I have and ordinal inputs (students' grade) that is students' grade before and after joining a tuition. The problem is the grade is not obtained from the same test but rather two different test. For example, the pre tuition grade comes from exam at primary school, and the post tuition test come from secondary school.
I opt to Mann Whitney but I am not sure whether it is suitable or not.
Edit: the marks are not available, only grades
If grades are available so will be the marks so you can use t-test (parametric) or wilcoxon (non-parametric). Yet, if you want to use grades then the non-parametric option will be Chi Square goodness of fit test.
I would like you to see its articles, I hope it useful.
https://www.sciencedirect.com/topics/medicine-and-dentistry/parametric-test
Each of the parametric tests mentioned has a nonparametric analogue. For example, the nonparametric analogue of the t-test for categorical data is the chi-square. The chi-square test (chi2) is used when the data are nominal and when computation of a mean is not possible. This test is a statistical procedure that uses proportions and percentages to evaluate group differences. Thus, in computing it, differences between observed frequencies and the frequencies that can be expected to occur if the categories were independent of one another are calculated. If differences are found, however, the analysis does not indicate where the significant differences are.
Thanks Rohana Syamsuddin
Since data collected in two stages, pre and post tuition. Therefore, this may be treated as two groups. as data is parametric for comparison of two groups the most suitable test is t-test.
Respected, Muhammad Niqab, your suggestion is applicable when data is in numbers. If the available data is in form of grades, applying t-test will be difficult. Better option will be to use chi-square test of association.
The following link could be helpful:
https://www.statisticssolutions.com/selecting-between-parametric-and-non-parametric-analyses/?__cf_chl_jschl_tk__=pmd_ztRyfOKwdEeDVDsfTc2cfb2n3D3gpnJzeKKJxfxmcQ0-1633459355-0-gqNtZGzNAhCjcnBszQi9
- Badges
- Science topic