russ%prism@gatech.edu (Russell Shackelford) (06/30/89)
From: russ%prism@gatech.edu (Russell Shackelford) [ crossposted to : sci.edu,comp.edu,sci.psychology,sci.astro,sci.bio, sci.lang,sci.math,sci.med,sci.misc,sci.philosophy.tech,sci.physics, sci.research,sci.space,sci.logic,sci.military,sci.chem. Followups to sci.edu ] The following is in response to several inquiries I have received over the net, suggesting that I provide some information about "OPTIMUS, The Teaching Information System". --------------------------------------------------------------------------- OPTIMUS: History and Dogma --------------------------------------------------------------------------- OPTIMUS grew out of a 3-year study (1983-86) in the Computer Supported Instruction (CSI) Project at Georgia Tech. The purpose of the study was to see how we might best use CHEAP technology to make an ACTUAL contribution to TRADITIONAL education RIGHT NOW. We were NOT blue-skying about what we could do if we had 10 years and $100 million to play with. We were trying to develop a model that could be applied in typical settings, college and pre-college, not just in "big-buck" universities. To make a long story shorter, we a arrived at a coherent perspective. A summary of conclusions relevant to OPTIMUS (I am responsible for the following statements; Georgia Tech is not): 1. Programmed Instruction, i.e., CBI, is neither cost-effective nor realistically do-able, at least not in this century. It has conceptual and practical holes big enough to drive a bus through, both technically and pedagogically. A key problem is the inability of CBI technology to support the processing of long/complex student responses, the kind that are crucial to education (We don't need MORE multiple choice tests!). 2. CBI fails to address the core problem directly. The crux of the educational problem is that we no longer have adequate teacher/student interaction (if we ever did!?!). We observe that students do not benefit from adequate individualized feedback and intervention attention from faculty. CBI attempts to COMPENSATE for this lack by substituting "computer attention" for "teacher attention". This does not CORRECT the problem, it only attempts to COMPENSATE for it. An alternative approach is to deploy computers to increase "instructional productivity" by allowing teachers to increase the quanitity and quality of individualized attention that they CAN give within the confines of existing resources. 3. Like any other system, Education requires adequate feedback loops in order to self-regulate. Education has been attempting to function in the Information Age with self-regulatory feedback loops that date from the Industrial Age. Since these loops are obsolete and inadequate, widespread failure of self-regulation has occurred (and made headlines: "Incompetent Teachers!!", "Illiterate Graduates!!!", etc.). 4. An appropriate role for computers is to SUPPORT education by providing the information processing functionality necessary to ALLOW educators to self-regulate. Given appropriate information, educators will NATURALLY self-regulate, which will inevitably work better than efforts to CONTROL, i.e., bureaucratize, the system. 5. For this to occur, it is necessary that computers be incorporated into the PROCESS of education, i.e., it must achieve the status of an EDUCATOR"S BASIC TOOL. As long as computers are "in a room down the hall", they will not have significant impact. As long as they are deployed primarily for use by the CLIENT (student) and are not used by the PROFESSIONAL (teacher), they will not have significant impact. Observe: (a) Education is the only profession that deploys its limited computer resources primarily for Client use, not Professional use; (b) Education is the only profession where computers haven't done lead to much of consequence. 6. For Professionals to USE computers, computers must be a productivity aid, i.e. they must DO something useful FOR the Professional on a daily basis. Thus, we require computer systems that: a. Do something FOR the Teacher in terms of daily practice. b. Provide USEFUL information as required for self-regulation. c. Are cheap enough to be ubiquitous. Specifications for such a system were developed in 1985-86. GT is not in the microcomputer software development business, and thus no system was developed there. A few people who worked with that project decided it was worth doing, and have implemented the idea over the last 2 years (1987-88) as "OPTIMUS, The Teaching Information System". I served as Coordinator of Software Development in the CSI Project, and I have had extensive input into the design of OPTIMUS. I have not done any of the actual implementation myself, having been busy working on an unrelated research grant and finishing my Ph.D. However, the idea is near and dear to me, and I am very pleased with the OPTIMUS program. I think it is a very important thing for Education. It allows us to have a "window" into important aspects of the teaching-and-learning process and it has the potential to be very important in terms of BOTH research AND practice. -- Russell Shackelford School of Information and Computer Science Georgia Institute of Technology, Atlanta, GA, 30332 russ@prism.gatech.edu (404) 834-4759
russ%prism@gatech.edu (Russell Shackelford) (06/30/89)
From: russ%prism@gatech.edu (Russell Shackelford)
The following is in response to several inquiries I have received over the
net, suggesting that I provide some information about "OPTIMUS, The
Teaching Information System".
