Ensuring every student has high-quality postsecondary options is essential for maximizing their likelihood of success. No matter a student’s path, quality should be at the forefront of an organization’s advising strategy. Unfortunately (or fortunately),
there is not a one-size-fits-all definition of “high-quality,” but there are some best practices for setting high expectations for students and putting them on a path that increases their likelihood of postsecondary attainment.
The word “likelihood” is important because not all postsecondary institutions offer students the same chance of attaining a degree. Previous work from The Vela Institute and National College Attainment Network (NCAN) demonstrated the considerable variation in graduation rates across institutions, even within the same sector, and these rates can also vary widely across race and ethnicity.
Important to following the rest of this blog post are two key assertions:
All institutions don’t offer students an equal likelihood that they will graduate; and
All institutions aren’t equally easy to get into.
Strategies that work to connect students with high-quality postsecondary options must keep both assertions in mind. To know where to advise a student to go that will offer them the highest likelihood of graduating, advisors also need to know where students
are likely to be admitted. To academically match students to institutions (or helping them overmatch) that increase their likelihood of graduating, advisors also need to be realistic about students’ enrollment prospects while driving them toward the
highest quality institutions into which they could likely be admitted.
College access and postsecondary advising work is all about making gradual, incremental improvements from cohort to cohort. Programs want to set ambitious, but achievable, benchmarks for their students that build upon progress made with previous classes.
Considering traditional postsecondary pathways through two- and four-year institutions, data-informed postsecondary advising should look at institutional graduation rates as a proxy for the postsecondary option’s quality. For example, if a program
sends 100 students to University A, which has a 40% six-year graduation rate, and 100 students to University B with a 60% six-year graduation rate, that program should expect to see about 20 additional students graduate from University B when compared
to University A in any given cohort. All else equal, programs focused on students’ outcomes should try to get as many students as possible to University B.
With the above simple example, it’s easy to see how important institutional grad rates are for predicting outcomes. That’s fine for University A and B, but what about in the real world where there are thousands of potential destinations for matriculating
students? This is where programs need to dig into their data.
Step One: Build a Data Set
The first step will be to obtain postsecondary enrollment data from the most recent graduating class or cohort. Ideally, the data set will include, by student:
Unweighted GPA
Highest ACT or SAT score
First matriculation destination (where did the student enroll, according to National Student Clearinghouse or other postsecondary outcome data – this is likely the first fall enrollment destination following their high school graduation date)
The ethnicity-specific six-year graduation rate for Bachelor’s degree programs and the three-year
grad rate for Associate’s from IPEDS (here’s a file to get you started)
Step Two:Identify Groups of Students with Similar Academic Profiles
In advising, we want to serve all students well, and we know our students have varying levels of academic achievement. Reiterating from above: to know where to advise a student to go that will offer them the highest likelihood of graduating, advisors
also need to know where students are likely to be admitted. Saying, “everyone from this cohort should only apply to schools with a 70% graduation rate” would, unfortunately, be unrealistic. For some students, perhaps even a large swath, the typical
admissions requirements for a school with a 70% graduation rate would be unattainable. This would leave students either with few or no realistic targets to which to apply or spending time applying to schools with a low likelihood of admissibility.
Given this reality, segmenting students to understand where they have enrolled based on their academic achievement can provide a good roadmap to advising future cohorts.
There are a lot of ways to group students, and organizations and programs can use whichever one feels right to them, but the below is one approach that uses a standardized segmentation to assign a student to a group based on their ACT score and their
unweighted GPA.
The margins of these groups are a little arbitrary. Depending on where a program puts its boundaries, a student grouped as a “2” could be a “1” or vice versa, but based on students’ academic achievements, a student grouped as a “2” is very likely to have
a very different academic profile than a “7.” Students within groups have similar academic profiles, and students in nearby groups have more similar academic profiles than groups that are further away on the chart.
Assign each student to a group. For example, a student with a 23 ACT and a 3.2 GPA would be in Group Four. Do this for every student who is currently enrolled in a postsecondary institution.
Step Three: Understand Groups’ Matriculation Patterns and Destinations
After assigning each student to a group, it’s important to understand where students matriculated and the graduation rates of the institutions to which they matriculated. After this, find the median graduation rate (ideally by ethnicity, if you have a
large enough sample of students) for each group.
Medians aren’t available in something like a Microsoft Excel Pivot Table but are available as a Microsoft Excel formula, Tableau, and other data packages. Why use a median here? Because it’s likely that in any given group you have some students who have
under-matched (enrolling at an institution where the student’s academic profile exceeds the institution’s) or overmatched (where the institution’s typical academic profile exceeds the student’s). In a pinch, you can also take the average by group,
but keep in mind that outlier data like these under- and overmatches affects (“skews”) average more than medians.
After taking the medians, the output should look something like this:
Group #
Median Grad Rate (%)
1
83
2
64
3
60
4
59
5
55
6
40
7
36
8
21
9
21
10
21
11
21
If you have done this with your own data set, congratulations! You have just built a data-driven framework to use while advising your current students. What’s next? How do we actually use this framework to advise students? Step four awaits!
Step Four: Postsecondary Advising Using a Target Graduation Rate
The table above that you (hopefully) just built is what we at Overgrad like to call a Target Grad Rate (TGR) Framework. For a student prospectively considering which group of colleges to apply to:
Compute their group number using their ACT/SAT score and GPA
Look up the median graduation rate for that group number (in your version of the table above)
Advise the student to apply to universities with a graduation rate (based on the IPEDS data linked above) that meets or exceeds their TGR
Focusing on the basket of institutions at which previous students with similar academic profiles have enrolled serves two purposes. First, it helps to increase the likelihood of admissibility for prospective students (we already know students with similar
academic profiles have been admitted and enrolled there). Second, by applying at institutions at or above the median graduation rate, we are trying to help students select againstand toward increasing the likelihood that they complete based
on graduation rates.
Working toward 100% compliance for college-bound students to enroll in a school that meets or exceeds their TGR will hopefully yield a continual increase in the percent of students going to and graduating from college from cohort to cohort. By improving
the selection decisions toward institutions with higher completion rates, the median grad rate of enrolled schools should continue to increase year-over-year across all groups (up to a limit).
It's fair to ask whether the TGR should be a hard and fast target for all students. The answer is no, but it should be a starting point. This is the science of high-quality counseling, but the art of counseling still has equal importance.
You know your students, so be sure to advise them to the best options you can. Grad rate is not meant to be a be-all, end-all for everyone. Typically, we see teams targeting 85% of students hitting TGR goals.
TGR doesn’t necessarily need to be for seniors only. The two primary ways students have to access higher quality institutions with higher grad rates is to increase their GPA or increase their ACT/SAT score. Showing the grouping table and accompanying
grad rates is an excellent early lesson to have with students. Freshmen and sophomores in high school have far more control over their GPA than a junior or senior does.
Step 5: What’s Next?
There are a lot of different approaches to using data for postsecondary advising, and this is one that our partners have appreciated and succeeded with. As with any approach in the college access and attainment field, we encourage finding ways to adopt
or adapt this to your needs and context. This post is part one of a three-part series on high quality, data-driven consulting. Part two will explore the remaining two pillars of high-quality data-driven advising: affordability and admissibility. In
part three, we will explore leveraging return on investment to help students compare and contrast college and non-college pathways.
Ryan Hoch is the Co-Founder and CEO of Overgrad. Overgrad is an all-in-one postsecondary access and success platform that guides students to and through high quality postsecondary pathways. From advisors to administrators, Overgrad streamlines processes and ensures data-informed practices for improved student outcomes.
Email to a Friend
