New Collection Topic

For the past couple of weeks, I have been deciding on the new topic for my statistic project. Last week, I found some interesting data in the Louisiana Almanac that was about the ACT. The book gave the results of the ACT by district from 2001 to 2007. My teacher gave me a good idea to do the project on: How has Hurricane Katrina affected the ACT scores? I would compare dates before 2005, the year of Katrina, and after Katrina. I cannot include 5005 because there were two or three ACT test before the storm and at least one after. It would be interesting to see if natural disasters cause more than mental and physical devastation to the citizens. Have the level of education decreased after a major named storm?

I will also be able to tell whether northern parishes/districts are able to better prepare their students for standardized test, in this case the ACT? Since only Southern parishes/ districts were truly affected by the storm, the scores could tell which region has the best scores. I would prove this by completing a two sample t-test where µ_a > µ₀ and see if it proved true or false. Some of the graphical displays that I will use are back to back stem plot, scatter plots, histograms, pie charts, and bar graphs.

There are many lurking variables with my topic. Many of the students that were living in Southern Louisiana relocated to northern parishes which can help them or lower them. Also, since not all the parishes felt the results of the hurricane, I would have to show the results by groups. Another problem is there are 66 different districts in Louisiana school system and there is too many for graphical displays and numerical summaries. The results would have to be averaged by region and then analyzed.