International Journal of Research Studies in Biosciences (IJRSB) Volume 5, Issue 5, May 2017, PP 33-37 ISSN 2349-0357 (Print) & ISSN 2349-0365 (Online) http://dx.doi.org/10.20431/2349-0365.0505005 www.arcjournals.org ©ARC Page | 33 How to Choose the Statistical Technique in the Data Analysis Hari Prasad Upadhyay Lecturer of Biostatics, Department of Community Medicine, College of Medical Science, Bharatpur- 10, Chitwan Abstract: The main objective of writing this paper is to provide the algorithm of choosing the statistical tools in the data analysis. Choosing an appropriate test one of the most important task in research. So, that right test will give the valid conclusion and wrong test may give the misleading inference. In order to choose the right statistical test we should be familiar with the different variable and their nature (for parametric, test of normality) Keywords: Statistical too, Parametric, Test of Normality 1. INTRODUCTION Nowadays statistics is the one of the most important part in all sectors. Like in the medical sectors it plays the vital role for the inference. With out of the knowledge on the tools and technique of statistics nobody can write the quantitative research paper. There are the various rules in the statistics for the data analysis. Knowingly or unknowingly there is no proper use of the statistics in data analysis in the research writing for the. Most of the researchers use the wrong statistics for the significant results. In statistics there are various rules and condition for the choosing of suitable test statistics. Most of the researcher find the mean and standard deviation but they don’t know the concept of normality and rational for choosing of the particular test. Why they calculate only mean and standard deviation, it becomes a grate issue in research. So, the result and inference are not highly valid. In most of the Intuition only few of the researchers follow the rule during the data analysis. In various research works published in biomedical journals journal we can see and observed the use of wrong or inappropriate statistical test used in the data analysis [1,2]. In order to do the statistical calculation and analysis there are lot of software. But the problem is that there software cannot chose the suitable test statistics. So, if we follow the suitable way or guideline for the data analysis our result becomes well and inference become valid. Researcher chose the test statistics depending upon the need and type of objective not by the nature or kind of collected data. Variable: A variable is any characteristics, number, or quantity that can be measured or counted. A variable may also be called a data item. Age, sex, business income and expenses, country of birth, capital expenditure, class grades and eye color and vehicle type are examples of variables. It of two type of variable depending upon relationship. Dependent Variable: A variable that may depend on the other factor is term as dependent variable. For e.g. Exam score as a variable may change depending on the student’s genders. Independent Variable: A variable that does not depend on the other factor is term as independent variable. For e.g. gender doesn’t change depending upon t he exam score. Qualitative Data: Those data which cannot be measure but can be count are called qualitative data. These data are also called the categorical variable. For e.g. number of child vaccinated, number of patients cured and number of person died. In descriptive statistics results obtained are presented in the form of percentage, rates, ratios and proportions. In inferential statistics the method employed in the analysis of such data are Z-test for proportion and Chi- square test. Quantitative Data: Those data which can be measure but not count are called quantitative data. These data are also called continuous data/variable. For e.g. height, weight and B.P., Hb%. In descriptive statistics depending upon their nature (Normally distributed or not), different, measure of central tendency (mean, median and mode) & dispersion (SD, range) will be use. For normally