Police Training Institute

University of Illinois at Urbana-Champaign

Illinois Law Enforcement Training and Standards Board

University of Illinois at Urbana-Champaign

(Prepared for delivery at the Meeting of the Academy of Criminal Justice Sciences in Albuquerque, New Mexico USA, March 12, 1998.)

Distance learning is one way to increase quality of instruction while lowering cost. But practical experience with distance learning has revealed many drawbacks. The authors share their three decades of experience in overcoming most of these drawbacks; discuss the virtual classroom; show criminal justice educators, trainers and practitioners how to develop a presence on the World Wide Web featuring Live Web Computer Based Training (CBT); and describe their distance learning materials - used successfully since 1978 in training over 20,000 sworn police officers and allied personnel with multiple learning preferences. The authors are available at: http://www.staff.uiuc.edu/~royw/.

TABLE OF CONTENTS

Page

ABSTRACT ……………………………………………….… i

TABLE OF CONTENTS ………………………………….… ii

______________________________________________________

COMPENSATING FOR HISTORICAL DRAWBACKS

OF DESIGN .…….………………………………….. 1

Human-Machine System Design …………………….. 2

Task ………………………………………….….. 2

Procedure …………………………………….….. 2

Human …………………………………………… 3

Machine ………………………………………….. 4

Modularization ……………………………………….. 4

The Fallacy of Subsystem Independence …………….. 6

The Fallacy of System Function ………………….….. 7

The Fallacy of Human Perspicacity ………………….. 8

The Fallacy of Human Memory ……………………… 9

The Fallacy of Human Patience ……………………… 10

The Fallacy of Human Homogeneity ………………… 11

Some Conclusions ……………………………………. 12

COMPUTER-BASED DISTANCE LEARNING…………… 13

Our Pioneering Background …………………………. 13

The Dawn of 21^st Century Internet Training …………. 17

Getting Connected to the Internet ……………………. 19

Network on the World Wide Web (WWW) …………. 19

World Wide Web (WWW) Addresses ………………... 20

Getting Connected to the World Wide Web (WWW) ... 20

The Virtual Classroom ………………………………... 21

Conclusions …………………………………………… 23

______________________________________________________

APPENDIX

A LESSONS …………………………………… 24

B SUGGESTED ARRANGEMENT OF

LESSONS BY MODULE ….…………. 25

LIST OF REFERENCES ……………………………………. 30

Computer-based Distance Learning has come a long way since its first public demonstration at the University of Illinois in 1961. Yet we can still find practitioners making the same mistakes in design and implementation of distance learning hardware and software that were being made almost four decades ago. Further, these mistakes can be traced to the continued use of a set of common misconceptions noted almost two decades ago (Avner & Friedman, 1981).

For one of the authors (A.A.), developing distance learning materials has been a major research effort during the past 30 years. Participatory observations have been made of over 140 courseware production teams and their products in a wide variety of settings, using all major instructional media, for a wide variety of student populations and a wide variety of subject-matter.

The problems met by the growing use of computer terminals in police departments and allied agencies are seldom unique. In fact, the experience of the authors over the past three decades suggests the existence of a set of design problems that turn up whenever interactive computer terminals are used, whatever the setting. These drawbacks emanate from inappropriate dependence on a few simplifying assumptions that make design easier at the cost of lowered effectiveness. We will outline six of these fallacious assumptions, describe the reasons for their beguiling attractiveness, and suggest alternative views that should lead to better design for distance learning and other applications of interactive computer systems.

First let us define some basic terminology; Human-Machine System Design and Modularization.

Human-machine interactions may be considered to consist of four major elements: task, procedure, human, and machine. The good system designer does not assume that any of these elements is a static, unchangeable factor that can be ignored. A thorough analysis may even reveal alternative approaches which eliminate the need for a special design.

