This project is motivated by the idea that data can and should be maintained and controlled by experiment participants. An initial focus of the project was to develop an “app store” for such research.
Joshua Pritkin and Yang Wang developed software through which subjects can sign up for paid experiments using their smartphones.
Using this software, Maintained Individual Data: Distributed Likelihood Estimation, or MIDDLE, will provide a way for researchers to collect and analyze the data they need, while ensuring that subjects’ personal data remains on their own smartphones and is never shared with researchers. This is possible because, using the app, researchers will send candidate models to subjects’ smartphones, which will evaluate how well the model fits their own data. Only the model fit metric will be communicated back to researchers. In this way, statistical analysis can be conducted without revealing any personal data.
MIDDLE will provide automatic management of opt-in and opt-out consent; lower research costs; and deliver faster results. Robust privacy protection will reassure participants and permit research on topics that are sensitive or involve data that must legally be kept private.
Joshua Pritkin is a PhD student in the Department of Psychology. He has several years of experience working on the project OpenMx, which has required him to develop diverse skills as a researcher and which provides much of the programming functionality that will support the MIDDLE project.
Yang Wang is a PhD student in the Department of Systems and Information Engineering. His interests lie in the area of statistical analysis and software development, particularly in how to use and analyze big data.
