STAT 217 Spring 2011 Exam 2 Preparation Logistics: • Mon May 16, 8:10-9:15 • Open-notes, open-handouts o You may bring anything that I have provided or that you produce yourself • Bring calculator, normal table, t-table o No computer use • Handouts 10-18, Quizzes 10-18, HW8-13 Overview: • We have analyzed studies that involve comparing two groups, both with a binary categorical response variable and with a quantitative response variable. o We have used both simulation methods and more conventional methods (z-test, t-tests) for assessing the strength of evidence that the groups differ. • We have also considered issues of how studies are designed (observational vs. experimental) and how this affects the scope of conclusions that can be drawn (cause/effect). • We have also studied descriptive methods (graphs and summary statistics) for analyzing quantitative data. Outline: • Handout 10: Statistical Significance for 2×2 Tables o Segmented bar graph, conditional proportions, simulating randomization test for comparing two groups with binary categorical response variable • Handout 11: Observational Studies o Observational study, confounding variable, (lack of) cause/effect conclusion1 • Handout 12: Experiments o Experiment, treatment, random assignment, cause/effect conclusion, blindness, double blindness • Handout 13: Two-Proportion z-Test o Two-proportion z-test, two-proportion z-interval, impact of order of subtraction, impact of sample sizes • Handout 14: Quantitative Response: Graphical and Numerical Summaries o Dotplot, center, variability, shape, outliers, symmetry, skewness, mean, nedian, standard deviation, inter-quartile range, five-number summary, boxplot, histogram, resistance • Handout 15: Statistical Significance for Quantitative Response o Simulating randomization test for comparing two groups with quantitative response variable• Handout 16: Two-Sample t-Test and t-Interval o Two-sample t-test, two-sample t-interval, impact of difference in sample means, impact of sample sizes, impact of within-sample variability, impact of confidence level • Handout 17: Paired Data o Paired data, paired t-test, paired t-interval, matched-pairs design • Handout 8: One-Sample t-Test and t-Interval o One-sample t-test, one-sample t-interval, misconception of prediction interval Advice: • Organize notes for efficient retrieval of information/formulas • Don’t plan to use notes too much o Prepare as if exam were closed book/notes o Focus on understanding, not memorization o Be cognizant of time constraint • Expect similar questions to what we answer in class every day, in quizzes, on HW, on exam 1 • Be prepared to think/explain/interpret o Not just plug into formulas o Be ready to explain process of how you would do calculations • Be ready to interpret computer output o Possibly exclude irrelevant output • Read carefully o Be sure to answer the question asked • Take advantage of information provided o Perhaps including computer output • Relate conclusions to context • Justify/explain your answers o Unless you are explicitly told not to bother • Arrive on time! • Prepare as thoroughly as you would for a closed-book exam o Re-read handouts o Re-work in-class examples o Re-work quiz questions o Re-work HW questions o Come to Wed class prepared with questions o Bring questions to office hours (Wed 1-2, extra office hours on Fri from 8-10) o I’ll also be available in our classroom from 7:30-8am on
View Full Document
Unlocking...