Week One: Part One

If you’re reading this, welcome back. As a word of warning, this post will probably be a little more dull than these posts normally are, but I feel it is still quite necessary in order to get all the background you need to understand my methods in the rest of these posts.

In other words, I feel this post is something of a necessary evil in order to understand what the heck I am talking about.

In my study, I will use regression analysis to test the impact of location, race, and wealth on educational outcomes on the school district level in Virginia.

“Why Virginia?” I hear you asking. The answer to that question is simple. There are two real reasons why I chose to examine Virginia. The first is that, well, at this point I’ve lived in Virginia for most of my life. The second reason why Virginia was utilized for this project was due to the availability of data such as SOL scores, teacher salaries, and government measures of economic well being.

All sources of data used in this project are derived from state and federal sources. All data used in this project is accurate to the best of the author’s knowledge.

Educational outcomes were measured via pass rates on the grade seven English reading and grade seven Math Virginia Standards of Learning (SOL) exams, which are available on the Virginia Department of Education website. The reason why this study focuses entirely on pass rates in grade seven, is due to the fact that above grade seven it appears that students generally split into different groups, with the more advanced students taking higher level exams than their peers. Grade seven was tested in order to attempt to minimize this issue. The Virginia Department of Education website also provides data on the number of students who took an exam in a given year, which is used as a proxy measure for school district enrollment size.

The wealth of an area is, for the purposes of this study, measured in two ways. The first is the Virginia Department of Taxation’s measure of local ability to pay, which can be found on the Virginia Department of Education’s website. Local ability to pay is an index consisting of land value, adjusted gross income, and the amount collected in sales tax. The second measure of wealth is median household income which can be found on the United States Census Bureau website.

The Isserman classification system is used to divide municipalities into categories based on location, as Isserman outlines in his 2005 article “In the National Interest: Defining Rural and Urban Correctly in Research and Public Policy.” Information on the Isserman classifications of counties can be found in Virginia government reports.

Similarly, data on the median salaries of teachers can also be found in Virginia government reports.

The data on race was collated by members of the University of Virginia’s Weldon Cooper Center for Public Service from United States Census Bureau reports and publicly released onto the internet.

This study examines educational outcomes from 2015-2019. The reasoning behind this time period is that it is a long enough time period to showcase long term trends in education throughout Virginia, and this time period is recent enough that the trends shown in this study are still important today but 2019 occurred long enough ago that datasets are now becoming available. Additionally, the use of a five-year time period reduces the risk of certain years being examined simply being outliers, and thus less accurate.

The online data analysis tool known as Stata will be utilized in order to analyze the gathered data to determine the level of correlation between separate datasets. The reason why Stata was used was due to its user-friendly nature, as well as its accessibility.

Before I end this (doubtlessly riveting) blog post and let you get on with your day, I feel I need to clear up one minor question which might have come up. The purpose of this study is not to examine all municipalities in the state of Virginia, rather, the purpose of this study is to examine as many as possible of the counties in the state of Virginia, and then trying to draw conclusions from them. Counties or cities ceased to be considered for the purposes of this study if they met one of the following two conditions. The first condition that caused counties or cities to cease to be considered was if data was not available for that county in one of the datasets utilized for this project. The second condition for exclusion was if the school district had 30 or fewer children taking either the English or Math exams in any year examined. Of the 133 counties and cities in the state of Virginia, 117 of them met these conditions and were included in this study. No “towns” in the state of Virginia were considered. These municipalities only possess data relating to their school district, with no data on other conditions being available. For the purposes of this project, Virginia is only considered as possessing Counties and Cities.

That wasn’t so bad, right? Don’t worry, this should be the last of these incredibly dull updates, and, assuming nothing terrible comes up, I should be releasing a second, much more interesting, post later this week