Task: The task is the problem that is to be solved. A common error is the failure to understand that past views of the problems may have been limited by what was possible with tools and procedures then available. New tools may make it possible to solve a larger problem. Where this is the case, the task should be defined to include factors not addressed by former approaches. An example common in computerization is that a computer brought in to simplify paperwork also turns out to be able to solve a part of the management process that the paperwork supported. Thus, a computer brought in to automate the production of the FBI=s Uniform Crime Reports would also be able to automate many other standard administrative operations. However, this added capability is likely to be included in the design of the system only if the designer is aware of the total task.

Procedure: Procedures are the methods used to complete a task. A common error is the confusion of procedures with tasks. A procedure (such as filling out police report forms) comes to be seen as a required part of an operation, rather than as simply one of several means of performing the actual task (gathering information to be used for effective and efficient management of police resources). This particular confusion often results in technological misapplications that meet task needs by the simple, but usually inefficient, expedient of mimicking old procedures such as requiring an officer to type information that could more accurately and efficiently be provided automatically by a computer (e.g., the decoding of a Vehicle Identification Number).

Another form of confusion of task and procedure results in the endowment of a procedure with almost magical powers. Thus, computerization may be cited as the reason for success of a new approach. That success is then used as the reason for blindly adopting computers in other situations without taking the trouble to determine what alternatives might be available. One result of such blind adoption of computers is the growing number of cases where an administrator computerizes an operation, shows great savings in time and money, and is promoted or retires. Enter a new administrator who is miraculously able to eliminate the computer without losing the advantages of computerization! If the first administrator had taken the time to examine the actual task and the possible alternatives, it would have been obvious that all that was really needed was a restructuring of the task. The apparent gains derived from computerization in such cases really result from the task restructuring that accompanies the unnecessary addition of a computer. The computer can be dropped from such an implementation with minimal effect on working efficiency and substantial savings in cost. Needless to say, cost savings would have been even greater if the computer had never entered the scene.

Human: Humans appear in many roles in human-machine systems. They may help the machine carry out procedures, or they may be clients served by the system. The major error lies in ignoring the human element. A designer may assume that since he or she is human, any system design will automatically include all human factors. That is not so. Humans vary enormously in training, motivation, and ability. Not only do they vary individually, but they vary with time. A system designed for naive users may not be efficient for experienced users. Users who begin as naive users will seldom remain that way with time. A system that is comfortable for brief human use can be an unbearable burden when used continuously for long hours. Good design demands a clear view of the nature of the humans who will interact with the system and the nature of that interaction.

Machine: The word "machine" is used here because we happen to be discussing computers. A more accurate word would be Atool.@ The machine or tool is a technological aid to application of the procedure. A pencil or an instructional technique is just as valid a form of technology as a computer. It is a mistake to assume that some complex form of technology is needed for every task. As was noted above, careful restructuring of a work situation can result in substantial improvements in productivity without any need for a computer or other expensive technology.

Once the major components of a human/machine process have been identified, good design practice dictates that the resulting system be further broken down into functional modules that cut across these major components. Each functional module performs a single distinct function in the solution of the overall task. A particular module is defined by the portion of the task that it covers, the procedures needed to address that subtask, and the human and/or machine carrying out those procedures. Each module has well-defined inputs and outputs, and most modules interact only with other modules.

Modularization is not done simply in response to an innate drive of system designers to categorize things. Modularization allows concurrent development of the many parts of a complex system by different groups of designers operating in relative independence, thus greatly reducing the length of time between initial planning and putting a system into operation. Modularization also has advantages in the completed product.

Both hardware (the terminals, computers and other devices associated with the system) and software (the computer programs which guide the hardware and interpret interactions between hardware and humans) can be modularized. Modularized hardware is easier to maintain, since modules that serve only a single, well-defined function are easier to isolate should they malfunction. Modules are also amenable to quick, inexpensive repair by substitution. Properly designed modular hardware is also more easily altered or expanded to meet changing needs of a given installation. For example, needs for additional terminals or increased information storage capability can be met simply by adding the needed equipment and the control modules required to interface it with the original system. The same expansion in a non-modularized system might require major redesigning of both hardware and software. Modularized software has similar advantages in identification and repair of problems and in modification of existing installations.

