After downloading this data and looking at the vtable, I picked the four variables that I found the most interesting and decided to explore how they relate to eachother. The variables that I chose were Total Income, Total Expenses, Tuition, and Research. Tuition was the variable that I thought the most interesting, and I thought that the other three would surely effect it.

Before making any graphs, I decided to turn these variables into per capita values, because otherwise the enrollment variable would lurk and skew the relationships of my variables. I created the per capita variables by dividing each variable by Total Enrollment, this turns the value from a total to a per-student value, giving us truer insight into the nature of the variable.

From here I created scatterplots which explore the relationships between each of my variables:

• I was surprised that this relationship was not more positive, as surely the schools with a higher income are the ones that charge more for tuition. I wonder if tuition counts as part of the schools income?

• Another relationship that I thought would be more positive, you would think that schools with higher expenses would charge more for tuition to compensate.

• Research is an expense so it makes sense that it would have a positive relationship with total expenses.

• This is the clearest positive relationship that we have. Clearly these institutions must be able to pay off their expenses by having enough income to do so.

• I thought that there would be a good correlation between the amount of money a school spends on research and the tuition that they charge, but that was not the case as this is one of our weakest relationships.

• Schools with a higher income have more money to spend on research!

The variable that I was interested in, tuition, was the one that had the least positive relationships with the others, which is disappointing. Perhaps there were other variables that would have had a greater effect on tuition and I picked wrong. However we have found that tuition has a slightly positive relationship with income, expenses and research, mostly so with income. One thing that concerns me about the results of my data is the extreme outliers present for almost every value. On every graph there are a couple of dots which are very high and don’t make sense to me as a per-capita value. If I were to go further with this investigation I would definitely choose to analyze those particular values and decide their validity, removing them if necessary.