1 / 15

Class 17

Class 17. Review of Hypothesis Tests and Scales of Measurement The special case of success/fail outcomes. Use Pivot Table to create. Use Pivot Table to create. Create from data in the question. You calculate this. Excel Data Analysis. T-test: Two-sample for means, equal variances .

booth
Télécharger la présentation

Class 17

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Class 17 Review of Hypothesis Tests and Scales of MeasurementThe special case of success/fail outcomes

  2. Use Pivot Table to create Use Pivot Table to create Create from data in the question You calculate this. Excel Data Analysis

  3. T-test: Two-sample for means, equal variances. Pivot table Data analysis

  4. T-test: Two-sample for means, equal variances. Pivot table Data analysis Or…the question may give you the summary stats.

  5. T-test: Two-sample for means, equal variances. Pivot table Data analysis Or…the question may give you the summary stats. BEWARE!It switches tails based on sign of t. Negative because it does F-M

  6. T-test: Paired 2 sample for means(book calls it matched samples) • If the data are paired, use the paired test. • The paired test is EQUIVALENT to a one-sample test of the mean of the differences. • Weights before and after treatment. 72 patients • BA rookie and sophomore years. • Sample mean was lower sophomore year, in part, due to REGRESSION TO THE MEAN. • Extreme outcomes are followed (on average) by less extreme outcomes.

  7. ANOVA: single factor. Pivot table Data analysis

  8. ANOVA: Single Factor • Inference about the equality of several means. • If 3 or more groups, no choice but to use ANOVA • I’ll give you the raw data • Either in two columns (like for Height and IT) • Or in 3 or more columns (one for each group/sample)

  9. The Alternative Hypothesis • Chi-squared goodness of fit • P’s not all equal • Chi-squared independence test • Not independent • T-test of μ=100 • μ≠100, μ>100, μ<100 you decide! • T-test: Two Sample (H0: μM=μF or μM-μF=0) • μM ≠ μF, μM > μF, μM < μF you decide! • ANOVA single factor • Means not all equal H0 is not true

  10. With One-tail alternative • Ha: μM > μF • If sample mean for females was HIGHER than males, then DO NOT REJECT. No need to calculate anything or use DATA ANALYSIS • Excel Data Analysis t-test: two samples will always give you the 1T p-value that is the lower! • If sample mean for females was higher, it assumes that was your alternative….and gives you a misleading p-value.

  11. The special case of Success/Fail Outcomes It’s a categorical variable. It’s a numerical variable. It’s…..special. It’s where we started our course.

  12. Categorical Numerical Roulette IQ 904 spins 34 IQs s=19.8 n=33 Chi-squared GOF

  13. Categorical Success/Fail Numerical Roulette Wunderdog IQ 904 spins 149 Success/fail 34 IQs 149 1’s and 0’s 0.58 n=149 s=19.8 n=33 Chi-squared GOF Chi-squared GOF Binomial

  14. Categorical Success/Fail Numerical Roulette Wunderdog IQ 904 spins 149 Success/fail 34 IQs 149 1’s and 0’s 0.58 n=149 s=19.8 n=33 Chi-squared GOF Chi-squared GOF Binomial

More Related