Estimation of change in successive sampling

Successive sampling efficiently estimates population characteristics over time by leveraging data from previous surveys. While most research focuses on estimating parameters like means, ratio or totals, this study evaluates the estimator proposed by [1] for estimating change in two-occasion successive sampling, building on [13]. Through both simulations and an empirical analysis of undergraduate academic records, it demonstrates that the estimator [1] is nearly unbiased, exhibiting minimal bias around zero. Furthermore, its variance and mean square error consistently decrease with larger sample sizes. These results confirm the estimator’s consistent and efficient measurement of change, making it valuable for monitoring population dynamics over time.