Research on college student persistence is by now voluminous. Much of this research is based on testing and validating Vincent Tinto's (1975, 1987, 1993) highly-acclaimed model of student departure. The basic premise of Tinto's model is that social and academic integration are essential to student retention. Tinto's model (especially the 1975 and 1987 version) has certainly provided a workable and testable foundation for analyzing the multiple factors involved with student departure particularly employing quantitative methods. Quantitative researchers such as Nora and Cabrera (1996) note that there is sufficient empirical evidence establishing the validity of Tintoís (1975, 1987) model of student persistence. Others have modified and improved the model utilizing diverse study populations at different higher education institutions (Nora, 1987; Nora, Attinasi, & Matonak, 1990; Rendón, 1982; Nora & Rendón, 1990; Cabrera, Nora, & Castaneda, 1992; Nora & Cabrera, 1993, 1996; Cabrera, Castaneda, Nora, & Hengstler, 1992; Pavel, 1992; Cabrera & Nora, 1994; Pascarella & Terenzini, 1992; Pascarella, 1980; Terenzini, Lorang, & Pascarella, 1981). Yet, more remains to be done.

Braxton, Sullivan, and Johnson's (1998) assessment of Tinto's theory (based on the 1975 version) found that, in the aggregate, assessment of empirical evidence regarding 13 of Tintoís primary propositions indicated only partial support for the theory. The researchers cited problems with empirical internal consistency in multi-institutional or single institutional assessments, in both residential and commuter universities, and across female and male college students. Further, Tierney (1992), Attinasi (1989, 1994), and Kraemer (1997) have questioned the validity of the model to fully and appropriately capture the experiences of non-White students, given that the model is based on an assimilation/acculturation framework. 

It is worthy at this point to note the linkage between Tintoís interactionalist theory and the assimilation/acculturation perspective. Interactionalist theory is concerned with the impact of person-and institution-related characteristics on a particular phenomenon (Caplan & Nelson, 1973, Braxton, Sullivan, & Johnson, 1998). Tinto (1993) notes that his persistence model is an "interactional system" (p. 136) where both students and institutions (through social and educational communities) are, over time, continually interacting with one another in a variety of formal and informal situations. Key to the interactionalist view is that persistence is contingent on the extent students have become incorporated (integrated) into the social and academic communities of the college. 

Interactionalist theory may be linked to the acculturation/assimilation perspective that was prevalent during the 1960s when social scientists from various fields studied how members of minority groups became integrated into the dominant white society. It was believed that minority individuals were engaged in a self-perpetuating cycle of poverty and deprivation and that they could avoid societal alienation by becoming fully absorbed (assimilated) or adapted (acculturated) into the dominant culture (Hurtado, 1997). Assimilation required a process of separation, a cultural adaptation that required minority individuals to break away from their traditions, customs, values, language, etc. in order to find full membership in the predominantly White American society. However, during the 1970s and 1980s, critics contested this perspective citing problems such as the use of mainstream cultural norms as evaluative criteria, as well as the problematic assumption that minority group norms and cultural patterns were inferior, deviant, and self-destructive when compared to those of the majority culture (de Anda, 1984). 

Along these lines, Caplan and Nelson (1973) provided important distinctions between person-centered and situation-centered problems, noting that the way a problem was identified gave way to specific solutions. For example, researchers focusing on person-centered problems would focus on individual characteristics as the root of the issue and the target of the solution, while ignoring situationally relevant factors. In the case of studying why minority cultures experience alienation, a person-centered definition would identify the pathology as residing with minority group characteristics. Conversely, Caplan and Nelson noted that situation-centered problems have a system change orientation. Here, the context in which individuals operate is examined and remedies are proposed to change the system. 

Once in effect and legitimated, irrespective of their validity, these definitions resist replacement by other definitions or perspectives. For example, the idea that minority students are not motivated to learn or have low expectations has been around for decades and ignores how systemic inequities, racism, and discrimination have worked against minority populations. Within the past 20 years, there has been greater emphasis on examining the interactions between individuals and systems. Yet, Caplan and Nelsonís (1973) view that to the extent that problem definitions conform to and reinforce dominant cultural myths and clichés, as indeed most definitions must in order to become widely accepted, their change or replacement will be stubbornly resisted. People tend to conform to public definitions and expectations, even when there are doubts regarding their accuracy.

Because interactionalist retention theory adheres to some of the basic premises of the acculturation/assimilation framework, such as separation and incorporation, several researchers have challenged the way these processes have been conceptualized in relation to explaining minority student retention in college. In particular, the assumption that minority students must separate from their cultural realities and take the responsibility to become incorporated into the collegesí academic and social fabric in order to succeed (with little or no concern to address systemic problems within institutions or to the notion that minority students are often able to operate in multiple contexts) becomes central to the critique of Tintoís student departure model.

At the same time, emerging scholarship which is beginning to take root not only in education, but in fields such as psychology, anthropology, and sociology is revolutionizing the way we conceptualize different phenomena and the selection of empirical tools to guide this understanding (Hurtado, 1997; Rosaldo, 1989). For example, Hurtado (1997) explains that much feminist research advocates a multidisciplinary and multimethod approach that is non-hierarchical (i.e., one dominant group is not favored over another) and reflexive (i.e., invites critique and further analysis). Given these developments, we believe that revisionist models and theory refinements are needed. Also needed are new models that consider the key theoretical issues associated with the experiences of minority students in higher education. 

