Dear All,
I just wanted to remind you at there is a final project required for Biostat 202. There are 3 components to the project, the first is due next Monday, August 14 and the following components (2 & 3) are due the following Mondays. The stages are to prevent you from getting too far behind on the project. Individual components are grade 0/1/2 and are based mainly on effort. The final combined project will be graded based upon content (letter grade).
There are 4 datasets this year that we ask you to choose from. The CLE website has guidance on each project as well as a description of the variables contained in each dataset. Recall that in general, the aim of this project is to exercise what you have learned about supervised and unsupervised learning algorithms via classification or clustering algorithms. It may be easier if you decide which one of these you want to tackle and develop your "research question" from there.
If you would like to use the NSQIP datset, the original file is approximately 1.2 GB. We have uploaded a 1% sample (6,500 observations, 6.7 MB) which may be more manageable for slower/older computers. You may always use the 1% sample to set up your SPSS stream, clean, and debug your stream before importing the larger file (or you may only use the 1% sample if you wish). All other datasets are < 3 MB and should not cause too much issues. Please let us know if you do have any issues opening any of the datafiles.
Although collaboration is allowed for lab and homework, please do not collaborate on your final project submission.
Good luck, and as usual, send us an e-mail if you have any questions!
John & Zara