Measuring the Graduation Rates of Hartford Promise Scholars at 4-Year Private Institutions

by Savannah Brooks and Emma Kozak

Last updated on December 8, 2023

for Data Visualization for All
with Prof. Jack Dougherty
Trinity College, Hartford CT, USA


As students in Trinity College’s Educational Studies course, Data Visualization for All, we are tasked with creating a data story with our community partner to help them better understand a data problem they are facing as well as for us as students to learn about the importance of data visualization. Our partner for this project is Hartford Promise , a Hartford-based non-profit with a mission that states they aim to “transform the lives of thousands of Hartford students through an integrated college success model that provides them the financial resources, support services, & key relationships that lead to college success and economic mobility.” Hartford Promise provides a college scholarship of up to $5,000 ($2,500 for two-year schools) to any Hartford public school student who is a Hartford resident, has a 93% attendance record or better, and has at least a 3.0 GPA. Hartford Promise wants its students to succeed, so understanding where Hartford Promise Scholars have had the best graduation rates is important to the organization. Through analyzing the dataset from Hartford Promise, we will be able to identify where students finish their education and compare Hartford Promise Scholars to other students at different institutions. Hartford Promise, with this information, will be able to better advise and mentor its students to help them succeed in college.

In the dataset we are analyzing from Hartford Promise there are 254 students who enrolled in a higher education institution in the 2016 and 2017 cohorts, so the dataset only includes students who graduated from high school in 2016 and 2017. Of the 254 students who enrolled in higher education institutions in these cohorts half earned bachelor’s degrees. We are more specifically looking to analyze the success of 92 Hartford Promise Scholars who started at private 4-year institutions. To achieve this, we will first look to see how many students graduated with a bachelor’s degree and identify where students have been most successful. We will also look at graduation rate data from the individual universities that Hartford Promise Scholars attended to compare their graduation rate to the general study body.


Explore the interactive chart

50.4% of Hartford Promise Scholars earned a bachelor’s degree from a 4-year college, but how does that percentage break down between public versus private colleges? To answer this question, we first compared graduation rates for Hartford Promise Scholars in public and private 4-year colleges to the graduation rates of the entire student body at colleges where Hartford Promise Scholars enrolled. We found that Hartford Promise Scholars graduated at lower rates compared to their peers at private 4-year colleges, but graduated at higher rates compared to peers at public 4-year institutions. For private 4-year, of the colleges the Hartford Promise Scholars enrolled in, the average graduation rate is 51%, compared to 86% of Hartford Promise scholars that start at a 4-year public institution earn a degree.

Deducing that Hartford Promise Scholars performbetter at public colleges compared to private colleges is important. Success for Hartford Promise is not just measured in graduation rates and students may not have the option to continue their education. However, understanding where students earn degrees will help support Hartford Promise in knowing what schools are equipped to support their student's needs for them to succeed. As the data shows, Hartford Promise Scholars are most successful at public colleges. The data also allows Hartford Promise to further examine why students who attend private 4-year institutions tend to graduate at lower rates than their peers. Hartford Promise then can use this to better support and advise students who attend private 4-year colleges.

Explore the interactive chart

Another way to gauge the success of Hartford Promise Scholars is to compare their graduation rates against all college students who were eligible for federal Pell Grants. Students who are eligible for Pell Grants display exceptional financial need and have not earned a higher degree than a high school diploma. This comparison is meaningful because Hartford is a low-income city with a median household income of $37,477, which is less than half the median household income in the United States. Analyzing the success rates of students who have Pell Grants is a good comparison to Hartford Promise Scholars, as students in the program come from Hartford which is a low-income community, creating a more equitable comparison to examine how low-income students succeed at colleges where Hartford Promise Scholars enrolled. We found that graduation rates at private 4-year institutions for Hartford Promise Scholars (55%) are 13.99 percentage points lower than the graduation rate for students with Pell Grants (68.99%). This is a smaller percentage point difference than comparing Hartford Promise Scholars to the general student body, demonstrating that students on Pell Grants at private 4-year colleges perform more similarly to Hartford Promise Scholars. When looking at public 4-year colleges Hartford Promise Scholars (86%) outperform students on Pell Grants (47.47%) with a 38.53 percentage point difference. Pell Grant students at public colleges perform, on average, worse than the general student body, creating an even larger difference between the graduation rates of Hartford Promise Scholars and Pell Grant recipients. Through this comparison, one can see that Hartford Promise Scholars still are not graduating at rates that meet other comparable student groups at 4-year private colleges.

