Free Multivariate Analysis Essay Sample

Very often in research we come across experiment in which we need to measure more than one variable from the same subject or experimental unit. Sometimes we may not get useful information when these variables are separated out and analyzed individually. This is because; one variable is influenced by the other. For example, let us consider a situation where changes in variable A and B are influenced by changes in each other, and also by changes in a confounding variable C (Rencher, 2002). Univariate analysis helps to analyze one variable at a time and it does not analyze relationship between the variable and thus its purpose is more towards descriptive statistic rather than explanatory. Example of univariate quantitative analysis is mean, frequency distribution, etc. Bivariate analysis is useful when two variable within a group needs to compared. It is a useful social research aid, as it gives an idea of the relationship between two variables. Example of bivariate analysis is percentage, correlation, t- test, e.t.c (Newman & Newman, 2006)
Multivariate analysis can be used to analyze both qualitative and quantitative data. In quantitative data we use numerical values to represent data. Quantitative data can be continuous (e.g. income, age) or discrete (e.g. gender, marital status). Multivariate analysis can be applied to both types of quantitative data. Multivariate analysis helps to analyze more than two variables simultaneously. We can use this analysis to explore the joint performance of the variables and also the effect of one variable in the presence of another. E.g. ANCOVA, Multiple Regression, e.t.c (Rencher, 2002)
One of the multivariate tests which I may find useful in my research is One-way MANOVA. This test is used when we want to study the influence of one or more independent variable on two or more dependent variable. For example if a clinician wants to study the influence of 3 different drugs on blood lipid profile. Each of the patients is allotted to 3 groups and the level of triglyceride; HDL and LDL are measured in each of them. In this situation, triglyceride, HDL and LDL are the dependent variable and the three drugs are the independent variables. Thus multivariate analysis can be useful for in-depth research and core data analysis.

References

Newman, I., & Newman, I. (2006). Conceptual statistics for beginners (pp. 183-7). Lanham, Md.: University Press of America.
Rencher, A. (2002). Methods of Multivariate Analysis (2nd ed., pp. 1-5). NY: John Wiley & Son.