Unfortunately, the advantages of modularization can cause a designer to downgrade the importance of other design considerations. For example, modularization is easiest when the task structure is relatively simple and when there are few interactions between tasks or procedures of different types. In seeking such simplicity, a good system designer tries to eliminate extraneous elements from the task that the system is to perform. Under time pressure, however, such commendable parsimony can lead to oversimplification. Oversimplification results either from failures of commission (misinterpreting a user=s description of the task) or failures of omission (failing to verify that an interpretation of the task actually leads to an acceptable final result). The fault of failures of omission is not always with the system designer alone. A user who is not familiar with the computer=s slavishly literal interpretation of directives may fail to specify the crucial decisions that are often made by a human faced with ambiguous information. A human is able to make common sense interpretations that may result in the job=s completion despite less than ideal information. A computer programmed with an oversimplified procedure for handling the same ambiguous data will merrily grind out enormous amounts of absolute rubbish. Eventually such failures will come to light, of course, but it is far more efficient to identify them at the time the system design is being specified. In the early design stages, no amount of experience in computer system design can replace the knowledgeable guidance of a person who has actually carried out the original task under real-life conditions.

The person following this erroneous design principle assumes that any component of a system can be designed effectively without any knowledge of the rest of the system. It is both convenient and useful to handle design of a system by breaking the major system into component modules. This does not mean, however, that the final system is intended to function as a set of independent modules.

Problems of the Fallacy of Subsystem Independence show themselves most frequently in hardware interactions. At the lowest level, the user might encounter massive delays in accessing or storing information at a terminal that is mismatched to a communications channel or storage device. The fact that two such components can be made to communicate with each other does not necessarily mean that the interaction will be efficient.

At a more complex (and, unfortunately, more often observed) level, a system might perform a variety of functions quite well when the system is supporting only one user. However, a system which has not been designed as an integrated whole may show severely degraded

performance when asked to support different operations simultaneously by several users at different interactive terminals.

The motto of the believer in this fallacy is AIf it works, the design is okay.@ No designer intentionally designs a system that is difficult to use, performs inefficiently, or assumes an inordinate amount of skill on the part of the user. Nevertheless, under pressure to produce a functioning system within time deadlines and funding limitations, designers are too frequently willing to accept almost anything that actually gets the intended task done. They may have started out with far grander intentions and an abiding desire to produce a system that would be both a joy to use and a paragon of efficiency. But in the cold, hard dawn of reality (and corporate solvency), they may have been willing to compromise with something that met the minimum specifications of the contract.

Given the tendency of humans to compromise when under pressure, it is wise to make sure that minimal contractual specifications will actually provide acceptable levels of performance in the finished system. To insure that minimal specifications are adequate, one must go beyond simple statements of input information and output products. One must identify important conditional factors such as speed and ease of operation, and specify the operating condition under which these performance levels are expected.

A system must be expected to perform differently under different usage loads. In recognition of this fact or life, the levels of performance required under a normal load and under the most severe load anticipated should both be specified. We must always remember that even if a system works (i.e., produces the desired results under ideal conditions), it will not necessarily be acceptable to users (i.e., produce the desired results in a real life setting).

This fallacy is committed by most persons involved with the design of interactive systems. We assume that all humans think exactly the way we think (and that they will automatically understand our intent at each step of an interaction) is woefully common. The more involved a designer becomes with the mechanics of getting a system to work, the more he or she grows accustomed to the idiosyncratic way it happens to operate at that moment. After a while, the designer forgets that everyone will not come to the system with a full understanding of the intent behind each human-machine interaction.

