A Model-based Clustering Approach for Analyzing Energy-related Financial Literacy and Its Determinants

Abstract

Recent research highlights the role of consumer's energy-related financial literacy in adoption of energy efficient household appliances in order to reduce the energy-efficiency gap within the household sector. The computation of an indicator for such a literacy measure has followed a somewhat less refined approach though. This paper demonstrates the use of a model-based clustering strategy in order to differentiate the population based on the level of energy-related financial literacy. Using a Swiss data with 6, 722 respondents, we are able to identify three latent groups that represent low, mid and high levels of literacy. We use this new measure within an ordered logit setting with the goal of explaining the determinants of the level of energy-related financial literacy and compare empirical results using classical indicators and approaches. The empirical findings suggest a significant gender-gap among the Swiss population, i.e. females, even those with university education, are less likely to possess a high level of energy-related financial literacy. Individuals who display strong concern for free-riding on their own energy reduction behavior, are also found to have higher odds of belonging to the low literacy group. The results show that it is possible to identify latent classes that have a general and intuitive meaning and provides support to the model-based clustering approach as a sophisticated alternative. This could be a useful approach when empirical researchers are interested in (attribute-based) latent groups of consumers. The identification of latent classes also provides a possibility to target consumers belonging to these classes with specific policy measures in order to increase their level of literacy.

Energy-related financial literacy and bounded rationality in appliance replacement attitudes: evidence from Nepal

“…One of the most important findings of this paper, which has been confirmed in previous studies from developed countries as well, is that on average, women fare much worse than men in these surveys, with scores of energy knowledge, skills, and energy-related financial literacy being lower for them than for male respondents (Lusardi and Mitchell, 2014;Blasch et al, 2018;Kumar, 2019). The variable is found to be significant at the 1% level in each model of Table 8.…”

“…Kumar (2019) also look at the determinants of energy-related financial literacy (represented as latent clusters with low, mid and high level of literacy instead of a numeric score) for a large sample of Swiss respondents.…”

Global Patterns of Contemporary Welfare States

“…Cluster analysis is an unsupervised learning method used to examine homogenous groups of observations within a multivariate dataset (García-Escudero et al, 2010; Kumar, 2019). In unsupervised learning, hierarchical clustering, partitioning methods, and model-based clustering are the most popular methods.…”

“…In unsupervised learning, hierarchical clustering, partitioning methods, and model-based clustering are the most popular methods. In this study, I used model-based clustering (or Gaussian Mixture Model), a formal and sophisticated method that relies entirely on statistical models and creates the prospects to make formal inferences (Kumar, 2019; Fraley and Raftery, 2002). Recently, model-based cluster analysis has advanced considerably in terms of methods, software, and interpretation of the output (Fraley and Raftery, 2007).…”

Knowledge economy classification in African countries: A model-based clustering approach

“…4. Recent applications of mixture models in different contexts can be found in Csereklyei et al (2017), Sulkowski and White (2016), Alfo et al (2008), Seo and Thorson (2016), Kumar (2019), and Clements (2020). 5.…”