Selection of Statistical Test in Educational Research and Assessment

Abstract

Statistical analysis has vital and significant roles in educational research and assessment. In today’s era of science and technology research should be rational with logical reasoning so, with flow of time educational assessment has gradually shifted from traditional assessment towards alternative assessment where more attention has been paid to the core research methodology, finding and its presentation. This article will try to update the reader with the basic research tools that are utilized while conducting various educational research and assessment in Nepal. For choosing right statistical tests on the basis of study design (univariate, bivariate, multivariate), level of measurement, and distribution of the data in the population. First researcher should deal with data types and its analysis. As per objectives descriptive or inferential or both analyses can be done. Likewise, it will be helpful in selection of parametric and nonparametric statistical tests. The relationships between two variables (bivariate) by means of parametric tests (t-test, ANOVA, Pearson´s correlation coefficient, simple linear regression etc.) will be more suitable. For interval or continuous, normally distributed data the non-parametric tests will be considered. Then, multivariate analyses (e.g. multi-way ANOVA and multiple linear regression) for determining the independent contribution of different factors to a single outcome, with tests chosen on the basis of the nature of the outcome variables and on the hypothesized relationship between variables.