The day of reckoning arrives with the first use by people who were not involved with the original design. If the designers are wise, such people will be brought in well before the design of human-machine interactions has been frozen. Careful study of the problems encountered by naive users will greatly aid in reducing potential errors and in increasing the ease of interaction. If the designers are not wise, they will delay exposure of the system to realistic field testing until final delivery of the system. Systems designed under this blind approach are most notable for their literary contributions - a frantic, last-minute effort to compensate for poor human engineering by provision of a stack of instruction manuals. In general, the less often a user interacts with a computer terminal, the less voluminous the printed instruction manuals should be. A well-designed interactive computer system provides complete interactive prompting for the new or infrequent user. Ideally, one should be able to start a new user by simply saying, "Follow the directions on that screen." Assuming reasonably literate users, anything less than this should be taken as a possible sign of limitations either in hardware capabilities or, more likely, in software design.

A fallacy which is particularly prevalent in the design of information and instruction displays for interactive terminal systems is the assumption that humans can remember every detail of information encountered several minutes or even seconds before. This fallacy is a special case of the Fallacy of Human Perspicacity and is perpetrated for the same reason. After many days of working with the structure of a system, one forgets that someone seeing the system for the first time will actually be using instructions as sources of new information and not simply as mile markers on a familiar path.

In the brief exchanges characteristic of use of interactive computer terminals, humans depend mostly on short-term memory. This is the same type of memory that we use for tasks such as remembering a telephone number from the time we look it up in a directory to the time we dial it on the telephone. Short-term memory normally has a very limited capacity (about three to four simple items or groups of items) and is easily overwritten by new information. It is not reasonable to expect people to take the time to memorize directions on a display which they have little intention of using frequently. Nor, given the limitations of short-term memory, is it reasonable to expect a human to remember an item of information from one display and combine it with information from another display. While this is a task that can be done, it is a task that forces a human to do something a computer can do far better.

Well-designed interactions provide directions appropriate to the needs of the user at the moment they are needed. Well-designed interactions also keep track of information acquired by the user (e.g., in a search procedure) and permit easy recovery of that information. For example, after completing a complex search operation, the user should be able to make a minor change in specifications and start a new search without having to retype the full set of specifications.

Time is probably the most frequently overlooked incidental factor in system performance. Designs that ignore the effect of delays in system response on user acceptance or performance are implicitly assuming that such effects do not exist. Rest assured, they do. When a user types a letter and nothing appears immediately on the display, most users will assume that the system has not seen an input. Delays as short as a quarter-second can lead users to make repeated inputs. The repeated input leads in turn to errors (a double letter where a single letter was intended) or to wasted resources (two requests for recovery of data when only one was desired). Longer delays can lead to user frustration.

Most frustrating of all are delays of random duration. One moment the user receives almost instantaneous service and the next moment the user must wait for what seems to be an eternity. Variable delays are generally the result of variations in load. Instant response is available when a single user is present, but delays become noticeable as more users attempt simultaneous use of system resources. In a well-designed system, loading effects should not be perceptible for rapid sequential operations (such as typing the separate letters of a name) and should be minimal for major operations (such as the delay between requesting information and first seeing the results of that request). Response time for rapid sequential operations should always be shorter than the time it would take a touch-typist to repeat a missed key (about 0.1 to 0.2 seconds). Response time (time elapsed to the beginning of responses) for more lengthly instructions should be a small fraction of the time taken to specify the operation, and should never exceed about three seconds. Note that it is only necessary that the response begin within that time.

Finally, the battle may not be won even if a system provides excellent interactive prompting for a naive user. The needs of a new user are rarely the same as the needs of an experienced user. Interactive prompts that are a necessity for a new user may be a frustrating waste of time for an experienced worker who is using the terminal extensively. As a further complication, the type of display device in a terminal may affect people=s acceptance of instructions which are superfluous to their needs. The relatively slow output rate of a printing terminal, for example, can be particularly exasperating if most of the printing consists of instructions the user does not need. The same instructions on a video display might be perfectly acceptable since the rapid rate of display would outweigh the fact that some of the instructions were superfluous.

