Predicting motivation and intention to participate and recommend Food & Drink groups on Facebook via eWOM

Abstract

Objective: Using an ANN-based analysis, this research aims to predict motivation and intention to participate and recommend Food & Drink groups on Facebook. Method: Data were collected from 345 individuals who participated in at least one Food & Drink related group. For data analysis, the non-linear method of ANN was used to predict occurrences within the same sample. Using this prediction method to test the theoretical model proposed, using scales adapted for the study, is relevant to the research. Originality/Relevance: Given the importance of the eWOM theme in social networks, being one of the prominent themes in the area, this study evolves the theme and contributes to expanding knowledge in non-linear methods.  Results: Based on model 1 reviews, ‘pleasure for helping’ (44.8%) is the most important predictor of ‘eWOM motivation’. Based on the analysis of model 2, the ‘sense of belonging’ (42.7%) is the most important for the intention to recommend via eWOM. In addition, model 1 and model 2 presented fair values ​​and observations for their validation. Theoretical/methodological contributions: A theoretical model was fitted using scales adapted for the study. With that, a survey was carried out and based on the results obtained in the sample, an approach of the ANN method was used.  Social/Management Contributions: This study helps participants, administrators, moderators, and others interested in Facebook Food and Drink groups understand how they work and take advantage of the information exchanged to design strategies that meet the needs of the community.