Melanie V. answered • 04/07/19
Patient and Effective Ph.D. College Statistics Professor!
It is common for a student to misunderstand the requirements for performing either a chi-square test of association or a t-test. A simple way of remembering is as follows:
Chi-Square Test of Association: appropriate when the independent and dependent variables are categorical (nominal/ordinal). For example, when determining if a relationship exists before gender (male/female) and political party (republican/democrat/independent). When a Chi-Square Test is conducted the null hypothesis is as follows: gender and political party are independent (not associated) and alternate hypothesis is as follows: gender and political party are dependent (associated)
T-Tests: appropriate when the independent variable is categorical (males/females) and the dependent variable is continuous (mean scores). For example if you were trying to determine if a mean difference of statistics exam scores exists between males and females. Here, the null hypothesis would state that the mean scores are equal for males and females. Conversely, the alternate hypothesis would state that the mean scores do significantly differ for males and females.
