by Liam O'Connor and Ben Bejoian
Last updated on 12/06/22
for Data Visualization for Allwith Prof. Jack DoughertyTrinity College, Hartford CT, USA
Connecticut school districts use their funding for school resources, such as, smaller class sizes, competitive teacher compensation, and other additional supports. When used wisely, school districts are more likely to have higher test scores, increased graduation rates, and other improved indicators of student achievements. In other words, the more money given spent on an individual student in the school district, the better educated the students will be. With the 169 towns in Connecticut, there is a large disparity between race and wealth due to preexisting discriminatory policies and prejudicial practices. Focusing on the 33 towns in the Hartford Region, we wanted to take a closer look, seeing if there was a correlation between the enrollment rates by students’ race and ethnicity as well as how the district-level per-student spending vary by students’ race and ethnicity.
To visualize our first question, asking how enrollment rates by school type vary by student race & ethnicity, refer to Chart 1. We decided on combining the school types (Technical Education and Career, Program, Shelf Magnet, Public, Open, Public Charter, and College Affiliated) in each town to help focus on the ethnic and racial divide between the towns. To best visualize this data, we chose a stacked bar chart with White ethnicity in descending order. For visual purposes, Additional Combined and Hispanic/Latino are moved to the middle. Some cautions to be aware of when viewing Chart 1 are the grouping of the schools, and “Additional Combined” ethnicity. As stated before, the schools are combined to focus on the racial and ethnic divide between the towns, however, regional schools and other magnet schools are not included in this list due to some towns not having them or the school population being too low. American or Alaskan Native, pacific islander, and 2 or more races are classified as Additional Combined due to the limited data on the students. While the chart features different races and ethnicities broken up by each race and ethnicity, for the remainder of this, we will be utilizing the "Black, Indigenous, and people of color" grouping or BIPOC for short. This is because the School and State Finance Project for Connecticut uses this grouping as well, so we are simply keeping the data consistent.
Figure 1: Sorted by white ethnicity, school types are combined to show the student's race and ethnicity overall enrollment rate.
For Chart 2, we will be using DataWrapper compiled by Net Current Expenditures per Pupil (NCEP). The NCEP application provides the latest net current expenditures (NCE), average daily membership (ADM), net current expenditures per pupil (NCEP) and the Special Education Excess Cost grant basic contributions. We decided to use a choropleth map to highlight the current expenditure by each town in the Hartford Region. On top of that, we also included the percentage of White and BIPOC students in each town. To summarize the totals of each town, there were five towns in which the NECP was below $17.8k. Moreover, out of those five, three of them were under $17k with $16.2k, $16.2k, and $16.4k respectively. There were 16 towns that fell in the $17.8–$19.5k category. There were seven towns that fell in the $19.5k–$21.2k category. Interestingly, there were no towns in the range of $21.2k–$22.9k, and the NCEP jumped from the $19.5k–$21.2k category to the ≥$22.9k category. Finally, there were five towns that fell in the ≥$22.9k category. What makes up the top NCEP with over $22.9k is an interesting set of towns. These five towns are Hartford, Bloomfield, East Windsor, Windsor Locks, and East Granby, all five of which have very different demographics. East Windsor and Windsor Locks have an almost even split of White and BIPOC students, whereas East Granby has almost 75% White students. Hartford and Bloomfield swing in the other direction as both have over 90% BIPOC students.
After compiling the data, we found that there is in fact a correlation with the percent of BIPOC students and per student spending. We determined that on average the towns with a higher percent of BIPOC students will have a higher amount of funding. This data has a correlation coefficient of 0.272. Although this is a positive correlation between the variables, the closer to zero, the weaker the correlation. Because of this, there is a correlation between the funding of each school district and Percent BIPOC students, however, the relationship is not significantly strong. The scatterchart visualize these relationships with a linear best fit line. We chose a scatterplot to allow easier interpreation for the two variables. For example, Newington is one of those towns that fall onto the linear best fit line making it one of the average towns in this example. Towns such as East Hartford, Bloomfield, and Hartford would be considered outliers due to straying from the best fit line.
The question that must be asked now is if the Hartford Region is an accurate representation of the state of Connecticut. To answer this, we must look at all the school districts. Similar to previous, we collected the data of percent BIPOC of students enrolled and the NCEP of the school district. Due to data being under review, Shelton and Danbury are not included on this list. We found a correlation coefficient of -0.28 making the two relationships not related. This could be because of the higher percentage of BIPOC students enrolled in the Hartford Region vs the entirety of Connecticut.
After further analysis, we have found in that in the Hartford Region, those towns with a greater percent of BIPOC students will recieve more funding. However, when broadening our scope past the Hartford Region, we find the opposite of what we found. To highlight this, according to the Greater Hartford Interfaith Action Alliance (GHIAA), Connecticut spends $720 million more to educate school districts that are 75% white students than they do on school districts that are at least 25% students of color.For further analysis, looking at data focusing on individual student and school funding (instead of school districts) would provide a further insight into the question proposed.
A Segregated Connecticut. School+State Finance Project. (n.d.). Retrieved November 26, 2022, from https://schoolstatefinance.org/issues/segregated-connecticut
Center for Leadership and Justice. The Center for Leadership and Justice. (2022, November 21). Retrieved December 6, 2022, from https://cljct.org/
Net current expenditures per pupil used for excess cost Grant Basic Contributions. CT.gov. (n.d.). Retrieved November 30, 2022, from https://portal.ct.gov/SDE/Fiscal-Services/Net-Current-Expenditures-per-Pupil-used-for-Excess-Cost-Grant-Basic-Contributions