Fitting a Multinomial Logistic Regression (MLR) Model to Need Assessment Survey on E-learning in Kenya
Lameck Ondieki Agasa *
Research and Extension Office, Kisii University, P.O.Box 408-402000, Kenya.
Anakalo Shitandi
Research and Extension Office, Kisii University, P.O.Box 408-402000, Kenya.
Wycliffe Cheruyout
Kenya Revenue Authority, P.O.Box 8240-00100, Kenya.
Wycliff Ombasa
Kenya Revenue Authority, P.O.Box 8240-00100, Kenya.
Onsongo Wycliff Nyaundi
Kenya Revenue Authority, P.O.Box 8240-00100, Kenya.
*Author to whom correspondence should be addressed.
Abstract
Information Communication Technology (ICT) advances have reduced the world to a small village. The use of technology to meet societal need has increased in developing countries. It’s from this perspective that technology should find its way to developing countries classroom through e-learning to replace the traditional methods of teaching .This research aimed at addressing the needs an e-learning class should have by fitting a model. The study was based on the Taskforce Report on Implementation of Electronic-Learning at Kenya, Kisii University, Faculty of Education and Human Resource Development (FEHRD) 2013. Both primary and secondary data was obtained from Kisii University and was entered in SPSS version 22.0 for analysis. A Multinomial Logistic Model (MLM) showed a relationship between the level of education, level of preparedness, readiness and computer literacy of students. The adequacy of the model was tested using Deviance method which proved the model to be adequate. It was established that certificate and diploma holders need to be prepared with the necessary computer skills to enable them undertake an e-learning class. E-learning can be rolled out to degree holders and masters holders as it was established that they are prepared, ready and have the computer skills to undertake an e-learning classes.
Keywords: E-learning, multinomial logistic regression, readiness, computer literacy, preparedness