---------------------------------------------------------------------------
OPTIMUS: Goals
---------------------------------------------------------------------------
OPTIMUS integrates Structured Hypertext, Database, Spreadsheet, Reporting,
and Graphing capabilities into a program that supports EXISTING teaching
practices and methods. It was designed explicitly for Professional
Educators. The broad goal is to increase "Instructional Productivity",
i.e., to help the Educator to accomplish more in less time. It's several
specific goals include:
1) Save the Educator time in Grading student work.
2) Increase the Educator's ability to give QUALITY feedback to students.
3) Dramatically increase Data Capture with respect to student performance.
4) Process student performance data to give the Educator instant access to
information necessary for Intervention with students "before failure".
5) Process student performance data to give the Educator instant access to
information about his own Teaching Effectiveness; allow the Educator to
monitor changes in Teaching Effectiveness after modifications to teaching
organization, approach, content, etc.
6) Process student performance data to provide information that allows
effective Strength-Sharing among colleagues.
7) Process student performance data to provide information that allows
effective Curriculum Integration.
---------------------------------------------------------------------------
OPTIMUS: The Basic Idea
---------------------------------------------------------------------------
OPTIMUS is a "processing engine" which integrates:
(a) a teaching plans and materials integration system,
(b) a student work evaluation environment,
(c) a performance feedback development and delivery system,
(d) an intervention management system,
(e) a teaching-effectiveness evaluation system,
(f) a system-effectiveness analysis system.
It is adaptable to any subject matter or grade level. It provides no
subject matter content itself, nor does it assume computing expertise on
the part of the user.
OPTIMUS is designed for use by educators, requiring the availability of one
computer per teacher, not one per student; it thus allows the benefits of
computer support while minimizing hardware requirements. It is also
designed to support the natural, incremental, and inevitable increase in
access to computers by both schools and students. As the student
population achieves the capability to submit work on-line, OPTIMUS offers
progressively greater convenience and capability.
OPTIMUS does not provide any course-specific data itself. Rather, it is a
"shell" which the Educator uses to develop a "Script" for his/her course.
Thus, OPTIMUS is both grade-level and subject-matter independent.
Script development can occur iteratively in the natural flow of teaching a
course, or it can be performed en masse prior to the teaching of a course.
The contents can be created with the text editor within OPTIMUS, or can be
imported into OPTIMUS from any ascii textfile. Linking of this information
must be done within OPTIMUS.
While any teacher can use OPTIMUS to develop a Script for a course, it is
anticipated that Scripts will be developed by "master" teachers, motivated
professors, textbook authors, or other on-the-ball-people-who-give-a-damn.
Once a Script has been developed, it may be distributed to others for their
use. Thus, an experienced teacher can provide a well-developed Script for
textbook support, departmental support, and/or new teacher support.
Regardless of source, a Script can be modified at any time. Thus, an
Educator might receive a Script from a professor or publisher, then modify
it to suit his own teaching style, preferences, etc.
Assuming the existence of a suitable Script for a given course, OPTIMUS is
used as follows:
1. COURSE ORGANIZATION:
Course organization takes the form of an outline. The Course Outline is
the central data structure to which all other Script data is linked.
OPTIMUS includes an Outline Processor (analogous to Ready, Thinktank,
etc.) to facilitate Outline development. Thus, the user can "expand"
the Outline to see it in complete detail or "contract" the Outline to
hide levels of detail and see the general organization. The Course
Outline might start out as just a simple course syllabus and gradually
be defined more specifically, topic by topic.
A developed Script would include a detailed Outline for the course.
Like all other Script information, such an outline can be modified at
any time. An Educator receiving a Script developed by someone else
would likely begin by modifying/rearranging/extending the Outline to
adapt it to his own course organization.
2. ASSIGNMENT GENERATION:
Assignments can be created within OPTIMUS or imported from ascii files
into OPTIMUS. OPTIMUS maintains a database of Problem and Assignment
Banks, so that any assignment or problem need be input only once. Each
problem is linked to one or more items in the Outline. Once it exists
within OPTIMUS, a given problem can be "picked off" to be incorporated
into a new assignment. With assignments "composed" within Optimus, then
printed out by or distributed electronically from Optimus, the clerical
tedium of preparing assignments is completely eliminated.
