Public School Characteristics

Members:

Kate Weibel, Alton Thoroughgood, Jack Scott, Astrid Heifner

This Dataset:

We chose to focus our project on the Public School Characteristics dataset from catalog.data.gov. This dataset involves every public school in the United States, with variables focused on basic directory and information for schools and districts, as well as characteristics on student demographics, number of teachers, school grade span, and various other administrative conditions. In the dataset itself, it covered over 101,000+ entries.

Our Problem Statement and Research Methods:

We chose to focus our main topic on bias and equality, so we decided to make our problem statement: how would this dataset imply any sort of inequality among public schools and what might this dataset be leading out that lets us see inequality more?

Focusing on these variables can highlight any possible educational or health impact to students in these districts and bring more recognition to educational inequality or bias. To find any conclusions for our problem statement, we decided to use Poirier’s deconstructive reading method and Koopman’s meso-level method for research. Using Poirer’s method elements will help us inquire into the literal, critical and interpretive readings of the data and what is missing or there. We did this by focusing on a critical investigation of the datasets' meaning, like what’s being rendered absent or externalized. We also used Koopman’s method to help us to inquire more about how we group individual variables together to see how they show inequality in the dataset. We did this by looking at the data-defined features in the dataset, rather than just the data inputs and configurations themselves.

Our Findings:

After using Koopman’s meso-level research method, we found that free and reduced lunch numbers are some of the few indicators of inequality, which we can use as a metric for wealth in the schools. Those variables can be determined by whether or not a family makes below a certain level of income, relative to the poverty line. With this information, you can also determine how many of the students need free lunch out of the total. This also shows what states provide students with more free lunch than others. We found it peculiar that the dataset presents info about free lunches, acknowledging the disparity in wealth between schools, but fails to address it directly. We also figured that race variables are a large portion of the variables present. With this, you can also see discrimination and inequality when comparing schools with a higher diversity count.

Using Poirier’s deconstructive research method, we formulated some variables we believe would benefit the datasets findings that are missing: School Income, Average Test Scores, Quality of Food Given to Students, Graduation  Rate, Average Teacher’s Salary, GPA of Students, Average Income of Students after High School (if applicable), and more. Having these variables would allow us to better understand and find conclusions focusing on our problem statement on any inequality among public schools.