In the summer of 2016 I taught ENG 099: an English as a Second Language (ESL) Massive Open Online Course (MOOC) using Open Educational Resources for the third time. The first two times under 30 students registered and I taught live video lessons, but last year over 100 students from the Ivory Coast signed up. As a result I changed my approach to more of a self-serve model using Wikiversity as the course home. Each week I published Listening, Reading, Grammar, and/or Mechanics assignments on the course page with Writing and/or speaking homework to be submitted directly to me via e-mail. By the end of the six week course, seven students completed all the assignments and received certificates of completion. All three sequences of the course used only OER for the learning materials.
Following the course I collaboarted with another student at the University of Illinois at Chicago (UIC) to analyze data from the course to understand what were the statistically signficant predictors of student success. Success here was measured as whether or not students finished all the assignments to complete the course. Using a multilevel Bayes Binary Probit Model we found that eagerness (defined as how quick they were to register), whether or not each week's lesson included specific vocabulary words to study, whether or not each week had optional work and the learning period (how much time passed since registartion opened [i.e., how long students had to study]) were the signifcant predictors of student success.
In my presentation I will detail how I created and administered the course as well as the results of the statistical analysis. It will be useful for:
* English as a Second Language teachers looking for examples of how to use OER in their classes,
* Educators hoping to run a MOOC on their own,
* Educators already running MOOCs wanting to learn what determines successful student completion, and
* Educators hoping to reach students in a developing country.
