18 - Statistical power - 04-01-08 - webct

18 - Statistical power - 04-01-08 - webct - Statistical...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
4/1/2008 1 Psyc110 April 1, 2008 Statistical Power Outline Power what it is, some historical background Effect size and power Pooled standard deviation for effect size with multiple groups Visualizing power Media reports of psychological research Review: Decision Matrix for Null Hypothesis Significance Testing (NHST) Basic logic of NHST Start with assumption that the null hypothesis is true Null hypothesis = H 0 : hypothesis of no difference or no relationship Create a sampling distribution (assuming H 0 is true) Collect some data Compare data to sampling distribution
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
4/1/2008 2 Review: Decision Matrix for Null Hypothesis Significance Testing (NHST) From 2/19/08 lecture: True State of Nature H 0 True H 0 False Statistical Decision Reject H 0 Retain H 0 Type I error α Correct decision Correct decision Type II error β Review: Decision Matrix for Null Hypothesis Significance Testing (NHST) Our concern now: True State of Nature H 0 True H 0 False Statistical Decision Reject H 0 Retain H 0 Type I error α Correct decision POWER = 1- β Correct decision Type II error β Defining Power Power = (1 β ) = Probability of rejecting H 0 when it is false (so, a correct rejection) Formal definition: ―The probability that a statistical significance test will reject the null hypothesis for a specified value of an alternative hypothesis‖ Note the language an alternative hypothesis needs to be specified (e.g., how much of a mean difference, or correlation, you expect)
Background image of page 2
4/1/2008 3 Power and Error Rates Type I error rate = α = set by researchers / conventions .05 , .01, .001 Cannot really influence once you’ve set it for a given study Type II error rate = β = (1-power) Increasing power will decrease Type II error rate, vice versa So, increasing power is a welcome strategy to help make more correct NHST decisions When Should Researchers Use Power? Power is most important in the
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 9

18 - Statistical power - 04-01-08 - webct - Statistical...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online