Introduction

The transformations resulting from the adoption of democratization policies in higher education have increasingly highlighted the need to discuss and think in a more consistent way about the retention process in this stage of education. Expansion and democratization are the two words that describe the changes that Brazilian higher education has undergone in recent years. With regard to this context of expansion, Neves et al. (2018) affirm that themes that aim to understand its effects predominate in studies currently developed in Brazil. The theoretical perspective of these studies has been directly or indirectly linked to the relationship between educational inequality and social stratification. Other themes, however, have been gradually introduced in the Brazilian context and, among them, the authors listed some that still require further theoretical deepening, and the themes of access, retention and dropout in higher education are among them.

The need for further deepening of this subject had already been pointed out previously, both in relation to the processes that students go through in order to become competent members of the university community (Teixeira, 2011; Carneiro & Sampaio, 2011) and regarding to the role of the institution in this process (Heringer, 2013). Thus, addressing this dimension of democratization – the persistence of students – helps us to think about in ways to deal with issues related to dropout, retention, increasing completion rates, as well as the university experience and the interaction between students and institution.

In order to contribute to the debate on retention in higher education, we propose to present the theoretical contributions of the international literature. Most of the explanatory models about retention and dropout in higher education, as we will see, are thought from or in opposition to the model proposed by Tinto (1975; 1993). The theoretical perspectives of Vincent Tinto and Alain Coulon will have special emphasis on the discussion about retention in higher education in this work, since Tinto's model has maintained, over the years, a “quasi-paradigmatic” status (Braxton et al., 2000, p.569) in the researches that deal with the subject, and the Coulon's model due to the growing space that it has gained in the researches developed in recent years in the Brazilian context (Sampaio, 2011; Heringer et al., 2014; Santos et al., 2017).

Tinto and Coulon bring in their models the processes of estrangement of the academic space, familiarization and subsequent involvement or integration, which are analyzed from different perspectives to explain, sociologically, how this process develops. We will see that the international literature helps us to think about the process of persistence from the perspective of the institution and, in this sense, we use the term retention, which concerns the actions practiced by a higher education institution with the objective of avoiding the dropout of students and, in this way, support them to complete their respective courses.

Afterwards we will present Vincent Tinto's theoretical model, the theoretical discussion that precedes it, Coulon's model, the limitations and potentialities of the mentioned models and, in this regard, how they support reflections and actions that consider the role of those involved in the process of retention in higher education in Brazil.

Methodology

The reflections proposed in this work are structured from the analysis of a specific literature that favours the two main theoretical models used to think about issues related to retention in higher education. The choice of these two authors over others was due to the fact that when we did the literature review, we identified that the notions of social and academic integration (Tinto, 1975) and student affiliation (Coulon, 2008) are recurrently used in the texts that deal with the theme of retention in higher education. Sometimes the referred concepts were not properly deepened in the texts and regarding Tinto's propositions as there are still no texts translated into Portuguese, the difficulties in the appropriation of his concepts become even greater.

The theoretical effort to bring together in a single text two authors who thought about retention models in different contexts and considered different audiences is justified: [1] by the relevance of their theoretical propositions – already discussed for decades – to think about contemporary contexts of retention in higher education in Brazil; [2] by the proposal of facilitating access to the theoretical models of these authors through a synthetic exposition, highlighting their main limitations and potentialities.

Among Tinto’s texts that we choose (1975; 1982; 1988; 1993; 1999; 2006; 2017) we privilege those that bring greater details about the notions of social and academic integration, the temporality of the integration process, the institutional role in this process. Recently the author indicated that the actions of retention by universities need to consider the students' perspective. In Coulon (2008; 2017), the author brings the notion of student affiliation, the temporality inherent to this process and the defense of a pedagogy of affiliation.

The discussion preceding Tinto's model

Tinto's (1975) model of dropout makes an application of Durkheim’s theory of suicide in order to explain the stages of academic and intellectual integration that may or may not lead to dropping out of higher education. However, the first researcher who used Durkheim's theory to think about dropout in higher education was William G. Spady (1970). In order to try to understand this phenomenon, Spady (1970) considers some literature reviews (Chart 1) performed in the 1960s and assesses that the studies developed, until then, lacked theoretical and empirical coherence. The author refers to the generalizations regarding the relationship between dropout and family background, ability and academic performance as "comfortable". He suggests that future work could be more "eclectic" in its approaches and more explicit in establishing relationships between student attributes and those of the institutional environment.

