Mini-Assignment 3: Recognizing Correlation in Data
| Assignment Day | Monday (evening): September 30, 2013 (assignment will updated until Assignment date, after that there will only be only clarification updates) |
| Due Date: | Friday: October 11 before class (hardcopy and email) |
| Format | Email to TA see home page for address and cc Instructor You will also need to hand in a Hardcopy of your Data Sheets |
In this assignment you will explore correlation, we will also work on this same data in class. We will work on two data sets: (1) smoking occupation and lung cancer mortality rate, and (2) smoking instances and instances of lung cancer from CDC.
Some Data is available in this workbook: AS3.xlsx
Question 1: Using the data under the tab containing "Occupation" in the above workbook (AS3.xls). Test whether Smoking is correlated to Mortality. Is it a strong correlation? (create a correlation plot with trend line in a sheet in a workbook and computer the correlation using =CORREL(), or PEARSON() functions in excel.
Smoking is the Explanatory variable (X-axis), and Mortality is the Response variable (Y-axis).
Question 2: Using the data from CDC, describe the correlation of people (men, women, both genders) who smoke and have lung cancer. Do the correlations differ?
CDC Cancer Data: http://apps.nccd.cdc.gov/uscs/cancersbystateandregion.aspx
CDC Smoking Data:
http://www.cdc.gov/tobacco/data_statistics/tables/trends/cig_smoking/index.htm
You will need to FETCH the data from the CDC website (see links above).
(create a correlation plot with trend line in a sheet in a workbook, compute the correlation coefficient, and R2 for both women and men, and the data combining both men and women)
Question 3: Answer the below and how they relate?
a) What is R2, What is R?
b) How doe R2 differ from R?
Resources:
http://www.cdc.gov/tobacco/data_statistics/sgr/2001/index.htm
http://en.wikipedia.org/wiki/Lung_cancer