Using Machine Learning Techniques to Examine the Relationship between Money, Personality, and Well-being
This research replicates and extends findings that the degree to which peoples’ personalities match their purchases (psychological-fit) positively predicts well-being. In study one we introduce a novel data-driven measure of psychological-fit, and compare it to the crowd-sourcing technique previously used. Multiple linear regression models were constructed from personality questionnaires for 9,933 individuals matched to 2,474,011 transactions within loyalty-card data. The two methods of calculating psychological-fit both confirm that psychological-fit is more important for predicting well-being than total spend. To extend this research, study two constructs a decision-tree model, a novel approach in the psychological study of well-being. This allowed us to unpack non-linearities and dependencies between age, income, gender, total spend, the Big 5, psychological-fit and well-being. The intrinsic ability of this non-linear approach to partition the dataset reveals new relationships between predictors overlooked by traditional models and provides fresh insights into the most important variables influencing well-being.
Bags of Impulsiveness: Psychological and Demographic Predictors of Plastic Bag Consumption in Big Data
January 3, 2019
Responsible Innovation: Governance of Big Data in the ‘Institutional Void’
January 10, 2016
Building a Recommender System to Improve Employment Outcomes in Portugal