Chart 1

Literature review on higher education dropouts in the United States – 1960s

Author	Comments
Knoell (1960; 1966)	He classified the surveys conducted until then into 4 categories: [1] census studies (record of dropout, transfer and retention rates both by institution and among them); [2] “autopsy” studies (a survey of self-reported reasons by students leaving the institution); [3] case studies (usually, longitudinal studies with students identified at the time of admission as “potential risk” of dropping out); [4] predictive studies (use a range of admission variables in order to generate predictive equations for a variety of measures of academic “success”).
Marsh (1966)	He classified the same literature into 3 categories: [1] philosophical and theoretical studies (usually include recommendations for action, assuming that dropouts must be combated); [2] descriptive studies (they describe characteristics of dropouts, how they lived as students and the reasons they gave for dropping out); [3] predictive studies (similar description presented by Knoell).
Sexton (1965)	The variable “motivation" is considered crucial in explaining dropout. The student's "maturity" would be a critical aspect as it represents the ability to control the “irrational anxieties” that hinder motivation and, therefore, the completion of the course.
Summerskill (1962)	He also gave considerable importance to motivation in his literature review. He pointed, however, to the difficulty of operationalizing this variable in the analyses undertaken on dropout. The author also considers that those who drop out are generally less “adaptable” to the academic environment, “non-conformists”.

Source: Spady (1970). Data organized by the author.

Spady (1970) points out that there are two operational definitions of dropout that are accepted: [1] a dropout is anyone who has left the institution in which he or she was enrolled; [2] a dropout is anyone who has not received a higher education degree from any institution. The first definition is limited to the level of the institution, specifically identifies what happens at the local level and does not allow us to think about how this phenomenon occurs in the system. The second definition, on the other hand, requires many data to be collected and would be more applicable to longitudinal research that "follows" the individual's journey for a given time in order to identify which institutions someone may have gone through and whether was able to complete their course at some point.

In Spady's (1970) model, in addition to factors concerning to the students' social background (socioeconomic and previous academic experiences), the author considers other factors to explaining the dropout phenomenon in higher education such as: the influence of the student's gender on the definition of educational goals and interests, the relationship between personal dispositions and the student's “maturity” to continue the course and the nature of interpersonal relationships developed in the academic environment.

The interactions that a student experiences occur in two systems that the author names as academic and social. There are two important elements in each of these systems that may influence the decision to continue or not in higher education. In the academic system are the grades and intellectual development, and in the social system, the first is a condition called normative congruence and the second is peer support.

In the academic system, grades are a practical reference and easier to appreciate regarding the student's performance. The perception regarding intellectual development, however, can vary depending on the educational goals of each one, also changing the meanings attributed, such as those who see higher education as an important part of their personal development and those who do not see it as something so significant.

As for the social system, normative congruence has to do with a notion of “success” in which the student's individual dispositions (attitudes, interest, and personal characteristics) are compatible with the attributes and influences of the academic environment. Peer support is the establishment of close relationships with other people who are also part of this system. These two elements of the university social system are pointed out by Spady (1970) as the components of social integration present in Durkheim's suicide theory.

Although departure from higher education is clearly a less drastic way of rejecting social interaction, Spady (1970) believes that the social conditions that generate dropout are similar, that is, lack of close and consistent interaction with others, maintenance of different values and orientations from peers, and lack of sense of compatibility with an immediate social system.

Vincent Tinto's theoretical model

Spady's (1970) contributions were a great influence in the development of the theoretical model of dropout proposed by Tinto and Cullen (1973). In the early 1970s, the Office of Planning, Budgeting and Evaluation in the U.S. Office of Education asked Vincent Tinto and John Cullen to elaborate a report containing a review and theoretical synthesis of research on dropouts in higher education.

Among the objectives of the work done by the authors, they should develop a theoretical model on dropout that would allow to synthetize the research that they were doing and explain, in longitudinal terms, the dropout process in higher education as well. Regarding this report, we extracted from it only the discussion of the proposed theoretical dropout model by the authors.

Tinto and Cullen (1973) start from the same operational definition presented by Spady (1970) in which dropout refers [1] to people who leave the institution in which they were enrolled; [2] only to people who have never received any degree/diploma from any higher education institution. For the development of their own theoretical model, the authors choose the first concept – even knowing its limitations – which considers the institution's perspective and not the student's. The authors intend to develop a more appropriate definition of dropout. They state that these conceptions of dropping out have two main limitations: the tendency to focus on the efficiency of the institution (use of resources) instead of being more concerned with effectiveness (achieving the desired result) and the fact that both ignore the student's perspective.

