You can do that but you can also use Mann-Whitney U test and see if you get the same answer. If your sample size is greater that 30 it is at least in Finland considered to be normally distributed. If sample size

Dear D.A.R.K Dilrukshi, I agree with most what S. Béatrice Marianne Ewalds-Kvist wrote. One thing you should do before is to use the Kolmogorov Smirnov test (K-S test) to test the normality of distribution on variable(s).

If variable scores are distributed normally then use a parametric test (in your case t-test), and if the requirement of the normality of distribution is not fulfilled then use a non-parametric test (in your case Mann-Whitney U test).

The central limit theorem (CLT) states that the distribution of sample means approximates a normal distribution as the sample size gets larger, regardless of the population's distribution. Sample sizes equal to or greater than 30 are often considered sufficient for the CLT to hold.

ref: Central Limit Theorem (CLT) Definition - Investopedia

No, it is a better option to check the percent change .

Paired t test will come to existence if you have quantitative data

Likert scale data is 5 or 7 point qualitative data.

thank you all

As applies to so many of these questions, it makes a difference if you have a single Likert-type item or if you are summing several Likert-type items into a scale. To me, the question isn't clear on this point.

In the case of a pre/post survey, it also makes a difference if you have recorded the data in a way that you can identify a pre score and a post score for any given individual. Or if you haven't recorded data in this way. To me, the question also isn't clear on this point.

To be honest, I would discard any advice on which analysis is appropriate if you haven't clarified these points. That's one reason that why you have received several and contradictory suggestions.