The location of developmental education in community colleges: a discussion of the merits of mainstreaming vs. centralization.

This study compares mainstreaming and centralizaton, two ways in which community colleges organize developmental education. Based on previous literature, the two models are compared in terms of instructional quality, ancillary services, teacher characteristics, student reactions, and reputation of remediation. Pending empirical evidence for the superiority of one model to another, recommendations are offered to college administrators and state policy makers for maximizing the effectiveness of each one.

**********

Ineffective high school education and increasing ethnic and linguistic diversity are combining to make developmental education critically important for individuals who wish to participate in postsecondary education. Developmental education has become an integral part of the community college mission (Carnevale & Desrochers, 2001; Levin, 2001). With their open admissions policy and commitment to serving a wide range of students in local communities, community colleges have historically played an important role in higher education by offering instruction in basic reading, writing, and math skills to enable academically underprepared students to master the college curriculum. As Levin (2001) stated,

   For many students in either large or small communities, the community
   college is the only public educational institution that will accept them
   for college-level studies given their high school academic performance.
   Furthermore, of the many types of postsecondary institutions facing
   students who are unprepared for college-level studies, the community
   college is the only institution whose legal and social mandate is remedial
   education. (p. xii)

Community college students display a number of academic and personal risk factors that are associated with low rates of persistence and achievement (McClenney, undated). In response, the colleges attempt to increase student preparedness for the college curriculum in a variety of ways including precollege level reading, writing, and math courses (variously termed "developmental education" and "remediation"); academic tutoring in learning assistance centers while students are enrolled in college-level courses; and instructional modifications such as writing-across-the-curriculum in discipline classrooms. Content-area remediation is also provided in some institutions in the form of supplemental instruction, a peer-tutoring model where students who have earned high grades in discipline courses (e.g. biology and history) lead study groups for students who are failing in those classes. However, among this complex array, developmental education courses are the most visible form of remediation in community colleges because these courses are clear catalog offerings in which basic skills instruction is formalized.

Organizational Approaches: Mainstreaming and Centralization

Given the importance of developmental education courses, the question has arisen as to whether they should be integrated into regular departments, here called mainstreaming, or housed in separate organizational units, referred to as centralization. The distinction between maistreaming and centralization is an important issue for college policy because the organization of developmental education may have direct impact on its quality (Boylan, Bliss, & Bonham, 1997). When developmental education is mainstreamed, precollege level remedial courses are offered in academic departments, such as English or mathematics, whose main purpose is to offer college-level courses applicable to associate's degrees or certificates. Courses are numbered as part of a sequence that begins with noncredit, remedial level instruction and continues through advanced associate-level preparation. Instructors are all considered faculty of the department in question and are paid through its budget. Working in close proximity in a departmental context permits developmental education instructors to mingle with colleagues who teach college-level courses. In fact, some faculty teach both developmental and credit-bearing courses simultaneously. On the other hand, when remediation is centralized, the remedial courses are offered in a separate department whose sole function is to offer precollege-level courses. Course numbers reflect the separateness of the department, and the faculty may communicate more often with each other than with instructors from academic departments. In addition to courses, the centralized department may offer ancillary support services such as counseling and tutoring. Most of the instructors will be paid from the centralized department's budget although some may have joint appointments with academic departments and teach courses in both (see McKay et al., 1998).

The term "mainstreaming" is used here in the context of stand-alone developmental education courses. The current question is whether such stand-alone courses should be offered in a regular college department or in a separate remedial department. Arendale (1998, cit. Damashek, 1999) uses the term to refer to the replacement of stand-alone courses with a comprehensive system of support available to all students via a learning assistance center. Boylan et al. (1997) use the term "decentralized" for what we are calling mainstreamed developmental education.

Although there have been strong statements in favor of centralization (e.g., Roueche & Roueche, 1999), there is little direct evidence to support this policy. The purpose of this article is to consider the relative merits of each approach. Because there is a shortage of empirical evidence in the previous literature, the discussion is speculative, relying on practitioners' views and relevant data reported in the existing literature. Sources of information for this discussion were journal articles, book chapters, and technical reports on community college developmental education identified in a search of the ERIC and Educational Abstracts electronic data bases, as well as bibliographies, conference presentations, and personal communications with experts in developmental education. Some of the studies relate specifically to two-year colleges and others to four-year institutions hut the organizational issues in both settings are identical.