Ignoring the student's perspective neglects two points: the fact that individuals entering institutions have a variety of skills, interests, motivations and levels of commitment to the goal of completing the course and that higher education, of any nature, may be inadequate to the needs, desires and/or interests of a certain number of individuals who still go to college. Ignoring the perspective of the individual in this process implies an idea of inferiority of those who do not proceed. Another aspect, more comprehensive, involves the discussion that should be made regarding the notion of higher education as the only space for high-level training after high school and, therefore, reinforces the trend of expanding higher education rather than reconsidering this status (Tinto & Cullen, 1973).

The model presented by Tinto and Cullen (1973) is based on Durkheim's theory of suicide and the theoretical perspective that considers the analysis of costs and benefits in the development of action. The authors are interested in explaining and understanding how individual, social and institutional characteristics relate to the process of dropping out of higher education. Regarding suicide theory, Tinto and Cullen similarly use Durkheim's perspective on social integration, or better, the lack of it as an important factor in understanding this process of social disconnection. For Durkheim, the chances of suicide become greater when there is insufficiency in both moral integration and collective affiliation – the first regarding the sharing of social values and the last to the interaction with others.

For the authors, dropping out from higher education would be the result of both the lack of “consistent and rewarding” interaction with other peers and the maintenance of values incompatible with those shared with most of the academic community. From this understanding, it is assumed that the lack of social integration in the college would result in low commitment to the institution and consequently, it would increase the probability of dropping out.

At the institutional level, it is important to differentiate the social dimension and the academic dimension in order to understand the possible institutional dropout types – in which the student decides to leave or when the institution dismiss the student – and the different types of interactions and social and intellectual demands that students are exposed to since they can be successful in one dimension and face difficulties in the other.

The application of Durkheim's suicide theory to the phenomenon of school dropout does not, by itself, produce a theory that helps to explain how different individuals adopt various forms of school leaving behavior. On the contrary, it is a descriptive model that specifies a longitudinal process of interactions that can produce different forms of persistence and departure behaviors (Tinto, 1975).

The perspective of cost-benefit analysis applied to the discussion of dropout in higher education, adds to the theory of Tinto and Cullen (1973) the influence of “external events” to the academic environment. From this point of view, the authors recognize that the decision to withdraw from college may not have a direct relationship with the interactions that occur in the university. In this perspective, it is crucial the individual's perception of the "reality", the way the student interprets it as more or less beneficial, as more or less compatible with his journey in higher education. The perception of the individual, however, varies according to his own characteristics and characteristics of the academic environment that he attends. In the interaction of the individual with the institution is that he or she evaluates the possibilities to continue or not.

Thus, the authors propose a theoretical model of multidimensional dropout that results from the interaction between individual and institution and it is influenced by the characteristics of both. This model considers the individual characteristics of students, their family background and previous educational experiences and how they influence expectations and motivations regarding the college experience. A point considered central to a student's decision to drop out of college is the “commitment to the goal” [of graduating] in which, it is judged, the higher an individual's level of commitment to the objective of completing college, the lower the probability to drop out. This commitment to the completion of the course is integrated with the commitment to the institution (Tinto & Cullen, 1973).

The way in which the phenomenon of dropout in higher education is dealt with may vary depending on who is the party interested in the theme. Tinto (1982) points out the three main interested parties in this process: students, the institution and the state. The act of dropping out of higher education can, in this sense, be interpreted in several ways, depending on the profile of the student, who will be affected and how students and institution will be affected after the event.

This withdrawal, however, can be interpreted in several ways. The meanings attributed to dropout by students and institution may diverge and go beyond the notion of “failure”. The student's goals and intentions when entering higher education must be taken into consideration. Depending on the objectives, students may have different patterns of interaction with the institution. The differences also extend to the departure processes experienced by different groups of students and in the different areas of knowledge in universities (Tinto, 1982).

The dropout at the beginning and at the end of the course can also have different characteristics and motivations. It is necessary to know if it is linked to difficulties in the process of transition from high school to higher education or to “path problems” that originated inside or outside the academic space. Tinto (1982) says that the transition to higher education is difficult for all students, whether they are considered "typical" – those who can dedicate themselves exclusively to their studies –, whether they are considered "foreigners" or "non-typical", that is, those who do not have exclusive dedication to their studies, who need to work and who are part of disadvantaged social groups.

Regarding to the institutional perspective on dropout, in practical terms, it is much simpler to report institutional withdrawal as abandonment. However, to know how the institution interprets its “losses”, it is necessary to know its educational objectives, and which is the student profile “valued” by the institution, that is, to know how much “interests” the institution to retain certain types of students.

Tinto (1982) claims that it is not clear whether all departures from higher education require equal attention or require similar forms of action on the part of the institution. The difficulty in defining what would be abandonment is to discern which types of withdrawal – among all that can occur – should be considered abandonment in the strict sense and which types should be considered the normal result of the functioning of the institution.