Design of human-machine interactions must, in short, take into consideration the needs of the novice user (who will require aid at every step), the experienced occasional user (who will need minimal prompting), and the experienced user with a substantial workload (who will be mostly interested in rapid response and minimal hindrance in getting the job done). These three levels of experience are frequently telescoped in time for a given individual who sits down at a terminal as a novice and stands up (several hours later) as an accomplished user. The well-designed system must accommodate all of these levels of expertise by an appropriate mixture of optimal paths, self-selected help sequences, and careful human engineering. The human engineering must, above all, minimize idiosyncratic forms of interactions that simplify the work of a computer programmer at the expense of the convenience of users.

We have concentrated on viewpoints rather than details of technique for two reasons. First, the physical design of systems is rapidly changing as new components become available. For example, years ago it would have been reasonable to list the advantages and disadvantages of making the application characteristics of a particular system hardware-resident rather than software-resident (e.g., a keyboard designed for a specific application versus a general keyboard with software prompts). Changes in types of memory and display devices available are now blurring such distinctions. In general, specific suggestions about system configurations simply do not age well in times of rapid technological change.

Second, our experience has shown that the real source of problems in most design efforts has been failure to identify clearly the goals and procedures that define the system. In the absence of clear goals, computer-design specialists must substitute their own view of what is intended or needed. To the extent that these specialists have specific experience in the practical problems of a given application, their views may lead to successful designs. To the extent that these specialists rely on the fallacies described here, the designs may be dramatically unsuccessful. As in any change in work procedures, there is a need for direct, active input by those who have experienced the reality of the task environment. When the change in work procedures requires investments in time and money of the magnitude demanded by selection or development of computer systems, that need becomes crucial.

The fallacies of system design described here can be averted most easily by continual, careful cooperation between design specialists and those who are thoroughly familiar with the ultimate application of the system. More than any other form of computer system design, systems that provide interactive terminals for occasional use by minimally trained persons demand careful design to insure that expected performance occurs under realistic conditions. Systems that make unrealistic demands on user training, memory, or ability will not be truly successful even though they may function under ideal conditions.

COMPUTER-BASED DISTANCE LEARNING

More detailed guidelines may be given for specific applications of networked computer systems. The application that we will focus on for the remainder of this paper is Distance Learning. Computer based materials produced since 1978 by Dr. Walker and the faculty of the Police Training Institute, University of Illinois at Urbana-Champaign, as certified by the Illinois Law Enforcement Training and Standards Board, Springfield, Illinois USA, form the experience base for this section.

On computer based training (CBT), Florida A and M University's Professor M.D. Roblyer, in a 1989 review based on a telephone survey of police education and training organizations in all 50 states in the USA, reported that the Police Training Institute of the University of Illinois was by far the most active in developing and using CBT resources. There is reason to believe that this pattern has not changed since that assessment was made..

Author Dr. Walker, an Associate Professor at the Police Training Institute (PTI), University of Illinois at Urbana-Champaign (UIUC), in a 1992 Macmillan encyclopedia review and subsequent presentations with colleagues at the 1993, 1994, 1995, and 1997 Meetings of the Academy of Criminal Justice Sciences in Kansas City, Chicago, Boston, and Louisville USA respectively, reported empirical research findings and new developments in the usage of microcomputer lesson materials. Among these are the microcomputer lessons listed in Appendix A of this paper which serve as a basis for developing the modules described in Appendix B. These modules use a combination of computer based training (CBT) and formal classroom training designed to provide student officers with a variety of skills that will assist with field-site response patterns. They were developed with the practitioner in mind by organizing and identifying fundamental functions that may be performed in the field.

Present day CBT technology allows each student to learn at his or her own pace. The underlying premise of this methodology is that highly motivated individuals can master a subject given adequate time and carefully designed materials. Through the use of innovative computer technology, designed specifically for education and training, trainees set their own pace for Aiterative@ learning, thus enabling them to master each lesson, program, unit, and subject before moving to the next. Trainees who need to repeat the instruction can do so with the press of a key. Students who learn quickly can progress rapidly through a review of basic knowledge/skills toward more challenging learning at their own pace. Despite individual differences in learning capacity and background, each student receives individualized instruction which produces maximum interest and challenge to his/her learning potential.