It is important to note that researchers (primarily White) began studying student retention prior to the time that minorities had become a critical mass on college campuses. Few minority students resulted in small sample sizes or total exclusion from the sample. Consequently, much of the most widely-acclaimed research guiding theories of studentsí transitions to college, departure, involvement, and learning was often based on White male students (Tierney, 1992; Belenky, Clinchy, Goldberger, & Tarule, 1986). This research produced a monolithic view of students devoid of issues of race/ethnicity, culture, gender, politics, and identity (Hurtado, 1997). 

The research on minority college students is relatively young, and the majority of it focuses on African American and Hispanic (primarily Mexican American) students. Especially fertile territory is research on American Indians, Asians, Pacific Islanders, Filipinos, Puerto Ricans, Cubans, and immigrant students from Asia and Central and South America. As our society becomes more multicultural and complex, the experiences of multiracial students will merit careful investigation. In the 1970s only a few studies, such as Gurin and Epps (1975) and Olivas (1979), focused on minority students. It is only within the past 15 years that researchers, many of them non-White, began to study minority students (Nora & Cabrera, 1994; Nora & Rendón, 1988, 1990; Rendón, 1982, 1994; Jalomo, 1995; Tierney, 1993; Wright, 1988; Allen, 1984; Ogbu, 1978, 1987; Thomas, 1984; Harvey & Williams, 1989; Attinasi, 1989; Fleming, 1984; Nettles, Gosman, Thoeny, Dandrige, 1985; London, 1978, 1989; Weis, 1985; Hurtado & Garcia, 1994; Tierney, 1992; Kraemer, 1997; Nora, Attinasi, & Matonak, 1990; Cabrera, Nora, & Castaneda, 1993; Lowe, 1989; Melchior-Walsh, 1994; Galindo & Escamilla, 1995; Gandara, 1993; Wycoff, 1996; Valadez, 1996; Mow & Nettles, 1990). This relatively new research not only lifts the knowledge base of student retention and development theories, it advances policy and practice and calls to question the predominant ways of structuring student development services employing research that included few, if any, minority students. 

Much of the research that provides important modifications to the problem definition, introduces new variables to the retention equation, and attempts to refine traditional paradigms of student retention, is scattered and unconnected. Consequently, a new, coherent vision of minority student persistence has failed to evolve. Researchers and practitioners alike tend to view issues related to the retention of minority students as similar, if not identical to those of majority students. What transpires is an almost universally entrenched view that Tintoís (1975, 1987) departure model, with all of its assumptions, is complete, appropriate, and valid for all students regardless of their varied ethnic, racial, economic, and social backgrounds. To his credit, Tinto (1993) elaborates on the importance of supportive student communities for students of color and adult students who may experience difficulties making the transition to college and becoming incorporated. Tinto (1993) also notes the need to build inclusive campuses, explaining that "to be fully effective, college communities, academic and social, must be inclusive of all student who enter (p. 187)." Yet researchers such as Hurtado (1997) would argue that linear models based on an assimilation/acculturation framework leave many questions unanswered, especially with regard to multiple group identifications and how both minority and majority groups change when they come into contact with each other.
 

The purpose of this chapter is to: 1) provide a critical analysis of Tintoís student departure theory (1975, 1987, 1993) with a specific focus on the separation and transition stage, 2) to critique Tintoís concepts of academic and social integration, and 3) to present future directions designed to take retention theory to a higher level. The main concern is not whether the Tinto theory works for minority students. Rather, the emphasis is on the kind of theoretical foundation and methodological approaches that are needed to more fully understand and facilitate the retention process for minority students in an increasingly complex and multiracial institutional environment. Our critique is not meant to assault or discredit the work of researchers who have devoted their careers to studying how students become engaged in college. Rather, we offer alternative perspectives that seek a similar aim: To more fully understand student retention in college. We believe scholars ought to periodically reassess their work and how they apply their empirically-based perspectives to new contexts in order to advance knowledge. Indeed, even the ideas advanced here should be taken further, and we encourage researchers to do so.. 
 

Tinto's (1975, 1987, 1993) model of student departure has been extensively employed to study how majority and minority students become academically and socially integrated into institutional life. To help develop his theory on student departure, Tinto employed the rites of passage framework of Dutch anthropologist Arnold Van Gennep (1960). Van Gennep was concerned with the movement of individuals and societies over time and the rituals designed to move individuals from youth to adulthood in order to ensure social stability. To facilitate a discussion of theoretical issues on the concepts of separation, as well as academic and social integration, a brief summary of Van Gennep's theory is presented.

Conceptual Issues In Van Gennep's Rites of Passage

The rites of passage as described by Van Gennep (1960) included a three-phase process of separation, transition, and incorporation. In stage one, separation, the individual became separated from past associations and a decline occurred in interactions with members of the group from which the individual originated. Specific ceremonies marked outmoded views and norms of the old group. In stage two, transition, the individual began to interact in new ways with members of the new group in which membership was being sought. Rituals such as isolation, training, and ordeals were used to facilitate separation which ensured that the individual acquired the knowledge and skills of the new group. In the third stage, incorporation, the individual took on new patterns of interaction with members of the new group and established competent membership. Though able to interact with members of the old group, individuals now did so only as members of the new group. In this stage, individuals became fully integrated into the culture of the new group (Tinto, 1987). Tinto (1987) stressed that it was "possible to envision the process of student persistence as functionally similar to that of becoming incorporated in the life of human communities..." (p. 94).

