Statistics 101 Introduction to Quantitative Methods for Psychology and the Behavioral Sciences Fall 2009 Course Objective: Statistics 101 introduces the basic concepts of statistics and the statistical methods that are most commonly used in psychology and in the social and behavioral sciences. There is no prerequisite. The first half of the course introduces descriptive statistics and the inferential statistical methods of confidence intervals and significance tests, applied to means and proportions. The second half introduces bivariate and multivariate methods, emphasizing contingency table analysis, regression, and analysis of variance. Instructor: Alan Agresti, Science Center 300b, e-mail AA@STAT.UFL.EDU, phone (617) 495-8711 Office Hours: Tentatively Tuesday and Thursday 11:30-12:30, Tuesday 4-5 pm, and by appointment. Teaching Fellows: Jonathan Hennessy (jhenness@fas.harvard.edu) and Joseph Kelly (kelly2@fas.harvard.edu) Office hours and help sessions to be announced Course web page: You can access this at my.harvard.edu, and also at www.stat.ufl.edu/∼aa/harvard/ Text: Statistical Methods for the Social Sciences, by A. Agresti and B. Finlay, Pearson Prentice-Hall 4th edition, 2009 Exams and grades: Two exams during the term and the final exam each contribute to 20% of the final grade. Make-ups will not be given unless arrangements have been made prior to the exam, and then only for illness or family emergencies. A project (details later, for the reading period) also counts 20%, and the remaining 20% is based on homework exercise solutions. Exam Dates (tentative): Exam 1 Exam 2 Final Exam October 1 November 5 December final exam period Homework: Exercises from the textbook will be assigned in 10 homework assignments. These are designed to help you to improve your understanding of the course material and to help you prepare for exams, which will contain similar questions. In the Thursday lecture you will be assigned a set of exercises. You should give your solutions to your teaching fellow by 4 pm, at the latest, on Friday of the following week. (You can leave this in the course mailbox in the Statistics Department office on the 7th floor of the Science Center.) Late homework solutions will not be accepted. You are welcome to discuss the homework exercises with other students and your teaching fellow, but you must write your final answer yourself, in your own words. Any software output that you submit as part of your solutions must come from work that you have done yourself. Selected exercises will be graded by the teaching fellow and will count for 20% of your final grade. Solutions will be posted at the course home page. Teaching fellows will have class sessions and office hours at which you can ask questions about course material, including homework exercises and software, and their sessions will present extra examples to supplement the basic ones presented in the class lectures. Software: Some examples in class will use SPSS statistical software, especially for the more computationally complex methods in the second half of the course. Students are encouraged to become familiar with the use of SPSS or another software package (such as Stata, SAS, or R) and use it in homework exercises where the use of software is appropriate. SPSS is available through FAS computing (www.fas-it.fas.harvard.edu). Outline: The class lectures will cover topics presented in the following text material. On exams you are responsible for all material presented in lectures plus any extra material that you are assigned to read. Chapter and Topic Text Pages Homework Problems (optional parenthesized) 1. Introduction to Statistics 1-7 2, 3, (5, 17) 2. Sampling and Measurement 11-21 1-3, 7, 13, 14, 27, 34, 36, 37 (5, 24, 38) 3. Descriptive Statistics Univariate description 31-55 Bivariate description 55-61 6, 10ab, 11, 19, 22, 24, 25, 28, 30, 35 (8, 33, 34ab) 39, 41, 59, 68, 69, 70, 72, 73, 74, 78, (43, 64, 66) 48, 49, 50, (47, 51, 52) 4. Probability & Sampling Distributions Probability, normal distributions 85-99 8ab, 9ab, 10abc, 11ab, 12ab, 17, 21, 23, 54, 55, (3, 8c, 9cd, 10de, 11cde, 12cd, 19, 47) 27, 29, 33, 41, 42, 46, 50-53, 57, (36, 54, 55) 5. Inference: Estimation Confidence interval for proportions Confidence interval for means Sample size determination 107-116 116-123 123-129 4, 7, 12, 13, 47, 48, 66-68, 77, (8, 18, 76) 21, 24, 25, 32, 69-71, 73, (22, 28, 54) 35, 41, 44, 62, (34, 38, 39) 6. Inference: Significance Tests Mean: Steps of significance test Proportion: Steps of significance test Decisions and types of errors 143-155 156-159 159-166 166-169 169-173 1-3, 9, 11, 39, 52-55, (6) 15, 40, 42, (14, 16, 21, 65) 17, 18, 24, 25, 45, 50, 51, 59, 61, (22, 23, 26, 27, 46, 63) 29, 30, 56, (28, 31, 57) 33, 34 7. Comparison of Two Groups Comparing proportions Comparing means Matched samples 183-190 191-193, 197-201 193-197 1, 9, 11, 16, 17, 59, 62, (8, 12, 14, 17, 47, 65) 21, 23, 32b, 49, 50, 60, 61, 63, (5, 6, 19, 22, 55) 26, (27, 28, 31) 8. Association for Categorical Variables Contingency tables, Chi-squared Residuals Summarizing association 221-229 229-233 233-239 1, 3, 5, 9, 10, 29a, (2, 4, 40, 44) 11, 16, (14) 18, 20, 22, (17) 9. Linear Regression and Correlation Regression model, least squares Correlation and r 2 Inference, assumption, influence 255-268 269-276 276-289 1, 3, 7, 10, (25, 63, 66) 11-12, 18, 20, 38, 58-61, (13, 67, 68) 29, 32, 42, 50-55 10. Intro. to Multivariate Relationships Association/causality, statistical control Relationships, statistical interaction 301-307 307-313 3, 5, 7b, 33, 34, 40, (7a, 39) 11, 14, 16, 32, 38, 42-44, (41, 45) 11. Multiple Regression Multiple regression and correlation Inference, interaction, comparing models 321-335 335-346 1, 4-7, 19, 46, 49, 66, (67) 25, 44, 48(a-f), (48(h-k,m,n)) 12. Comparing Groups: Anal. of Variance Comparing means: F test Multiple comparisons ANOVA using regression Two-way and factorial ANOVA 369-376 376-378 378-381 382-391 2, 5, 36, 38, 39, 51, 52, (1, 3, 6, 57) 10, 14, 42, (12, 40, 53) 15, (16) 17, 22, 24, 41, 45, 48, (23, 44) 13-15. Advanced topics (brief intro.) Quantitative and categorical predictors Modelling nonlinearity Logistic regression 413-422 462-473 483-493 Sampling distributions Finding power and P(Type II error) Binomial test for small n 73-85