Another plus for individualized CBT is that, once materials have been produced, quality standards are assured with continuous updates for currency, and the instruction can be presented effectively anywhere in the world there is a compatible personal computer, telephone and modem for Internet delivery -- such as a college or university, police or fire department, home, elementary and secondary schools, or library. This training is best followed by opportunities to apply what the student has learned through self-study in an organized exercise or other form of practical application. In advanced training, supervised, pre-planned applications of learning in real-life situations assure greater learning and better performance in less time. One of our findings has been the importance of use of carefully designed pretests. Well-designed pretests insure that students have the prerequisite skills and information assumed in the design of the learning materials, serve as motivation for the instruction to follow, and provide opportunities for gathering validation data for test items to be used in future posttests. Knowledge of the general characteristics of the student population is an absolute necessity in design of effective pretests. For example, a pretest that is too long can be demoralizing and a pretest that includes items that do not appear to be job-related can demotivate students.

Our intent in design of posttests has been to build a massive pool of items that completely covers the subject area. Posttests are made up by sampling from this pool. Thus, students know that a particular posttest may contain items from any part of material taught and that, should they fail the test one time, the posttest they see the next time will probably contain many different test items. Computer-presented test items may also be automatically varied from presentation to presentation, so even the same core item may have a different appearance from one presentation to another. In short, we can be very sure that a student who performs well on a posttest has in fact achieved the required level of learning.

The CBT materials featured in the Appendices are designed to provide the police officer with a variety of competencies intended to insure appropriate development of effective skills involving critical operational problems. All of these materials are developed and continually updated to reflect results of the latest research and field experience. The unique aspect of this CBT approach is the computer-based pretesting and posttesting interactively coordinated with tutorials, intermittent questioning, drills, and final examinations. After use of the microcomputer, direct classroom instruction may culminate with hands-on activities to insure application of selected course content and ultimate learning certification. Thus, the microcomputer materials in the Appendices are designed to permit the best learning outcomes to occur within an environment that motivates and challenges student involvement. Author Dr. Walker published a review of the microcomputer modules in the Appendices of this paper and their associated materials on the Internet in Spring 1996. For his presentation of a paper at the 1997 Meeting of the Academy of Criminal Justice Sciences (ACJS), Dr. Walker coined the term Web CBT to introduce demonstrations of interactive, computer based training (CBT) lessons running live via the Internet from his home page. His microcomputer, Internet lessons run unit-by-unit providing a fast and secure method of remotely delivering instruction without download delays often associated with other CBT delivery systems. The power of Internet delivery of CBT is to essentially broadcast live to an enormous world-wide audience. An excerpt of this 1997 ACJS paper can be found on-line in the Instructional Microcomputing newsletter (Walker, Janssen, & Avner, 1997) (URL: http://www.oir.uiuc.edu/etag/news/issue/may97.html) published bimonthly on the Internet by the Educational Technologies Assistance Group (ETAG), University of Illinois at Urbana-Champaign, Illinois USA.

Since 1978, some solutions to problems important to education and training have been primarily developed by Dr. Walker and other PTI faculty including Professors Chris Flammang, Albert Johnston, Frank Manella, and Cliff Van Meter; and PTI staff associates including Allen Avner, Richard Dennis, Darlene Chirolas, John Janssen, and Paul Tenczar. Our pioneering "state of the audience" microcomputer lessons are based on aforementioned instructional techniques proven effective in training over 20,000 sworn police officers of city, town, county, state and federal agencies. Dr. Walker's experience in computer courseware development began with central computer systems. In the past 13 years, he has logged over 5,000 hours of computer programming using the TenCORE Language Authoring System (1998). TenCORE is a microcomputer software product of Computer Teaching Corporation, 1713 S. State Street, Champaign, Illinois 61820 USA; E-mail pt@tencore.com. Dr. Walker's Internet home page address or universal resource locator (URL) is: http://www.staff.uiuc.edu/~royw/.

