College of Public Health and Human Sciences H220: Introduction To Health Data Analysis Spring 2016 Instructor: Harold Bae, PhD, MS (Assistant Professor, Biostatistics) TA: Erin Buth Score: ______ / 20 Due Date: Apr. 21, 2016 Homework 3 1. A study is conducted to evaluate the association between binge drinking and cancer of various types. 100 individuals with cancer and 100 individuals that are cancer-free are selected for the study. 70 participants with cancer reported binge drinking at least once in their lifetime, and 27 cancer free participants recall binge drinking at least once in their in their lifetime. a) What type of study is this? (2pt) b) What is the appropriate measure of association (odds ratio or relative risk)? Why? (2pt) c) Compute the measure of association in the study and briefly interpret the quantity. (4pt) d) Can we conclude that binge drinking causes cancer? Why? (2pt) 2. A study is run to estimate the mean total cholesterol level in children 2-6 years of age. A sample of 9 participants is selected and their total cholesterol levels are measured as follows. 185 225 240 196 175 180 1 194 147 223 College of Public Health and Human Sciences H220: Introduction To Health Data Analysis Spring 2016 Instructor: Harold Bae, PhD, MS (Assistant Professor, Biostatistics) TA: Erin Buth When answering the questions below, please show all of your work. Answers with no work shown will not receive any credits. If more space is needed, you may attach another sheet of paper showing your work clearly. a. Compute the sample mean. (1 pt) b. Compute the sample standard deviation. (2 pt) c. Compute the median. (1 pt) d. Compute the first and third quartiles. (2 pt) e. Which measure, the mean or median, is a better measure of a typical value that represents the overall value of the total cholesterol in this data? Justify. (4 pt) 2