Introduction

Higher education serves as a gateway to professional and social success in the United States, with the unemployment rate of college graduates at around 2.5 percent, half of the national average. However, segregation in early education has led to an increase in gaps between people of color obtaining a Bachelor’s degree and white people obtaining a Bachelor’s degree. The U.S Department of Education created a list of practices that colleges should employ to increase POC student enrollment, which can be found here. On July 27, 2020, Elon University introduced ten diversity, inclusion and equity initiatives aimed at abolishing the systemic racism in predominately white institutions. The initiatives are detailed in depth, here.

Hypothesis

How does Elon’s student enrollment compare in diversity to other colleges in the South East Coast? How does Elon’s tuition costs (in-state and out-of-state) compare to other colleges in the South East Coast? Is there a trend between the tuition costs and increased diversity in a school?

I am interested in comparing Elon’s diversity rate (percentage of minority students out of total enrollment) to every other college in Florida, Georgia, South Carolina, North Carolina, Virginia and Maryland. I chose the South East Coast since Elon is a part of this region and the schools in this region are similar to Elon, yet have enough variability from each other to be somewhat representative of all colleges in the country. I am interested in tuition costs for Elon Vs. other schools on the South East Coast because Elon has been named a “Best-Value” college in Kiplinger’s Personal Finance. I expect to see a relationship between schools with cheaper tuition and a higher diversity rate, due to systemic challenges that people of Color face when working or earning an education.

Working with Data

The first step in this project was to load the packages I would need for this report:

library(tidyverse)
library(tidytext)
library(ggthemes)
library(wordcloud2)
library(textdata)
library(knitr)
library(shiny)

Next, I found two data sets from Kaggle detailing every college in the United States’s total minority population, total enrollment, in-state and out-of-state tuition. Next, I imported these datasets to R in order to filter the states to only include the South East Coast:

read.csv("~/Desktop/project_data/diversity_school.csv") -> diversity

diversity %>% 
  filter(state %in% c("Georgia", "Florida", "South Carolina", "North Carolina", "Virginia", "Maryland")) -> diversity_southeast

write.csv(diversity_southeast, "diversity_southeast.csv")

read.csv("~/Desktop/project_data/tuition_cost.csv") -> tuition

tuition %>%
  filter(state %in% c("Georgia", "Florida", "South Carolina", "North Carolina", "Virginia", "Maryland")) -> tuition_southeast

write.csv(tuition_southeast, "tuition_southeast.csv")

Once my data was filtered, I went back into Excel and calculated each college’s diversity percentage by dividing total minority enrollment by total enrollment, as well as added cities for each school to make my Tableau map more precise. Next, I made a new spreadsheet with only the 200 biggest schools in the South East Coast. I wanted to exclude the schools with less than 4,000 students, since Elon is around 6 or 7 thousand students.

Visualizing Diversity

I wanted to first visualize how Elon’s diversity percentage compares to other schools on the Southeast Coast, so I created a map in Tableau that color-coded by diversity percentage, with darker colors indicating a higher percentage.

I labeled Elon on this map, and as we can see, Elon’s point is lighter in color indicating that Elon has a relatively lower diversity percentage than other schools in this region. From this map, we can also see that the higher diversity percentages tend to be in large cities like Atlanta and Baltimore. This is not out of the ordinary, since those cities have a higher minority population than the rest of the region. I added filters to the map in the side legend, and if we filter the map to only include ‘Private’ colleges (because Elon is a private university), we can see that the only school less diverse than Elon is High Point University. Elon has a diversity percentage of 15.35% and High Point has a percentage of 15.19%.

Tuition and Diversity

Before I created maps to visualize both tuition costs and diversity to compare Elon with other schools, I wanted to get an idea of which schools were the most diverse and which schools were the least expensive. I predict that there is a negative relationship between tuition costs and diversity for both Elon and other colleges, so these tables will allow me to look for any overlap between the two. Since I’ve established that Elon is less diverse and more expensive, I don’t expect Elon to show up in either table and will look at Elon’s data separately.

read.csv("~/Desktop/project_data/diversity2.csv") -> diversity_clean
diversity_clean %>%
  arrange(-minority_enrollment) %>%
  select(name, minority_enrollment) %>%
  head(10) %>%
  kable(caption="Schools in the Top 200 Biggest Schools on the Southeast Cost with the Highest Minority Enrollment")
Schools in the Top 200 Biggest Schools on the Southeast Cost with the Highest Minority Enrollment
name minority_enrollment
Atlanta Technical College 0.9656702
Bethune-Cookman University 0.9379327
Savannah State University 0.9324517
Hampton University 0.9216936
Florida A&M University 0.9145591
Bowie State University 0.9064091
Morgan State University 0.8965965
Norfolk State University 0.8954704
Prince George’s Community College 0.8874104
Virginia State University 0.8798010 
read.csv("~/Desktop/project_data/tuition2.csv") -> tuition_clean
tuition_clean %>%
  filter(degree_length %in% "4 Year") %>%
  arrange(in_state_total) %>%
  select(name, in_state_total) %>%
  head(10) %>%
  kable(caption="Schools in the Top 200 Biggest Schools on the Southeast Coast with the Lowest In-State Tuition Costs")
Schools in the Top 200 Biggest Schools on the Southeast Coast with the Lowest In-State Tuition Costs
name in_state_total
University of Maryland: University College 7416
University of Baltimore 8958
University of North Carolina at Pembroke 12342
Fayetteville State University 13151
Dalton State College 13196
Western Carolina University 13902
Savannah State University 14556
University of North Georgia 15212
North Carolina Central University 15257
Winston-Salem State University 15364 
tuition_clean %>%
  filter(degree_length %in% "4 Year") %>%
  arrange(out_of_state_total) %>%
  select(name, out_of_state_total) %>%
  head(10) %>%
  kable(caption="Schools in the Top 200 Biggest Schools on the Southeast Coast with the Lowest In-State Tuition Costs")
Schools in the Top 200 Biggest Schools on the Southeast Coast with the Lowest In-State Tuition Costs
name out_of_state_total
University of Maryland: University College 12336
University of North Carolina at Pembroke 16342
Strayer University at Arlington (Va.) 16750
Strayer University at Atlanta (Ga.) 16750
Strayer University -RTP (N.C.) 16750
Western Carolina University 17902
University of Baltimore 21076
Dalton State College 21618
Middle Georgia State College 23000
Bethune-Cookman University 24226

