Mining Teachers' MOOC TPD Enrollment Analysis based on their Engagement with Student-Centered Teaching Strategies

Analytically speaking, demographics information is part of the enrollment details to be examined in any research. This study explores the most important predictive variables for the engagement of schoolteachers in the MOOC TPD environment. Six demographic variables and two corresponding student-centered teaching variables from the pre-entry survey of 484 schoolteachers of one course titled ‘Becoming a 21st Century Teacher’ were used. The study identifies the best algorithm based on the accuracy of the proposed models from the variables described above. Experience and age range show the highest correlations with the target variable (active strategies course). Classification algorithms were the best for the model formulation, with a Decision Tree having the highest accuracy followed by a Random Forest. However, it is evident that in the formulation of predictive models, the demographic variables' influence depends on their importance to the targeted variable.