Want to know whether your college is better than its peers? Want to boast that your college is better than its peers?

A long awaited new system of grouping colleges, released Thursday, won't give you definitive answers, and (if used properly) shouldn't yield boasting rights. But the new classifications by the Carnegie Foundation for the Advancement of Teaching might just help colleges identify true peers and look for weaknesses that need fixing.

The Carnegie Classifications have traditionally grouped institutions by degrees offered, so that doctoral institutions were in one group and community colleges in another, and so forth. The new classifications take a very different approach. Institutions are grouped (multiple times) based on what is taught, to whom, and in what setting. The old system -- with some revisions -- will still be used when a new list of institutional groupings is released next month.

Under the old system, once institutions shared certain characteristics about the top degree they offered, they were considered similar. So two institutions that both offered many Ph.D. programs would be grouped together -- even if they both primarily educated undergraduates and did so in very different ways. The many new classifications allow one to see whether colleges educate primarily undergraduates, whether the undergraduates are primarily seeking bachelor's or associate degrees, whether students are primarily studying the liberal arts and sciences or professional fields, whether most students have transferred or not, whether students are full time or part time, and on and on.

With these new rankings, institutions that might have been considered identical under the old rankings are suddenly not identical. Both Ohio State University and the University of Michigan have extensive research and graduate programs, and both are large public universities in the Midwest. But the new classifications point to key differences, such as the fact that there is more of an arts and sciences orientation among Michigan undergrads.

Lee S. Shulman, president of Carnegie, said that the old system, by itself, failed to reflect "the complexity and ambiguity" of American higher education. "One single classification simply can't bear the weight of all of the ways that the classifications are used and misused," he said.

While the classifications were created as a tool for researchers, they are commonly used to help states set policies (research institutions might get different appropriations formulas than teaching oriented colleges), to help foundations give out grants (programs might be set up for certain kinds of colleges), and -- to the dismay of many educators -- to help magazines develop rankings. While Carnegie has never ranked colleges in order, U.S. News and World Report uses the Carnegie Classifications to create the categories in which it ranks colleges, and there was much U.S. News bashing at the press briefing where Carnegie officials released the new system.

While colleges will no doubt continue to boast about their rankings in various places, Shulman and other Carnegie officials said their hope was that the range of comparisons would encourage colleges to find true peer institutions and then learn from them. And a college might find that it has one set of peers in undergraduate curriculum and another for transfer rates.

Alexander C. McCormick, a senior scholar at Carnegie who directs the Carnegie Classification projects, said that the old ratings were so successful that they "put blinders on us." As a result, he said, educators haven't compared their institutions outside of their overall, traditional Carnegie classification.

Carnegie officials said that one benefit of the new set of classifications was that it might encourage a more sophisticated look at the role of for-profit higher education, since the various measures used in the new system will group together for-profit and nonprofit institutions.

Using the Carnegie search engine, for example, a user might find that the University of Phoenix's Houston campus shares much (but not all) in common with the Indiana Institute of Technology (a private, nonprofit college) and the Colorado School of Mines (a public university). While the three institutions no doubt have many differences, they may not have previously thought of themselves as being peers in any meaningful way. But based on some enrollment and curricular measures, they are.

Peter Ewell, vice president of the National Center for Higher Education Management Systems, is a leading education researcher and he had earlier been invited to "test drive" the new classifications and to share his thoughts at the briefing. Generally, Ewell had high praise for the new system, saying that he found an amazing number of ways to search and group institutions and the most of the groupings he found made sense.

He said that he noticed limitations with two sets of institutions: those with "multiple identities," and community colleges. As an example of the former, he cited Regis University, an institution that combines a traditional undergraduate liberal arts program (primarily for full-time students) with an extensive online professional program (primarily with part-time students). When all of the Regis data is used in the system, you don't necessarily get great comparisons for either part of Regis, he said.

Community colleges, Ewell said, were not being divided into truly different groups by the new classification tools. On many of the sorting mechanisms used by Carnegie, most community colleges come out the same (their programs are all undergraduate, many of their students are enrolled part time, etc.).

McCormick acknowledged that this was a problem, and said he hoped that the project would evolve to develop better ways to sort community colleges. In fact, McCormick said that he considered the inability of the old Carnegie system to sort community colleges "the single biggest weakness" of the classification.

A problem faced by researchers, McCormick said, was that many of the tools for sorting institutions by curricular emphasis have data based on degrees awarded. Many community college students never earn degrees -- and never sought degrees in the first place.

When the traditional classifications are released next month, McCormick said that there would be some differentiation among community colleges, which have historically just appeared as a single group. They will be grouped instead as rural, suburban and urban community colleges.

When the revised traditional classifications are released, they will be accompanied by yet a third approach to the classifications. A new set of voluntary categories -- such as civic engagement -- will be introduced. Colleges from a variety of categories may decide to submit data to be grouped in these classifications and the foundation has been working with a small group of colleges over the last year to refine that set of groupings.

Carnegie officials were up front about the fact that they were delaying the release of the traditional classifications intentionally. Asked if they were holding back the traditional classifications to get people to think about the new ones, Shulman smiled and said Yes.