The following assignment is to be turned in over the web at the class turn-in space,

https://catalyst.washington.edu/webtools/secure/esubmit/turnin.cgi?owner=bmck&id=106

Taking into account the discussion in class on May 10 and any comments on your week 6 HW, please

• Using the data in crist.txt, choose categories of appropriate time variables, time-dependent exposure and confounding variables, and fixed-time confounding variables, and create a data set which contains one record for each cell in the cross-classification of these variables.  Each record should contain the number of deaths from lung cancer in the cell, the number of person years contributed to the cell, and values of the variables which identify which category of each of the categorical variables correspond to the cell. 

 
• Decide what statistical modelling approach might be used to relate the numbers of lung cancer deaths and person years for each cell to the explanatory variable values defining each cell.

To turn in:
 

1. Please present a  labelled table, giving the numbers of lung cancer deaths and person years at risk in each of the cells in your data.

 
2. Please write a  one-page-or-less description of the statistical model that will be used to analyze this data, with an explanation of what parameters give information about the scientific questions of interest.

Please turn in your report using either text, html, microsoft word or Adobe Acrobat .pdf format.  Be sure your report has your name on it, and be sure it clearly labels each of the sections of the report.
 

Last update:  5/16/01