Therefore, the author indicates how relevant it is for the understanding of the dropout phenomenon to know the more general institutional values and conceptions, since these inform the patterns of interaction between students and professors and between these and their own peers and the rest of the academic community and, more than that, whether this type of interaction and environment - constructed and shared - is favorable for retention and for what type of student.

It is also seen that the selectivity of an institution does not end in the admission process, and it should be considered that the planning and execution of retention actions take into account the standards [and values] of selectivity - which vary not only by institution, but also by area of knowledge.

Thus, it can be said that the internal selectivity of the institution is spread all over the academic journey through criteria for granting financial assistance and/or scholarships for scientific initiation, criteria for participation in leagues and academic organizations or other extracurricular activities, curricular organization and academic routines. In other words, it is advantageous to have time for dedication to studies and/or to invest in relationships or other activities -, among other possibilities for interaction that can create academic integration and intellectual development for the student.

Tinto (1988) provides further details of his theoretical model relying on anthropology, more specifically the perspective of rites of passage – becoming a member – in tribal societies proposed by Van Gennep (1960). He intends to highlight what he calls the temporal dimension of the institutional withdrawal process, which would unfold in different stages in the initial years of the course. These stages refer to the three subdivisions that Van Gennep makes of the category “rites of passage”: separation, transition and incorporation. These three stages represent the phases through which freshmen go until they become competent members of the university community and duly committed to the completion of the course.

The separation phase implies a dissociation, in several levels, of the communities which the student maintained some bond before college, typically their high school or place of residence. Usually, this process requires transformation that may involve the rejection of the knowledge and norms to which the student was previously attached. This phase can also be characterized as a disorientation phase. This process will be more difficult the further the environment and demands of the college are from the social and intellectual characteristics of their previous socialization.

The transition phase is characterized by the passage from the old to the new, from the old associations with the past to the expected associations with the communities of the present. It is, in fact, a stage of fragility, because the links with the past socialization are loose and the links with the current community are not yet consolidated. Students will react differently to this period of stress in their trajectory.

Tinto (1988) indicates that there is an “inevitable fact” at this point in the trajectory, that some students will not be willing to deal with this stress of transition because they are not sufficiently committed either to educational objectives or to the institution they have joined. Others, however, will be so committed that they will be able to do anything to stay. The author also points out that the institution should mobilize to help students deal with this stage.

After these two phases, comes the incorporation stage, in which the student needs to recognize and adopt the appropriate norms of this new community. Therefore, the student must establish active contact with other members, whether they are students or professors, in order to avoid isolation. New patterns of interaction with members of the academic community are being established and valued. Although the student will still maintain contact with his "old socialization", it will occur from the viewpoint of his new group. In this perspective, the process of institutional withdrawal and institutional persistence are two sides of the same coin, in that they point to the successful (or not) passage from one phase to another of the "rite of passage".

Despite making use of these three typical stages – rites of passage – to illustrate the trajectory of persistence and retention of students in higher education, it should be noted that, unlike what happens in traditional (tribal) societies, there are no formal rituals established at first to mark the transition from one phase to the other. In the context of higher education, academic orientation can be an important tool in this process of teaching perceptions and postures that are often not obvious to students who in most cases usually make their way through the "institutional labyrinth" in an individualized way, at their own risk (Tinto, 1988).

The author emphasizes that institutional actions aimed at students' retention in higher education should focus on the first year, right at the beginning of their trajectories. Students will continue to have needs to be met after the first year, however, this period has shown to be critical in terms of retention.

Tinto (1988) adds this perspective of rites of passage to his theory in order to increase the possibilities of analysis. He recognizes, however, that his previous work, based on Durkheim's theory of suicide, has been much more recurrent in research in the area. The vision that derives from Durkheim– present in Spady (1970) and Tinto (1975) – is a mapping of a general theory of dropout that proposes to explain how institutional forces in their social and intellectual (academic) dimensions shape the process of incorporation, or in other words, integration of the individual in academic life. The intention with the addition of Van Gennep's perspective (1960) is to provide a temporal dimension, describing the longitudinal stages of the academic integration process.

Tinto (1988) states that he has no intention of simplifying the complex and fluid process of institutional withdrawal. For the author, these stages are "abstractions" used to facilitate the process of analyzing this phenomenon that ranges according to each [group of] student[s] and area of knowledge. It may also be the case that some students are not aware of the transition required to integrate into academic life. Others may experience these stages in an isolated way or simultaneously. In any case, providing information about stages or common events to the university experience can be beneficial to students inserted in this process and is in this sense that the notion of rites of passage is used.

