Analysing change: measurement and explanation using longitudinal data
A guide to the analysis of quantitative longitudinal data, emphasizing the difficult conceptual problems of measuring change with social science data. Aids in the construction of causal models using 'real' longitudinal data from a broad range of contexts in the social sciences, including data relevant to educationalists, social and developmental psychologists, social administrators, sociologists, and political scientists.
What people are saying - Write a review
We haven't found any reviews in the usual places.
Ways of Measuring Change
Models for the Description and Explanation of Relative Change
Causal Models for Change
7 other sections not shown
analysis Appendix approach assumed assumption autocorrelation biased estimates categorical data categorical variables Causal diagram causal direction causal models cells Chapter cohort conditional model consider constant contingency tables control group correcting for measurement correlated covariances cross-sectional studies defined degrees of freedom dependent variable dichotomous discussed distribution EPVT equation 4.4 error terms error variance example explanatory variables Figure fixed given in Table gives hazard function health at occasion individual change interaction interest latent variable least squares linear log-linear model longitudinal data longitudinal studies marginal probabilities Markov model Markov process means measurement error methods obtained parameters Plewis poor mental health population possible post-test pre-test problem quasi-experiments reasonable regression coefficients relationship relative change relative odds reliability sample scale school type shows social sciences standard errors Suppose test statistic time-homogeneous transition probabilities treatment effect unbiased estimate unconditional usually values zero