LONGITUDINAL RESEARCH METHOD

Longitudinal Research Method

Longitudinal Research Method

Introduction

Longitudinal' is a broad term. It can be defined as research in which: (1) data are collected for two or more distinct periods (implying the notion of repeated measurements);(2) the subjects or cases analysed are the same, or at least comparable, from one period to the next; and (3) the analysis involves some comparison of data between or among periods (Menard, 1991: 4). There are a number of different designs for the construction of longitudinal evidence: repeated cross-sectional studies; prospective studies, such as household panel surveys or cohort panels; and retrospective studies, such as life and work histories and oral histories. In the social sciences, cross-sectional observations are the form of data most commonly used for assessing the determinants of behavior (Schulsinger 2001, p.12-19).

Pros and Cons of Longitudinal Research

Not only a thousand gifted children were followed over their life course but gifted adults also were studied backward to the period of childhood. Terman and his team carried on the study for over thirty years until his death in 1956. The five significant volumes published as a result of this monumental work bear evidence to what longitudinal research can do to the advancement of our knowledge. More elaborate discussions on advantages and disadvantages of longitudinal research had to wait for four more decades. Two important studies are worth citing here. And the other is the Report of the National Foundation for Educational Research in UK. This Report tried to identify the distinctive contribution of longitudinal studies to the advancement of the social sciences. After a careful and scrupulous evaluation of the advantages and disadvantages of longitudinal studies, both of them were not in favor of continuing longitudinal studies for reasons that are valid even today.

Complicating Factors in Longitudinal Studies

Before launching into any longitudinal research, one should be aware of some important factors that can complicate the analysis of longitudinal data. Most social science researchers rely on secondary longitudinal data, and unless these complicating are taken care of, they will seriously undermine the inferences that we make from statistical results, especially for policy purposes (Menard 2002, p.20-28).

Socio-psycho-dynamism

The major virtue of any longitudinal information is that it is inherently socio psycho- dynamic. Longitudinal studies are meant to uncover that dynamism. This essentially implies that for analysing longitudinal information, we need to reorient ourselves to using dynamic models in our studies and be ready to abandon static models of all types, however much we cherish them. This particularly applies to such techniques like multiple regression and path analysis. As Rogosa (1995) has shown, a regression analysis that simply considers “scores” from different waves of a longitudinal survey as “covariates” in the model is flawed because the estimated parameters depend only on the times at which the observations were taken and have nothing to do with the “scores” themselves.

So, the ever-increasing availability of longitudinal information offers us the best chance to examine the dynamic indeterminism that characterizes human behaviour. With dynamic observations through longitudinal surveys, we need then to ...