How generalizable are Van Gennep's perspectives and assumptions when studying minority college students? First, let us consider the concept of separation and the way that some scholars have interpreted the theory. One of the assumptions scholars have made is that individuals should disassociate themselves from their native cultural realities in order to assimilate into college life. The assumption made is that an individualís values and beliefs rooted in his or her cultural background must be abandoned to successfully incorporate the values and beliefs of not only the institution but of the majority population upon which they are based. Only in this way can the individual students become integrated into their new environment. Minority students must reconcile the fact that the world from which they come is distinctly different from the world of college and that they must leave the old world behind in order to find full membership in the new college world. A second assumption is that there is one "dominant" culture and that in order to succeed, members of minority cultures should become more similar to this dominant culture. A third assumption is that it will be relatively easy to find membership and acceptance in the new college world and that individuals who become integrated will have little or no contact with members of their old group. Indeed, the hallmark of Tintoís (1993) revised model is that students should find social and intellectual communities to attain membership and receive support. Even students who initially resist separation will later determine that leaving their group to succeed in college is appropriate and necessary. These assumptions are not entirely correct. Alternative views that challenge the three aforementioned assumptions are presented next.

A Critical Analysis of The Assumption of Separation

Scholars investigating how minority students make the transition to college should be familiar with the concepts of biculturalism and dual socialization which challenge the assumption of separation. In addition, scholars should note the problematic issues in relation to the assumption of a dominant culture and the membership assumption.

The Concept of Biculturalism: While conducting an ethnographic study of poverty and Afro-Americans in a large northern city, Charles A. Valentine (1971) found that accepted cultural deficit and difference models of the time neglected and obscured important elements of the Afro-American culture. Referring to cultural deficit models as an alternative to analyze non-mainstream cultures, the researcher argued that "any theory of class or racial deficits of biological origin is quite undemonstrable, indeed scientifically untestable, in an ethically plural and structurally discriminatory society" (p. 138). While not negating cultural distinctions between Black and mainstream cultures, Valentine (1971) observed that: "The central theoretical weakness of the 'difference model' is an implicit assumption that different cultures are necessarily competitive alternatives, that distinct cultural systems can enter human experience only as mutually exclusive alternatives, never as intertwined or simultaneously available repertoires" (p. 141). Valentine cited cultural difference models as incorrect and harmful when employed for establishing new educational policies and programs.

As an alternative to predominant cultural difference models, Valentine (1971) proposed the employment of a bicultural educational model. The researcher argued that since many Blacks were simultaneously committed to both Black and mainstream cultures, the two were not mutually exclusive of each other. Rather, Blacks could be simultaneously socialized in two different cultures. He relied on the findings of Steven Polgar (1960) who earlier found that individuals living on an Indian reservation regularly went through a process he termed "biculturation." Biculturation occurred when individuals were simultaneously enculturated and socialized in two different ways of life. In Polgar's example, teenage Mesquakie boys experienced a contemporary form of their traditional Amerindian lifeways and mainstream Euro-American culture. 

Valentine (1971) used Polgar's (1960) research to expand the concept of biculturation, the ability of a minority individual to step in and out of the repertoires of two cultures (Figure 1) that were seen as distinct and separate (de Anda, 1984). For Valentine, biculturation helped explain how people learn and practice both the mainstream culture and ethnic cultures at the same time. He indicated that: "the Black community is bicultural in the sense that each Afro-American ethnic segment draws upon a distinctive repertoire of standardized Afro-American group behavior and, simultaneously, patterns derived from the mainstream cultural system of Euro-American deviation. Socialization into both systems begins at an early age, continues throughout life, and is generally of equal importance in most individual lives" (p. 143). The concept of bilculturalism seriously challenges the first two assumptions (noted earlier) of the separation stage.

The Concept of Dual Socialization: Diane de Anda (1984) elaborated on Valentine's (1971) concept of biculturation, citing six factors that affect biculturalism. Unlike Valentine, she indicated that the bicultural experience was possible not because the two cultures were totally disparate, but because there was some overlap between the two cultures (Figure 2). For de Anda, "dual socialization is made possible and facilitated by the amount of overlap between two cultures. That is, the extent to which an individual finds it possible to understand and predict successfully two cultural environments and adjust his or her behavior according to the norms of each culture depends on the extent to which these two cultures share common values, beliefs, perceptions, and norms for prescribed behaviors" (p. 102). In short, de Anda's model is not about individual separation from an old world in search of membership in a new one. 
 
 
 
 

Instead, de Anda argued that converging the two worlds could allow individuals to function more effectively and less stressfully in both worlds. This requires changing, indeed transforming, the academic and social culture of institutions of higher education to accommodate culturally diverse students.

Kuh and Whitt (1988) suggested that culture, in the context of higher education, could be described as a "social or normative glue" that is defined by the shared values and beliefs that exist within a college or university while serving four general purposes: 1) conveying a sense of identity; 2) facilitating commitment to an entity, such as the college or peer group, other than self; 3) enhancing the stability of a group's social system; and 4) providing a sense-making device that guides and shapes behavior (p. 10). In addition, the researchers proposed that the culture of a college or university defines, identifies, and legitimates authority in educational settings. However, they caution that institutions may, perhaps even unwittingly, have "properties deeply embedded in their cultures that make it difficult for members of historically underrepresented groups to prosper socially and environmentally" (p. 15). In cases such as this, 

students already potentially at risk often find themselves decidedly at odds with prevailing social and cultural norms on campus.

