evaluation of statistical methods used in ph.d. theses of social sciences in indian universities

Research Article
Punita Govil, Mamun Ali Naji Qasem and Swati Gupta*
DOI: 
xxx-xxxx-xxx
Subject: 
science
KeyWords: 
Statistics, Parametric and NonParametric methods
Abstract: 

The study was aimed at finding out the suitability of statistical methods used in the analysis of the data in Ph.D. theses of social science faculty in Indian Universities. It was also the purpose of the study to find out the types of common errors committed by the researchers while using statistical methods to analyze the data in Ph.D. theses. For the purpose, a standardized tool developed by the researchers was used by the investigators to evaluate the statistical methods used in theses of social science faculty. The sample consisted of 20 Ph.D. theses of social science faculty of Indian Universities including 120 statistical methods. In the sample, 10 theses from Department of Education and 10 theses from Psychology Department in faculty of Social Sciences of AMU were randomly selected. The analysis of the study revealed that: (1)79% of the studies have employed appropriate method according to sample size and number of variables and 77% of the theses have employed appropriate method according to research design. However, only 47% of the researchers fulfilled the assumptions of the method used by them. (2) The common errors committed by the researchers while using statistical methods are of serious nature. None of the researchers calculated effect size in his/her thesis. 87% of the evaluated theses did not select the sample in appropriate manner and 36% of the researchers of social sciences used insufficient statistical methods in their theses.