Related Links

• Richmond Fed Survey of Community College Outcomes
• Results From the Survey of Community College Outcomes Extended Pilot, Regional Matters
• A New Approach to Measuring Community College Outcomes, Speaking of the Economy
• Topics: Community Colleges

Transcript

Tim Sablik: Hello, I'm Tim Sablik, a senior economics writer at the Richmond Fed. My guest today is Laura Ullrich, a senior regional economist at the Charlotte branch of the Richmond Fed. Laura, welcome back to the show.

Laura Ullrich: Thanks so much for having me. Happy to be here.

Sablik: You've been on the show a few times now to talk to us about community colleges, which is a research focus for the Richmond Fed that you've been leading for a couple of years. The last time you were here, you discussed a new community college survey that you were just launching. So, I'm excited to have you back on now to talk about the results of that survey.

In that last episode, you discussed how the currently available measures of higher education, enrollment, and achievement are not the best suited to community colleges because their student body tends to look quite a bit different from the student body of a four-year institution. Can you refresh us on some of those key differences?

Ullrich: Google "graduation rate X University" and "graduation rate Y community college" and you'll probably see there's a very big difference between those rates. Community college rates are going to be considerably lower. The way those rates are calculated are limiting for community colleges.

The basic story is that the group of students that are used to create that graduation rate, which is called a cohort, is full-time, first-time students at a university. These are students that take at least 12 credit hours a semester and they're in college for the first time. That fits well for a lot of universities. At Washington and Lee, a prestigious liberal arts university in Virginia, 96 percent of their students that started last year are full-time, first-time. At Tidewater Community College, it's 35 percent.

Sablik: That's a big difference.

Ullrich: Yeah. At Virginia Tech, it's 85 percent. So, that cohort does a pretty good job capturing students at four-year colleges, not a good job at two-year colleges because they have so many part-time students and so many students who tried college before somewhere else, it didn't work, and are coming back.

So that's the denominator. The other issue is the numerator of that graduation rate, which looks at the percentage of students who graduated with a degree, added to their schooling via an associate degree, or got a certificate that was more than 16 credit hours. Again, that works well at a four-year school because everybody that started Washington and Lee last year is there to get a bachelor's degree. That's why you go to Washington and Lee.

But at a community college, there are students that have all sorts of different goals. If a student, for example, went to Reynolds Community College for one year and then transferred to VCU, in the traditional graduation rate measure that student is counted as a failure for Reynolds. As we dug into the data, we just felt like that wasn't really the right way to be looking at student success. Also, there are a lot of students who go to community colleges to get shorter-term credentials and licensures.

We're basically broadening both the numerator and the denominator to redefine student success.

Sablik: Thinking about these national enrollment figures, you've written about how the way those are measured can disadvantage community colleges when it comes to receiving state or external funding. Can you talk a little bit about how that works?

Ullrich: As I started talking to community college leaders, they immediately say, "That data doesn't do a good job explaining what we do. Just don't use that data." The reality is this is the data people are using. Students are using these data, teachers are using these data — and counselors, policymakers, legislators, funders — to try to see how schools are performing. If we're not measuring the full value of the community college, then the full value may not be getting managed.

Our measure isn't perfect. I don't think there's a way to create something that's truly perfect. But we do feel like we're looking at this through a workforce lens [and] through an employment lens because of our employment mandate. And we feel like our measure better exemplifies really what these schools are doing.

Sablik: The new Richmond Fed Survey of Community College Outcomes that you've launched seeks to overcome some of these challenges by capturing the different types of students who attend community colleges, as well as different measures of success beyond just getting a traditional degree.

When we spoke last year, you had just received responses from an initial pilot cohort of 10 schools. How many schools ended up participating in the final survey and what states do they represent?

Ullrich: This year, we're in what we're calling "extended pilot mode." We're still working to get our survey tool exactly how we want it and we're working with the people providing data to make sure that all of that is in alignment. So, everything we're releasing this year is kind of like experimental data subject to change.

We were really excited to get four of our five states to participate. We have all of the community colleges from South Carolina, Virginia, and Maryland and all but one school in West Virginia. We have a total of 63 schools in our extended pilot.

Sablik: Great. So, what does success at these community colleges look like?

Ullrich: At the 63 schools that participated in our extended pilot survey, 28.2 percent of students attain an associate degree or a certificate of more than 16 credit hours within three years of beginning school.

Sablik: And that's the previous kind of traditional metric?

Ullrich: Yeah, that's the traditional metric. With our success rate, we look at it four years after [a student begins school]. The reason we did four years is because we're including part-time students — it's important to give them a little bit more runway. We include students in that metric that either graduated with an associate degree; got any industry recognized certificate, credential or licensure; successfully transferred to a four-year school; or persisted — and persisted means they're still enrolled, in good standing and still working towards degree completion.

We found our success rate to be 51.8 percent. So, 28.2 percent compared to 51.8 percent. It's quite a big difference.

