Is it possible to use or apply F-test(or T-test) to observe the difference among(between) different socio-demographic variables such as age of differernt groups, sex (male and female), education (illiterate, literate, primary......), occupation (business, job, ....)??
F and T test are for continuous variables, so only applicable to such variables (age). Sex, education and occupation are categorical variables and you should look at tests such as binomial test (sex) and chi-squared goodness of fit. You need to read carefully what each test is designed to do and make sure it answers the exact question you need to ask.
In order to decide what statistical test is appropriate, you must first decide what type of variable you have for the dependent variable, and what type of variable you have for the independent variable. That is, is each: nominal, ordinal, or continuous?
From there, you might look at a list of potential tests:
http://rcompanion.org/handbook/D_03.html
https://stats.idre.ucla.edu/other/mult-pkg/whatstat/
The question is not clear on what variable you wish to treat as the dependent variable and which as the independent variable. For example, if you want to compare age (continuous) between sex (nominal), a t-test may be appropriate, although you need to understand what assumptions about the data the t-test makes.
Hi,
Dear James and Salvatore
Thank you for your inputs. It is to test whether disability adjusted life years (DV; that is continuous) differ according to sex (IV; that is categorical; male or female); and according to SES(IV: categorical eg., low, medium, high)??
In that case, for sex you might consider: two-sample (Student) t-test; Welch's t-test; Mann-Whitney U test.
For SES you might consider: one-way analysis of variance; Kruskall-Wallis and Dunn tests.
For any you plan to use, be sure to understand the hypothesis that the test tests and the assumptions of the test.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India