This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: We use both tails of the sampling distribution Our alternative hypothesis is non directional – H : = μ IV. One tailed tests o A directional test is called a one-tailed test We use only one tail of the sampling distribution Out alternative hypothesis is directional – H 1 : > μ IV. Why use a one tailed vs. two-tailed tests? o Only if we have a theoretical reason in advance. o One-tailed test gives us more power o But we might miss unexpected results IV. Using the t-distribution o For smaller samples, we use the t-distribution instead of the z-distribution o Formula is the same: T calculated = (- μ)/ (s/ n ) o But our critical value of t might be different than our critical value of Z Z calc μ = 2.5 Z calculated = (- μ)/ (s/ n ) = 2.8 = (2.8-2.5)/(6/ 100 ) S x = 6 Z calculated = 5 n = 100 α = .05 H : = μ H 1 : ≠ μ...
View Full Document
- Spring '09
- Null hypothesis, Statistical hypothesis testing, Life expectancy