Another aspect that deserves to be noted - and is also pointed out by the author - is that this theory is thought considering a profile of young students, recent high school graduates. It does not mention adults or transferred students with some prior experience in higher education. However, he believes that these groups are susceptible to the same phases, and may only experience them in different ways, due to their characteristics.

Involvement, both social and academic, is considered fundamental for retention, however, according to Tinto (2006), what is still not so clear is how to promote this integration in different contexts and with different students in order to promote retention, because the most recurrent strategy to promote integration of students in the university or to meet eventual training demands are the "additional courses" that are created according to emerging demands (Tinto, 1999).

Tinto (1999) also points out that some institutions need to face the retention problem more seriously, due to the small (structural) changes and lack of mobilization in conducting institutional actions and policies in this direction. He points to four institutional conditions that are evidenced to be important to promote retention: [1] information/guidance; [2] academic support; [3] engagement; and [4] learning.

Thus, the clearer and more consistent the information regarding institutional demands, the more likely that students will persist and graduate. This happens because students need to understand the "guide to completion" and know how to use these instructions in order to decide and achieve personal goals. Academic support should be available to students and integrated into other spaces of interaction that they have with the institution. As for involvement or "becoming a member", this point is related to the frequency and quality of interactions with professors, other students and other members of the academic community, which has been shown to be an important predictor of persistence. And the last point concerns learning, because, according to the author, students who learn are students who stay (Tinto, 1999).

"Active involvement" is presented as a key in this process, which would be able to promote learning, especially when this experience takes place through interaction with peers, which seems to be uncommon for most freshman students, who usually have their learning experiences - social or academic - in isolation from others (Tinto, 1999).

There is no doubt that there are many challenges for the institutions to promote these conditions of persistence, especially if we consider that many students need to combine studies and work. For them, the classroom may be the only place where they meet their professors, their classmates, and get involved with the curriculum.

For this reason, Tinto (1999; 2006) says that retention actions must not only include but must begin in the classroom. In this context, the institution and the teaching staff become fundamental in the execution of these actions, even though, according to Tinto (2006), there is a challenge of another order, because the professors at universities and in higher education, in general, are the ones who do not have specific training to teach their students.

Alain Coulon's theoretical model

Alain Coulon's research took place in France with the University of Paris 8 students, with whom he conducted processes of listening, conversations, and observation during his first months at the institution. In addition, he guided freshman students enrolled in his course to keep a diary during the first three months of their college journey.

Coulon (2008, p. 31) considers that after entering university "learning the student's occupation" is the most important task, meaning the relationship that new students establish with the university rules and knowledge. To explain how this occurs, the author describes the affiliation process, relying on the formulation of Van Gennep (1960). Coulon (2017) justifies that Van Gennep's work helped him to think about and classify the large amount of data that he had obtained through his research with Paris 8 students, their experiences could be interpreted in the light of rites of passage.

Coulon (2008) says that the affiliation process starts from the admission at the university, where the passage from pupil status to university student occurs, since, for the author, a person who arrives at the university does not automatically acquire the status of student, this initial stage would be that of a claimant to higher education. For him there is a clear distinction between "being a pupil" and being a student. Pupil is the one from basic education, while the student would be the young person who enters the higher education level. Entry into university life would in this case be a passage from one state to another. The competence of being a student is acquired through an initiation ritual into this new universe.

Learning the student's occupation, even if it is a temporary status, is an essential task for the student's persistence at the university. Affiliation, in other words, the acquisition of a new social status, is a process that occurs in three stages: (i) the time of estrangement, (ii) the time of learning, and (iii) the time of affiliation as such.

The first year is the one for learning the institution, and in this year, Coulon (2008) defends that the applicants to higher education should join their efforts around one goal: becoming a professional student. These efforts, however, would only make sense if there were a project whose future perspectives would justify a significant use of time and, in addition to this, the institution would provide conditions that favor retention.

The students would basically go through three stages that correspond to the three times of the affiliation process. The first is the stage of separation from the past at school or the time of estrangement, in which the student still sees the university as an unfamiliar environment and the process of articulation between the university and the student's future needs to be started. The second is the margin stage or the time of learning. In this second moment, the student's journey is still unstable, because he has no longer a past, but he still has no future, it is a period of meaning attribution, of learning a (new) perspective of the future. Finally, we have the admission stage or the time of affiliation. The duration of this stage varies according to the assimilation of the rules, and from here the risk of abandonment becomes gradually more distant.

Affiliation is taken as an autonomy learning process that occurs by the active participation of the student in a collective task. Although it is not a fully finished process, it is possible to speak of affiliation when at a given moment, the apprehension of certain dispositions is sufficient for there to be mutual recognition among the individuals who are part of the same group or the same institution. In the journey that the individual creates for himself a habitus as a student, he experiences a process of progressive familiarization, of elaboration of strategies until finally he can incorporate routines that are the first signs of affiliation.

