Bootstrapping in educational studies

Most of the outcomes of real life events follow normal distribution. In educational research, researchers often ensure a distribution closely associated to the normal distribution in order to proceed on their analyses. Testing of the normality is normally done before any inferential analysis is carried out. However, assumptions about normality can be exempted if bootstrapped procedures are used. The computation in bootstrap is complicated. The advanced analysis through powerful computer facility enables the possible implementation of the computation. Hence, the application of computer contributes to the success of implementing bootstrap procedures. In this study, the focus is to estimate parameter and inference required about the mean of some observations. Generally, this study provides some guidelines in using bootstrap procedures in educational research. Specifically, the purpose of this study is to demonstrate the construction of the procedures of bootstrapping for the common usage in educational research in finding confidence intervals.