Sablik: Do you also consider noncredit programs, like for workforce development?

Ullrich: That is a great question. You'll hear community colleges call them "sides of the house" — the credit side of the house and the noncredit side of the house. All this data we've been talking about so far is the credit side of the house. These are traditional college courses where you receive credits.

Many of the workforce programs — so if you think about things like commercial driver's licensing, welding certificates, phlebotomy — tend to be housed on the noncredit side of the house. Those data are not collected at the national level. We did collect them in our data. The numbers I just gave you does not include noncredit, but we do have noncredit.

It is important to note that those data are very messy. We knew it was going to be messy because noncredit is not funded in the same way credit is in four out of our five states — Maryland is the exception. So, in most states, the data for noncredit are not collected in the same system as the credit students.

Sablik: Okay.

Ullrich: In some states, it's not collected at all, really. They might just have Excel spreadsheets. It's not in a database management system. So, collecting the data is very, very messy.

But we were able to get data from all for the states that participated. And we can see that noncredit students do make up a sizable portion of the total enrollment at Fifth District community colleges.

Sablik: In our last conversation, you also mentioned that you're going to be collecting data on high school students who attend community colleges. Why is that important to you, and what did you end up finding?

Ullrich: I knew when I started this work that dual enrollment — high school students taking college classes — was something that all community colleges did. I knew it for my own children — two of my three sons have now taken dual enrollment classes. But I didn't realize that it makes up a sizable share of their total student population.

In Virginia, for example, 21.7 percent of all students enrolled in credit programs are high schoolers. In our survey, the highest percent was 48 percent.

Sablik: Wow. So nearly half.

Ullrich: Yeah. The data that are collected nationally ... this is really interesting and it gets so confusing, Tim, but I'm going to give you one little example.

You cannot be counted in a school's graduation rate as getting an associate degree unless you already have a high school diploma. However, there are a lot of students that are getting their associate degree at the same time they graduate high school. But because college graduation tends to be before high school graduation, they do not count in the data. So, there are all these very convoluted reasons why these high schoolers are not showing up in the data fully.

Also, you are not able in IPEDS [Integrated Postsecondary Education Data System] to identify the high school students directly. You can sort by those under age 18, but, of course, not all high school students are under age 18. So, we are collecting data on these students completely separately.

Sablik: You also mentioned last time that you were hoping to learn more about some of the additional services provided by community colleges, such as health care or food pantries. Did you learn anything more about that from the full survey?

Ullrich: We did. These are broadly called wraparound services. These are services beyond the kind of traditional educational mission of the college and are used to support students and their success.

It's funny, when we decided to collect this [data], we did it because schools told us they weren't reporting this anywhere and they kind of felt like, "Hey, this is something we're doing that's important. We should be reporting it." We didn't realize how much interest there would be in these datasets since they haven't been reported previously and they're not collected. The states really don't collect them either.

It's been a little difficult to collect the data. So, of the 63 schools right now, we just have this data on 41 of them. But we're working to get the rest of them.

There are some wraparound services that almost all schools have: 100 percent of the schools in our survey have academic counselors and 95 percent refer students to offsite mental health counseling if they need it, but only 22 percent have childcare and only 12 percent have on-site health services; 80.5 percent of schools have offices where they connect current students to other social service programs. So, if you qualify for SNAP benefits or TANF or other things, they'll help you access that application process.

These data are not going to be super interesting until we have a panel of data. But once we have that built, we will eventually be able to observe through research did adding childcare help enrollment at a school, right, or did it help success? Or, are the schools that have food pantries, for example, seeing higher success rates amongst their low-income students? We'll be able to do more research on that. I'm really excited about what these data can tell us a couple years down the road.

Sablik: Well, that raises a good point. What are your next steps for this project?

Ullrich: You know, there are some question marks around how big we could get with this because we've gotten such good feedback from community colleges, also stakeholders — anywhere from funders to other researchers, academics, state agencies — a lot of different groups.

Our goal right now is to have all 122 Fifth District community colleges in our survey next year. North Carolina has 58 colleges and North Carolina didn't participate this year. That's just because they had some turnover in the state system office, so I'm very hopeful that that'll happen next year.

Our other goal is to build an internal system where it's easy for schools to submit their data, or some of the data are processed automatically. Then, on the back end, you're able to get the data by school.

There have been some other groups that have tried to do similar things, but it was "pay to play" — you either had to pay to participate in the survey or the data costs money. This will be a public good. We want to have it available on our website by school next year so that the stakeholders can use it.

Sablik: Well, Laura, thanks so much for joining me today to update us on this exciting project.

Ullrich: Thanks so much for having me. Always a pleasure.

Sablik: Listeners can find the full data release on our website, as well as all the other work we've published on community colleges on our topics page. We'll also link to that survey result in the show notes. And if you enjoyed this episode, please consider leaving us a rating and review on your favorite podcast app.