Last changed 25 Feb 2008 ............... Length about 8,000 words (53,000 bytes).
(Major first draft completed 7 May 2003.)
This is a WWW document maintained by Steve Draper, installed at You may copy it. How to refer to it.

Web site logical path: [] [~steve] [localed] [dropout] [this page]


Go over papers for method criticism
Pursue wider lit. e.g. via Lynn chapter.
*Find measures used by others

Tinto's model of student retention

By Stephen W. Draper,   Department of Psychology,   University of Glasgow.

These notes are a very personal view, not well researched, and possibly severely flawed.

The first topic is what determines whether students stay on or drop out at universities. Various terms may reasonably be used for this area. The negative-looking ones are failure, dropout, attrition; the positive-looking ones are retention, persistence. Tinto offers a theory for understanding this. Elsewhere I also have some notes on basic comparative dropout rates.

The follow-on topic is about the school-university transition. In it I argue that this is in fact a sub-part of the Tinto issue.

Some other pointers related to Tinto are on another page including surveys related to dropouts at this university, a review of the literature related to Tinto, and a variety of diagrams expressing Tinto's theory.

Contents (click to jump to a section)

Tinto's model

The most commonly referred to model in the student retention/dropout literature is Tinto's. It was first offered in a literature review (Tinto, 1975), and so began with the support of being broadly consistent with a considerable range of other people's research, as well as having a theoretical derivation by analogy to Durkheim's model of suicide. It probably gains most support though because it immediately appeals to people's commonsense with its central notion of "integration". It is less clear whether there is much direct empirical support for it, and certainly it is hard to find direct empirical tests of and challenges to it. The literature claiming to support it seems to be about reporting weakly consistent evidence: not controlled experiments, nor comparing alterantive theories against Tinto's with respect to data.

Diagram of Tinto's model

This is adapted by me from Tinto,V. (1975) "Dropout from Higher Education: A Theoretical Synthesis of Recent Research" Review of Educational Research vol.45, pp.89-125.

Its central idea is that of "integration": it claims that whether a student persists or drops out is quite strongly predicted by their degree of academic integration, and social integration. These evolve over time, as integration and commmitment interact, with dropouts depending on commitment at the time of the decision. A first pass might perhaps try to measure these by:


Tinto was keen for studies to measure / distinguish different reasons for departure: being thrown out for failing exams vs. voluntary leaving. In reality there is a middle category where you can't tell if students are marked fail because they stopped attending (voluntary but didn't tell the university), or did badly and although not told they must leave this removed their commitment and they then decided to leave.

Common failings in papers reporting studies of Tinto's model

Such papers seldom report the actual questionnaire items used to operationalise the theory. This means that they may not test the theory at all, but the reader cannot know this. For instance Borglum & Kubala (2000) seem to have used a questionnaire designed for a quite different purpose, simply measuring student satisfaction with the college, yet assumed that all items could be classified post hoc as measuring a Tinto-related variable.

More perniciously, by never discussing the design of questionnaire items, major theoretical issues are ignored. For instance does "social integration" mean integration within that institution, or generally? Probably Tinto meant the former. Yet a student with no friends anywhere, and a student with plenty of friends who however are not enrolled at the same college are likely to show different tendencies to dropout. Scrutiny and discussion of individual questionnaire items is a good way to identify theoretical issues, and conversely eschewing such discussion furthermore makes it likely that no two studies are measuring the same thing, yet are unable to determine this.

Standard possible methodological failings in studies of dropout

Need 100% samples especially of the dropouts: any less, and self-selection must be likely to distort it by losing those ashamed in some way (or leave only those with the most distorted rationalisations).
Furthermore, face to face as opposed to paper instruments (i.e. interviews not postal questionnaires) may be very important for quality all round. Certainly comparing face to face persisters with paper dropouts could be bad.