A developed Script would include banks of Problems and Assignments for
the course. An Educator would create an assignment by some combination
of (a) browsing the banks to select existing problems, and (b) creating
new problems "on the fly".
Performance data is maintained for each problem. Thus, that the
Educator knows both (1) when a given problem had been assigned to
classes, and (2) the overall performance of students on the given
problem (in total and class-by-class).
3. EVALUATION OF STUDENT WORK & DETAILED DATA CAPTURE:
Student work provides a wealth of information about student competence
and understanding (or lack thereof). The challenge is to capture and
utilize this information. Typically, we record one grade for each
student submission, jot a comment or two in the margin of the student's
paper, return the paper to the student ... and effectively THROW ALL
THE INFORMATION AWAY.
Research at Georgia Tech led to the conclusion that the Evaluation
process is both (a) crucial to effective instruction, and (b) currently
the most time-consuming, unproductive, and wasteful of all instructional
activities. The Grading Environment within OPTIMUS is the key to both
increasing effective data capture and decreasing grading time. The
larger goal is to allow vastly improved feedback to both students and
educators.
The Teacher (or other Grader) uses the Grading Environment in
conjunction with the normal process of grading student work. It
includes facilities to support the evaluation of essays, compositions,
or other student work which requires subjective evaluation, as well as
short answer, multiple choice, or other student work which requires only
"answer scanning" for grading.
OPTIMUS acts automatically to construct and maintain a database of all
information available during the grading process. This includes both
(a) "objective" data (points correct for problems linked to various
topics), and (b) "subjective" data (the various phenomena in student
work that is deemed by the teacher to be worthy of comment). Optimus
thus provides an accurate objective record of all the phenomena noticed,
objectively and subjectively, in student work. This database is indexed
by course topic, by assignment and problem, and by individual student.
4. PERFORMANCE FEEDBACK TO STUDENTS:
The student receives more than just brief comments on his paper. The
Grading Envieonment allows the grader to insert detailed feedback
comments into student work via a mouse and a fast "point-and shoot"
evaluation procedure. The grader uses Optimus and a mouse to simply
"click" on on-screen "comment codes". Each comment code is "linked" to
a textual message defined in the Script. When a comment code is
"clicked" with the mouse, the detailed textual message is automatically
inserted into the student's work at the appropriate location.
Educators can create new Comments (codes and textual messages) "on the
fly" whenever something new is noticed in student work which is not
addressed by an existing Comment. This allows Educators can say as much
or as little as they wish in response to student work.
Research in both the English Department and the School of Information
and Computer Science at Georgia Tech resulted in findings which
surprised teachers who anticipated that a database of feedback comments
would be too "forced" or "impersonal" to be viable. When these very
teachers used OPTIMUS, they found that approximately 90-100 comments
covered everything that they encountered while grading for a given
course.
The effect is that graders no longer respond with just a word or two in
the margin. Instead, thought can be given to what the teacher "would
like to say" to a student who evidences a given problem. With OPTIMUS,
appropriate messages can be created once, in whatever length and detail
is appropriate. Once created, these detailed feedback comments can be
"written' on student work with just the "click" of a mouse button.
This allows experienced faculty to provide guidance to their graders.
TA's can be provided with the specifics "what to look for" in student
work, and Faculty can define the actual feedback messages to students
which their graders apply.
The result is that
(a) students receive better feedback;
(b) teachers have greater control of the grading performed by TA graders;
(c) grading is done very rapidly;
(d) a detailed database of student performance strengths and weaknesses
is automatically maintained.
NOTE: The Grading Environment and its associated database present a
panorama of research opportunities. Which feedback comments have best
effect? Does a change in classroom teaching result in fewer occurrences
of a given problem/comment? And so on. This technique has already been
utilized at Georgia Tech to discover a universal error in the teaching
of Pascal. The opportunities for blending teaching and research seem
unlimited. Think about this!
5. INTERVENTION:
Use of the Grading Environment results in a detailed database of student
performance. The availability of such data allow new efficiencies with
respect to intervention.
* Students who have a particular deficiency with respect to a topic can be
directed to remedial material early enough for their to be a good chance
of success. This can be done automatically: feedback Comments can
easily include Study Guidance components.
* When a student comes to the office for assistance, there is no need to
fumble, trying to assess the student's problem areas: OPTIMUS can draw
a graph of student competence, topic by topic, either in isolation or in
reference to the class as a whole; both student and teacher can
literally "see a picture" of the student's strengths and weaknesses.