Current Practices

Roueche and Roueche (1999) have reported that the majority of community colleges across the country mainstream their developmental education programs. Some community colleges have a separate remedial division that teaches their lower level developmental courses, with an academic department teaching the higher level courses. In other colleges, a mixed model may be used, for example, mainstreaming writing and math in their respective content departments, while offering reading courses in a separate developmental education department. According to NCES (1996, Table 10), 54% of community colleges mainstream remedial reading, 59% mainstream remedial writing, and 62% mainstream remedial math courses.

A study of 15 states by the Southern Regional Education Board (Abraham, 1992) found that in community and four-year colleges combined, most developmental education was mainstreamed: 41% of institutions delivered reading remediation in an academic department, (57% in writing, 58% in math) while only about one-third centralized remedial education. Since separate percentages were not provided for community colleges, it is difficult to compare these figures with the NCES data.

Similarly, Boylan, Bliss, and Bonham (1997) examined the organization of developmental education in two- and four-year colleges combined. In a national sample, they found that 52% centralized developmental education. In contrast, the findings of a national survey conducted by the American Association of Community Colleges (Shults, 2000) indicated that only 25% of community colleges centralized their remedial courses, while 15% mainstreamed these courses within academic departments. Although a further 61% of institutions surveyed offered their courses "within their respective subject areas," a term not defined, it is notable that Shults' (2000) findings differ from the NCES (1996) data in finding a low incidence of mainstreaming. Similarly, Grubb and Associates (1999) reported that remediation tended to be centralized, stating, "Within community colleges, remediation is usually organized as an activity separate from the core purposes, isolated in a jigsaw puzzle of developmental reading and writing departments and tutorial programs" (p. 171). Taking into consideration all of these studies, it is not clear whether the trend is for colleges to mainstream or centralize remediation.

Drawing on NCES (1996), Roueche and Roueche (1999) note that community colleges with high proportions of minority students are more likely than are low-minority institutions to centralize rather than mainstream their developmental courses. People from ethnic and racial minority groups may predominate among the segment of students who are academically underprepared. In a national study of students who completed remedial programs, McCabe (2000) found that among individuals whose skills were "seriously deficient" (i.e., students who tested into reading, writing, and math remediation, including at least one lower level course), minority groups were overrepresented (56%). In particular, 51% of all students in this category were women from minority groups. McCabe found that only 20% of the seriously deficient students in his sample completed remediation, compared to 43% for higher functioning students. The tendency of minority-dominated institutions to provide centralized rather than mainstreamed developmental education suggests that this approach may be particularly helpful for lower achieving remedial students.

Beyond the prevalence of the two approaches, Boylan et al. (1997) studied the relation between organizational structure and student outcomes. Based on an analysis of academic data on a random sample of 6,000 developmental education students attending 300 community and four-year colleges, it was found that students attending institutions where developmental education was centralized had significantly higher first-term grade point averages, cumulative grade point averages, retention rates, and math and English grades, compared with students in colleges where remediation was mainstreamed. Unfortunately, the means were not reported so that it cannot be ascertained whether the group differences were in fact educationally meaningful or only an artifact of statistical power associated with large sample size.

Because other empirical evidence is lacking, the remainder of this discussion relies on descriptive studies and practitioner commentary that, although not providing direct evidence, help weigh the advantages and disadvantages of the two approaches. We frame the discussion in terms of a number of educational components that are frequently mentioned in discussions of remedial education: quality of instruction; availability of ancillary support services; teacher motivation and experience; students' reactions; and the reputation of developmental education in the larger college structure. How do the two models compare in those areas?

This discussion considers the organizational structure of remediation from the perspective of student learning. The issue could also be considered from philosophical, political, or budgetary perspectives. For example, Klicka (1998) claims that centralization protects program philosophy, makes remediation more visible, and ensures budget allocation and administrative representation in the college. The issue of student learning is considered here in terms of transfer of skill from the remedial context to college classrooms. However, whatever measure of learning is considered important, there are no empirical studies in the literature that determine the impact on student learning through direct comparison of mainstreamed and centralized developmental education.