Even then, it will be like interviewing people about their divorces: everyone will have a story, but it is a story they can live with, scarcely a dispassionate account. Rationalisation by each student, particularly dropouts, may mean that what they say about causes is not useful. They will be very likely to describe cause as external factors (the classic Social Psychology attribution error?). So for this, should attend only to data on external factors, and get it equally for persisters. In fact the Brown and Harris method of collecting descriptions of external factors for all, and getting a panel of experts to rate their seriousness "blindly", may be essential.

Similarly for "internal" and all "ask them" measures of attitude, Tinto integration etc.: we should ask all students before as well as after external events, and before exam results, and before dropouts. I.e. do prospective studies.


  1. Prospective studies, with measures (especially subjective/internal ones) taken before (as well as after) dropout events such as failing exams, and collecting these measures for both persisters and dropouts.
  2. For external events (always collected retrospectively), collect these for both persisters and dropouts.
  3. Get a panel to assess the seriousness of external events; don't trust subjective assessment. (And hence, don't use dropouts' own opinion on why they dropped out.)
  4. Must get 100% or random samples especially of dropouts (not self-selected samples).

New methods required

The literature to date seems not to provide strong proof or even good tests of the theory. However to do so would require a large programme of research with multiple methods, particularly to address the extensions to Tinto's model discussed below. The simplest approach is to generate large questionnaires, with items relating to parts of the model, and use correlations. The trouble with this is that it implicitly treats all factors as independent and as adding linearly. Thus it fails to test the structure of the model, and similarly cannot deal with quite simple aspects. For instance consider vitamins: eating more of one vitamin does not compensate for having too little of another; and furthermore, if most subjects have enough of one vitamin, correlations will be low and tell you as much about the population as they do about the importance of the factor. Techniques such as path analysis, and structural equation model testing try to addrss this to some extent, but probably not sufficiently. It seems extremely likely that learning is determined in some places by conjunctions like vitamins: a learner has to have all of a set of factors, and will fail if any one is deficient (e.g. must have both motivation and adequate study skills and adequate learning resources). In other places it is probably determined by disjunctions: a learner may either learn from lectures or from textbooks, but may well need only one of these alternatives to work well for them.

At the opposite end of a spectrum from statistical treatment of multiple factors at once, would be case studies: looking for cases where a particular feature of the model is crucial, for instance personal staff contact as essential for adequate "social integration" which in turn is an important pre-requisite for whether nor not a student seeks help when they need it. Slightly beyond case studies might be surveys measuring just this factor every 2 weeks (say), then when a jump is seen in an individual's measure on this, following up with an interview to identify what critical incident caused this shift. Such an approach might both operationalise and establish parts of the overall model, piece by piece.

Beyond the original model: my extensions/interpretation

Recently a different sociological approach claiming to be a rival has appeared (Braxton; 2002). A key phrase is "social capital" (see also in the Liz Thomas section below). However, perhaps we could see this as a part of what is implied in Tinto's model, but with more emphasis on integration between the student and social groups (and forces) outside the university.

Another notion is Bordieu's "habitus", which Thomas (2002b) explicates as "the norms and practices of particular social classes or groups". For me, the issue this indicates can be construed as to do with how the role of student has aspects to do with fitting the academic institution, with fellow students, and with external social groups and their views of the place and value of students.

What follows is my proposed extension of Tinto or synthesis of Tinto's original model with additional concepts. A further development of the concept(s) might expand the notion of "integration" in the following way. Firstly, consider it as a measure of fitting the role of student. Does the student feel that they fit happily into the role of student? Fit has two aspects: internally, do they feel it fits them from a personal perspective, and externally, do they feel happy in how others view them in this role. Fitting is about any causes of friction or dissonance, even those too slight to be consciously noticed and spoken about. We can see the role as having two major aspects, academic and social. The academic is about learning, and the activities necessary for that. The social is about fit with the groups the student cares about, both inside and outside the university. A person who identifies totally with being a student will care only about their place with other students, ignoring the values of any outside groups; someone who comes from a family that expects a university qualification will probably make friends in the university but also expect family and employers to regard being a university student as an expected and worthwhile stage in life; but someone from a family or group unused to university may have trouble with the mismatch between being a student and markers of respect such as a job, current income, an expensive car, children of one's own, etc.

