Fake Banner
    Optimizing Science Education: What We Will Need For The University Of 2020
    By Carl Wieman | July 28th 2008 09:23 AM | 23 comments | Print | E-mail | Track Comments
    In Optimizing The University - Why We Need a New Educational Model for a New Century I discussed some of the issues facing post-secondary science education.

    In particular, that piece focused on how changing student demographics, modern faculty responsibilities and the new knowledge gained from advanced assessment techniques showed us that we need to fix fundamental aspects of science education if students are to receive the high quality education that is becoming increasingly important to individual and societal success.

    Now I am going to discuss what this new optimized university - our university of 2020 - will look like. While academics might wish for an ideal university with unlimited resources and no real constraints, it is more useful to ask what is possible in the real world.

    So I will discuss what we need for an attainable university that provides the best undergraduate education possible within certain basic constraints on resources and organizational structures - a truly “optimized” university. The constraints are based on the pragmatic assumptions that resources in support of higher education will not dramatically increase and most of the long standing structures such as disciplines and departments will be largely intact, as will the current broader faculty responsibilities.

    I do not believe that it is possible to avoid these particular constraints, for two reasons:

    First, there is no indication that dramatically higher levels of resources are forthcoming for public education.

    Second, where attempts have been made to create universities with dramatically different organizational structures, such as the new University of California campuses without discipline-based departments, they have effectively reverted to largely traditional structures. It is difficult to see how anything else is possible given the complexity and extensive scale of modern natural and applied science and the limitations of the human mind.

    There must necessarily be some organizational unit at the level of “extent of material that intelligent human (i.e. faculty member) can reasonably master” that serves as the basic unit of educational organization.

    While there are many new interdisciplinary areas of activity, from a longer-term perspective, these are primarily a continuation of the historical evolution of disciplines to remain appropriately aligned with the directions in which science and engineering is developing.

    Therefore, while I assume that the labels and orientation of departments will change, there will remain entities on the size and scale (intellectual and number of faculty) of departments.

    These will continue to be the basic organizational structure of the faculty members and the primary educational unit within the university.

    While these external features of the current university and my transformed optimized university will look the same, there will be some dramatic differences. Education in the optimized university will focus on the desired student educational outcomes and these outcomes will be measured and achieved through a structure of pervasive thoughtful use of both research on learning and information technology.

    This focus on learning outcomes is in contrast to current practice of focusing on processes, such as number of students taking certain number of courses covering particular list of topics. If properly implemented, this switch from processes to outcomes will ultimately lead to dramatically improved educational results and improvements in educational efficiency.

    Another subtle but important difference of the optimized university are the roles of the student and the faculty member in the learning process. Currently the implicit roles are that the faculty member simply transfers their expertise, as if it were bits of information, to the receptive students, much like pouring water from a large jug into a set of small receptive cups.

    This model is inconsistent with what we know about how people learn science.

    In the optimized university the role of the faculty will be as “educational designers,” utilizing their knowledge of the discipline and how best to learn that discipline to design optimized educational environments, activities, and assessment. Within those environments, students will have the role of effortful constructors of their understanding.

    A critical part of the educational design will be the ongoing formative assessment, by which the instructor, assisted by technology, will assess each student’s development of mastery and ensure suitable targeted feedback and challenges are provided to him or her to optimize their learning.

     Characteristics of the optimized university educational environment

    The student will first encounter a choice of academic programs, each of which has a clearly delineated set of educational goals. These goals are created collectively by the relevant faculty in consultation with other stakeholders such as industry, educational systems, and government, and will encompass the full set of skills, knowledge, and ways of thinking that are part of a post-secondary education.

    Each academic program will then have a series of courses that are carefully aligned and sequenced to progress toward the program goals. Each course will have its own explicit learning goals that identify what students should be able to do at the completion of the course and these will directly relate to the program goals.

    These learning goals will be also be established by a consensus of the department faculty members, and will be maintained, regularly reviewed, and updated in the normal functioning of academic departments.

    In each class, the student will encounter pedagogical approaches, materials, and technology all based on careful research and testing.

    The student’s learning will be measured and guided on an ongoing basis using a variety of tools and technology. The development and improvement of these measurement tools will also be seen as a basic departmental responsibility and will reflect the values of the faculty. Faculty teaching evaluations will be linked to these measures of student learning.

    The entire educational process will be driven by these clearly established and measured outcomes of student mastery of detailed educational goals. While it will take substantial investment to produce meaningful measures of outcome to make this possible, the knowledge base and technology now exists to make this feasible on a large scale, and the ultimate returns on this investment will be enormous.

