Multivariate Statistical Analysis and Required Sample Size

Since most statistical techniques are sensitive to sample size, sample size has an important place in data analysis. There is a flexible approach for sample size in multivariate analysis; however, determining the sample size correctly is an important problem for researchers. Multivariate statistical techniques are used in many scientific fields, especially health sciences, biology, economics, education, and engineering. In this review, multivariate analysis of variance, multivariate regression analysis, multivariate logistic regression analysis, factor analysis, principal component analysis, structural equation modeling, cluster analysis, discriminant analysis, canonical correlation analysis, multidimensional scaling analysis, and conjoint analysis and techniques are discussed. The necessary sample size criteria for each analysis technique are presented.