 |
Basic Classification Technical Details
Basic Classifications are time-specific snapshots of institutional attributes and behavior based on data from 2002-03 and 2003-04. Institutions might be classified differently using a different timeframe.
Associate’s Colleges
Institutions were included if their highest degree conferred was the associate’s degree or if bachelor’s degrees accounted for less than 10 percent of all undergraduate degrees (according to 2003-04 degree conferrals as reported in IPEDS). As in previous editions, these categories were limited to institutions that were not eligible to be classified as Tribal Colleges or Special Focus Institutions.
The Associate’s Colleges categories are based on a classification scheme developed by Stephen Katsinas, Vincent Lacey, and David Hardy at The University of Alabama. Katsinas and Hardy conducted the analysis and provided the institutional classifications. The following criteria determined category assignment in this analysis:
Rural-, suburban-, or urban-serving. Urban-serving and suburban-serving institutions are physically located within Primary Metropolitan Statistical Areas (PMSAs) or Metropolitan Statistical Areas (MSAs), respectively, with populations exceeding 500,000 people according to the 2000 Census. Institutions in PMSAs or MSAs with a lower total population, or not in a PMSA or MSA, were classified as rural-serving.
Size (public rural-serving categories). Institutional size is based on full-year unduplicated credit headcount, where small is defined as less than 2,500; medium as 2,500 through 7,500; and large as greater than 7,500. Size is based on IPEDS data for 2003-04.
Single-campus. Suburban- and urban-serving colleges were identified as single-campus if they have one primary physical campus under the institution’s exclusive control and governance, at which the institution provides all courses required to complete an associate’s degree. A single-campus college may offer educational services at more than one site if the other sites are not under the institution’s exclusive control and governance, or if all courses required for the associate’s degree cannot be completed at the other sites (examples include leased spaces; shared sites or regional education centers that provide offerings of multiple institutions; or satellite locations that do not have the full range of an institution’s programs and services).
Multicampus. Suburban- and urban-serving colleges were identified as multicampus if (a) they have more than one primary physical campus under the institution’s exclusive control and governance, each of which provides all courses required to complete an associate’s degree, or (b) they are part of a district or system comprising multiple institutions, at any of which students can complete all requirements for an associate’s degree, and that are organized under one governance structure or body. Institutions were not classified as multicampus simply due to control by a single statewide governing board. Multicampus institutions may report their data as separate entities in the IPEDS system, or they may participate as a single reporting entity.
Special Use. Colleges were identified as special-use institutions if their curricular focus is narrowly drawn and they are not a part of a more comprehensive two-year college, district, or system.
Other considerations. Public 2-year institutions under the governance of a 4-year university or system are included in the "Public 2-year Colleges under Universities" category. Baccalaureate-granting institutions where bachelor's degrees account for fewer than 10 percent of undergraduate degrees are designated as "Primarily Associate's" colleges.
Doctorate-granting Universities
Institutions were included in these categories if they awarded at least 20 doctorates in 2003-04. First-professional degrees (J.D., M.D., Pharm.D., etc.) were not counted for the purpose of this criterion, nor were doctoral degrees in audiology or rehabilitation and therapeutic professions, except when we were informed that the degrees were research doctorates (the IPEDS Completions data do not identify degree titles, so professional-practice doctorates such as Aud.D. or D.P.T. cannot be explicitly identified). Institutions with lower levels of doctorate production can be identified by using Custom Listings to intersect categories of the Basic and Graduate Instructional Program classifications. As in previous editions, these categories were limited to institutions that were not identified as Tribal Colleges or Special Focus Institutions.
Level of research activity. Doctorate-granting institutions were assigned to one of three categories based on a measure of research activity. It is important to note that the groups differentiate solely with respect to level of research activity, not quality or importance.
The analysis examined the following correlates of research activity: research & development (R&D) expenditures in science and engineering (S&E; “science and engineering” is defined by NSF to include the social sciences); R&D expenditures in non-S&E fields; S&E research staff (postdoctoral appointees and other non-faculty research staff with doctorates); doctoral conferrals in humanities fields, in social science fields, in STEM (science, technology, engineering, and mathematics) fields, and in other fields (e.g., business, education, public policy, social work). These data were statistically combined using principal components analysis to create two indices of research activity (each index was a component score for the first principal component). One index was based on aggregate levels of these factors, and the other assessed per-capita research activity using the expenditure and staffing measures divided by the number of full-time faculty whose primary responsibilities were identified as research, instruction, or a combination of instruction, research, and public service. The values on each index were then used to locate each institution on a two-dimensional graph. We calculated each institution’s distance from a common reference point, and then used the results to assign institutions to three groups based on their distance from the reference point. Thus the aggregate and per-capita indices were considered equally, such that institutions that were very high on either index were assigned to the “very high” group, while institutions that were high on one (but very high on neither) were assigned to the “high” group.
Before conducting the analysis, raw data were converted to rank scores to reduce the influence of outliers and to improve discrimination at the lower end of the distributions where many institutions were clustered. Index scores based on the raw data and on the rank-order data had correlations of .881 (aggregate) and .813 (per-capita).
For more information on this methodology and descriptive information for each of the categories, see Related Files at the end of this page.
