Understanding and predicting the motivators of mobile music acceptance – A multi-stage MRA-artificial neural network approach

Abstract

The adoption level of digital music is still at its formative stage although the adoption renders advantageous to consumers. Therefore, the study develops a model to predict on the motivation leading to consumer’s intention to adopt mobile music services by extending Perceived Cost (PC), Perceived Credibility (PCr), Social Influence (SI), and Personal Innovativeness (INNO) with Technology Acceptance Model (TAM). 160 Respondents were tested using a multi-stage Multiple Regression Analysis (MRA) and Artificial Neural Network (ANN) approach. A non-linear non-compensatory Multi Layer Perceptron (MLP) ANN with feed-forward back-propagation algorithm and ten cross-validation neural networks was deployed in order to capture the motivators of mobile music adoption. All predictor variables were found to have relevance to the output neuron based on the non-zero synaptic weights connected to the hidden neurons. The RMSE values indicated that the ANN models were able to predict the motivators with very high accuracy. The ANN models have out-performed the MRA models as they are able to capture the non-linear relationships between the predictor and criterion variables. While the study found that TAM is a significant predictor, the insignificance linear relationships of PCr and INNO requires further investigation. The music industry can use the findings from this study beneficially to the development of mobile music adoption.