Another dimension is to distinguish goals, methods, and effectiveness or achievement. Clearly a person may love an objective but dislike some method necessary to achieving it: may like writing essays but be bored by the preparatory reading (or vice versa), just as someone may love tropical holidays but be afraid of flying to get there. Treating the achievement as distinct from the goal is in a way redundant, but provides an opportunity to examine the gaps there can be -- for instance due to the problems of assessment -- between the measures used and the aims they are supposed to assess, and also between a student's aims and their actual achievement. A person can sometimes feel they love a subject and yet be hopeless at learning it. Another reason (for looking at both goals and achievement) is that a person may not have thought much about a goal, yet on failing to achieve it they feel a problem e.g. not getting on with staff or fellow students may not have been an aim one way or the other, but can subsequently be felt as a problem anyway.

The third distinction, between internal / external aspects of fit, comes from the standard distinction between intrinsic and extrinsic motivations for learning: whether you do it for personal reasons (interest, enjoyment, curiosity, "for its own sake"), or for extrinsic reasons (means to another end, to get the qualification, to be admired, ...). But in principle this can be applied not just to the motivation (the goal), but to activities/methods and results. For each activity may have some positive or negative inherent value for an individual apart from the goal, and this may be for intrinsic or extrinsic reasons. For instance asking questions in class might draw dislike from other students (extrinsic negative value), but be useful for the individual in checking whether they have understood (a standard personal learning technique with positive intrinsic value). In general, for methods and intrinsic/extrinsic, we should ask a) are there any things others (staff or students) require of you but you hate (or love); e.g. tutorials make you nervous, computer use is compulsory, hours in the lab are tedious. b) are there any things (learning methods) that you require or find important for your learning, but which others obstruct; e.g. you need to ask questions (but there's not time); you want time to think, but the lecture always rushes on; you want to discuss an idea, but everyone has to leave and there is no place or time to do it.

Summary of dimensions

Putting these together, we have [A] three fields for integration: academic, and social inside and outside the university. Fit (or conflict or dissonance) [B] might be divided into arising from goals (or wishes, or desires), from methods (or skills or capabilities or habits), and from effectiveness (or measurable achievement). And [C] there are always two aspects of fit: with the individual's own internal self, and with external demands on them. Multiplying these together would give us the set of questions and issues below: {Academic, social within university, social outwith the university} X {Goal, method, effectiveness} X {Intrinsic, extrinsic}. (I shall interpret the combination of methods with intrinsic/extrinsic as follows. Method-intrinsic: deals with the student's own existing methods and asks the question: are they allowed/used or obstructed /not useful. Method-extrinsic: deals with the methods / activities externally required by the course, and asks the question: does the student like or hate them, are they good or useless at them?)

For each resulting element I indicate one or more draft phrasings for a corresponding questionnaire item. These items may be in an open-ended form (asking the participant to tell us if there's anything that might be an issue in this category), or sometimes specific where experience suggests examples of specific things that have been a problem for some students. Additionally, for some I indicate a remedial intervention (abbreviated below to "fix") that might be tried if the aspect seemed a particular problem in a given context. Words in [square brackets] are pointers to other theoretical concepts.

The questions

Here is how the dimensions play out into parts, each with a draft questionnaire item.

Unresolved issues with this scheme

Goal - strategy - but also opportunities.

