Week 5: Data Analysis

Hi all,

After we finished cleaning up the data, we made a lot of graphs, so they can be interpreted easier. Here are a few examples.

In this one, you can see the number of patients spread across each age group.

Here you can see what percentage of the patients belonged in each race.

There are of course many more graphs with more detailed information about the actual usage of alcohol, tobacco, and drugs, but you'll have to read my paper for those.

With the data, we used excel to find the regression statistics. In order to do this, we had to make a lot of adjustments to the data. For example, every time the data said "Yes" or "No" for alcohol or drugs, we had to change it to a numerical number. We used 1 for yes, and 0 for no. Here is a small portion of the data, to give you an idea of what it looks like.

Through this, we obtained an equation which tells us a prediction of what age a person might expect a below-knee amputation. The equation involves two variables: the HbA1c level of the person, and whether or not the person had a long-term use of aspirin. These variables were the most common among the patients in our data set who did get a below-knee amputation, which is why they can be used for predictions. Additionally, their P-values were below 5%, meaning they have statistical significance. I can't tell you what the equation is yet; you'll have to find out in my paper.

Since this week was purely for analyzing data, there isn't much else to say. Check back next week for updates!