Feeling Around the Data: An Exploration of Paradata as Indicators of OER Utility
Open educational resources (OER) are heralded in a global movement toward high quality, affordable, accessible, and personalized education. However, stakeholders have expressed concern about scaling OER use due to a lack of means to ensure fit between learner, resource, and task. A possible way determine fit is through examination of transactional data, i.e., paradata, such as reviews, ratings, views, downloads, and “favoriting.” We examined National Science Digital Library (NSDL) paradata, the largest extant accessible corpus, for the extent to which resource fit can be determined from user- and system- generated data. We conducted sentiment analyses of user reviews and correlations between the sentiment scores and data elements. While some relationships between paradata elements were discernible notions of resource fit, reliable research in this area depends on access to larger and more robust paradata sets. We conclude with observed data trends and further research directions.