Data sources. Doctoral conferrals by field were based on IPEDS Completions data reporting 2003-04 degree conferrals. Faculty counts were from the IPEDS Employees by Assigned Position (EAP) data for fall 2003. R&D expenditures came from the NSF Survey of Research and Development Expenditures at Universities and Colleges for fiscal year 2003. Research staff data came from the NSF Survey of Graduate Students and Postdoctorates in Science and Engineering for fall 2002. These were the most current and complete data available at the time of our analysis, and we judged currency to be more important than temporal alignment of all data sources.
Reporting of non-S&E expenditures was mandatory for the first time for the FY2003 survey, but not all institutions reported these data. Of 277 institutions analyzed, 53 did not report non-S&E expenditures. For these institutions, we implemented a regression-based imputation scheme to fill in the missing data.
In some cases, the NSF data were reported at a higher level of aggregation than is needed for classification purposes (i.e., a university system comprising multiple campuses that are distinct entities for classification purposes, but that are reported together as a single entity in the NSF data). Because the Graduate Students and Postdoctorates data were reported by department, we made manual changes to create campus-level records. For the R&D Expenditures data, it was not possible to disaggregate the data at the campus level. We adopted the allocation scheme developed by TheCenter at the University of Florida (now housed at Arizona State University; see http://mup.asu.edu/DataNotesIntroText.html [accessed March 15, 2007]).
Doctorate-granting institutions that did not participate in the NSF data collections were automatically assigned to the "Doctoral/Research Universities" category.
Master’s Colleges and Universities
Institutions were included in these categories if they awarded at least 50 master’s degrees in 2003-04, but fewer than 20 doctorates (as defined above). Exclusively graduate-level institutions that awarded fewer than 50 master’s degrees and fewer than 20 doctoral degrees were also included. (Institutions with lower levels of master’s degree production can be identified by using Custom Listings to intersect categories of the Basic and Graduate Instructional Program classifications.) As in previous editions, these categories were limited to institutions that were not identified as Tribal Colleges or Special Focus Institutions.
Program size. Master’s program size was based on the number of master’s degrees awarded in 2003-04. Those awarding at least 200 degrees were included among larger programs; those awarding 100–199 were included among medium programs; and those awarding 50–99 were included among smaller programs. The smaller programs group also includes institutions that awarded fewer than 50 master’s degrees if (a) their Enrollment Profile classification is Exclusively graduate/professional or (b) their Enrollment Profile classification is Majority graduate/professional and they awarded more graduate/professional degrees than undergraduate degrees.
In December 2006, some institutions that had been classified among Master's Colleges and Universities were given the option of classification among Baccalaureate Colleges based on their overall profile (see below).
Baccalaureate Colleges
Institutions were included in these categories if bachelor’s degrees accounted for at least 10 percent of all undergraduate degrees and they awarded fewer than 50 master’s degrees (2003-04 degree conferrals). In addition, these categories were limited to institutions that were not identified as Tribal Colleges or as Special Focus Institutions.
Among institutions where bachelor’s degrees represented at least half of all undergraduate degrees, those with at least half of bachelor’s degree majors in arts and sciences fields were included in the “Arts & Sciences” group, while the remaining institutions were included in the “Diverse Fields” group.
Institutions where bachelor’s degrees represent at least 10 percent but less than half of undergraduate degrees were assigned to the Baccalaureate/Associate’s category.
The analysis of major field of study is based on degree conferral data (IPEDS Completions). Up to two majors can be reported, and both were considered for this analysis. Thus for an institution with 1,000 bachelor's degree recipients, half of whom completed double majors, the analysis would consider all 1,500 majors. The mapping of fields of study to arts & sciences or professions is documented in an Excel file available on the Downloads page.
In December 2006, some institutions that had been classified among Master's Colleges and Universities were given the option of classification among Baccalaureate Colleges based on their overall profile. These institutions met the following criteria:
- FTE enrollment of fewer than 4,000 students
- Highly residential (Size & Setting classification)
- (a) Enrollment Profile classification of Very high undergraduate or High undergraduate, combined with No graduate coexistence or Some graduate coexistence (Undergraduate Instructional Program classification), or (b) Enrollment Profile classification of Majority undergraduate combined with No graduate coexistence.
Special Focus Institutions
The special-focus designation was based on the concentration of degrees in a single field or set of related fields, at both the undergraduate and graduate levels. Institutions were determined to have a special focus with concentrations of at least 75 percent of undergraduate and graduate degrees. In some cases the percentage criterion was relaxed if an institution identified a special focus on the College Board’s Annual Survey of Colleges, or if an institution’s only recognized accreditation was from an accrediting body related to the special focus categories.
Tribal Colleges
Tribal colleges are defined as members of the American Indian Higher Education Consortium, as identified in the IPEDS Institutional Characteristics data.
|
|
| Related Files
Research Activity Index |
This spreadsheet file provides more details on the research indices used to classify doctorate-granting institutions. It includes documentation of the methodology; descriptive statistics of the variables analyzed, by classification category; and plots showing the distribution of institutions according to the indices.
|
| research activity index documentation (XLS) » |
|
|
|
|
|
|
|
|
| Answers to questions you may have about the Carnegie Classifications. |
|
|
|
|
|