Comparison of Mainstreamed and Centralized Developmental Education in Terms of Critical Educational Components

Quality of instruction. The main purpose of remedial education is to prepare students for the college-level academic demands. Therefore, the quality of remedial instruction can be considered in terms of its alignment with the college curriculum. Specifically, the skills and content taught in developmental reading, writing, and math classrooms should be related to those that students will later encounter in their subject-matter classrooms. From a cognitive perspective, close alignment of developmental and college-level instruction should promote students' generalization of learning beyond remediation to the college-level classroom. Transfer from learning to application is one of three major types of cognitive generalization (Simons, 1999) and is a central goal of education (Bereiter, 1995). An important factor in the transfer of learning is the reinforcement of students' original learning through the use of multiple examples in numerous contexts. As Haskell (2001) states, "Teaching that promotes transfer ... involves returning again and again to an idea or procedure but on different levels in different contexts, with apparently `different' examples." (pp. 26-27). The remedial classroom is where academic skills are learned and the various college-level classrooms that the student attends as he or she moves through the discipline program are the settings in which these skills are applied. Transfer of skill is more likely if learning and application occur close in time.

Remedial programs described as exemplary include the "integration of coursework within and beyond the developmental program" (McCabe & Day, 1998, p. 25). There are several ways to accomplish the integration of remediation and higher-level instruction, for example, through paired courses that create formal links between precollege developmental and college-level courses in discipline areas (Badway & Grubb, 1997). These pairings provide immediate opportunities for the application of newly learned reading, writing, and math skills. The alignment of remedial with occupational courses seems useful, since many remedial students plan to pursue career-related degree programs (McCabe, 2000). However, most alignment between developmental and college level curriculum involves general education courses such as freshman composition, history, and psychology, rather than specialized technical courses (Perin, 2001).

Irrespective of students' interests, course pairing may be ruled out when state or institutional policy mandates remedial completion prior to enrollment in college-level courses. That is, if students are prohibited from enrolling in college-level courses before they complete their remedial requirements, they are necessarily barred from participating in a paired-course model since one of the courses bears college credit.

In situations where policy allows formal connections between remedial and college-level classes, is this innovation more likely when developmental education is mainstreamed or centralized? Instructional reform requiring the interdisciplinary collaboration necessary for course linking depends on positive working relationships among instructors (Perin, 2000). Centralizing developmental education may serve to marginalize it within the college, reducing the likelihood of regular interaction between developmental and college-course instructors. If this is the case, curricular alignment in the form of paired courses may be more likely to occur when developmental education is mainstreamed.

Apart from course pairing, instruction can also be aligned by matching exit levels of developmental education to entry levels of the college-level courses. Lining up these levels, at least for college composition and mathematics courses, seems more feasible when developmental education is mainstreamed, because in principle, at least some instructors who teach college-credit classes would also have as part of their teaching load some developmental-level classes.

In practice, as discussed below, discipline area instructors may decline developmental teaching assignments, and when taught in academic departments, the instruction of remedial courses may be left to part-time, adjunct faculty who may or may not also be teaching college-level courses. If this problem can be overcome administratively, mainstreamed developmental education may have better potential than centralized departments to align curriculum, at least in the subject areas of English and mathematics, thus facilitating the generalization of student learning.

The benefits of the greater use of full-time instructors in centralized rather than mainstreamed developmental education programs may be undermined by the lack of awareness of the academic demands and content of college-level study that such instructors may have as a result of isolation from the academic departments. The danger in this case is that even at the highest level remedial courses, students considered ready to exit remediation may actually remain underprepared for academic study in the content areas (Perin et al., in press).

Availability of ancillary support services. Overall, community colleges have a strong reputation for providing assistance to support learning and are perceived by students as more nurturing than four-year colleges (Carlan & Byxbe, 2001). A report by the Institute for Higher Education Policy (1998) suggested that the effectiveness of remediation in higher education could be improved by including "support services that rely on multiple intervention strategies" (p. 23). Roueche, Ely, and Roueche (2001) linked community colleges' effectiveness in educating remedial students to the provision of supplementary tutoring, mandatory participation in learning labs, and "case management models" (p. 33) that permit individualized attention. Programs described as exemplary by McCabe and Day (1998) provide support services including tutoring, academic and career advisement, and workshops in areas such as time management and study skills (Moriarty et al., 1998).