The incorporation of the student habitus is in this way, a process of individualized and collective construction, because, on the one hand, the student needs to decide how much effort to employ in this process, and on the other hand, he or she needs to know how to follow instructions. In other words, to count on other’s help in the experience and interpretation of the most fundamental rules.

The hypothesis of academic success is linked to the measure of the student's affiliation, knowing how to operate with the proper administrative rules and with the intellectual work demanded by the University. Regarding the intellectual aspect of the affiliation process, Coulon highlights three operations considered fundamental for the realization of this, such as: reading, writing and thinking. In this perspective, “a competent student, from an intellectual point of view, knows how to identify the contents of intellectual work and at the same time, the implicit codes that organize them, hears what is not said and sees what is not designated"(Coulon, 2008, p. 256).

Becoming intellectually affiliated is also to establish links between private experience and university experience, in order to build an inside and outside connection. There is an ongoing process of intellectual affiliation when the perceptions, practices and habits previously valued by and in the university environment stop being external to the student and become incorporated, ceasing to be an effort and becoming something “natural”.

Coulon (2017, p. 1247) conceives and proposes a "pedagogy of affiliation" based on two activities: writing and learning documentary methodology. The daily writing or the "affiliation diary" is considered an important exercise for the appropriation of the symbolic dimension of the academic experience, and of the student's relationship with knowledge. Building the habit of writing would promote affiliation, insofar the reflections registered in the diary would bring clarity about the individual trajectories of the students. The documentary methodology would be a way - a course - for students to learn how to use the library's resources, as well as improve their reading, memory, and organizational skills with their studies.

Potentialities and limitations of Tinto and Coulon models

The theoretical models of Tinto (1975; 1993) and Coulon (2008) try to explain sociologically how the process of university integration/affiliation happens. In both models, the temporal dimension has an important place, since both emphasize that becoming a member of the university community happens gradually and the person who starts in higher education needs to know how to make a good use of the time available to experience this new environment. This temporal dimension is anchored in Van Gennep's (1960) notion of rites of passage, in which both authors subdivide it into three stages that go from estrangement, familiarization until reaching an "optimal point" in which the risk of departure becomes more unlikely.

We see in the literature that many authors (Astin, 1999; Bean, 1982; Braxton et al., 2000; Milem & Berger, 1997; Pascarella, 1980; Pascarella & Terenzini, 1979; Terenzini & Pascarella, 1980; Terenzini, 1982) who propose to discuss about retention in higher education establish a dialogue - directly or indirectly – with Tinto's theory. In general, these authors expose models of student dropout/persistence. In its constructs, the student and institution interaction (institutional experience) and the academic integration processes gain several approaches. Several propositions are presented regarding which factors (institutional or external to the institution), characteristics (of the student or the institution) or actions (institutional or students) would be more or less relevant both in explaining dropout and in increasing the probabilities of student retention.

There are studies that consider the centrality of the learning process in promoting student involvement (Astin, 1999), others privilege aspects of the interaction between student and professor and how this influences decisions to stay or leave higher education (Terenzini & Pascarella, 1980) and still others try to consider as many variables as possible (Bean, 1982), taking into account organizational aspects (referring to the institution), environmental variables (external factors and social context), attitudinal and outcome variables. Each model proposes to be as comprehensive as possible and points out the variables that it considers most relevant in explaining the phenomenon.

The hypothesis of academic and social integration proposed by Tinto's model - widely cited and accepted in the academic environment - has been tested, discussed, and opposed by several authors. In this sense, there are those who seek to identify which variables would be more or less significant in this integration process and those who oppose this perspective.

Braxton et al. (2000; 2008), besides the authors previously mentioned, also dialogue with Tinto and support his integrationist proposition, more specifically, regarding the importance of the classroom environment and consequently the student-professor interaction in the retention actions and in the construction of meaningful processes for students.

There are also criticisms to the use of the notion of "rites of passage" to think about the process of students' retention. Tierney (1992) says that there is a conceptual inadequacy in Tinto's model and that this notion is no longer appropriate, especially when it comes to underrepresented student profiles in higher education, who must to a certain extent abandon their culture and appropriate another dominant culture. The author opposes the idea of integration and proposes an alternative model in which universities would be multicultural entities and would promote difference rather than conformity.

In general terms, both models bring with them a notion of adjustment and conformation of the entering student to the university environment. The authors try to explain the process but not properly problematize institutional aspects that do not corroborate the efforts of certain groups to persist and finish their courses.