Does "integrated" mean:

Does social integration mean:

There aren't just 3 points on the dimension of {Goal, method, effectiveness}. Instead there are at least 5 points, maybe an arbitrary number. If so then should multiply out the schema above by 5 not 3 points. The basic idea is that some goals correspond to large external motivations, others are simply means to an end serving larger goals. And similarly, a large method like a lecture requires component skills from the student to benefit from it much. My suggestion for an expanded dimension might be:

  1. Goal
  2. Subgoal e.g. learning statistics as a subgoal of learning another topic. Learning to touch type as a subgoal of the whole course. Learning mind mapping as a study technique for the whole course.
  3. Large scale M-acts e.g. seminars, tutorials, lectures
  4. Small scale M-acts e.g. bringing a personal agenda to each tutorial; reviewing notes after every lecture; ...
  5. Effectiveness

Similarly, perhaps should split the goal and method points above and multiply all by {(don't know), Know, can do, fit/like}. As well as asking questions that presuppose they KNOW what is needed for methods (say), we should test this assumption by questions about whether they know what is needed. That is, do they HAVE:

I.e. it is not only fit between the students' methods and the required methods, but also the issue of knowledge of what method is needed, and then possession of that method.

Liz Thomas: 5 spheres of integration

Thomas (2002a) suggests 5, not 2 or 3, types of integration / spheres.

What do I think of this? well it is true that all of these have typical university structures associated with them, so if I want to explain the LTP perhaps I do need to expand to cover them? On the other hand, they are probably important to dropouts, but maybe not otherwise to learning.

She is interested in a) dropouts b) "widening access" i.e. getting and retaining a wider set of types of student. And argues with evidence that maximising these means attending more to all 5 spheres.

She uses, and partly explains, the notion of social capital. (Her paper gives some explanation of the concept and a number of references such as Shuller & Bamford (2000).) But perhaps it actually isn't necessary except broadly to think of this broader set of spheres, and the general idea (already in Tinto) that weakness for a student in one can be compensated by strengths in others. But in fact maybe her data (Thomas; 2002b) really partly goes against this: i.e. she found that money wasn't an important reason for presistence or dropout, and so isn't the same kind of predictor as, say, social integration.

Social capital (seems to) mean: prior acquisition of contacts substitutes to some extent for present knowledge. This is both learning but also actually connection to people/resources i.e. not just internal learning but connection. Actually consistent with Unix expertise: you can substitute knowing how to learn for already having learned; and consistent with socially distributed knowledge. I don't know if the metaphor of capital helps; but it is in another way a smple extension of the idea of pre-requisites from facts and skills in the chosen topic to other things.

So what do I take from it?
a) To predict dropouts, we may need all 5 spheres.
b) And they are all definitely about integration (e.g. economic: learning to live on that amount, and this is eased by living with others using the same constraints).
c) "Capital" does signal the advantage of pre-adaptation or prior preparation, and how it can be traded on to solve new problems rather than be the pre-solution.
d) And how it is not just about individual knowledge so much as working contacts: having access to the socially distributed resources important to being a student.

I'm dubious because:
a) The support and democratic spheres don't seem to affect all or even most students; but the others do. The capital metaphor may help in understanding the preconditions for these spheres too to work well; but I don't believe they are so important?
b) The economic sphere affects all students; yet her research apparently suggests it isn't as important in determining dropouts. So Tinto was right after all? focus on academic and social spheres.

There are really 3 possible views of this:

Tinto interventions

This section is to collect and list educational interventions that can be explained by Tinto-like theories (but often not by other theories). They may also be designed to increase poor scores on some Tinto-related variable e.g. "integration".

This is a crucial section because:

Classics / majors

Summer schools: to widen participation by increasing integration for targeted groups in advance.

The whole business of school qualifications as preparation for HE. Commonsense says that this is about knowledge pre-requisites: knowing facts and skills that will be necessary and presupposed at university. However it may really be a case of "pre-integration": of giving students previous experience of what the subject mattter, and its associated study patterns, feels like so that they can make an informed choice about what university course they may like and be competent at. A relevant study would be to measure prior conceptions of both subject matter, higher education, jobs, ... etc. as tacitly creating a pre-integration level.

