Peer Grouping: The Refinement of Performance Indicators

Community colleges operate under much scrutiny these days, and these institutions have experienced a growing emphasis on performance indicators as paths to institutional accountability. California's system of 109 community colleges recently developed and implemented an innovative accountability program that used peer group comparison as one of its elements. This article describes California's use of peer grouping in terms of its development, mechanics, and implications for the future. Introduction For purposes of evaluation and planning, higher education institutions have often tried to compare themselves to other institutions on selected performance indicators. This common practice gives college officials some idea of "where their institutions stand" with respect to a certain performance indicator. But college officials generally recognize that such a comparison is only valid and fair if it involves colleges with similar characteristics. In fact, the Integrated Postsecondary Education Data System (IPEDS) tries to address this historic need by integrating a "peer selection" procedure on its website (US Department of Education, 2007). The Carnegie Foundation has provided a framework for identifying peer institutions since 1970 (Carnegie, 2007; McCormick & Zhao, 2005). A study from the National Center for Educational Statistics (NCES) actually identified groups of similar colleges through a cluster analysis of various postsecondary institutions in the U.S. (Phipps, et al., 2001) and a new analysis extends that work (Goan & Cunningham, 2007). The National Survey of Student Engagement (NSSE) system has attempted to help an institution to find its institutional peers in its data system (National Survey of Student Engagement, 2007). Finally, one publisher even devoted an entire volume of its series to the topic of peer identification (McCormick & Cox, 2003). Hurley (2002) documents the various efforts that have focused upon finding peer institutions for community colleges. Postsecondary educational institutions have company when it comes to this need to compare themselves for evaluative purposes. California's secondary education system essentially administers its accountability program, with its Academic Performance Index (API), in such a way that each school's performance has a set of similar schools (its "peers") to facilitate a fair comparison (California Department of Education, 2007). In any case, to determine which institution is a "similar school" in the postsecondary environment, an analyst can undertake an extensive analysis to categorize or classify a specific population into groups so that each group contains institutions that closely resemble one another. Researchers often label such an analysis as peer grouping. Researchers use peer grouping to "level the playing field" or to avoid the problem of "comparing apples to oranges." In essence, peer grouping is a method of arranging data so that people can compare institutions while "controlling" for specific, measured factors (often referred to as "confounders") that would otherwise lead to unfair and invalid comparisons of performance. How California Constructed Its Peer Grouping California used peer grouping to satisfy an assortment of legislative, political, and technical criteria. California's central office for the state's 109 community colleges (henceforth known as the System Office) had responsibility for implementing state-level mandates that affect these institutions. The System Office therefore had authority to plan and implement a peer grouping plan that the state legislature had passed in 2004 (Chancellor's Office, 2007). Historically, state oversight bodies (such as the California Department of Finance and the Legislative Analyst's Office) have preferred straightforward forms of evaluation such as institutional rankings. Rankings can give analysts quick, if simple, insights into relative performances of institutions. …