    This is the only way to ensure that good pedagogical methods and environments are replicated and improved upon, and poor ones are eliminated.

    When a student starts a class in the optimized university, they will first complete a detailed diagnostic examination that accurately determines their preparation/knowledge-state. This will examine their content and conceptual knowledge of the subject and those subjects that the course builds upon, such as mathematics and related science disciplines.

    This will also diagnose their beliefs and epistemologies about the subject and how it is best learned.

    Before they have ever seen an instructor, the instructor would have a profile of their strengths and weaknesses, and the computer would have already flagged serious deficiencies. If these deficiencies are widespread, the student will be guided to enroll in a more appropriate course. Where the deficiencies are localized and not severe, the computer would provide the student with feedback and suitable exercises that they must complete to remedy these deficiencies.

    This will ensure that the course will begin with all students at roughly the same level of knowledge and competence, and the instructor will have an accurate profile of that level. This will make it possible to design learning environments that are explicitly matched to the population of students; something that currently is very seldom the case.

    This initial extensive diagnostic exam will be the first of regular ongoing evaluations throughout the course of the student’s thinking and learning. These evaluations will be linked to targeted timely feedback to both student and instructor. Such a scale of evaluation and feedback will only be practical with widespread use of information technology.

    This evaluation and feedback will be largely provided by online homework systems that include intelligent grading and tutoring programs. In this way, much of the individualized evaluation and feedback that has been lost, in moving from a personal expert tutor model to a single instructor with many students in a class model, will be reestablished. This will be done in an economically practical manner by using IT to greatly extend the instructor’s capabilities.

    Utilizing modern educational technology

    The college classroom is primarily pre-computer in its level of technology use. However there are many new educational technologies that have been demonstrated to be highly effective and will be used widely in the optimized university. These new technologies have the capability to transform the higher education system, in much the way the high technology industrial setting has been transformed from what it was in the 1960s.

    The type of technology required for these purposes has been demonstrated in certain specific areas and has been shown to be highly effective under limited experimental conditions, and in a few of cases, fairly large-scale experiments. However, it is used in an extremely limited fashion in education.

    The quantity and quality of what exists has barely scratched the surface of what is needed and what could readily be created, if there was support to do so. For it to be created there must be a viable business model (which does not currently exist) driving its development by private industry, or governmental support. There are major problems with the creation of such a viable model, as long as there is no link between educational outcomes and resources, as discussed below.

    These are much the same factors that have given us textbooks that routinely and blatantly conflict with well-proven pedagogical principles. Assuming resources can be found to carry out the development of these valuable educational technologies, their development must be guided by knowledge of the specific disciplines and research on how people learn. A clear understanding of the educational capabilities and limitations of IT and careful testing of the products are also essential.

    The untapped potential educational applications of IT range from the mundane but time (and hence money) saving to the highly sophisticated new methods for learning.

    It would require a far longer paper to do justice to this subject, but some of these applications include technology for new teaching methods (interactive simulations, intelligent tutors, sophisticated diagnostic capabilities, student in-class personal response systems “clickers”), improved class organization and management systems, archival systems for educational materials and data, and new modes of presenting material and enhancing communication by linking students with each other and faculty in novel ways.

    Moving toward research-based instruction

    The faculty of the optimized university will have sophisticated “pedagogical content knowledge”, in addition to the usual content knowledge for every course they teach. This “pedagogical content knowledge”, as characterized by Lee Shulman, means knowing: how the content and skills are best learned, what common student difficulties are encountered in learning it, what approaches are most effective in helping students overcome those difficulties, and how best to motivate students to master the subject.

    This also requires knowing the relevant research on learning, and assessment of learning, as it specifically applies to the subject in question.

    In the optimized university, a general knowledge about how people learn science will be part of every faculty members’ basic competence, and the many subject specific pedagogical and assessment issues will be fully researched, and detailed information on them will be readily available to every faculty member. When a faculty member is starting out to teach a course, their first step would be to study these course-specific pedagogical content materials.

    While the student will likely still see an instructor in charge of each particular course in the optimized university, the relationship between the course and instructor will be rather different. The department will have the basic responsibility for each course and what students are learning in it. The instructor of a course will thus be working as part of a collective enterprise to optimize the course within the goals, guidelines, and assessments established by the department.

