Prediction of Socio Emotional Wealth in Ecuadorian Family Businesses Using Artificial Intelligence Techniques

Résumé fourni par la source

The purpose of the research was to provide a prediction of the level of socio -emotional wealth of Ecuadorian family businesses, using artificial intelligence instruments and certain family, organizational and technological variables. It was carried out wit hin the framework of a quantitative research, with a non-experimental, cross-sectional, explanatory and predictive research design. The sample was made up of 220 family businesses from the commerce, services, manufacturing, agro-industrial and construction sectors. A structured questionnaire with a Likert scale was used, which was validated by expert judgment and showed adequate internal consistency, using Cronbach's alpha. The data were processed with descriptive statistics, correlations and logistic regression machine learning models, decision tree, Support Vector Machine, artificial neural network, Gradient Boosting and Random Forest. In general, the results indicate that there is a high level of socio-emotional richness, especially in emotional attachment, family identification and family control. The model that had the best predictive performance was the Random Forest model with an accuracy of 0.89 and an AUC of 0.92. The variables that had the most weight were family participation in management, succession planning, family identification and administrative professionalization. It can be deduced that artificial intelligence allows accurate prediction of socio-emotional wealth and provides a useful tool to guide strategic decisions in Ecuadorian family businesses.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.