November 7, 2014

I am a Research Associate at Teachers College (TC), Columbia University in NYC, where I work closely with members of the new graduate degree program in Learning Analytics (the first in the country).  I also maintain an affiliation with Worcester Polytechnic Institute (WPI), and I am the Director of BROMP USA, the organization which oversees training and certification of coders who wish to use the BROMP method to study student engagement. I am primarily interested in how educationally relevant emotions emerge and influence learning, and I am especially interested in how those emotions vary from one population to the next.

My PhD (Sociolinguistics, Michigan State University under Dennis Preston) provided me with extensive training in a variety of fieldwork techniques which are able to capture both micro and macro-level variation. In my current role, I have used this training to refine and adapt BROMP for new uses. In its current instantiation, BROMP is primarily used to collect observations of student engagement during the use of educational software.  These observations are then used to develop models of engaged/disengaged behaviors and affective states that can run in real-time within the educational software using techniques developed by Ryan Baker and Ma. Mercedes T. Rodrigo. More recently, we have adapted BROMP for use in non-technological classrooms, and efforts are underway to expand its internationally. We now have over 150 BROMP-certified observers in the US, the Philippines, India, and the UK.

Since becoming the director of BROMP USA, I have used this method to explore student engagement, motivation, and learning in fine-grained detail in over a dozen published research studies. We have shown, for example, that confusion and frustration have similar effects on learning, and that those effects are not unidimensional (Liu, Pataranutaporn, Ocumpaugh, & Baker, 2013). We have also explored intercultural differences within the US, demonstrating that  affective states trigger different patterns of behavior among urban, suburban, and rural students using ASSISTments software, which provides automated middle school mathematics instruction to over 50,000 students  (Ocumpaugh et al., 2014).

I have completed projects with several systems already, and I am currently working on several other funded research projects. So far, these projects have included efforts to model and study student engagement within:

  • Newton’s Playground (funded by the Bill & Melinda Gates Foundation)
  • vMedic (funded by Army Research Laboratories)
  • ASSISTments (funded by the Bill & Melinda Gates Foundation and by several different grants from the National Science Foundation)
  • EcoMUVE (funded by the Bill & Melinda Gates Foundation)
  • Reasoning Mind (funded by the Bill & Melinda Gates Foundation)

I am also working on a project to study the relationship between design patterns identified in kindergarten classroom decorations and student engagement (funded by the Institute of Education Sciences at the USDOE). More research is forthcoming as we continue to explore data collected within these established partnerships, and we are working to expand our research to Latin America and India. Future research will also work to show converging validity for this method, since gold standards are not available for many constructs associated with engagement.