Kleroterion Forum

When a lottery is the right way to share, select, decide


  • Blog Stats

    • 1,422 hits

from 2003 (29 Oct): Astin recommended choosing students by lottery in 1969

Posted by kleroterion on Monday, 21 December 2009

>Dear Professor Astin
>I have just discovered a letter of yours to 'Science' dated 1969 (21st November). 
In passing, you advocated selecting students for college courses by lottery
>Brilliant! I suppose you are aware that that is what they have been doing in the Netherlands
 for more than 25 years. I'm following up the idea at Swansea University, Economics Department -
 That in all fairness, there ought to be much more use made of random selection. (some more details on
the website below)
>(Just to jog your memory I've attached your original 1969 letter and 1970 reply scanned from Science)
>Do you still think, for the reasons you gave then, that some form of lottery for college entrants 
would be a good idea? Have you followed it up?

>Many thanks (by the way, I'm old enough to remember the 1960's and the idealism 
that existed in those days!)

>Conall Boyle
>Margam Park Village, West Glamorgan
>Website: http://www.conallboyle.com

From Alexander W. Astin
Allan M. Cartter Professor of Higher Education, Emeritus, &
Founding Director, Higher Education Research Institute
Graduate School of Education & Information Studies
University of California, Los Angeles

Dear Dr. Boyle:

Thanks for your note. I had long forgotten about the exchange in Science
The most through exposition of the rationale for the lottery concept (or
other radical changes in admission policies) is contained in my book,
Achieving Educational Excellence (Jossey-Bass, 1985). I haven't pursued the
lottery idea much in recent years, but to me it still represents the berst
way to achieve real "equity" under conditions where there is an imbalance
between supply and demand for slots in any institution.

Best wishes,


Science letters 20 Feb 1970 1075-6                                           BXRS4

from: Science 21 November 1969 from alexander W. astin    (note  italic section  by CFB)

original letter:---------------------------------------------------------------------------------

