Life events: A classification study and an exploration of consumer behaviour by cluster

A life event is a major incident that changes the status or circumstances of a recipient, such as marriage, divorce, or the birth of a child. Life events tend to cause changes in consumer behaviour and provide companies with opportunities to respond to those changes. Much of the research into consumer behaviour is based on the cognitive perspective, but an important stream of research looks at situational influences. Life events are a form of situational influence that can impact consumer behaviour. This thesis addresses three key research questions, which are outlined below:

RQ1. What is known about life events with respect to consumer behaviour, and what remains unexplored and needs to be investigated further?

RQ2. What are the different characteristics of life events, and how could these characteristics be used to classify life events, identifying their commonalities and differences?

RQ3. What is the impact of different types of life events on consumer behaviour?

Consistent with these research questions, this thesis aims to contribute to the life events literature by first developing an in-depth and systematic understanding of the extant literature on life events with respect to consumer behaviour, which is a highly fragmented interdisciplinary area. Second, it aims to identify different characteristics of life events and use them to classify them where life events that share commonalities will be clustered and their differences and similarities highlighted. Third, it aims to determine the cluster-wise impact of life events on consumer behaviour by highlighting the effect of varying life events on consumer behaviour using the well-established Life Course Model (LCM).

Research question 1 was addressed through a Systematic Literature Review (SLR). After identifying the most suitable articles using the PRISMA framework, analysis was performed on the titles and abstracts of the articles using NVivo 14. The codes generated v through the auto-coding wizard were manually reworked and recoded to generate a meaningful platform for the identification of gaps in the literature. The clustering was then performed on the basis of coding similarities using Jaccard's coefficients. The SLR (chapter 2) highlighted six gaps in the literature, and the major gap identified was then addressed by the rest of the thesis (chapters 3-6).

Research question 2 was addressed by first refining and validating a list of life events through an expert panel study, where scholars in the field were surveyed about their perceptions of the list compiled by using two commonly used lists in the literature and adding life events that suit the modern lifestyles of the consumers. A list of thirty-six life events was validated and then used in the main study, where the classification was done through a systematic procedure. Clustering the life events was performed after developing the updated list of life events and using the important themes and characteristics of life events identified from the literature. A national survey in the UK was conducted where consumers were asked about their perceptions of the life events, they had experienced. The survey was informed by characteristics of life events found during the review, consumer traits and the LCM. The survey was designed using the Qualtrics platform and administered via a Toluna panel. The analysis was performed by presenting the data's demographic attributes and assessing the measures' reliability and validity prior to the cluster analysis. A hierarchical cluster analysis was then performed to classify life events, and MDS was used to generate geometric representation. This was followed by MANOVA, where attributes of each cluster were further studied by comparing means.

Research question 3 was then addressed by analysing the LCM for each cluster and illustrating its differences and similarities. The LCM predicts that life events affect consumer behaviours when mediated by variables that are based on three major perspectives, including normative, stress and human capital. The outcome variable of LCM is ‘change in consumption vi activities’. ‘Change in consumption activities’ was measured through two broad categories of consumption activities (leisure and necessities). This analysis was performed using mediation analysis (Model 4) by installing the PROCESS macro in IBM SPSS Statistics 27.

The findings of the analysis suggested that life events can be classified into four groups including ‘Riding the Swell: Unobtrusive, Low Impact Life Events’ (cluster 1), ‘Choppy Waters: Major Transitions and Midlife Melodramas’ (cluster 2), ‘Calmer Waters: Mixed Emotions and Anticipated Transitions’ (cluster 3), and ‘Rough seas: Personal and Financial Crises, Emotional Turmoil's (cluster 4). While determining the cluster-wise impact of life events using LCM, the finding revealed a consistent direct effect of the first two clusters on both forms of purchase. The varying direct effect of cluster 3 and cluster 4 on two forms of purchase can be used to foresee a change in consumer behaviour.

This thesis made several theoretical and methodological contributions. After highlighting several gaps through SLR that could advance research in the area, a major theoretical contribution is the classification of life according to its similarities and differences. Four clusters of distinctive groups were identified and described in terms of their key characteristics, LCM, and consumer traits. The findings contribute to the literature by explaining the varying effects of different life event groups on two major categories of purchases. In the future, researchers can use these findings to help design their studies and find out if this behaviour changes when other factors are included in the model. The life event characteristics used in this study were not empirically measured in past life event literature. The list of life events developed through expert judgment is an updated version for researchers who are concerned about life events that are particularly important to consumer behaviour. The prevalence of each life event in the data collection provides researchers with an estimate of the prevalence of each life event in the UK population. vii

The findings further have implications for policy and practice. The updated list of life events compiles the life events important for consumer purchase, and life events that are not crucial for consumer behaviour were sifted. The list could, therefore, be treated as a compilation of life events that are specifically useful for decision-makers within a business's marketing department. The classification performed in this study brings about an empirical understanding of the ways some life events share certain attributes. Practitioners can use these clusters to target customers. The cluster-wise effect of life events was determined on change in the purchase of leisure and necessities by using LCM as a framework. Managers can use the findings to understand customers better and target them based on the findings