Introduction to Statistics for the Social Sciences SBS200 - Lecture Section 001, Spring 2017 Room 150 Harvill Building 9:00 - 9:50 Mondays, Wednesdays & Fridays. Welcome http://www.youtube.com/watch?v=oSQJP40PcGI http://www.youtube.com/watch?v=oSQJP40PcGI
A note on doodling
Schedule of readings Before our fourth and final exam (May 1st) OpenStax Chapters 1 – 13 (Chapter 12 is emphasized) Plous Chapter 17: Social Influences Chapter 18: Group Judgments and Decisions
Homework on class website: Please complete homework worksheets #25 and #26 Multiple Regression Worksheet and Test Review Extended Due: Friday, April 28th
Test review and tutoring Lab sessions Everyone will want to be enrolled in one of the lab sessions Optional Test review and tutoring
By the end of lecture today 4/26/17 Review for Exam 4
Review Sheet
As variability goes down, it is easier to reject the null go down narrower As variability goes down, it is easier to reject the null ANOVA 99.18%
.9918 .4918 .5000 40 z = 52-40 5 z = 2.4 Go to table .4918 Add area Lower half .4918 + .5000 = .9918 also fine: 99.18% .9918 .4918 .5000 z =2.4 40
As variability goes down, it is easier to reject the null go down narrower As variability goes down, it is easier to reject the null ANOVA 99.18% Interval True experiment
r2 Income Education has the largest correlation coefficient of 0.85 Yes No Age IQ 0.91 No Income x Education is a significant correlation, p < 0.05 None r2
75% because .52 = .25, so 25% is explained so 75% is not explained Standard error of the estimate because it is a measure of the amount of error in the regression line (average of residuals) 81% because .92 = .81 19% The correlation between the heights of mothers and their daughters is moderate, positive and statistically significant, r(28) = 0.60; p< 0.05 36% because .62 = .36 64% because so 36% is explained so 64% is not explained - 100 – 36 = 64 75% because .52 = .25, so 25% is explained so 75% is not explained
r = 0.92 r2 = 0.8464 b = 6.0857 84.64% 15.36% b = 6.0857 55.286 b = 6.0857 residual r r2 b r2 b a -1.0 +1.0 0 +1.0 anything anything 0 +1.0 anything anything anything anything 0 any positive number
They are both difference from expected value Residual is difference from score to predicted score (y – y’) Deviation score is difference from score to mean (x - µ) Over-performing The standard error of the estimate is the average of the residuals just like standard deviation is the average of the deviation scores zero That there is no significant difference between these groups
Y’ = 480.94 - 4.89 (temp) - 14.76 (insulation) + 3.06 (age) 1 3 480.94 -4.89 -14.76 3.06 Y’ = 480.94 - 4.89 (temp) - 14.76 (insulation) + 3.06 (age) Y’ = 480.94 - 4.89 x1 - 14.76 x2 + 3.06 x3 1 3
Decrease level of confidence from 99% to 95% Yes Decrease variability (by increasing sample size or minimize variability due to error) Decrease level of confidence from 99% to 95% Easier Narrower Easier Common and rare scores
didn’t make a difference it did Get smaller Type of cartoon Level of aggression Two-tail True 48 No difference in level of aggression based on type of cartoon watched Type of cartoon did make difference in level of aggression did make a difference it didn’t didn’t make a difference it did Mean approaches true population Shape approaches normality Variability goes down
58 3.5 12 3 25 100 4.0 84 percentile
Thank you! See you next time!!