Dual socialization does not occur naturally in a college environment that contains values, conventions, and traditions that are alien to first-generation students, many who are minority. Jalomo (1995) substantiated de Anda's dual socialization model in a study of Latino community college students who had completed their first semester. For these students the transition was not linear. Rather, Latino students were found to operate in multiple contexts: The Latino culture, comprised of four subcultures (family, work, barrio, and gang), and the prevailing culture of the community colleges they attended. In the study, students who conveyed difficulty in their transition to college spoke of the growing incongruence between their native environment and the newly encountered college arena. These students indicated that they had maneuvered a number of social domains in their native environment while attempting to meet the growing 
 
 

demands associated with college life. Upon transiting to college, Latinos experienced the downside and upside of college attendance. They expressed some tension and loss associated with separation, although at the same time they experienced excitement at learning new things and making new college friends. They also experienced culture clash given the differences between their home life and the world of college. To diminish this disjuncture, Jalomo proposed that individuals not totally separate but instead be supported to transit between two cultures.

Levy-Warren (1988) provides additional perspectives to the concept of separation. In analyzing disruptions that people experience in separating from their cultures of origin (culture loss, culture shock), Levy-Warren indicates that the passage involves cultural dislocation and relocation, a disjunctive process that is both internal and external. The internal level involves identity formation - an individual is shedding a part of the self and assuming a new, redefined identity. The external level is the actual move from one geographical location to another and involves the loss of familiar objects and people. Cultural relocation may be highly traumatic if the move is made before the individual has established mental representations of culture. That is, individuals must be able to distinguish differences between their own world and the new world. Consequently, Rendón (1996) argues that rather than asking students to disassociate themselves from their culture, they should be assisted to make modifications in their relationships. The passage to college needs to be gradual, giving students time to slowly break away and move toward healthy individualization. 

Theoretically, the concept of dual socialization seriously challenges the assumptions of separation. In addition, there are retention policy considerations. Navigating two landscapes, one of which is almost entirely different from the student's home realties, requires both individual and institutional responsibility. To this end, the critical role of the institution cannot be overstated, yet is often diminished in retention and involvement studies. Tinto highlights the importance of the classroom as a learning community in his 1993 model. However, as noted earlier, Tinto's revised model has yet to gain widespread attention of the higher education research community. Connecting the world of the student to the world of college means that students must be able to find animate and inanimate objects in the new college culture that might evoke a sense of comfort that originates in their early cultural upbringing. That is why events and programs such as Black History Month, Women's Studies, and Cinco de Mayo celebrations are so important. Converging two worlds requires the use of cultural translators, mediators, and role models to 1) provide information and guidance that can help students decipher unfamiliar college customs and rituals; 2) mediate problems that arise from disjunctions between students' cultural traits and the prevailing campus culture; and 3) model behaviors that are amenable with the norms, values, and beliefs of the majority and minority culture (Jalomo, 1995; Rendón, 1996; de Anda, 1984).

The Assumption of a Dominant Culture: Tierney's (1992) critique of the Tinto model substantiates de Anda's convergence perspective. Tierney argued that Tinto did not consider an important point in Van Gennep's theory. The point was that Van Gennep used the term "ritual" to speak of rites of passage within the same specific culture, i.e., some Indian cultures have puberty rituals designed specifically for girls and others specifically for boys. Here, the ritual is nondisruptive. However, Tinto employed the use of ritual in a way that "individuals from one culture, such as Apache, are to undergo a ritual in another culture, such as Anglo" (p. 609). In this case the transition constitutes a disjuncture - in effect, students from a minority culture enter a majority world that is vastly different from their sociocultural realities. 

However, in his 1993 revised model, Tinto argues that the majority of colleges are made up of several, if not many, communities or "subcultures." Rather than conforming to one dominant culture in order to persist, students would have to locate at least one community in which to find membership and support. Further, Tinto notes that membership does not require a full sharing of values. Instead, only some degree of consensus is necessary. Consequently, Tinto explains that the use of the term "membership" is more applicable than "integration." Of course, safe havens and enclaves have their benefits and drawbacks. One of their key benefits is that they help students break down the institution into manageable parts. However, special communities and programs do not address the real challenge of todayís institutions: The total transformation of colleges and universities from monocultural to multicultural institutions. This requires more convergence between the minority studentís world and the college world. Further, Hurtado (1997) notes that research based on assimilation/acculturation views group contact as unidirectional in nature (i.e., the ethnic group changes to reflect the mainstream/dominant/White group). Hurtado argues that "[t]his type of approach effectively blocks the possibility that cultural contact can indeed bring change in both the minority and majority groups" (p. 305). In some areas of the nation and in some colleges and universities, minorities are the majority or are rapidly on their way to acquiring majority status. While White students have normally been viewed as the dominant or majority group, future research will have to take into account how group identities and power relationships are changing and the overall impact on student persistence.

The Membership Assumption: It is also important for scholars to consider that de Anda's concept of "dual socialization" - where individuals both develop and sustain membership in new and old cultures - has been the reality of behavior for most Americans who maintain an ethnic identity while coexisting within the dominant culture. In short, many minority students are not likely to give up their affiliation and lose contact with their cultural group in order to find membership in a new college world. Just like a great deal of Latino immigrants maintain extensive and frequent contact with Mexico (Hurtado, 1997), many minority students experience multiple associations with their own culture and their new college realities. Tierney (1992) indicates that American Indian students value group membership over the individualized process of separation. In many Latino cultures separation is often not a viable option, as family is a source of rootedness and strength. This view that both minority and majority groups co-exist and actually hold similar attitudes, values, and perceptions has been obliquely (indirectly) established by Nora and Cabrera (1996). Within this context, similarities between White and non-White students were noted in attitudes regarding the influence of family, perceptions of discrimination on campus, and educational goals and commitments. de Andaís (1984) notion that groups (or minorities) do not need to break all ties with past communities in order to attain membership status in a new or alien culture, as in the case of minority students who retain their sense of identity and cultural values while integrating in a predominately majority educational environment, is validated in Nora and Cabreraís (1996) research. 