    Faculty members will work in teams to first establish clear educational goals from large scale to specific topic level, and then collectively develop and refine approaches, materials, and assessment tools. The products produced by these collective efforts will be routinely reused and improved upon to provide ever more efficient and effective instruction. Members of the faculty team will each share their strengths to achieve a whole that is greater than the sum of its parts, and, in the process, expertise will be shared so that younger, less experienced, faculty will rapidly gain teaching expertise.

    This is in stark contrast with the current system where teaching is an isolated activity in which faculty set their own agendas and goals for the courses they teach, and they struggle in isolation to teach the subject effectively. They seldom know what students have mastered in previous courses or exactly what students will be expected to know in their subsequent courses, and all of those change with each new instructor in every course.

    Although the collaborative approach described above is highly unusual in teaching, it is not unprecedented. Also, all of these activities and modes of operating are the norm in the modern scientific research lab. Hence the problem is not one of convincing faculty to function in a radically new manner, but rather the lesser challenge of getting them to see how approaches that they know and recognize as very effective in one setting (the research lab) can be equally effective in another (teaching).

    A common claim is that such a collective approach to teaching would fail because what are effective or ineffective teaching and learning styles are totally or largely dependent on the individual personalities of the teacher and learner. Such a claim is quite inconsistent with a large amount of research data.

    All normal human brains function in the same basic way, and research has clearly established that there are very general features of effective teaching and learning. While there of course are individual distinctions, particularly in the learners, these distinctions are small compared to the range of teaching approaches for which there are advocates.

    For example, there is extensive physics education research literature examining the effectiveness of various teaching practices.

    This consistently shows that practices that increase the average learning for a class also increase the learning for each of the subgroups of low, medium, and high achieving students in the class. The individual student distinctions with respect to effective teaching styles are evident primarily only at the much finer level of the student thinking on specific topics.

    Thus, they are best addressed by the careful evaluation of thinking and providing appropriate feedback as described above, rather than trying a wide variety of teaching approaches in the hope that what fails for some students might be successful for others.

    Next timem I will discuss those aspects of education; getting away from thinking that education quality is determined by the personality of the instructor and student and that some instructors and students just don't 'have' it. In our optimized university of 2020, research-proven practices make everyone a good instructor and nearly every student a good learner.


    He wants to turn universities into correspondence courses:
    just like the old RCA Institute of Electronics Courses--with goals in each unit, and tests of the goals at the end of each unit.
    This is exactly the sort of thinking that destroyed our educational system in the first place!

    It was inspired by the low level technician training of
    the world wars--and based on B.F. Skinner's studies of rats.

    University Education is not TRAINING--it should be all about
    getting the students to be creative and interested in a burning way in being creators--especially in science.

    Once a student is strongly motivated to be original, she will learn whatever she needs to do that. No structure is necessary.

    I suggest that he stop thinking so abstractly about education-all that he suggests has been tried, and it failed
    miserably--consider the history of programmed learning.

    Instead, he might think about what sort of experiences inspired him to become a Nobel Laureate.

    p.s. The root stem of "education" is the same as that of the word "educe" : to bring out.

    Goal based Training is for dogs, not future scientists.

    I will also that these kinds of "methods" studies are exactly what departments of education do. And, it was departments of education--in dishing out these "effective"
    methods that destroyed American Education--once the finest in the world.

    Such methods were used in reforming math and reading education for the last few decades--result: a nation of innumerate illiterates.

    In the days before "grade based world lists"
    and chapter goals, most Americans could read quite well--the Declaration of Independence was posted on tavern walls. The old McGuffey readers were used in one room school houses for part time farm boy students--and by the sixth grade ( NOT SIXTH FORM) the reader included selections from "Paradise Lost".

    I withdraw my previous comments--having read the author's bio at wikepedia--he is clearly far smarter than I, and clearly far more qualified to make statements about science education
    than I am.

    Fools rush in where Angels fear to tread--and what a fool I am.

    A gnat shouldn't offer opinions to an elephant!

    I look forward to his next post to learn what I can from
    a clearly far superior mind on this important issue!

    It's sort of funny that his Nobel prize, his 20 years of formal science education and his later experience as a professor didn't count for anything with you but his Wikipedia entry did.

    Lots of people are Professors, Lots of People have decades of science education, and Nobel Prize winners have been known to say very silly things about education. As a strong example,
    William Shockley had a Nobel Prize in Physics and said that
    the black race is genetically intellectually inferior, and
    Watson--with a Nobel prize in physics ( whose prize largely depended on the work of Rosalyn Franklin) has said that females are genetically inferior in math and science ability.