These ancillary services may be necessary to increase the persistence and performance of academically low functioning students, many of whom experience not only the family and financial difficulties typical of community college students in general but may also suffer from low self-esteem related to academic difficulties. Remedial students can feel lost in a college environment that they may perceive as impersonal. Support services seem especially important for students at the lower remedial levels who test into three or more remedial courses. In particular, students who enter the college with reading difficulties are at severe academic risk (Adelman, 1998; Roueche & Roueche, 1999).

Centralized developmental education departments may be more likely to recognize the need for support services for at-risk students. Since the sole purpose of a centralized department is remediation, chairs may be more willing than heads of regular academic departments to allocate funds to provision of support services. Further, because their teaching staff may be more attuned and sympathetic to the needs of academically low performing students, centralized departments may be more likely to implement an "early alert" system (Hebel, 1999) that identifies and refers at-risk students for counseling or other support services. However, taking into consideration both the need for curricular alignment and provision of support services, one can speculate that the lower level remedial student, marked by the need for reading instruction, is best served in a centralized department while the higher functioning student may benefit most from developmental courses in a mainstreamed department.

Teacher motivation and experience. Faculty in centralized developmental education departments see the teaching of remedial students as a primary task, while academic discipline instructors may view developmental teaching as a low status assignment and even a punishment. Developmental teachers seem more likely to be able to identify both strengths and weaknesses, rather than only deficiencies, in remedial students. Additionally, hiring criteria in centralized departments are more likely to include commitment to teaching remedial reading, writing or math. Professional development activities are more likely to focus expressly on remedial issues in a centralized than in a mainstreamed department. Thus, on the dimension of teacher motivation and experience, centralized departments seem superior to mainstreamed developmental education.

Students' reactions. Developmental education courses have been criticized as causing feelings of discouragement by reinforcing students' sense that they are at risk and forcing them to take longer to finish their degrees (McCusker, 1999). Alternatives to traditional remedial courses include tutoring and adjunct courses directly connected with regular college-level courses (Commander & Smith, 1995; Maxwell, 1997, both cited by McCusker, 1999). These options provide opportunities for academically underprepared students to interact with their higher achieving peers and participate more fully in college life. Locating remedial education in a regular academic department may hold similar promise. Course numberings indicating that remedial reading, writing, and math courses are part of a larger departmental sequence including college-level English and math may also have positive effects on students' feelings about education. In terms of student reactions to developmental education, mainstreaming appears to be superior to centralization.

In fact, when the mandate to attend remedial classes is weak, students may take it upon themselves to mainstream their remediation within their own programs of study by taking developmental education and credit-bearing courses simultaneously, even where remediation is centralized administratively within the college. Since developmental education courses are intended as preparation for postsecondary-level study, it is surprising that students are rarely required to complete remediation prior to matriculating in college-level programs. NCES (1996) reported that only 2% of higher education institutions (community and four-year colleges combined) prohibited simultaneous enrollment in remedial and credit courses of any type. Among the other 98% of institutions, practices varied across the remedial areas of reading, writing, and math. Between 29% and 35% of institutions placed no restrictions on simultaneous course taking in any area, and between 64% and 69% of institutions imposed some restrictions in one or more remedial areas (NCES, 1996, Figure 4).

While NCES provides some information at the institutional level, little has been reported about state policy regarding completion of remediation prior to matriculation, although at the time of writing this report, the Education Commission of the States was in the process of surveying states on this among other issues (Boswell, 2001, personal communication). That state policy tends to be weak or nonexistent on this topic is suggested by findings of Boylan et al. (undated) that only one half of all states require remedial placement based on initial assessment. Where remedial placement itself is not mandatory, it seems unlikely that completion of remediation would be required for entry to college-level courses. In contrast, within institutions and states where remediation must be completed prior to program matriculation, developmental education serves as a vestibule that must be exited in order for credit-beating college work to begin. Students may react to the long wait by simply dropping out. The mainstreaming of remedial courses either organizationally within the college or programmatically within students' own course selections seems more likely than centralization models to create positive student reactions.

