Types of institutions in the data set
The data set includes descriptive information about higher education institutions outside the United States and its territories that are (or were) American in some way. An institution is included in the data set if it has met one or more of the following criteria:
Beyond these initial criteria, other requirements for inclusion are:
We primarily developed the data set through publicly available information. We identified most institutions for inclusion via web searches. A team of graduate students from Teachers College, Columbia University developed the initial data set, focusing exclusively on currently operational independent American higher education institutions. That team conducted country-by-country searches for “American university of [country],” leading to a list of approximately 80 institutions. The Global American Higher Education research team later expanded this data set by drawing from Long (2020) and Lutz (1971) to include non-operational independent institutions established by American citizens, viz. defunct missionary colleges in the former Ottoman Empire and China.
We then expanded the dataset to include other institutional types. We brought in qualifying American institutions from the international campuses list compiled by the Cross-Border Education Research Team (C-BERT). Next, we used lists on accreditors’ websites to identify regionally accredited overseas institutions. We consulted the U.S. Department of Education’s Closed School Search File to identify U.S. institutions that operated branch campuses or microcampuses abroad. Finally, we referred to a Government Accountability Office (GAO) report that translated official Chinese sources for lists of U.S. universities approved by the Chinese Ministry of Education to operate cooperative education institutions or programs. For each of these sources, we conducted web searches to determine if a potential institution qualified for inclusion. This process resulted in snowball sampling that led to discovery of yet additional qualifying institutions.
After identifying institutions, we scanned their websites for relevant information. Institutional websites are the most common source of information. In some cases, though, we make use of other secondary sources. Where not noted, source material is available by request.
Types of information in the data set
The data set includes a wide range of information for each institution (where applicable), such as:
We note of each institution whether it is active or inactive. Active means that it offers at least one relevant bachelor’s degree-granting program at present. Inactive means that it is either a closed institution or no longer offers a qualifying program. For example, London Metropolitan University (est. 2002) was regionally accredited in the United States from 2007-2020. It still operates today, but not as a qualifying institution in our database. We therefore mark it as inactive.
The data set tracks two sets of dates for each institution: 1) the years it was established and, where applicable, closed as well as 2) the years its tertiary programs started and, where applicable, ended. Such distinctions are useful for many institutions in the data set. For example, Lebanese American University was first established as a girls school in 1835, but did not offer tertiary level programming until 1924. For institutions that qualify for the data set on accreditation status alone, the university’s first year of accreditation is used as the tertiary start date. Abu Dhabi University was established in 1976, but did not achieve U.S. regional accreditation until 2016. It is only on this latter date that the institution qualified for inclusion.
We use a binary profit/non-profit variable for classifying an institution’s approach to revenue. This variable is complete for all institutions in the data set. For branch campuses, international joint universities, and microcampuses, we ascribe the revenue model used by the U.S. university involved. For independent institutions, we ascribe non-profit status to those with websites that indicate it operates a non-profit model. We regard independent institutions that do not make this claim as proprietary. For accredited institutions, we list the revenue model posted on accreditor websites.
We obtained geographic coordinates for campuses using Google Maps. In most cases, coordinates reflect the precise location of the campus. In instances where precise coordinates were not available, we used city coordinates as proxies. We use the United Nations member list for our country framework and the Sustainable Development Goals for regional groupings.
Wherever possible, we use institutional websites to source the most up-to-date enrollment figures (all degree-seeking students). However, many institutions do not publish this information. In such instances, we sought out estimates from other websites. In general, it is best to regard enrollment figures as approximations.
The dashboard allows users to filter by a variable called accredited degree (yes, no, N/A). This variable is inclusive of but distinct from institutional accreditation. The purpose of the filter is to show institutions where a student could obtain a degree accredited in the United States. Some independent institutions are not accredited in the United States, but host microcampuses that enable students to earn accredited degrees. American University of Phnom Penh is not accredited, but its students are eligible to earn degrees from the University of Arizona. Conversely, some microcampuses are operated by regionally accredited institutions but do not award accredited degrees. Troy University is accredited in the United States, but the degrees it awards at its microcampuses in Vietnam and Malaysia are not. The same is true of some international joint universities. Yale-NUS College has a foundational relationship with Yale University, but not with its accreditation status. Inactive institutions are listed as N/A.