Moreover, the membership option often represents a false choice given the lack of acceptance of racial/ethnic minorities in many dimensions of college life. In fact, many minorities leave college due to "cultural assaults" (Zambrana, 1988) to their sense of identity and self-esteem that lead to stress and tension. Though some students may leave, others may exhibit differential patterns of behavior. Some will become subservient to the codes of others, and others may deny their cultural heritage. Still others will manage to turn a negative into a positive - developing a strong sense of ethnic consciousness, i.e., pride in their cultural heritage and awareness of racism, discrimination, sexism, and elitism prevalent in higher education systems (Zambrana, 1988). 

While more research is needed to substantiate the multiple varieties of external and internal group memberships and their impact on retention, research has substantiated the view that different forms of encouragement and support from family and friends from the students' past communities not only continue to influence students during college enrollment, but are very instrumental in affecting persistence. These associations negate discriminatory experiences, enhance the social and academic integration of students, and positively affect the students' commitments to earning a college degree (Nora & Cabrera, 1994, 1996; Nora et al., 1996; Nora, Kraemer, & Itzen, 1997; Cabrera et al., 1993; Cabrera, Castaneda, Nora, & Hengstler, 1994).
 

Separation and transition are the initial processes associated with Tintoís student retention model. The next stage involves incorporation in institutional life. Tinto (1987) noted that "eventual persistence requires that individuals make the transition to college and become incorporated into the ongoing social and intellectual life of the college" (p. 126). Incorporation is analogous to integration. The term "integration can be understood to refer to the extent which the individual shares the normative attitudes and values of peers and faculty in the institution and abides by the formal and informal structural requirements for membership in that community or in the subgroups of which the individual is a part" (Pascarella & Terenzini, 1991, p. 51). To learn more about how incorporation came about, Tinto turned to French sociologist Emile Durkheim (1951) and the study of suicide. Durkheim was interested in the character of the social environment, including social and intellectual characteristics, and its relationship to individual behavior, such as suicide. To draw an analogy between suicidal and dropout behavior, Tinto employed Durkheimís view of "egotistical suicide." Individuals who committed egotistical suicide were unable to become socially or intellectually integrated within communities of society. As noted earlier, interactionalist theory, when used to study minority student retention, has been called into question given that some of the theoryís premises are based on an assimilation/acculturation framework. 

Problems With the Use of An Acculturation/Assimilation Framework

Hurtado (1997) notes that during the 1960s an assimilation framework was the impetus of research on ethnicity, especially for Mexican Americans. Since assimilation was contingent on the minority group becoming incorporated into the life of the majority group, assimilation scales were developed to measure the quickest and most efficient ways to assimilate immigrants. In 1964 Gordon advanced the notion that individuals could be highly acculturated in one group but remain unassimilated in other dimensions of society. In short, biculturality became a viable option. But the assimilation/acculturation framework had multiple problems:

1. A Focus on Academic Failure As Opposed to Success. Hurtado (1997) indicates that assimilation/acculturation research has focused on the left tail of the normal distribution curve, "that is, it has focused on cultural adaptations that are not particularly healthy, ones for which the only solutions are to assimilate to the dominant mainstream or spend a lifetime of psychological and social alienation" (p. 312). Some of the recent research (primarily within the past five to ten years) has called for a focus and examination of a variety of adaptations leading to academic success in addition to the repository of traditional research approaches examining academic failure (Hurtado & Garcia, 1994).

2. The Exclusion of Contextual and Historical Forces. Zambrana (1988) notes that during the 1960s and 1970s, studies on racial/ethnic communities were in large part descriptive and a historical. These descriptions were in line with the cultural deficit model that emphasized the problems or "deficiencies" within these communities, for the most part pathological and disruptive. It was believed that these deficits led to a self-perpetuating cycle of poverty and deprivation. When applied to education, the cultural deficit model suggests that cultural patterns of marginalized groups are essentially inferior and predispose students within those groups to poor academic performance. de Anda (1984) notes that the discrediting of the cultural deficit model led to the ascendance of the cultural difference model that emphasized the uniqueness of each minority culture. An inherent assumption with both of these models was that upward mobility and acculturation were possible to the extent that minority groups approximated the values and norms of the dominant society. However, a model known as "internal colonialism" (Almaguer, 1974; Blauner, 1972) argued that cultural adaptations of ethnic/racial groups were due more to the organization of economics, labor, and power in the capitalist way the nation was structured than to differences in traits between minority and majority groups. As such, groups that had the experience of conquest were subject to race discrimination in every dimension or organized society, including education (Hurtado, 1997).

In her review of cultural studies, Zambrana (1988) notes that "the most limiting aspect of the majority of studies has been their neglect of the relationship of racial/ethnic groups to the social structure" (p. 63). The lives of racial/ethnic minorities are shaped by social forces such as racism, sexism, and discrimination, yet many researchers tend to view people of color as if they have all the options and privileges of White, middle-class Americans. But this is not often the case. 

