This paper outlines the design features, data collection methods and analytic strategies of the International Tobacco Control (ITC) Four Country Survey, a prospective study of more than 2000 longitudinal respondents per country with yearly replenishments. This survey possesses unique features that sets it apart among surveys on tobacco use and cessation. One of these features is the use of theory-driven conceptual models. In this paper, however, the focus is on the two key statistical features of the survey: longitudinal and "quasi-experimental" designs. Although it is often possible to address the same scientific questions with a cross-sectional or a longitudinal study, the latter has the major advantage of being able to distinguish changes over time within individuals from differences among people at baseline (that is, differences between age and cohort effects). Furthermore, quasi-experiments, where countries not implementing a given new tobacco control policy act as the control group to which the country implementing such a policy will be compared, provide much stronger evidence than observational studies on the effects of national-level tobacco control policies. In summary, application of rigorous research methods enables this survey to be a rich data resource, not only to evaluate policies, but also to gain new insights into the natural history of smoking cessation, through longitudinal analyses of smoker behaviour.