Field trips
Reading parties
Cheese and wine welcome parties

So called peer assisted learning (PAL) or supplemental instruction: student-student mentoring.

Classic but not easily recognised as Tinto-relevant

Personal staff-student contact; "Empathy".
Tutor assignments and contacts.
Advisor assignments and contacts.

Feedback: summative assessment information to tell students that they "are" a Geographer or whatever. Rank in the class?

Summer scholarship / working in a staff member's lab.

Groupwork (i.e. organised and made compulsory by the course).
Study groups (i.e. student-only peer groups).
Amount of discussing in L-acts (class, seminars, ...) BOTH personal contributions AND seeing what others think.

Other HEI standards implied by Thomas

There are other units and services, widely funded by universities, to do with student support, and presumably likely to reduce dropouts. Implied by Thomas' expanded list of spheres of integration.

Finance: bursaries, scholarships, hardship funds, etc.

Support services e.g. counselling, health

Students' unions. Student representation on committees etc.

Designing new, ideal interventions

What do they need to be or include?

Where do these models fit?

How do such theories fit with anything?

Senses of "social"

I really want to integrate Tinto and the Laurillard model of the learning and teaching process (LTP). It does address the "social" aspects so missing from the Laurillard model, and in so doing explain some frequent activities put on that don't fit the Laurillard model e.g. summer schools, reading parties etc.

Senses of "social":

We could ask, and perhaps even find empirical answers to, which of these levels most determines a student's success (i.e. is it external forces like money and social class, individual taste for learning, or what). However one of the ways in which Tinto's approach may be better than some other ways of talking about this area is that it doesn't align with the simplistic question of whether the student or the university should be "blamed", despite what Ozga & Sukhnandan (1998) suggest. The metaphor of integration is about fit; it is not about one party adapting to the other, but about whether they go together well. Even more than that, like other human relationships (but unlike whether a square peg fits a round hole), integration is clearly the current outcome of a relationship of sequential interchanges which progressively modify that relationship: hopefully for the better. As a student has more successful interactions with a tutor, for example, they are likely not just to be learning a few extra facts but to feel more integrated with positive knock-on effects for instance in how willing they are to ask for further help in future, and to ask for it in a way that gets results from that individual tutor.

Why is this synthesis, and Tinto's part in it, important?

So in the end we should be able to:

What practical use could these models be?

These models are basically sociological ones. Do they have any potential for actually improving things in practice? It won't be easy, because there are so many ways in which a student's "integration" might be low: or to put it another way, students drop out for diverse reasons, and having a general "explanation" doesn't tell you how to do something effective for each student. However in principle we could imagine first developing a detailed diagnostic instrument e.g. using the questions above, and using that to determine what the particularly bad issues are in each situation (each department of each institution); and then select a remediating intervention specific to that diagnosis (e.g. the possible remedies also listed above in the framework). We're a long way from demonstrating this, though.

Tinto (1982) has a striking fact illustrated in a graph: that for the last 100 years the dropout rate for universities in the USA has been constant at 45%, despite big changes in the participation rate and amount of public funding. (Dropout rate was here defined as the ratio of undergraduate degrees awarded to the first-time enrollment four years before.) The second world war causes the only big wobble in the flat graph, and yet averaged across 10 years even there the rate is near-constant (because positive and negative blips cancelled out). In the UK and again in Europe, rates are very different, but perhaps largely constant in each. (Thomas 2002b gives the UK rate as 13% in 1982/3 and 17% in 1997/8 after great expansion, attributing these figures to a House of Commons Select Committee report.) Tinto discusses how that implies that such research is probably limited to dealing with social and/or local inequalities, rather than to overall change in dropout rates.

School-university transition

This section is about the transition from school to university, with particular reference to computing science. There seems to be a problem.

A related point of view is expressed by Tony Jenkins here.

