How to analyze Likert scale through t test and interpret the results 

Noted with thanks!

You can compute the mean score after testing reliability of each variable

It depends on your research objectives and data transformation techniques

@@MohamedBenhimaPhD the validity of the test depends on these assumptions...

@@freemindas there are specific techniques to measure validity (convergent and discriminant validity)

T-tests are parametric tests that assume the data being tested is normally distributed and has equal variances. Ordinal data, such as that obtained from a Likert scale, is often not normally distributed and violates the assumptions of equal variances. Therefore, using a t-test on ordinal data can lead to incorrect results. One alternative to using a t-test on ordinal data is to use a non-parametric test, such as the Wilcoxon signed-rank test or the Mann-Whitney U test, which do not assume normality or equal variances. These tests are specifically designed for ordinal data and are appropriate for Likert scale data. Before conducting any statistical test, it's important to carefully evaluate the assumptions of the test and make sure that the data meets these assumptions, or to use a test that is appropriate for the type of data being analyzed. Sensitivity analysis is preferred instead by using both parametric and non-parametric tests.

@@MohamedBenhimaPhD good that's why you can't use t test on likert scale like you did in this video. It's misleading 😅and will make many poor students fail.

Your hypothesis determines whether you refer to 1 or 2 tailed. If your hypothesis is directional eg dogs are larger than cats, then it's 1 tailed. 2 tailed is non directional...eg there is a difference between dog and cat size.