Total View of Campus Unrest

I was pleased to see Kenneth Kenis-ton's letter (19 Sept.) on the 11 July Science articles about our study of campus unrest. One effect of the con­troversy is that some of the principal purposes of the project have not been clearly stated anywhere. The campus un­rest project is an extension of our on­going study of student development which we initiated 4 years ago, primarily because it seemed at the time that college administrators had for too long been avoiding the question of how stu­dents were really being affected by their decisions, and that students had for too long been choosing their colleges on the basis of an untested body of folklore. Our principal goals were thus to con­front the professional educators with some hard facts about the effects of their practices on students, and to pro­vide students with a better basis both for choosing an appropriate college and for bringing about meaningful changes in existing educational practices. Our be­lief was—and still is—that ignorance concerning the effects of colleges on stu­dents represents one of the biggest ob­stacles to the improvement of higher education. Some of the research from this larger program is already beginning to pay off; for example, we recently found convinc­ing evidence to suggest that most col­leges—including those that are highly selective—could greatly increase their enrollments of black or other minority group students without materially af­fecting their dropout rates. These and other findings suggest that the entire practice of college admissions needs to be reexamined, and that colleges, in the interests of putting the concept of "equality of educational opportunity" into practice, might want to consider abandoning altogether the use of grades and tests in admissions, and instituting instead a lottery system for choosing among their applicants. While this idea may be distasteful to many adminis­trators and faculty and even to many students, a few institutions—including some highly selective ones—are already considering such a change in their ad­missions procedures, primarily as a con­sequence of our research findings. Unfortunately most of the criticism to date of the campus unrest study is based largely on ignorance and misin­formation. With the exception of Robert Powell, former president of the National Student Association, critics have apparently not taken the trouble to find out what the research goals of this or the larger project actually are, how the studies are designed and being car­ried out, how we plan to disseminate the findings, how the security of the data is protected, or even who the research­ers are. Since the study of campus un­rest is part of the larger longitudinal study, one of our major research ob­jectives is to find out how the typical student is being affected by campus un­rest—a topic which has been largely ignored by social scientists in their pre­occupation with the characteristics of the radical left, the dynamics of con­frontation, and the tactics of adminis­trative response. It is both ironic and exasperating that critics who claim to be "protecting" students are—perhaps unwittingly—attacking a research proj­ect that offers some real hope of ultimately giving the student a better shake in his college experience. Some of the critics have implied that we are engaged in a kind of conspiracy against student radicals, and that the study represents a form of "counter-insurgency" research which involves the compilation of extensive "dossiers" on protest leaders. This is rubbish. While student radicals represent one of the groups being studied, the research is focused much more on other students— protestors and nonprotestors alike—and is concerned with their needs and de­sires for higher education and with how they are affected by campus unrest when it occurs. We have not prejudged any of the students, faculty, or administra­tors who are taking part, but are inter­ested rather in learning more about how they interact and how they are affected by campus unrest. In this regard, the ACE research staff is not a "commis­sion" that has been assigned the task of producing a report which attributes blame to various parties to the "prob­lem." As researchers we have not taken the view that campus unrest is a "prob­lem" in need of a "solution." Nor have we assumed that it represents a panacea for the ills of higher education. We claim no special expertise in making such value judgments. What we do claim to be expert in is the objective empirical study of higher education, and we assume that our findings will provide a better basis in fact for others to make such judgments. As for compiling "dossiers," we have gone to extraordinary lengths to protect the anonymity of all students, faculty, administrators, and institutions that pro­vide us with data. All identifying in­formation from our personal interviews has been destroyed. In addition, our longitudinal survey data on individuals are not accessible to any governmental agency, other institution, or individual. Recently we have instituted a data pro­tection system which makes it virtually impossible for anyone (including myself or any other member of the ACE re­search staff) to obtain access to data on any individual, even by means of a court order or congressional subpoena. Although this new system makes it very unlikely that we should ever be forced to do so, we are prepared to go to jail, if necessary, to make good on our prom­ise of anonymity. alexander W. astin American Council of Education, 1785 Massachusetts Avenue, NW, Washington, D.C. 20036       reply-------------------------------------------------------------------------------------- Should College Applicants Be Selected by Lottery? from allan P. gray Department of Pharmacology, University of Vermont College of Medicine, Burlington 05401 Alexander   W.   Astin    (Letters,   21 Nov.)   indicates  that  his  critics  have not taken the trouble to inform themselves of his research goals. Not being an educationalist, I readily confess to total ignorance not only of the goals but of the entire research project. Consequently, I would be fascinated to learn what "convincing evidence" his research has uncovered leading him to the incredible suggestion that it might be possible   "to   consider   abandoning . together the use of grades and tests in  admissions, and instituting instead a  lottery system (italics supplied) for choosing among their applicants." There  already  is  tremendous  pressure being brought to bear on the universities to abandon grading in favor pass-fail (with emphasis on pass, of course). In fact, this College of Medicine has just gone over to a pass-fail system,  which  may work  as long  as applicants are carefully screened. The abandonment of grading in the college admittance   process,   however,   would inevitably hasten the demise of grading throughout the universities, since if there were no need for considering grades in admittance  there  would   certainly  be  need for grading subsequent course work. Thus, Astin's implied prediction that applicants chosen by lottery would do as well (from the point of view of  dropout rate)  as applicants chosen by currently accepted practices would no doubt be fulfilled. In fact, in the absence of grading, dropout rates could surely be reduced. Students could then successfully  evade  being judged until they left the academic shelter and had to demonstrate their competence in the world. A  more   fundamental   objection   to Astin's proposal is that it would be a perfect example of reverse discrimination. Selecting applicants by lottery would clearly discriminate against those serious students who are talented, who are interested in being educated and can profit from education, and who also somehow manage to get good grades.        