Information This paper describes the first iteration of a design-based research project that developed an early warning system (EWS) for an undergraduate engineering mentoring program. Using near real-time data from a university’s learning management system, we provided academic mentors with timely and targeted data on students’ developing academic progress. Over two design phases, we developed an EWS and examined how mentors used the EWS in their support activities. Findings from this iteration of the project point to the importance of locating analytics-based interventions within and across multiple activity systems that link mentors’ interactions with an EWS and their interventions with students. Introduction Colleges and universities are increasingly aggregating and analyzing once disparate sources of data, such as a student’s admissions records, academic history, and use of campus information technologies, all under the rubric of “learning analytics” (Campbell, DeBlois, & Oblinger, 2007, Fritz, 2011; Goldstein & Katz, 2005). Learning analytics (LA) is a developing research area and a topic of increased conversation; yet, most studies are often limited to intriguing possibilities and frequently lack assessments for specific interventions paired with LA tools (Parry, 2011; Rampell, 2008). In this paper, we describe the first iteration of a design-based research project that developed an early warning system (EWS) for an undergraduate engineering mentoring program. The purpose of this iteration was to identify the necessary infrastructure for building an EWS and to understand the factors affecting how the EWS was used. The EWS described in this paper represents an application of LA that is gaining popularity across colleges and universities—the near real-time aggregation and analysis of