3. Lack of Focus on Systemic Barriers. Rather than focus on systemic issues in the lives of oppressed people, most past researchers employing an assimilation/acculturation framework have tended to focus on perceived cultural traits or differences (i.e., poor motivation; academic deficiencies) as the source of a group's ability or inability to succeed and be upwardly mobile. This view suggests that students who possess or adopt mainstream cultural norms are capable of moving farther along the educational pipeline than those who lack these cultural traits (Nieto, 1996). In opposition to this view, more recent educational research has documented that systemic barriers such as tracking, low expectations, and funding inequities, among others, play a critical role in hindering the educational achievements of ethnic/racial minorities (Oakes, 1985; Brint & Karabel, 1989; Nieto, 1996). Nonetheless, perceptions of minority student inferiority persist to this day and have even been used as a rationale to place restrictions on college access. 

4. Failure to Challenge Theoretical Assumptions and Paradigms. Even when minorities are studied, researchers often fail to challenge the philosophical assumptions made in traditional paradigms that are often grounded or developed from studies based on full-time, traditional age, residential, middle-class, White, male students and/or fail to consider current research that presents a more comprehensive and contextual view of minority student lives and educational experiences. The lack of a grounded historical perspective has led to the frequent omission of minority groups, or else they are identified as a source of their group's problems (a deficit perspective). Myths and stereotypes continue to prevail for racial and ethnic groups simply because there is a void in the incorporation of roles, characteristics, and perceptions of these subgroups. Many times variables are operationally defined in the same manner for all groups involved, thus excluding any cultural or racial differences in perceptions and attitudes. Rather than conducting culturally- and racially-based studies that can uncover new variables and that can offer insightful and meaningful findings to transform institutional structures that preclude academic success for minority students, invisible hierarchies are left intact. In these cases, minority students are measured simply by scales that reveal their level of acculturation and integration, or lack thereof (Zambrana, 1988; Hurtado, 1997).

5. Failure to Connect Theory to Practice. In his 1993 theory review, Tinto acknowledged the importance of policy-relevant research and the importance of the institution in enhancing retention. When theoretical propositions are not compared across different subgroups or when diverse and culturally-driven theoretical views are not incorporated in retention studies, institutional policies and practices cannot truly detect or address differences among student groups. Theories developed without using minority student perspectives and/or without "member checks" from the field may miss important details and nuances about the connection between student cultural realities and collegiate experiences. Tierney (1992) elaborates: "The search for an understanding about why students leave college is not merely of theoretical interest; if a model may be built that explains student departure then it may be possible for colleges to retain students" (p. 604). 
 

There are at least three conceptual problems with social/academic interactionalist theory as used in Tintoís (1975,1987,1993) student departure model. 

1. A Focus on Individual Responsibility as Opposed to Institutional Responsibility. The first conceptual problem is the over-emphasis on individual responsibility for change and adaptation. In Leaving College, Tinto (1987) emphasized the following points:

Nora, Kraemer, and Itzen (1997) and Nora and Cabrera (1993) argue that current quantitative models must include factors that are able to differentiate among racial and ethnic groups or must include measurement approaches (and techniques) that provide indicators of constructs that reflect racial, ethnic, and cultural differences. In a study of student persistence at an exclusively Hispanic two-year institution, Nora, Kraemer, and Itzen (1997) employed a different, and more culturally sensitive set of items that more closely reflected the manner in which the studyís Hispanic student became integrated on their campus. In doing so, the researchers reduced misspecification in the model. The researchers elaborate: "...The measures of academic integration used to form the scale not only [represented] possible academic interest and involvement with faculty and staff, but ... also [reflected] those circumstances (both financial and academic) that [were] prevalent among [the] Hispanic group" (p. 15).

2. Problems Associated With the Concept of Student Involvement. While interactionalist theory is concerned with the interaction among individuals and institutions, involvement is the mechanism through which student effort is engaged in the academic and social life of the college. In the 1993 model, Tinto explains that the model is "at its core, a model of educational communities that highlights the critical importance of student engagement or involvement in the learning communities of the college" (p. 132). Consequently, it becomes important to address problematic issues related to the involvement dimension implicit in the Tinto model. 

Alexander Astin's (1985) theory of student involvement is perhaps the most widely adopted college impact model of student development. According to involvement theory, "the individual plays a central role in determining the extent and nature of growth according to the quality of effort or involvement with the resources provided by the institution" (Pascarella & Terenzini, 1991, p. 51).Astin's involvement theory is based on the Freudian notion of cathexis, in which individuals invest psychological energy in objects outside themselves such as friends, families, schooling, jobs, and the like. Astin (1984) defined involvement as "the amount of energy that the student devotes to the academic experience" (p. 27). Indeed, research indicates that the more time and energy students devote to learning and the more intensely they engage in their own education, the greater the achievement, satisfaction with educational experiences, and persistence in college (Pascarella & Terenzini, 1991; Tinto, 1987).

While both Tinto and Astin would agree that the institution plays an important role in facilitating involvement, and in fact Tintoís 1993 revised model emphasized this point, practitioners have concentrated on the individual responsibility aspect. The result is that practitioners have resorted to offering programs to help students get involved, but have not focused on active outreach to students. Consequently, few dropout-prone students actually get involved. If practitioners accept the cultural separation assumption without understanding its inherent trauma for nontraditional students, then practitioners will tend to see involvement as a relatively easy task since they will also assume that all students, regardless of background, are ready, willing, and able to get involved. 