Prose argument about this

What should the relationship be between what is taught at school and at university? The naive, but apparently commonsense, relationship is: whatever schools teach, universities don't need to teach but should assume as pre-requisites. Once established in schools, then universities should a) require it for entry, and b) stop re-teaching it. This content is to be thought of as facts, or perhaps skills that are directly tested.

I wanted to suggest that part of the issue may be that that commonsense model of school-university pre-requisites is actually wrong for most subjects, and perhaps particularly wrong for computing science. Pre-requisites may be facts, may be specific skills (e.g. integral calculus, debugging a program regardless of language), or they may be still more general: an orientation to a way of learning that suits a particular subject. Facts are almost entirely useless as a pre-requisites in computer science, not only individually but also in the big "lumps" of programming languages and specific packages such as Excel. Syllabuses [?spelling] written in these terms will fail as worthwhile pre-requisite qualifications (even though assessment within and outwith university is usually reliant on knowledge of such facts). Actually, I argued, this is also true to a greater degree than is usually acknowledged in other subjects such as English and Physics. For instance (if you'll accept a decaying memory of how it was a long time ago in England for physics as any kind of evidence), specific A-level material in physics was hardly ever re-used, but the maths I'd had to do was almost all vital from early on, but most important probably was that learning school physics was indeed a good guide to whether I'd enjoy university physics AND to the kinds of skill and activity involved in learning university physics. Thus the real function of requiring school physics in order to do university physics may really, contrary to the commonsense model, not be the explicit curriculum of Newton's law etc. (i.e. of facts) but of getting experience of what learning physics feels like, and so allowing the learner to make an informed choice of university subject. Insisting on it may possibly exclude some who would actually have turned out to be able to cope, but because the requirement has existed for a long time, it disadvantages few.

The main complaint from staff, but more importantly from students, in computing science is that school computing does not prepare them for university computing. They do NOT in fact say they "already have a substantial understanding of the subject matter" (as Kenneth suggests) and that it is all too easy. That is what the commonsense model predicts, but it doesn't seem to be what is actually the case. That is why universities feel justified in ignoring school computing science. On the other hand, the failure rates mean universities wouldn't mind at all at all if schools found a way to do useful preparation: but the most useful preparation (I suggest) would be in expectations, to pre-select students who would turn out to enjoy (and cope with) university computing science. So from this viewpoint, the challenge is to redesign school computing to do a job comparable to that done implicitly by school physics (say), rather than the apparently commonsense requirement of learning some facts and skills. In other subjects these overlap enough not to have to recognise the difference, but in computing science we may just not be able to get away with the commonsense but wrong idea of the relationship between school and university learning.

Transition: bullet point summary of my view

This is a summary of my whole theory of school-HE connection.

There is no reason to think one subject (e.g. computing science) is going to be just like any other in the matter of what is important for teaching, and hence what is important for school fore-runners to university forms of the subject.

Historically, subjects probably migrate from research down to schools. Part of this is learning how to teach it better.

Should school and HE forms of a subject be coordinated in any way?


  • The real importance for HE of prior qualifications may be, I hypothesise, giving learners an accurate feel for what the subject is like in content, what it is like in required study activities, and whether they would enjoy studying it. Studies and theories of HE dropouts usually show that "match" of student and subject is an important predictor of persistence vs. dropouts.
    This probably is what is good about trying to introduce the "scientific method" in primary schools. This is probably the real way in which school qualifications are useful entry requirements for many subjects (rather than specific content known).
    This is what may be really bad about current mismatch of some school computing studies and HE computing science.
    Covering the "same" topics: may only be damaging in that some learners believe they know it when they don't to the new standard required.

    Transition is arguably, as far as theory as opposed to implementation detail goes, a) pre-integration (i.e. a subarea of Tinto). b) How to interest learners in a subject with simpler, smaller, versions of it.

    Summer schools

    Summer schools are part of transition: pre-integration interventions, done by HE rather than by school.