    But, after the Wikipedia article, I can see that C.W. is the real article!! He is not shooting from the hip, but has made serious time commitment to science education improvement--and he has won national level teaching quality awards, and he ALSO has a Nobel Prize in Physics.

    He is applying his Nobel Level intellect in a very serious
    and time-intensive way to improving science education with
    the resources of an Institute behind him

    He is going to change everything!!

    The point is that departments of education tried these things before but they haven't had someone of the intellectual quality of Nobel Laureate C.W. to do it properly!

    That should make all the difference. He is going to change

    I eagerly await his next post.

    I too eagerly await his next post. Meanwhile, Penny, have you read this one, by Vladimir Arnold? http://pauli.uni-muenster.de/~munsteg/arnold.html or http://pascal.iseg.utl.pt/~ncrato/Math/VladimirArnold.pdf Robert H. Olley Physics Department University of Reading England
    Robert H. Olley / Quondam Physics Department / University of Reading / England
    Watson, with a Nobel Prize in Biology.

    Dear Robert,
    Thanks for the link. I strongly disagree with Arnold--who I have heard lecture on his work on singularities--and it was terrible. Arnold has written a nice book on ordinary differential equations, though. But, I much prefer the
    book of Petrovsky, which was my text in sophomore ODE.

    He is a product of an education in a math and mechanics
    ( aka applied math) program--and he, though brilliant, and though he knows a lot of certain parts of pure math--was well
    brainwashed by the bias of his educators.

    Lots of people of his generation bemoaned the books of Bourbaki--but I was LOVED the Bourbaki books--which are wonderfully clear, and very carefully thought out. The generation of mathematicians educated in this style-by Bourbaki or similar texts ( which would include most math texts from 1960-1980) is arguably the most productive generation of mathematicians in history. Look at all the famous problems ( Poincare, Fermat, Hilbert etc.) that this
    generation has solved.

    Of course, educating the MOST talented is never a problem.

    What do you think of " The Feynman Lectures on Physics"?
    I very much enjoyed them, when I studied physics from them,
    but there is a serious circular reasoning error in Vol III.

    Feynman was also interested in education, but not at the
    system level--where he was frustrated by the professors of education.

    Carl W. is going to change everything. I have waited for someone like that for decades.

    Dear Robert,
    For math education
    at the high school and lower level--and for the most talented
    students--the Russians had a very good system--based on pure math. Israel Gelfand brought this program to the USA and he has been quite successful--but it was never adopted on a systemic scale.

    Again, teaching the most talented is probably pretty well understood. It is teaching the people a notch down that is
    where we need Carl's help the most. They are the majority of
    science users--what was once called " The solders of industry". Without them our technology cannot be sustained.

    Submitted by Anonymous (not verified) on 28 July 2008 - 6:41pm
    is me. I forgot to put my name down.

    My perspective is that of an Adolescent Medicine Specialist with a Princeton Univ degree in Engineering. I was educated through the Sputnik era and found quite quickly that I was not qualfied to compete with the Oppenheimers, Wigners, and Wagners of this world and thus switched from honors physics to engineering. I see numerous young people with talents equal to those of the successful physics students of my class and agree that their higher education in science is well constructed at Cal Tech, M.I.T., etc.
    The youth with just below that level of innate ability are rarely interested in the sciences as they perceive themselves, as did I, to be incapable of competing with the hyper-bright in ares such as theoretical physics, math, and advanced materials research. My children and my patients, rather than attempt the rigors (with respect to time and effort) of Medical or Engineering school, are uniformly opting for a liberal arts education with only enough Math and science so as to havee little difficulty with the requirements of an M,B.A.
    I remain clueless as to a solution while starting salaries from Tuck, Wharton, Kellogg, and Harvard Bus Sch remain above $100,000 and a Pediatrician may never make $130,000.

    I was quite disappointed that this plan does not express more respect for the individual interests of students. It still assumes there is One True Way when it comes to building a curriculum's content, even if it does not do that when it comes to its process.

    Why can't we stop seeing universities as places that dictate the ultimate knowledge base needed for the intellectual elite in specific fields? Why don't we allow and enable students to pick their own courses according to their interests as well as their skills? This would still require most advances as described here, but it would leave out the backwards restriction of clearly segregated curriculums.

    A great advantage of making university grade knowledge more fluidly accessible is that it could be applied and learned outside of universities. Businesses can be really slow when it comes to taking advantage of new research. Universities could do well to mitigate the brain drain they are currently submitted to because businesses feel they can't acquire that knowledge in a less hostile way.

