Null and alternative hypothesis

Null and alternative hypothesis

1. hypothesis---t-tests (numerical data)

   • alternative hypothesis is the research question
     • If the -value is less that the Type I error rate (p<⍺), then we reject the null hypothesis.
     • If the p-value is greater than or equal to the Type I error rate (p≥⍺), then we fail to reject the null hypothesis.

2. 1 way contingency chi square

   • Ho: no difference between observed and expected
   • Ha: there is a difference
   • df = #cells-1

3. 2 way contingency chi square(categorical data)

   • Ho: 2 variables are independent
   • Ha: two variables are not indep
   • Reject null ifꭓ2O>ꭓ2C;p<⍺
   • Fail to reject if ꭓ2O≤ꭓ2C; p≥⍺
   • df= (#rows-1)(3columns-1)

4. Correlation test: association

   • non-directional
     • HO: correlation coefficient is 0 — ⍴ = 0;HA: correlation coefficient is not 0 —- ⍴ 0
   • directional
     • depends on whether looking for a positive correlation (or negative) ;HO : ⍴ 0 (⍴ 0) HA: ⍴ > 0 (⍴ < 0)
       • Linear regression: 2 single sample t-test
         • Ho: intercept/slope not different from ref; Ha: are different
         • directional also possible
   • Reject the Null: to > tc; p < alpha
     • fail to reject the null: to ≤ tc; p ≥ alpha