Understanding the variety of factors that cause Hartford Promise Scholars to graduate at lower rates when looking at private 4-year universities is important because it will allow Hartford Promise to analyze the differences that occur with students when attending private colleges. Additionally, with these findings, Hartford Promise can investigate new ways to support Scholars to meet their needs as they continue on their journey of higher education.

Explore the interactive chart

One key insight we noticed when analyzing the Hartford Promise and College Scorecard data was that Hartford Promise students did not generally see as much success as their peers at private four-year universities when it comes to completion. Since there were only three private four-year universities where five or more Hartford Promise students have enrolled, we completed a case study on those three colleges: Springfield College, Goodwin College, and the University of Saint Joseph. At all three colleges, Hartford Promise students had at least a 10% difference between graduation rates from their peers. Goodwin College in particular seems to be a trouble spot for Hartford Promise students — while their graduation rate for all students is not particularly high at 29%, no Hartford Promise student that attended Goodwin College graduated from there, which is notable. This data can help Hartford Promise advise students when it comes to applying to and choosing colleges; for instance, Goodwin College likely does not provide an environment that produces successful strategies from low-income students. This could be due to the college accepting students that did not have the resources or education to succeed at their college.


Our two sources were the Hartford Promise dataset our community partners provided as well as the College Scorecard dataset found publicly online. First, we filtered the Hartford Promise dataset for only 4-year colleges and then further filtered the schools by public and private. We created a pivot table calculating the numbers of starters at private 4-year and public 4-year universities. We did this by using the filter instance = 1 which provided us with the school where Hartford Promise Scholars first enrolled. While this method does not account for transfers we felt that this was the best way to analyze the data because we want to help Hartford Promise advise students on the best colleges for their needs, and focusing on original placements without needing to transfer is an important part of that process. After finding the starters at private and public 4-year colleges we then created another pivot table finding how many bachelor's degrees were earned at public 4-year and private 4-year institutions.

We then began to analyze the College Scorecard data by recording the graduation rates in 150% of normal time for the schools where Hartford Promise Scholars enrolled. Next, we averaged the graduation rates for private 4-year and public 4-year colleges to find the graduation rate of the general student body at these colleges and compared them against the percentage of Hartford Promise starters who earned a bachelor’s degree. The same data collection process applied to students on Pell Grants. Using the College Scorecard data set we collected the graduation rate for students on Pell Grants who graduated in 150% of normal time and compared that to Hartford Promise Scholars. We broke it down even further, creating case studies for three private colleges that were well-attended by Hartford Promise students and comparing those particular students to their peers at the institutions they attended. A potential issue with our results is that the transfer students we include are not necessarily starters at the school where they earned their degree, which would affect the graduation rates for Hartford Promise Scholars as more people enrolled at a school than started. However, we believe that we should still include the degrees transfer students earned to better represent the data. Through our analysis using data from Hartford Promise and College Scorecard we were able to have a better understanding of the success of Hartford Promise students at private and public 4-year colleges.

Section 4: Sources

College Scorecard Dataset,

Hartford Promise - About,

Hartford Promise Dataset.*

*While HP provided individual de-identified data, we promised not to share the raw data with the public to ensure privacy. Other researchers who wish to examine the raw data should request it directly from HP.