* Clusters of students with similar weaknesses can be easily identified
and grouped, thus allowing better use to be made of limited Teacher and
TA time for intervention.
5. OPTIMIZATION TARGETING:
The same database that allows weaknesses of individual students to be
recognized and addressed allows similar advantages for the Educator with
respect ot teaching effectiveness. Performance data for the class as a
whole is always instantly available. The Educator can set his own
standards of performance, and OPTIMUS' Critique function will search the
database and identify all course topics where he (a) exceeds and (b)
falls below the standards he specified.
Thus, the Educator has an empirical basis for determining (a) where he
meets his own standards of teaching effectiveness and (b) where he is
challenged. If a problem area is identified, the Educator may make
modifications to his teaching approach or content, then use OPTIMUS to
monitor the results. In this way, OPTIMUS helps the Educator turn
teaching into a self-correcting activity, based on empirical data
gathered from his own teaching activities.
This is something that we believe is VERY important!
6. STRENGTH SHARING:
It is curious that Faculty share the benefits of their teaching
experience so rarely. A large part of the reason is that there is no
"medium of exchange" for teaching materials and teaching experience.
OPTIMUS is designed to change that.
The development of Scripts within OPTIMUS provides new opportunities for
Course Development. A Teacher who routinely teaches a course will, by
using OPTIMUS, naturally develop a detailed Script for the course. The
Script can then be given to other faculty who might only rarely teach
the course, thus offering them the benefit of their colleague's work.
New inexperienced faculty can benefit from the experience of senior
people.
A Teacher who identifies a weakness in their class' performance in a
given topic can consult with colleagues and borrow Script segments from
others who achieve better results.
7. CURRICULUM INTEGRATION:
With OPTIMUS a School or Department can systematically evaluate the
effectiveness of their curriculum empirically. Do students in a
4000-level course evidence weaknesses in topics that were addressed in a
2000-level course? If so, what changes to the 2000-level course might
reduce or eliminate the propagation of problems to later courses? Or,
might there be problem areas in a 4000-level course that were NEVER
addressed in an earlier course? If so, what curriculum changes might
solve this problem? OPTIMUS provides a medium by which such questions
can be formulated and by which the effectiveness of response can
be empirically evaluated. The effect is to move Educational Research to
the local level, allowing any Educator to systematically investigate
such issues. We beleive that this is VERY important!
---------------------------------------------------------------------------
OPTIMUS: Summary
---------------------------------------------------------------------------
OPTIMUS is designed to be a basic tool for Educators who wish to
systematically examine and refine their own teaching.
* It provides a window into the teaching-and-learning process by tracking
the wealth of information, both subjective and objective, that is
available to Educators as they evaluate the work of their students.
* It provides a medium by which the consequences of changes in teaching
approach or content can be empirically measured.
* It provides a medium by which Educators can exchange the benefits of
their teaching experience.
* It provides a standard tool which can be applied to any course,
regardless of subject matter or grade level.
* It provides a standard tool by which any Curriculum can be evaluated,
it's "holes" identified, and solutions tested for real-world effect.
---------------------------------------------------------------------------
OPTIMUS: System Requirements and Current Status
---------------------------------------------------------------------------
OPTIMUS runs on MS-DOS 3.xx. A hard disk and 640K are necessary. An
AT-clone and color monitor are recommended (suggested hardware
configuration can be had mail order from innumerable sources for about
$1500).
MS-DOS version is in Beta testing at Georgia Tech and should be released in
Summer, 1989. Both individual and site licenses will be available. Prices
are not yet set.
Unofficial best guess on pricing (this may prove to be off the mark):
$200-300 in quantity for site licenses; $500-700 single copy price.
Versions for the Mac and for Unix are in development.
It is anticipated that OPTIMUS will never be finished. We are routinely
developing new capabilities in response to requests from Educators. Thus,
we are most interested in your ideas for a USEFUL processing engine to
support you in the education work that you do.
If you are interested in obtaining a copy, write to:
Mindsight Corporation
2314 Pleasant Ridge Rd.
Bremen, GA, 30110
There is much in OPTIMUS that was not described above. I can answer most
functionality questions you might have. However, you should write to the
company, above, for purchasing information and other business matters.
Russell Shackelford (russ@pyr.gatech.edu)
--
Russell Shackelford
School of Information and Computer Science
Georgia Institute of Technology, Atlanta, GA, 30332
russ@prism.gatech.edu (404) 834-4759