Another important point in both models is the issue of time that must be dedicated to university experiences. Tinto says that his model was designed for "typical" students, which means that they don't work and consequently have time to dedicate to their studies. The students who participated in Coulon's research, on the other hand, can be considered "atypical". However, in none of the models, we see the proper problematization of what it means to "have time" to dedicate to studies and to apprehend new cognitive and behavioral schemes inherent to university life.

Final considerations

Affiliating or integrating to the groups that are part of the university environment, more than a personal and institutional expectation is synonymous of a successful trajectory. Coulon (2008, p. 261) is categorical in affirming that "the student who is affiliated” is successful, which means that he or she has understood the rules, knows how to interpret them, and how to activate them according to everyday demands. The idea of academic success is widely associated with a learning process. You learn the institution, you learn the content and specific knowledge of a field of study, you learn a new way of relating to your peers and to the academic community and you learn a new perspective on the world, through new insights.

The retention process is, in this sense, and in many ways, a learning process. In this context, the academic support or pedagogical support gains new contours and the university has a privileged space for action with the students, so they can have an academic trajectory full of opportunities and effective conditions for learning.

Both the learning process in higher education and the persistence process are a collective construction that requires adjustments by both students and the institution in achieving a common goal, the completion of the course. Pedagogical support as a retention action, far from being the only answer to the problem, is configured as a powerful resource in the construction of a university that should not exempt itself from the responsibility in the teaching and learning process of its students.

Tinto (2017), more recently, points out that retention actions cannot put aside the perspective that students have on the process of persistence. He highlights that it is necessary to pay attention to some aspects that can influence students' motivation – especially those historically underrepresented in higher education – such as the notions of self-efficacy, sense of belonging and perception of the curriculum's value[2].

This isn't about ignoring the institutional perspective in the retention process. On the contrary, it is to add to the institutional perspective the student's view. In this way, the institution is challenged to adjust its focus taking into consideration the students it receives. Thus, it is not only up to the student to assimilate the institution, but also up to the institution to comprehend its students.

References

Astin, A. W. (1999). Student involvement: a developmental theory for higher education. Journal of College Student Development, 40(5), 518-529. https://www.middlesex.mass.edu/ace/downloads/astininv.pdf

Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191

Bean, J. P. (1982). Conceptual models of student attrition: how theory can help the institutional researcher. Em E. Pascarella (Ed.). New directions for institutional research: studying student attrition (pp.17-33). Jossey-Bass.

Braxton, J. M., Jones, W. A., Hirschy, A. S., & Hartley III, H. V. (2008). The role of active learning in college student persistence. New Directions for Teaching and Learning, 115, 71-83. https://doi.org/10.1002/tl.326

Braxton, J. M., Milem, J. F., & Sullivan, A. S. (2000). The influence of active learning on the college student departure process: toward a revision of Tinto's theory. The Journal of Higher Education, 71(5), 569-590. https://doi.org/10.2307/2649260

Carneiro, A. S. C., & Sampaio, S. M. R. (2011). Estudante de origem popular e afiliação institucional. Em S. M. R. Sampaio (org.). Observatório da vida estudantil: primeiros estudos. (pp. 53-69). EDUFBA.

Coulon, A. (2008). A condição de estudante: a entrada na vida universitária. EDUFBA.

Coulon, A. (2017). O ofício de estudante: a entrada na vida universitária. Educação e Pesquisa, 43(4), 1239-1250. https://doi.org/10.1590/S1517-9702201710167954

Heringer, R. (2013). O próximo passo: as políticas de permanência na universidade pública. Em A. Paiva (org.). Ação Afirmativa em questão: Brasil, Estados Unidos, África do Sul e França. Pallas.

Heringer, R., Vargas, H., & Honorato, G. (2014). Assistência estudantil e permanência na universidade pública: refletindo sobre os casos da UFRJ e da UFF. 38º Encontro Anual da Anpocs. Caxambu, MG, Brasil. https://www.anpocs.com/index.php/papers-38-encontro/gt-1/gt25-1/9066-assistencia-estudantil-e-permanencia-na-universidade-publica-refletindo-sobre-os-casos-da-ufrj-e-da-uff/file

Herpen, S. G. A., Meeuwisse, M., Hofman, W. H. A., & Severiens, S. E. (2020). A head start in higher education: the effect of a transition intervention on interaction, sense of belonging, and academic performance. Studies in Higher Education, 45(4), 862-877. https://doi.org/10.1080/03075079.2019.1572088

Knoell, D. M. (1960). Institutional research on retention and withdrawal. Em H. T. Sprauge (Ed.). Research on college students. Western Interstate Comission for Higher Education.

