IOSR Journal of Research & Method in Education (IOSR-JRME) e-ISSN: 2320–7388,p-ISSN: 2320–737X Volume 4, Issue 1 Ver. V (Feb. 2014), PP 61-69 www.iosrjournals.org www.iosrjournals.org 61 | Page Quality enhancement in health research using biostatistics S.Valarmathi 1 , K.Kanimozhi 1 ,S.Kalpana 1 , Jasmine S Sundar 1 , Joseph Maria Adaikalam 1 , Parameswari Srijayanth 1 , D.Shantharam 1 1 ( Department of Epidemiology, The Tamilnadu Dr.MGR Medical University, India) Abstract: “Absence of evidence is not evidence of absence” –(Altman D G and Bland MJ ) ,in the present era where health research is being focused on more evidence based practice, it has become imperative to comprehend the research findings with the support of statistics. Since most researchers these days exacerbate their research findings in an attempt to prove them as statistically significant, dissertation submitted in the department of epidemiology between the years 2006 to 2013 were reviewed. This article, discusses on how biostatistics could be applied to enhance the quality of research in its every sphere thereby avoiding the aberrant research methodology, thus promoting all our pain staking research to give us the pleasure of worth taking. Keywords : Normality, presentation of data, testing of hypothesis, types of variables, value alignment. I. Introduction Statistics explores collection, organization, analysis and interpretation of numerical data. Biostatistics is the application of statistics in the biological and health sciences and it plays a major role in health research. In promotional materials for drugs and other medical therapies it is common to see statistical results which are quoted from research papers. And, number of statistics related to research investigations in medicine are now regularly used in medical literature. Numerical facts being more precise than words in communicating the scientific results, biostatistics does an effective role in exploring the truth and authenticating it as evidence based research. The misuse or inaccurate use of statistical methods may point the research in the wrong path and produce incorrect study results. [1] The critical aspects of research lies in the statistical analysis, and in the manner in which the findings are prepared and published. Hence, use of statistics is inevitable in medical research and the accurate interpretation of statistical results may pose a challenge for healthcare providers. With the increased availability of statistical software, it is easy to use and misuse the statistical methods, particularly when the user is not aware of the assumption that is to be satisfied before a particular statistical test is used for analysis. [2] The objective of this paper is to enhance the quality of health research through proper understanding of the statistical concepts and to present the results in the right way. 1 Methods of Preparing Results Scientific results are presented as a mixture of tables, graphs and diagrams. Always,summaries of data will be given as a support for the study and the raw data will not be given. Statistical analysis could be done in two ways, namely 1. Descriptive Analysis 2. Inferential Analysis The purpose of descriptive analysis is that, the data will be explained with a single value which will be the most representative value of the samples. An Inferential analysis enables to arrive at conclusions about the population based on the samples collected. There are different methods for drawing inference which depends on the type of variable and the distribution of the data. 2 Knowledge of Measurement Scales Statistical analysis of data is done through meaningfully coded variables and in which measurement scales plays a major role . To categorize or to quantify a variable, measurement scales are used. In statistics, we have four types of measurement scales namely Nominal, Ordinal, Ratio and Interval Scales. A Nominal scale is, making the data into categories without ordering and it also means that there is no distance between categories. Ordinal data is also making the data into categories but with ordering or ranking and there is distance between categories. Interval scale is a standard survey rating scale and it has meaningful equidistance between categories.