Once I foumd the top 10 most diverse schools and saw that they were mostly 4-year universities, I filtered the tuition data to only include 4-year universities, to get a better sense of the overlap. As we can see, the schools vary a little bit from each table, but two or three universities are consistent among the tables. This gives me an idea that there is a relationship between high diversity and low tuition.

Next, I wanted to compare Elon’s diversity and tuition costs to the average diversity rate and tuition costs of colleges on the Southeast Coast. First, I needed to calculate these averages:

mean(diversity_clean$minority_enrollment)
## [1] 0.4152173
mean(tuition_clean$in_state_total)
## [1] 17385.52
mean(tuition_clean$out_of_state_total)
## [1] 25219.77

The average minority enrollment percentage is 41.52%, the average in-state tuition costs are 17,385.52 dollars and the average out-of-state tuition costs are $25,219.77. There is about a ten-thousand dollar difference between in-state and out-of-state average costs for this region, but Elon will have the same in-state and out-of-state costs. Here are Elon’s minority enrollment percentage and tuition costs:

diversity_clean %>%
  filter(name %in% "Elon University") %>%
  select(name, minority_enrollment) %>%
  kable(caption="Elon University Minority Enrollment")
Elon University Minority Enrollment
name minority_enrollment
Elon University 0.1534783 
tuition_clean %>%
  filter(name %in% "Elon University") %>%
  select(name, in_state_total) %>%
  kable(caption="Elon In-State Tuition")
Elon In-State Tuition
name in_state_total
Elon University 47549 
tuition_clean %>%
  filter(name %in% "Elon University") %>%
  select(name, out_of_state_total) %>%
  kable(caption="Elon Out-of-State Tuition")
Elon Out-of-State Tuition
name out_of_state_total
Elon University 47549

Elon’s minority enrollment percentage is 15.19%, over 30% less than the average for the Southeast coast. Elon’s tuition costs are $47,549, which is about twenty-thousand dollars more than this region’s average out-of-state costs. Elon has a smaller diversity percentage and higher tuition costs, perhaps representing a negative relationship between tuition and diversity.

Visualizing Tuition and Diversity Simultaneously

Now that I’ve analyzed each variable separately, I’m interested in visualizing tuition and diversity at the same time to compare Elon to the rest of the region. I created two maps, one for in-state tuition and diversity and one for out-of-state tuition and diversity, and color-coded them by tuition with size representing diversity. The filters allow you to select by type or state.

In the in-state tuition map, we can see that most of the schools with higher diversity have lower tuition costs and vice versa. Elon is an example of this, as its’ point on the map is small and dark orange, representing low diversity and higher costs. Elon is not the most expensive or least diverse in North Carolina though, and it seems to be in mid price range and less diverse compared to the rest of the region.

In the out-of-state tuition map, tuition costs seem much more evenly distributed among diversity rates. There are more schools with higher diversity rates among all price ranges, whereas lower in-state tuition costs correlated with higher diversity rates. Elon’s price looks to be in the mid range in out-of-state costs, where it was closer to the higher range for in-state. Elon’s diversity percentage is still relatively small compared to the rest of the region. There also appears to be lower out-of-state tuition costs and higher diversity rates in major cities like Atlanta and Baltimore.

Looking for Correlations

Next, I wanted to plot minority enrollment against in-state and out-of-state tuition to explore a correlation or negative relationship between the two. I separated my trend lines by the type of school (public or private), and filtered out “Non-Profit” schools since there were only three in this data set.

read.csv("~/Desktop/project_data/diversityandtuition.csv") -> both
both %>%
  filter(type %in% c('Private', 'Public')) %>%
  ggplot(aes(minority_enrollment, in_state_total, color=type)) + 
  geom_point() +
  geom_smooth(method="lm")

both %>%
  filter(type %in% c('Private', 'Public')) %>%
  ggplot(aes(minority_enrollment, out_of_state_total, color=type)) +
  geom_point() +
  geom_smooth(method="lm")

From these scatter plots, I can see that there is a strong negative relationship between minority enrollment and tuition costs for Private schools, but not so much for public schools. In private schools, as tuition costs (both in-state and out-of-state since they’re the same) increase, minority enrollment decreases. In public schools, minority enrollment varies within tuition costs and there does not seem to be a strong relationship. Since Elon is a Private university, the school is likely to see a minority enrollment percentage of around 40% based on the trendline, however, Elon’s percentage is around 15%.

Conclusions

Elon’s new diversity initiatives were only implemented in the summer of 2020, so it’s hard to say whether they’ve had any effect on the results I’ve found here. With that being said, it is evident that more work should be done at Elon to increase diversity and inclusion, seeing that their minority enrollment is relatively low compared to the surrounding universities. These results also suggest that schools with cheaper tuition are more accessible to students of color, something Elon might want to consider in its new diversity policies, since Elon’s tuition costs appear to be relatively higher than the Southeast coast. If this negative relationship between diversity and tuition costs does exist, Elon should consider implementing financial aid programs for their students of color to make Elon more accessible to its minority students.