Dr. Vipul Patel 2 Continuous Variable v/s Categorical Variable Dependent Variable v/s Independent Variable 2 T tests 3 T test is the simplest test of whether two groups differ. One Sample T test Independent Sample T test Dependent Sample T test One Sample T Test 4 The one-sample t-test is used to compare the mean of a sample with a specified (known) test value. To conduct one sample t test, one variable measured on continuous scale and a known value is required. The unit of measurement of variable of interest and known value should be same. 3 Independent Sample T Test 5 An Independent Sample t test assume that two mean scores are independent of each other. It has exactly two variables. Independent variables must be categorical having two categories (i.e., groups) only. Participants in each groups must be different. Dependent variable must be continuous. The non parametric alternative to an independent sample t test is the Mann-Whitney U test. Paired Sample T Test 6 A paired sample t test assume that two mean scores are paired in some manner. It has exactly two variables. Independent variables must be categorical having two categories (i.e., groups) only. Participants in each groups must be same. Dependent variable must be continuous. The non parametric alternative to a paired sample t test is the Wilcoxon signed ranks test. 4 One Way ANOVA 7 The One Way Analysis of Variance (ANOVA) is used when you want to test whether the score of three or more groups differ statistically. It has exactly two variables. Independent variables must be categorical having three or more categories (i.e., groups). Participants in each groups must be independent. Dependent variable must be continuous. The non parametric alternative to One Way ANOVA is Kruskall Wallis test. Factorial ANOVA 8 A Factorial ANOVA is extension of One Way ANOVA. It has more than two variables. Two or more variables are categorical and they are independent. Exactly one variable is continuous and it is the dependent variable. There is no non-parametric alternative to Factorial ANOVA. 5 Correlation Test 9 Correlation looks at how two measurement vary together. There are exactly two variables. Both variables are continuous. The variables cannot necessarily be defined as independent and dependent, although they might be. Non parametric alternative of Spearmens Correlation test is Spearmens rank order correlation test. Multiple Regression 10 Multiple regression looks at the relationship among multiple variables to try to make a prediction about how some variables (the independent or explanatory variables) may predict scores on another variable (the dependent or response variable). It has two or more variables. All the variables must be continuous. One variable must be dependent and the others are independent. There is no non-parameteric alternative to multiple regression. 6 Chi-Square Test 11 A Chi-Square test is a non parametric test. A chi-square test investigates whether there is a relationship between two categorical variables. It has exactly two variables. All variables are categorical. Each variable has two or more levels (categories) within them. You cannot calculate average of the variables; you can only count how many are in each category. The variables cannot necessarily be defined as independent and dependent, although they might be. Factor Analysis 12 Factor analysis is a method of data reduction and summarisation: take many variables and explain them with a few factors or components or dimensions The common objective of factor analysis is to represent a set of variables in terms of a smaller number of latent variables. The primary function of factor analysis is to define the underlying structure among the variables in the analysis. 7 13 Best of Luck & Thank You!!!! vipulpat@gmail.com