Modeling the Role of Self-Directed Learning and Media Literacy Factors in Students’ Cyberloafing with Artificial Neural Network Approach

This research aimed to model the role of self-directed learning skills and media literacy in cyberloafing. This study was quantitative in terms of main strategy and correlational research in terms of analytical technique. The participants included graduate students and the sample were selected through available random sampling. The sample size was 620 according to the Krejcie- Moran model with considering the error of α=0.05. For data collection, cyerloafing questioner of Bella et. al. (2006) were applied with α=0.89 reliability, a research-made self-directed questioner with α=0.95 reliability and media literacy questioner (Palsafi, 2014) with α=0.86 reliability. The content validity of instruments was confirmed by the opinions of ten e-learning experts. The data were analyzed by artificial neural network approach sing Multilayer Perceptron (MPL) method. The results revealed that modeling of self-directed learning and media literacy skills in students’ cyberloafing has an input layer with ten nodes and a hidden layer with four nodes and the artificial neural network is well able to predict the jumps and students cyberloafing process and variables. All coefficients of influence of the hidden layer on the output layer in the neural network obtained negatively and, therefore the higher students self-directed and media literacy skills, the less their cyberloafing behaviors. It could be concluded that the level of students’ self-directed and media literacy skills can predict the level of their cyberloafing behaviors