Reputation of developmental education in the larger college structure. Centralizing developmental education in effect segregates it from the rest of the college (Eaton, 1994), which may make it difficult for remedial faculty to engage in discussion about curriculum and pedagogy that may occur in the rest of the college (Grubb & Associates, 1999, p. 206). Despite the sometimes unfavorable view of developmental education within academic departments, centralizing remediation may be worse by stigmatizing remediation in the whole college.

Summary. The following table summarizes the relative potential effectiveness of the centralized and mainstreamed structures in the five areas discussed above. Relative superiority is indicated as "+" and inferiority as "-."

Conclusions and Recommendations

Both centralized and mainstreamed developmental education models show advantages and disadvantages. Among five critical features considered, mainstreaming appears to have the potential for higher quality instruction and more positive student reactions. Centralized departments seem superior regarding ancillary support services and teacher motivation and experience. Both models seem to suffer from the low reputation of developmental education in higher education.

One issue that emerged in this discussion is that lower level remedial students may benefit from a centralized department while students closer to the college level of academic performance may be better served in a mainstreamed department. The tendency of institutions with higher proportions of minority students to centralize developmental education provides indirect support for this speculation, since minority students show greater academic risk than do nonminority students. However, at-risk students are also particularly prone to drop out of community college altogether. One wants to prevent remedial education from driving them away. Any evaluation of the relative merits of centralized versus mainstreamed developmental education should include data from both successful completers and drop-outs.

Pending the availability of comparative evaluation studies, colleges in the process of selecting between centralized and mainstreamed approaches must weigh the severity of each disadvantage. On a more positive note, it is possible to incorporate the beneficial features of both models in either a centralized or mainstreamed setting. The following recommendations could be implemented within either model given the necessary level of administrative commitment and financial resources.

Whether mainstreamed or centralized, developmental reading, writing, and math curricula should be aligned with content and skills found in college-level courses. Remedial literacy and math practices should be authentic, utilizing actual material and examples from the college curriculum rather than drilling in skills that fragment the literacy process (Grubb & Associates, 1999; Levin, 1999). While alignment of remedial reading, writing, and math curricula may be easier when developmental education is mainstreamed in English and mathematics departments, there is no reason in principle why teachers in centralized departments could not incorporate meaningful, content-based, college-level reading, writing, and math material. Ideally, whether centralized or mainstreamed, the content of remedial reading, writing, and math instruction should be closely connected to the subject matter students will later study in degree programs. Many would benefit from being exposed to specific technical and career-related knowledge in the context of reading, writing, and math remediation.

Individualized attention and supplementary tutoring are important sources of support for academically underprepared students. Borrowing practices characteristic of centralized departments, colleges that mainstream developmental education should ensure that appropriate support services are available to students who need them. This may require setting up the early-warning system referred to above. A major challenge concerns the allocation of funds for these services in departments that are also committed to a wide range of college-level activities, as well as the administrative attention of program heads whose primary commitment may be to degree preparation. To overcome these challenges senior administrators need to work with relevant academic program chairs to ensure that remedial students are adequately supported and monitored. Whether developmental education is centralized or mainstreamed, it should "create conditions for learning," and provide the advisement and support needed to help students overcome the fear of failure (McClenney, undated).

Crowe (1998) asked, "Do colleges train and support developmental instructors or just throw them in the breach?" (p. 15). Professional development, with appropriate incentives for participation and application, would help improve teaching ability and motivation in both mainstreamed and centralized developmental education. In the mainstreamed model, collaborations between remedial and college-level instructors may help the latter develop the passion that the former feel for helping students who have failed in the past. Further, mainstreamed developmental faculty need to learn systematic techniques for teaching reading, writing, and math typical of the learning disabilities field, with which centralized faculty are often highly familiar. On the other hand, instructors in centralized departments may not be adequately familiar with the literacy requirements and content of the college-level, subject-matter curriculum. Contact with college-level English and math instructors would give them an opportunity to examine discipline curricula in order to identify content and skills that could improve the effectiveness of developmental courses in preparing students for college-level work.