    Lynn Walker's 1996 thesis says they worked here at University of Glasgow except for science in raising "participation" from deprived areas to that of the average.

    Is summer school meant to be better than first year teaching (smaller groups, and take advantage of this by more interaction and better learning activities) OR should it be realistic and so prepare them.

    Functions of summer schools may be all of these:

    1. Academic, institutional, bureaucratic integration
    2. Study skill preparation
    3. Subject: get you interested in it
    4. Subject: get an accurate feel for what it's like studying it.
    5. Social integration at least with that group, with those staff.


    Braxton,J.M. (ed.) (2000/02) Reworking the student departure puzzle (Vanderbilt University Press)

    Tinto,V. (1975) "Dropout from Higher Education: A Theoretical Synthesis of Recent Research" Review of Educational Research vol.45, pp.89-125.

    See also:
    Tinto,V. (1982) "Limits of theory and practice in student attrition" Journal of Higher Education vol.53 no.6 pp.687-700

    Tinto,V. (1988) "Stages of Student Departure: Reflection on the Longitudinal Character of Student Leaving" Journal of Higher Education vol.59 no.4 pp.438-455

    Ozga,J & Sukhnandan,L. (1998) "Undergraduate non-completion: Developing an explanatory model" Higher Education Quarterly vol.52 no.3 pp.316-333

    Tinto,V. (1987) Leaving College (Chicago,University of Chicago Press).

    But for criticism see Brunsden,V. & Davies,M. (2000) "Why do HE Students Drop Out? A test of Tinto's model" Journal of Further and Higher Education vol.24 no.3 pp.301-310

    Bill Patrick (2001) "Students Matter: Student Retention: who stays and who leaves" The University Newsletter This report is based on a survey of all first year students at the University of Glasgow, and is an example of the implicit influence of Tinto.

    Claire Carney & Sharon McNeish (2001) "Students Matter: Study links part-time work to student ill-health" The University Newsletter

    See also Rosanna Breen's PhD at Oxford Brookes.
    Breen,R. & Lindsay,R. (1999) "Academic research and student motivation" Studies in Higher Education vol.24 pp.75-93

    Tony Jenkins (2002) "On the difficulty of learning to program" LTSN conference

    Schuller,T. & Bamford,C. (2000) "A social capital approach to the analysis of continuing education: evidence from the UK Learning Society research programme" Oxford Review of Education vol.26 no.1 pp.5-19

    Thomas,E.A.M. (2002a) "Building social capital to improve student success" BERA conference

    Thomas,E.A.M. (2002b) "Student retention in Higher Education: The role of institutional habitus" Journal of Educational Policy vol.17 no.4 pp.423-432

    Lynn Walker (1996) An evaluation of the pre-university summer school at the University of Glasgow, 1986-1993, and its effects on student performance PhD thesis [Faculty of Arts, Department of Education], University of Glasgow. [Level 12 Spec Coll Thesis 10493]

    Yet more Tinto-related references

    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" Journal of Higher Education vol.75 no.5 pp.569-590
    [Shows stat.sig. positive effect of "active learning" e.g. class discussions on student retention.]

    Thomas,S.L. (2000) "Ties that bind: A social network approach to understanding student integration and persistence" Journal of Higher Education vol.75 no.5 pp.591-615

    Bray,N.J., Braxton,J.M. & Sullivan,A.S. (1999) "The influence of stress-related coping strategies on college student departure decisions" Journal of College Student Development vol.40 no.6 pp.645-657

    Elkins,S.A, Braxton,J.M., & James,G.W. (2000) "Tinto's separation stage and its influence on first-Semester college student persistence" Research in Higher Education vol.41 no.2 pp.251-268

    Borglum,K. & Kubala,T. (2000) "Academic and social integration of community college students: a case study" Community College Journal of Research and Practice vol.24 pp.567-576

    Web site logical path: [] [~steve] [localed] [this page]
    [Top of this page]