This preview shows pages 1–8. 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 DocumentThis 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: Chapter 13: Interpreting the Results of Hypothesis Tests Significance vs. Importance Effect Size Errors in Hypothesis Testing Power of tests Significant only means unlikely to occur by chance if H O is true Doesnt necessarily mean important or useful This assessment depends crucially on the research question of interest Direct relation between effect size and pvalue Larger sample sizes will increase significance (decrease p) However we can still use statistics to determine whether the effects of are study are big or small n s X t X hyp /  = Effect Sizes Effect Size: Estimate of the degree to which the effect is present in the population Usually difference between hyp and true There are three standard measures of effect size: Cohens d Hedges g Pearsons r Cohens d Effect (difference between true and hyp ) in standard deviation units Cohens d 2200 true = sample mean; hyp = value when H is true You can use this when you know hyp true d = Hedges g You can use this when estimating using s x ( 29 1 2 = = n X X X s X g hyp X hyp m m Example Our developmental psychologist from Chapter 12 concluded that the infants in her region have a mean speaking age (14.3 months) that is different from the national average (13 months) How big is this effect?...
View
Full
Document
This note was uploaded on 04/17/2008 for the course PSY 201 taught by Professor Arthur during the Spring '08 term at Purdue UniversityWest Lafayette.
 Spring '08
 Arthur

Click to edit the document details