the sample size is 300 (thinking of using stratified sampling)
the context is among 3 courses of engineering: ChE, ME and CE between 2 public and 2 private universities within 5 year range.
the objectives are To evaluate the performance of public and private university passers and topnotchers of three engineering licensure examinations from 2015 to 2019. • To determine the factors and challenges that affect the Chemical, Mechanical and Civil engineering board passers and top-notchers from 2015 to 2019 in their licensure examination. • To describe the different learning styles utilized by the engineering licensure examination passers and top-notchers based on their lived experiences. • To distinguish and compare public and private university graduates' preparations for the engineering licensure examinations
You can use analysis of variance (ANOVA) to tell if the datasets are significant overall. ANOVA can tell if the results are significant, however, it won’t tell you exactly where those differences lie. A post hoc test is then required based on studentized range distribution.
There are a host of post hoc tests to deploy namely; Duncan multiple range test (DMRT), Tukey test, Dunnett test etc. But I would suggest u run a pairwise comparison test using Tukey method.
- Badges
- Science topic