This preview shows pages 1–4. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Null Hypothesis Significance Testing PSY 2801: Summer 2010 Hypothesis Testing Jeff Jones University of Minnesota Jeff Jones Null Hypothesis Significance Testing Null Hypothesis Significance Testing For the rest of the semester, we will attempt statistical inference. The major way that Intro Stats courses teach statistical inference is through Null Hypothesis Significance Testing (or NHST for short). The reliance of Psychology on NHST is problematic, but in order to critique the logic behind NHST, you need to first understand the logic of NHST . Jeff Jones Null Hypothesis Significance Testing NHST: Steps to Inference NHST Steps 1 Formulate Hypotheses 2 Set Level 3 Decide Test 4 Perform Test 5 Make A Decision 6 Write A Pithy Conclusion Jeff Jones Notes Notes Notes Null Hypothesis Significance Testing NHST: Steps to Inference NHST Steps 1 Formulate Hypotheses 2 Set Level 3 Decide Test 4 Perform Test 5 Make A Decision 6 Write A Pithy Conclusion Jeff Jones Null Hypothesis Significance Testing Formulate Hypotheses There are 2 types of hypotheses you will see in Statistics: Null Hypothesis : Generally, the Null Hypothesis is the one you are attempting to reject or refute. Null indicates that nothing happened in that our data are from our assumed distribution. Null does not indicate a zero effect hypothesis  you can have a Null Hypothesis where the number of the right side isnt 0. Null Hypothesis Symbols: H N or H Alternative Hypothesis : Generally, the Alternative Hypothesis is that the data do not come from the assumed distribution. Alternative Hypothesis Symbols: H A or H 1 Jeff Jones Null Hypothesis Significance Testing Formulate Hypotheses The Null and Alternative Hypotheses deal with theories about a parameter of interest , which is the parameter you are trying to make conclusions about. H : = 1 H : 1 H 1 : 6 = 1 H 1 : > 1 If you noticed, based on these simple equations, the Null Hypothesis always has an equals sign! Furthermore, the combined hypotheses are mutually exclusive and exhaustive (remember Probability Stuff). Jeff Jones Notes Notes Notes Null Hypothesis Significance Testing Formulate Hypotheses Heres an example of the process for constructing these hypotheses: Is the mean height of men who enlist in the army different than the average height of US men? For simple hypotheses, on the left side is our parameter of interest and on the right side is a number . In this case, what is our parameter of interest? What is our number? Is it directional or nondirectional?...
View Full
Document
 Summer '08
 GUYER

Click to edit the document details