Introduction

As mentioned earlier, in this book I use two longitudinal datasets and event-history methods in order to capture and explain changes across cohorts in Italian and British women's lifetime employment patterns. The analysis is conducted within an institutional rational-action framework, which recognises different forms of rationality, the role and heterogeneity of both preferences and constraints, and their complex interrelations. This chapter provides a description of the methods, datasets, variables and techniques used for the empirical analyses described and discussed in the rest of the book. More precisely, it starts, in the next section, by illustrating the nature and advantages of a lifecourse perspective and of longitudinal data compared to cross-sectional data. The section also briefly describes the different longitudinal designs in order to show when and why an event-oriented observation design is more suitable. The following section focuses on the specific method used in my empirical analyses, namely discrete-time logit models. It then addresses the problems of sample selection and unobserved heterogeneity and discusses how they are dealt with in my models. The next section then describes the two datasets, the sample of women used and the variables chosen. The final section links the previous two chapters with the current chapter and the following two by formulating hypotheses on what has changed across cohorts in the types and correlates of women's work histories in Italy and Britain.

The role of time: longitudinal versus cross-sectional analyses

The lifecourse perspective

At a very intuitive level, the concept of change comprises a temporal dimension. Thus, any data that track the same subjects over time and measure the relevant micro and macro variables at different time points should ‘by definition’ afford better understanding of social change than cross-sectional data. As has been well conceptualised by Elder (1975, 1985), a pioneer of lifecourse theory and research, individual biographies and social phenomena are time dependent in different ways. First, they develop across a person's lifetime, that is, through the chronological process of growing up and ageing. Second, they take shape within a specific generation and a specific historical period, with its cultural, economic and institutional arrangements. Third, they are influenced by normative definitions of the ‘proper’ timing, sequence and interrelating of the various experiences that mark the lifecourse.