A Multivariate Clustering Study of Consumer Protection Conditions in European Union

Abstract

Consumer protection is an important part of maintaining market order and enhancing consumer trust. Based on the data of the EU Consumer Conditions Survey in 2024, this study compares and analyzes the differences in consumer protection models of 29 European countries. The study uses principal component analysis (PCA) to build consumer trust indicators, and uses the Ward hierarchical clustering method to classify countries. The number of clusters is determined by the silhouette coefficient and the aggregation schedule, and then the cluster results are verified by linear discrimination analysis (LDA). The research results show that the first two main components can explain most of the differences in consumer trust. According to it, 29 countries can be divided into three categories: volatile trust type, high trust type and medium trust type. Discriminative analysis further shows that unfair business experience, financial pressure and consumer rights knowledge are the main factors that distinguish different types of countries. The study also finds that the level of consumer trust is not only related to the institutional environment, but also influenced by factors such as consumer risk, financial situation and consumers' own ability. In general, improving the consumer protection system, creating a fairer and more transparent market environment, and raising consumer awareness of rights are conducive to improving the level of consumer trust.