Researchers who have studied nontraditional students (Terenzini et al., 1994; Rendón, 1994; Jalomo, 1995) have contributed important findings and modifications to involvement theory. While the importance of involvement cannot be negated, these researchers note that many students, especially nontraditional students, find it difficult to get involved. Important differences between traditional and nontraditional students were not explained in the original conception of student involvement theory. Traditional students often come from upper- to middle-class backgrounds, are predominantly White, and come from families where at least one parent has attended college and where the expectation of college attendance is well established. For traditional students college attendance is a normal rite of passage and a part of family tradition. Consequently, they are more likely to understand and manipulate the values, traditions, and practices of college to their academic advantage. Involvement theory does not emphasize the fact that most two- and four-year colleges are set up to facilitate involvement for traditional students. 

On the other hand, nontraditional students often come from working-class backgrounds, are older, work at least part-time, are predominantly minority, and first-generation--the first in their family to attend college (Rendón, 1994; Terenzini, Allison, Gregg, Jalomo, Millar, Rendón, & Upcraft, 1993; Jalomo, 1995). Jalomo's (1995) study of Latino first-year community college students found that involvement was difficult for students who found the transition to college troublesome or whose background characteristics did not "fit" the traditional student profile found on most college campuses today. Table 1 portrays the characteristics of students who found college involvement difficult. Moreover, Jalomo (1995) found that students required the assistance of cultural translators, mediators, and role models in order to survive or succeed in their first semester in college.
 
 

Table 1

Characteristics of Latino students who found it difficult to get involved in their community college

Rendón (1994) found that validation, as opposed to involvement, had transformed nontraditional students into powerful learners. While it is likely that most White and traditional students can become involved on their own in an institutional context that merely affords involvement opportunities (i.e., tutoring centers, clubs and organizations, extracurricular activities, etc.), nontraditional students expect active outreach and intervention in order to become involved. Rendón explains: "It appears that nontraditional students do not perceive involvement as them taking the initiative. They perceive it when someone takes an active role in assisting them" (p. 44).

Presenting a model of validation, Rendón (1994) noted that what had transformed nontraditional students into powerful learners and persisters were incidents where some individual, either in- or out-of-class, had validated them. Validating agents made use of interpersonal and academic validation. Validating agents took an active interest in students. They provided encouragement for students and affirmed them as being capable of doing academic work and supported them in their academic endeavors and social adjustment. The critical role of the institution and its agents is underscored in Rendónís validation model. The role of the institution is not to simply offer involvement opportunities, but to take a very active role in fostering validation. Faculty, counselors, coaches, and administrators take the initiative to reach out to students and design activities that promote active learning and interpersonal growth among students, faculty, and staff (Rendón, 1994).

3. A Focus on the Negative Impact of the External Community. A third conceptual problem with interactionalist theory is that external forces and cultures are seen as distinct and having mainly a negative impact on student involvement. Tinto (1987) acknowledged that family and culture may play an important part in student decisions to depart from college. However, what Tinto (1987) stressed (and what scholars and practitioners are likely to emphasize) was that "[i]n some situations, external social systems may work counter to the demands of institutional life. When the academic and social systems of the institution are weak, the countervailing external demands may seriously undermine the individual's ability to persist on completion" (p. 108). Even in his 1993 revised model, Tinto argues that external elements are secondary to those in college, conditioning but not determining the character of the experience on campus. Researchers have validated some of the negative effects of the external environment. For example, Terenzini and others (1993) found that "friends who did not attend college could complicate the transition by anchoring students to old networks of friends and patterns of behavior rather than allowing them to explore and learn abut their new college environment" (p. 5). Similarly, parents who feel anxious about students leaving home may function as liabilities. Nora, Cabrera, Hagedorn, and Pascarella (1996) found that minority students who needed to work off-campus for financial reasons were 36% more likely than those who did not to drop out of college. Moreover, the researchers found that female students, as opposed to male students, who were required to leave campus immediately after class to help care for a family member were 83% more likely to withdraw. 

However, not everything external is a liability. For example, Terenzini and others (1993) found that pre-college friends performed a "bridge function," providing support and encouragement. And with few exceptions, students named family members when asked "Who are the most important people in your life right now?" Jalomo (1995) found that there were more out-of-class agents helping students to make connections on campus than in-class agents. Clearly, much more research is needed to assess the positive and negative influence of the external environment and how students negotiate external influences, not only during the first-year of college, but throughout the student's collegiate experience.
 

The conceptual issues presented in this chapter, based both on empirical evidence and conceptual critiques, substantiate that Tinto's college student retention theory needs to be taken to an even higher level of theoretical development. Tinto has done this through extensions and refinements of his theory (Braxton, Sullivan, & Johnson, 1998). However, Attinasi (1989, 1994) and Tierney (1992) would likely go as far as rejecting the theory and building another that is capable of reflecting subtle processes (particularly cultural and political and emerging from qualitative analysis) involved in persistence. We believe that with all that is now known about student retention, it is quite possible that a totally new theory is needed to take Tinto's theory to a different level. Moreover, knowledge from disciplines other than education can also be used to develop new theoretical perspectives regarding student retention. For example, Hurtado (1997) has developed a "social engagement model [that] takes into account gender as well as other significant social identities like ethnicity/race, class, and sexuality to study how groups change as they come into contact with each other" (p. 299).