There may be a trade-off between instructional quality and teacher motivation across the two models of developmental education. In the mainstreaming approach, instructors may dislike the assignment of teaching remedial courses, but within this same model there may be greater opportunities to link remedial instruction to college-level material. The challenge for institutions would be to raise instructor motivation within the mainstreaming model or to provide incentives for linking remedial and college-level content within the centralized model. Possible mechanisms for accomplishing this aim include incentive pay, caps on class size, and reduction in teaching load.

Whether in mainstreamed or centralized departments, developmental education students should be encouraged to participate in college activities, especially related to the majors and professions to which they aspire, to reduce their feelings of discouragement and self-perceptions as academic failures. Although their skill levels may preclude enrollment in college-level courses, developmental instructors could find ways to provide contact between developmental and college-level students that could raise the motivation of developmental students to persist in what may be a multiyear remedial endeavor. For example, developmental educational students could visit selected credit-level courses as guests, or peer-tutoring programs could be mounted in which students in credit courses work with developmental education students on basic academic skills needed in the degrees to which the latter aspire.

Learning experience is enhanced when students feel that they are connected with an endeavor that is respected in the college. Efforts should be made by academic departments and college administrators to integrate developmental education with the rest of the college program, rather than marginalizing it within departments or within the college. Colleges will have different ways of accomplishing this integration. Doing so seems appropriate given the extent of remedial need in the student body and the growing centrality of developmental education to the community college mission.

Although centralized models have been recommended by experts in the field, Boylan and his colleagues (Boylan et al., 1997; Boylan, 1999) suggest that it is not the centralization itself that might be responsible for superior outcomes but the fact that this structure makes it easier to coordinate services and promote communication among staff. Coordination and communication may come more easily in a centralized model but are, of course, entirely possible in a situation where remedial education is incorporated in a larger department. In conclusion, both mainstreamed and centralized models have good potential to prepare students for postsecondary academic work, as long as the college demonstrates commitment to the ongoing improvement of developmental education in whatever form is institutionally appropriate.

Table 1
Relative potential effectiveness of
centralized and mainstreamed structures

                                                       Mainstreamed
Educational Component               Centralized Model     Model

Quality of instruction                     --               +

Ancillary support services                 +                --

Teacher motivation and experience          +                --

Student reactions                          --               +

Reputation of developmental                --               --
education

References

Abraham, A.A. (1992). College remedial studies: Institutional practices in the SREB states. Atlanta, GA: Southern Regional Education Board.

Adelman, C. (1998). The kiss of death: An alternative view of college remediation [Online]. Available: www.highereducation.org/crosstalk/ct0798/voices0798adelman.html

Arendale, D. (1998). Trends in developmental education [Online]. Available: http://umkc.edu/cad/nade/nadedocs/trends.htm

Badway, N., & Grubb, W.N. (1997). A sourcebook for reshaping the community college: Curriculum integration and the multiple domains of career preparation, Vols. I and II. Berkeley, CA: National Center for Research in Vocational Education.

Bereiter, C. (1995). A dispositional view of transfer. In A. McKeough, J. Lupart, & A. Marini (Eds.). Teaching for transfer: Fostering generalization of learning (pp 21-34). Mahwah, NJ: Erlbaum.

Boswell, K. (2001, July 5). Education Commission on the States [personal communication].

Boylan, H.R. (1999). Developmental education: Demographics, outcomes and activities [Online]. Available: www.ncde.appstate.edu.

Boylan, H., Bliss, L., & Bonham, B. (1997). Program components and their relationship to student performance. Journal of Developmental Education, 20 (3), 2-9.

Boylan, H., Saxon, P., & Boylan, H.M. (undated). State policies on remediation at public colleges and universities. Paper prepared for the League for Innovation in the Community College. Boone, NC: National Center for Developmental Education, Appalachian State University.

Carlan, P.E., & Byxbe, F.R. (2000). Community colleges under the microscope: An analysis of performance predictors for native and transfer students. Community College Review, 28 (2), 27-43.

Carnevale, A.P., & Desrochers, D.M. (2001). Help wanted, credentials required: Community colleges in the knowledge economy. Princeton, NJ: Educational Testing Service.

Commander, N.E., & Smith, B.C. (1995). Development adjunct reading and learning courses that work. Journal of Reading, 38, 352-360.

Crowe, E. (1998, September). Statewide remedial education policies: State strategies that support successful student transitions from secondary to postsecondary education. Denver, CO: State Higher Education Executive Officers.