Our goal must be to admit to our colleges and universities all those in­terested in, and capable of benefiting from, higher education. This must, of course, be without discrimination against any minority group—black, white, or intelligent. "Equality of edu­cational opportunity" means an equal right to be considered for admittance to an educational institution; it does not mean equal right to be admitted. Granted that grades and tests are but imperfect measures of admittability, they have proved at least serviceable over the years. Any system offered as a replacement should be carefully worked out and thoroughly tested on a comparative basis before being adopted. The proposed lottery would, in my view, lead to utter chaos. allan P. gray Department of Pharmacology, University of Vermont College of Medicine, Burlington 05401   reply- see next page Astin’s reply: My comment about a lottery system has provoked reactions not only from Gray but also from several friends and colleagues, primarily because it pre­sents a seemingly outlandish conclu­sion without providing any of the premises. Most of the empirical data that led me to this conclusion are re­ported in two forthcoming books: The Campus and the Racial Crisis (Ameri­can Council on Education) and Predicting Success in College (Free Press). While there is not enough space here to adequately summarize the findings and related arguments as set forth in these books, it should be pointed out that the "educational" justifications for selective admissions simply are not supported by the data: (i) Highly selective institutions do not appear to enhance the student's intel­lectual development; (ii) the few aver­age or below-average students who manage to get into highly selective colleges do not have high dropout rates; and (iii) the intellectual development of the highly able student does not appear to be retarded if he attends an unselective institution. In other words, the "track" system that we have devel­oped in American higher education simply does not seem to have its in­tended effects. (my italics) A more basic difficulty with current admissions practices is that they seem to be modeled along the lines more of a business than of an educational in­stitution: Instead of searching for students who can be maximally bene­fited educationally, colleges simply compete for talent. If admissions were designed instead along the lines of, say, a hospital, then the whole procedure might be inverted—the poorest-per­forming students would be given the greatest opportunity. The basic problem here is that we know a lot about pre­dicting performance, but very little about how to influence performance.         It is important to point out that as the concept of "universal higher edu­cation" gains currency, the admissions process will become less a question of exclusion and more a matter of differ­ential sorting of students among insti­tutions. Consequently, the use of a lottery to adjudicate supply-demand imbalances at specific institutions will probably be much easier within large state and city systems of institutions that have "open" 'admissions than at individual private colleges. Neverthe­less, there are already a few private institutions that are seriously consider­ing the use of a random selection pro­cedure, at least for a portion of their vacancies.         Gray has reiterated one of the falla­cies that tends to perpetuate selective admissions: that academic standards are somehow determined by admissions standards. Not only is this not true, but if it were, colleges would have no educational function; they would simply be talent scouts and certification agencies for business, industry, and the graduate and professional schools. In my opinion, the sympathies to adopt pass-fail or to abandon grading alto­gether are generated by the selective admissions process itself: Some col­leges employ such high standards of admissions that even the poorest per­formers do not "deserve" low grades. (A much better solution to this prob­lem, it seems to me, would be for col­leges to abandon the use of local, rela­tive grading schemes and to employ comparable, absolute standards of per­formance.) In short, rather than obviating the need for evaluation, the use of an open or lottery system in admissions should create a need for more elaborate and improved methods of measuring the student's perform­ance. The surest way for colleges to avoid any responsibility for educating the student is to employ selective admis­sions: If only the brightest students are admitted at one end, then the high quality of the final product at the other end is virtually guaranteed. What hap­pens in between—the quality of the educational experience itself—need not be of concern since the secondary schools are suitably impressed with the college's high admissions standards, and the employers and graduate schools are suitably impressed with the "high quality of the graduate." My impression is that professors support selective admissions because they feel that bright kids are more fun (and easier?) to teach. Alumni, legis­lators, faculty, administrators, and probably many students support it be­cause having only bright students en­hances the prestige of the institution. Furthermore, the secondary schools support the track system that results from selective admissions because they see it as a reward or incentive system for motivating their students: "study hard so you can get into a 'good' col­lege." While each of these arguments may have merit, none really has much to do with the educational mission of the college. If the principal function of the college is to educate, then the ad­missions process ought to be designed to sort the students so as to maximize their educational development. Cur­rently, we are woefully ignorant as to how best to do this sorting. If nothing else, even a partial lottery would permit us as scientists to explore the possible advantages of many student-environ­ment combinations other than those that result from current selective ad­missions policies. alexander W. astin American Council on Education, 1 Dupont Circle, Washington, D.C. 20036        

Alexander W. "Sandy" Astin

Allan Murray Cartter Professor Higher Education and Work Director, Higher Education Research Institute 3005C Moore Hall -- (310) 825-8331 -- aastin@gseis.ucla.edu Ph.D., Psychology, University of Maryland, 1958 Areas of Interest Higher education policy in the United States; educational reform; values in education; impact of different types of institutions on student development; assessment and evaluation research in higher education. Publications
  • What matters in college? Four critical years revisited. Jossey-Bass, 1993.
  • An Empirical Typology of College Students. Journal of College Student Development, January 1993.
  • The Future of Higher Education: Competition or Cooperation? Learning, Spring 1993.
  • Diversity and Multiculturalism: How Are StudentsAffected? Change, April 1993.
  • College Retention Rates Are Often Misleading. The Chronicle of Higher Education, September 1993.
  • What Really Matters in General Education: Provocative Findings From a National Study of Student Outcomes. Perspectives, Fall 1992.
  • Assessment for excellence: The philosophy and practice of assessment and evaluation in higher education. MacMillan, 1991.
  • Competition or cooperation? Change, September/October 1987.
  • Achieving educational excellence: A critical assessment of priorities and practices in higher education. Jossey-Bass, 1985.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: