lecture11note

# lecture11note - Psych 100A Winter 2010 Lecture 11 Power...

This preview shows pages 1–3. Sign up to view the full content.

1 Psych 100A Winter 2010 Lecture 11: Power Analysis & Statistical Inference (cont.) 1. Relationships among , power and n 2. Power curves 3. Confidence intervals for when 2 is unknown 4. Properties of the t-distribution 5. Finding areas in t-distribution 1 Power – Probability of rejecting null hypothesis when a specified alternative hypothesis is true. Power = Area under the alternative distribution curve’s tail. If the alternative distribution mean is less than the null distribution mean then it is the left tail. 2 Directional z-test 3 4 5 6

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

View Full Document
2 2. Power curves The power of a test is the probably of rejecting the null hypothesis given that the alternative hypothesis is true. Power depends on 1. Type of test: H 0 versus one- or two-sided H A 2Sm p l s i z: n 2. Sample size: n 3. Meaningful difference: = | 0  A | 4. Type I error: -level 5. Standard deviation: [You can combine 3 & 5 to get effect size | 0 A |/ ] Power curves show relationships among power and parameters when one is varied and others are fixed. 7 Example, flex-time study: Compute power when varies from 0.1 to 3 days in a one-sided test of H A : < 6.3 days. Fixed: H O : = 6.3, = 0.05, = 3, n = 100. | 0 A | = 0.1 A 6.2 % Power 1  9 80 100 er 0.2 0.5 1.0 1.1 1.2 1.5 3.0 6.1 5.8 5.3 5.2 5.1 4.8 3.3 16 51 95 99 99.96 99.999 98 0 20 40 60 Percent Pow 34567 Alternative Means 8 80 100 r Flex-time study continued: Fixed: H O : = 6.3, = 0.05, = 3, n = 100.
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 03/15/2010 for the course PSYCHOLOGY 100B taught by Professor Firstenberg,i. during the Winter '10 term at UCLA.

### Page1 / 5

lecture11note - Psych 100A Winter 2010 Lecture 11 Power...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online