Damashek, R. (1999, fall). Reflections on the future of developmental education. Journal of Developmental Education, 23 (1), 18-20, 22.

Eaton, J. (1994). Strengthening collegiate education in community colleges. San Francisco: Jossey Bass.

Grubb, W.N., & Associates (1999). Honored but invisible: An inside look at teaching in community colleges. New York: Routledge.

Haskell, R.E. (2001). Transfer of learning: Cognition, instruction and reasoning. San Diego: Academic Press.

Hebel, S. (1999, May 7). Community College of Denver wins fans with ability to tackle tough issues. Chronicle of Higher Education, A37.

Institute for Higher Education Policy (1998). College remediation: What it is, what it costs, what's at stake. Washington, DC: Author.

Klicka, M.A. (1998). Developmental education services at Bucks County Community College. In R.H. McCabe & P.R. Day, Jr. (Eds.). Developmental education: A twenty-first century social and economic imperative (pp 37-52). Mission Viejo, CA: League for Innovation in the Community College.

Levin, H. (1999). Improving remedial education. Paper prepared for the Social Science Research Council, New York.

Levin, J.S. (2001). Globalizing the community college: Strategies for change in the 21st century. New York: Palgrave.

Maxwell, M. (1997). What are the functions of a college learning assistance center? (ERIC Document Reproduction Service No. ED413031)

McCabe, R.H. (2000). No one to waste: A report to public decision-makers and community college leaders. Washington, DC: Community College Press.

McCabe, R.H., & Day, P.R., Jr. (1998). Developmental education: A twenty-first century social and economic imperative [Online]. Mission Viejo, CA: League for Innovation in the Community College. Available: www.league.org

McClenney, K.M. (undated). Teaching in a future that ain't what it used to be. [Online]. In Celebrations: An Occasional Publication of the National Institute for Staff and Organizational Development (NISOD). Austin, TX: University of Texas at Austin. Retrieved May 24, 2001, from: www.nisod.com

McCusker, M. (1999). ERIC review: Effective elements of developmental reading and writing programs. Community College Review, 27 (2), 93-105.

McKay, S.E., Red Shirt, E.M., and Hickey, M.C. (1998). The Guilford Technical Community College Developmental Education Program. In R.H. McCabe & P.R. Day, Jr. (Eds.), Developmental education: A twenty-first century social and economic imperative [Online]. (pp 67-72). Mission Viejo, CA: League for Innovation in the Community College. Available: www.league.org

Moriarty, D.F., with Naigus, N., Wyckoff-Byers, N., Greenfield, T., & Mulligan, D. (1998). In R.H. McCabe & P.R. Day, Jr. (Eds.), Developmental education: A twenty-first century social and economic imperative [Online]. (pp 73-77). Mission Viejo, CA: League for Innovation in the Community College. Available: www.league.org

National Center for Education Statistics, Statistical Analysis Report (1996, October). Remedial education at higher education institutions in Fall 1995. Postsecondary education quick information system (PEQis). (NCES No. 97-584). Washington, DC: Author.

Perin, D. (2001). Academic-occupational integration as a reform strategy for the community college: Classroom perspectives. Teachers College Record, 103, 303-335.

Perin, D., Keselman, A., & Monopoli, M. (in press). The academic writing of community college remedial students: Text and learner variables. To appear in Higher Education.

Roueche, J.E., & Roueche, S.D. (1999). High stakes, high performance: Making remedial education work. Washington, DC: Community College Press.

Roueche, J.E., Ely, E.E., & Roueche, S.D. (2001). Challenges of the heart: Pursuing excellence at the Community College of Denver. Community College Journal 71 (3), 30-34.

Shults, C. (2000). Remedial education: Practices and policies in community colleges. AACC Report No. RB-00-2. [Online]. Washington, DC: American Association of Community Colleges. Available: http://www.aacc.nche.edu/initiatives/issues/Remedial.pdf

Simons, P. R. J. (1999). Transfer of learning: Paradoxes for learners. International Journal of Educational Research, 31, 577-589.

Dolores Perin is an associate professor of psychology and education in the Health and Behavior Studies Department and senior research associate in the Community College Research Center, Teachers College, Columbia University in New York.