International Journal of Education and Development using ICT

International Journal of Education and Development using ICT > Vol. 14, No. 1 (2018)

Building and evaluating logistic regression models for explaining the choice to adopt MOOCs in India

Sangeeta Trehan, Indian Institute of Foreign Trade, India
Rakesh Mohan Joshi, Indian Institute of Foreign Trade, India

Abstract
Logistic regression is a popular tool used to build and evaluate binary choice models. It has been applied in a variety of situations and contexts involving dichotomous choice. In the current paper, we apply it to explain and predict the individual choice of adopting online learning through a Massive Online Open Course (MOOC), a specific artefact in the domain of online, open learning. The MOOC holds promise for a developing country like India to scale quality education to keep up with an overwhelming demand of its large population. Hence the importance of the choice to enter MOOCs in the Indian context cannot be overemphasized. Factors possibly driving MOOC adoption were selected after an extensive literature study and a logistic regression model was applied to analyze their influence using data from a primary survey. A variety of different subsets of possible explanatory variables under consideration were experimented with. Popular information criteria like Akaike Information Criterion (AIC) and Schwartz Criterion (SC) were used to choose the most useful models. The predictive models were evaluated and validated using various tools and processes like the train-test process, the ‘ROC curve’ and ten-fold cross-validation and the ‘best’ model was identified and discussed.

Full Text: PDF

Research
Support Tool