    This also would enable a new business model, providing courses, structured knowledge and support for business employees. Since they can be much more critical than typical students when it comes to the results of their courses, this would also help transforming universities to more educational result oriented organisations. It also helps build a positive bond between the economic interests of businesses and the educational interests of universities, instead of just the research interests.

    Penny, I think I'd better write another article in a week or so, so we don't take up Carl Wieman's space. But for now, here's a couple of quotes: 1: from the introduction to "Geometry in a Modern Setting" by Gustave Choquet. "As soon as possible, the child must think of the set R of numbers as a totally ordered, commutative field: he must realize that when he is carrying out calculations, he is only using a small number of the properties of addition and of multiplication, the ones which mathematicians call the axioms for totally ordered commutative fields." 2: from: http://serge.mehl.free.fr/chrono/Dieudonne.html Dieudonné est aussi l'auteur du célèbre slogan “à bas Euclide”, non pour dénigrer le genial mathématicien grec, mais pour fustiger l'enseignement excessif de la géométrie du triangle au collège et au lycée, ce qui mena à l'avènement de l'enseignement des mathématiques dites modernes des années 1970 dont l'efficacité fut cependant très contesté et source de nombreuses polémiques. Les excès du “modernisme” furent d'ailleurs regrettés par Dieudonné lui-même. I had originally thought that those who propagated the slogan "Down with Triangles" in the UK were dervishes inspired by Osama bin Bourbaki, but with more knowledge it seems that it was others who didn't really understand the issues. Still, though you may disagree with the first part of the next Arnold bit, the second (in bold) is pretty certain: Since scholastic mathematics that is cut off from physics is fit neither for teaching nor for application in any other science, the result was the universal hate towards mathematicians - both on the part of the poor schoolchildren (some of whom in the meantime became ministers) and of the users. One of Robert's theorems is that when someone comes along with a new method, those in power leap at it, because it appears to justify their own lack of progress in the subject under the old method. Robert H. Olley Physics Department University of Reading England
    Robert H. Olley / Quondam Physics Department / University of Reading / England
    Dear Robert,
    Well then, just briefly: Most people have ALWAYS hated both math and physics--and putting math in terms of physics only really helps some future physicists--the ones who wouldn't just as easily become future mathematicians. The same can be said for engineers.

    What people hate is being forced to learn something that they view as beyond their capacity and beyond their practical needs. I had a similar feeling for French--cured only after living in France. Learning French took time from learning science, and it made me feel like a moron.

    IMHO, unless Carl has a way to change this--the problem is the compulsion put on students to learn things that they are not--and may never be--ready to learn.

    If everyone were forced to learn to draw portraits--with photoaccuracy to get through high school then everyone would HATE art and artists.

    I would be interested in Carl's take on all this, as he is the expert here.

    I also don't want to take up Carl's space.
    But, Robert, two things:

    First, Bourbaki was actually a committee and Dieudonne was in it. At one point, he chaired it.

    Two, The Bourbaki program was designed for French university math--for future math researchers at the grand ecoles--and it starts at about junior year uni honors level in the USA. It has nothing to say about elementary education.

    I like lots of what I'm reading, but...

    Apparently the University of the future will produce the least intellectual diversity possible. Why not harness the same techniques to make sure the backgrounds of scientists vary as much as possible? Feynmann benefitted greatly from a somewhat eccentric mathematical education, merely by being different. The whole history of science echoes that. A factory that builds only one type of widget ain't what we need.

    What is needed is Honest Science and to correctly teach children the scientific method.

    Right now Science is being suppressed in the name of those with enough money to control proper peer review. It is not science - it is political science.

    Young adults sense that they are being flim-flammed by "science" and are hungry for real scientific inquiry.

    Will Scientists, entrenched in their comfortable roles with perks, salaries and networks, offer honest science?

    I don't think so.

    Learning French took time from learning science, and it made me feel like a moron.

    Moi aussi, j'y suis!

    But then I discovered that language is an art, but linguistics is a science.

    I look forward to reading more from Carl Wieman.
    Los textos y demás recursos educativos deben diseñarse a fin de que el alumno pueda aprender aun sin docente o a pesar del docente.

    Dog clicker training modifies your dogs behavior by making use of a plastic device known as a clicker. The device gives a neutral sound when pressed and is used to make the dog know whether a certain behavior is desirable or not.