Our primary interest is to estimate cumulative value-added earnings (VAE), net of costs, for cohorts of entering students over an appropriate follow-up period after entry. We refer to this estimate as “cumulative net VAE.” 

A cohort’s cumulative net VAE can be summarized as follows:

Cumulative Net VAE = (Cumulative VAE) - (Cumulative Net Costs)

Cumulative Net VAE = (Cumulative Earnings - Cumulative Comparison Group Earnings) - (Cumulative Net Costs)

Students’ value-added earnings (VAE) — in a particular year or cumulatively over our follow-up periods — are the differences between students’ actual earnings and their earnings in a counterfactual scenario in which they do not enroll in the institution in question or pursue postsecondary education elsewhere during the relevant follow-up period.

To estimate comparison group earnings, we assemble, and observe the earnings of comparison groups of individuals who closely resemble students in our entering cohorts, minus the choice to enroll in the institution in question or pursue postsecondary education elsewhere.

For more information on our approach to estimating students’ value-added earnings, see the Baseline Sample and Data Sources, Treatment Cohorts, and Comparison Groups sections of the website.

We estimate net costs for each student in our treatment cohorts, in each semester of their enrollment. 

Our tabulation of net costs includes estimates of students’ expenses for tuition, fees, books, and education supplies, net of students’ federal and state tuition grants, waivers, and exemptions.

For our tuition and fees estimates, we use students’ annual enrollment intensity to assign to them a pro-rated portion of the relevant IPEDS-published tuition and fee schedule at their institution. We determine which tuition and fee schedule at an institution (in-state, out-of-state, or in-district) is relevant to each student based on student-specific residency information.

We do not include in our net cost calculations estimates of students’ expenses related to on-campus or off-campus housing, meals, transportation, or childcare. We make this choice partly because they are difficult to observe and partly because students often incur these expenses whether they enroll or not.

In our net cost calculations, we also do not include costs that arise for students if they elect to pursue degrees or credentials during the relevant follow-up period other than the one they are pursuing in the institutional cohort in which we place them. For example, if an associate’s degree-seeking student in one of our institutional cohorts elects at some point during the follow-up period to pursue an associate’s degree or a bachelor’s degree at another institution, we do not tally costs related to that choice. 

We study three types of entering cohorts in an institution in a given academic year: institutional cohorts, programmatic cohorts, and demographic cohorts. We sometimes refer to these entering cohorts as “treatment cohorts.”

Institutional cohorts include all students who enroll at a given institution in the same academic year to pursue the same type of degree (i.e., a certificate, an associate’s degree, or a bachelor’s degree).

A single institution can have one, two, or three institutional cohorts arise in a given year (one for each degree type it offers).

We classify students at the time of their entry as certificate seekers, associate’s degree seekers, or bachelor’s degree seekers based on their declarations of program and course enrollments in their first semester.

Programmatic cohorts are subsets of institutional cohorts that include students who pursue the same program of study or major (“program”). 

We assign certificate seekers and associate’s degree seekers to a program upon entry, based on their program choice at entry.

We assign programs to bachelor’s degree seekers, who often do not declare programs at entry, in the second semester of their second year after enrollment, at which point most bachelor’s degree-seeking students have settled on a program of study. 

If bachelor’s degree-seeking students exit an institution before the end of their second year, we assign them to a program based on their last declared program and their course selections while enrolled. If bachelor’s degree-seeking students switch programs after the second semester of their second year, we do not change their program assignment.

We use the National Center for Education Statistics’ 2-digit or 4-digit Classification of Instructional Program (CIP) codes for our list and classifications of programs.

When reporting findings for programmatic cohorts, we collapse programmatic cohorts based on 2-digit and 4-digit CIP codes into 17 general program categories. We make the choice not to report findings for programmatic cohorts organized by 2-digit or 4-digit CIP codes because it would be unwieldy (they are numerous) and would frequently trigger measurement problems associated with small cohort sizes.

We also report findings for all STEM programs as a group and for all non-STEM programs as a group. 

Demographic cohorts are subsets of institutional cohorts that include students who share a demographic trait.

In this study, we report findings on demographic cohorts organized by students’ household income, by students’ math achievement on standardized tests in high school, and by students’ age at entry into the institution.

We create household income-related demographic cohorts based on students’ status as “economically disadvantaged” in high school, a designation that depends on students’ eligibility in high school for the US Department of Agriculture’s free or reduced-price lunch program and on students’ eligibility for other public subsidies, including the US Department of Health and Human Services’ Temporary Assistance for Needy Families program.

We create math achievement-related demographic cohorts based on the quartile into which students fell on standardized tests of high school math in Texas.

We create age-related demographic cohorts based on three age ranges at the time of entry into the institution: under 20 years of age, 20-24 years of age, and 25+ years of age.

We study entering cohorts for different follow-up periods after entry, as follows:

• We study cohorts of bachelor’s degree-seeking students for 15 years from entry.
• We study cohorts of associate’s degree-seeking students for 10 years from entry.
• We study cohorts of certificate-seeking students for 5 years from entry.

We recognize that our choice of follow-up periods, while reasonable, is debatable and elective. 

One of our rationales for choosing the follow-up periods described above is that they allow students a reasonable amount of time, within the follow-up period, to complete their degree or certificate and to enter the labor market.

In our sample, average time-to-completion for bachelor’s degree-seeking students, associate’s degree-seeking students, and certificate-seeking students was 5 years (in the 15-year follow-up period), 4.25 years (in the 10-year follow-up period), and 1.5 years (in the 5-year follow-up period), respectively.

We expect, in future iterations of our research, to expand our follow-up periods.