Knoell, D. M. (1966). A critical review of research on the college dropout. Em L. A. Pervin, L. E. Reik, & W. Dalrymple (Eds.). The college dropout and the utilization of talent. Princeton University Press.

Marsh, L. M. (1966). College dropout: a review. The Personnel and Guidance Journal, 44(5), 475-481. https://doi.org/10.1002/j.2164-4918.1966.tb03549.x

Milem, J. F., & Berger, J. B. (1997). A modified model of college student persistence: exploring the relationship between Astin’s theory of involvement and Tinto’s theory of student departure. Journal of College Student Development, 38(4), 387-400. https://core.ac.uk/download/pdf/32438027.pdf

Neves, C. E. B., Sampaio, H., & Heringer, R. (2018). A institucionalização da pesquisa sobre ensino superior no Brasil. Revista Brasileira de Sociologia, 6(12), 19-41. https://doi.org/10.20336/rbs.243

Pascarella, E. T. (1980). Student-Faculty informal contact and college outcomes. Review of Educational Research, 50 (4), 545-595. http://rer.sagepub.com/content/50/4/545

Pascarella, E. T., & Terenzini, P. T. (1979). Interaction effects in Spady and Tinto’s conceptual models of college attrition. Sociology of Education, 52(4), 197-210. https://doi.org/10.2307/2112401

Sampaio, S. M. R. (org.). (2011). Observatório da vida estudantil: primeiros estudos. EDUFBA.

Santos, G., Vasconcelos, L., & Sampaio, S. M. R. (orgs.). (2017). Observatório da vida estudantil: dez anos de estudos sobre a vida e cultura universitária, percurso e novas perspectivas. EDUFBA.

Sexton, V. S. (1965). Factors contributing to attrition in college populations: twenty-five years of research. Journal of General Psychology, 72(2), 301-326. https://doi.org/10.1080/00221309.1965.9710701

Spady, W. G. (1970). Dropouts from higher education: an interdisciplinary review and synthesis. Interchange 1, 64-85.

Summerskill, J. (1962). Dropouts from college. Em N. Sanford (Ed.). The American College (pp. 627-657). Wiley.

Teixeira, A. M. F. (2011). Entre a escola pública e a universidade: longa travessia para jovens de origem popular. Em S. M. R. Sampaio (org.). Observatório da vida estudantil: primeiros estudos (pp. 27-51). EDUFBA.

Terenzini, P. T. (1982). Designing attrition studies. Em E. Pascarella (Ed.). New directions for institutional research: studying student attrition (36), 55-71. Jossey-Bass.

Terenzini, P. T., & Pascarella, E. T. (1980). Toward the validation of Tinto’s model of college student attrition: a review of recent studies. Research in Higher Education, 12(3), 271-282. http://www.jstor.org/stable/40195370

Tierney, W. G. (1992). An anthropological analysis of student participation in college. The Journal of Higher Education, 63 (6), 603-618. https://doi.org/10.2307/1982046

Tinto, V. (1975). Dropout from higher education: a theoretical synthesis of recent research. Review of Educational Research, 45 (01), 89-125. https://doi.org/10.3102/00346543045001089

Tinto, V. (1982). Defining dropout: a matter of perspective. Em E. Pascarella (Ed.). New directions for institutional research: studying student attrition (36, pp. 3-15). Jossey-Bass.

Tinto, V. (1988). Stages of student departure: reflections on the longitudinal character of student leaving. Journal of Higher Education, 59 (04), 438-455. http://doi.org/10.1080/00221546.1988.11780199

Tinto, V. (1993). Leaving college: rethinking the causes and cures of student attrition. University of Chicago Press.

Tinto, V. (1999). Taking retention seriously: rethinking the first year of college. NACADA Journal, 19(2), 5-9. https://doi.org/10.12930/0271-9517-19.2.5

Tinto, V. (2006). Research and practice of student retention: what next? J. College Student Retention, 8(1). https://doi.org/10.2190/4YNU-4TMB-22DJ-AN4W

Tinto, V. (2017). Reflections on student persistence. Student Success, 8(2). https://doi.org/10.5204/ssj.v8i2.376

Tinto, V., & Cullen, J. (1973). Dropout in higher education: a review and theoretical synthesis of recent research. Office of Planning, Budgeting, and Evaluation. Department of Health, Education and Welfare.

Van Gennep, A. (1960). The rites of passage. University of Chicago Press.

[i] PhD in Education, Federal University of Rio de Janeiro (UFRJ) (2021). Pedagogue at Federal Institute of Education, Science and Technology of Maranhão.

[2] Regarding the notions of self-efficacy and sense of belonging, see, respectively, Bandura (1977) and Herpen et al. (2020).