Employing a social psychology perspective, Hurtado (1997) advocates that in order to understand cultural transformations in an increasingly complex and multicultural society, as in the case when students from one group enter the sphere of social engagement of another group, requires not an assimilation/acculturation framework, but a social engagement model. Hurtado (1994) has employed a social engagement framework to study the participation of Latino parents in school. Hurtado's analysis of Latino parents' participation in school is quite similar to how one might analyze college student retention. 

For example, research findings illustrate that working-class Latino students are not as likely to get involved in the academic and social domains of the college as often as Whites do. Engagement is usually defined as participating in clubs and organizations, meeting with faculty in- and out-of-class, etc. Hurtado would argue that this narrow definition of student engagement is predominately based in the dominant group's perspective and not from the Latino students' view of what is possible and desirable for them. Indeed, Rendón (1994), Jalomo (1995), and Terenzini et al. (1994) have found that involvement is not an easy thing for nontraditional students from working-class backgrounds and that both in- and out-of-class validation were essential to their engagement and persistence. Validation is a powerful, interactive process involving a student and a validating agent. Much of the validation occurred out-of-class (with friends, parents, spouses, etc.), substantiating that there are other forms of engagement that can have a positive impact on persistence. These researchers employed qualitative methods that allowed students to express who and what was making a difference in their academic lives and why this was so. 

If these researchers had relied only on an assimilation/acculturation framework (i.e., narrowly measuring student traits that restricted minority group involvement), then they most likely would have reached the following conclusions: 

1) Latino students from working-class backgrounds are not as academically and socially integrated in college as are White students, leading to their higher dropout rate. 

2) Traditional, primarily White students from upper- and middle-class backgrounds are more engaged in college than Latino students and that accounts for their higher levels of educational adjustment and attainment. 

3) Consequently, we need to encourage working-class Latino students to avail themselves of services and opportunities that can increase their college retention rates. Further, because White students score the highest on scales of college involvement, they are the models that all students should emulate. 

An assimilation/acculturation framework would not allow Latino students to contribute their own perceptions and definitions of all that constitutes integration. Nor would their definitions influence the views of White students. We would also not be able to discuss the internal variations of each group, i.e., there are Hispanic students who exhibit very high achievement and engagement levels, and there are White students who do not. Many studies may not capture much of the variability in withdrawal decisions because of the misspecification of important constructs. Findings may turn out to be statistically significant, even though very little of the variance is explained. In these cases what may be most interesting is not what was statistically significant. Rather, the most important finding could be that there are other multiple, unaccounted factors that may be influencing retention.

Hurtado (1997) explains that a social engagement model, which has at its core a definitional approach to differences in social adaptations, would yield different results. Besides standard measures of college integration, there would be measures that allowed different groups of students to provide their own definitions of what they consider to be engagement and why. It could very well be that Latino students would report that they considered cultural activities, external relationships with family and friends, and race-based programs as essential and vital to their personal and academic development. Students could also identify the systemic barriers to integration. Similarities among the different ethnic/racial groups in terms of engagement and in terms of barriers related to involvement could also be identified. These variables could then be incorporated into quantitative models for statistical testing. Strategies for facilitating in- and out-of-class involvement for both minority and majority students could be generated from these findings. The key issue is that the sole use of an acculturation/assimilation framework to study retention does not go far enough.

Taking existing retention/involvement theory to a more sophisticated level will require a thorough, thoughtful, and critical analysis of all of the quantitative and qualitative data that have been generated to date. Rather than operate in isolation, quantitative and qualitative researchers should be open to each other's methods, share findings, and probe further into the meaning of their results. We should also be open to theory developments in fields other than education. Multi-method approaches to the study of retention are likely to lift the current corpus of college persistence research. In short, we believe that the future of college student retention research offers exciting and viable possibilities to both uncover the dynamics involved in retention and to use data to shape practice and policy.
 

Researchers employing quantitative models based on Tintoís (1975, 1987) depiction of student persistence have conceptually advanced some of the factors and interrelationships postulated in Tintoís model (i.e., Nora & Cabrera, 1993, 1996; Nora et al., 1996; Cabrera & Nora, 1995; Cabrera, Nora, & Castaneda, 1992, 1993). Qualitative studies also provide some support to Tintoís propositions (Terenzini et al., 1993; Jalomo, 1995). But while traditional theories of student retention and involvement have been useful in providing a foundation for the study of persistence, they need to be taken further, as much more work needs to be done to uncover race, class, and gender issues (among others) that impact retention for diverse students in diverse institutions. Certainly, the theoretical issues regarding separation, transition, and incorporation we have presented in this chapter provide avenues for conducting future research. Yet we stress that the ideas presented here are intended to go beyond stirring intellectual discussion that will lift theory and research. 

Minority students are altering the nature of higher education in many ways. Over the past 20 years we have witnessed dramatic changes in the classroom and the curriculum (with the inclusion of ethnic/racial perspectives and the use of learning communities), in student services (with race-based programs), and in faculty and staff composition, among others. While we believe that theory-building is important, out of scholarly discussions and research should come advances in the development and the dramatic transformation of academic and student services. Assuming that good social scientists are also caring humanitarians, the goal of student retention research transcends making conceptual modifications in theoretical models. In the end, students will elect to stay or leave college not so much because of a theory, but because college and university faculty and administrators have made transformative shifts in governance, curriculum development, in- and out-of-class teaching and learning, student programming and other institutional dimensions that affect students on a daily basis. Consequently, connecting retention research to field practitioners and policy makers in new and creative ways that involve collaborative relationships and mutual learning experiences can take student retention research to a whole new level of theoretical accuracy and applicability.