Dr. Walker's 1996 account of how he got started on the Internet and World Wide Web (WWW) was reported to the 1997 Meeting of the Academy of Criminal Justice Sciences in Louisville, Kentucky USA. By January 1997, Dr. Walker ran his first computer based training lesson via the Internet without downloading. This new TenCORE Runtime feature runs files direct without going through a browser. Only units or screen displays that are needed get transferred over the Internet, thus, screen displays run very fast. No lessons are downloaded. Instructional materials exist in the local machine only for the temporary period that they are executed. A demonstration of this new Internet Runtime software can be obtained from Dr. Walker's Internet home page (URL: http://www.staff.uiuc.edu/~royw/), or at the following TenCORE Internet URL address: http://www. tencore.com/tcrunnet.htm. This new TenCORE Runtime software is also a product of Computer Teaching Corporation (CTC). Dr. Walker produced the following lesson titles which demonstrate Runtime via his aforementioned Internet home page without the need for the end user to download: Crisis Intervention, Variant Behavior; Problem Personnel Management; The Arson Crime Scene Search; Community Policing; Emotional Stability for Police Officers; The Functional Components of Interrogation; and Sexual Harassment in Policing. These microcomputer lessons will run via the Internet allowing users to study at any location of choice where there is a telephone. The telephone costs would be no different than the local calls made for connecting to the Internet.

Walker, Charles, and Avner (1994) with Avner (1992) reported that student officers who successfully completed selected microcomputer lessons demonstrated exceedingly strong learning treatment effects, and that treatment gains were all substantially above the 3 standard-deviation effect size sometimes mentioned as an ideal goal for instructional treatments. See Appendix A for a complete listing of lesson titles, and Appendix B for the lesson modules. A selection of these lessons also serving as the basis for a study of the effectiveness of computer-based education at the undergraduate university level was reported by Johnson, Van Meter, and Walker (1995). Study guides were developed to give added support for notetaking and learning. The modules containing the microcomputer lesson titles in Appendix B and associated materials including study guides and curricula form the basis for over 150 hours of traditional classroom instruction. They are designed for police officers. Alternative modules can be made up from these same lesson titles for allied criminal justice personnel including security officers, correctional officers and human service workers.

Most college and university students, professors, staff, etc., can get connected to the Internet by contacting an on-campus representative. They may connect to the Internet from any location where there is a telephone, e.g., school, library, work and home. Usage charges are paid or absorbed by student fees, etc.

Even persons without direct connection to the Internet via a college, university, or other organization may still get connected to it via telephone line and a commercial information service provider like Microsoft, America Online, etc. These providers supply software to configure a computer in such a way that it can talk through the modem to the Internet. Everything a person needs to get started already has been worked out to make the process user friendly. A modem connects the computer to the Internet via a telephone line. Behind the scenes, telephone lines, satellite connections, fiber-optic wiring, coaxial lines, microwaves of various kinds and infrared beams, etc., link networks and individual computers of the Internet.

The most recent development in information management is the World Wide Web (WWW). As stated earlier in this paper, the Internet has become more accessible because of the organizing system of the World Wide Web (WWW) which makes it easy to establish links between computers around the world. Links between various types of information appear seamless across time and space. Word-processed files, graphical images, video, sound, program files or data files such as spreadsheets can be inserted into the Web with ease. Any kind of file can be processed and transmitted on the Web. Specific words or phrases in a document can also be linked to specific words or phrases in other local or remote Web sites. This is a hypertext environment of diverse, unrestricted linking of information. Thus, users can search for words or phrases embedded in any WWW document instead of being restricted to searching the subject file of a book, its table of contents, or its index (Handbook of the United Nations, 1997).

Navigation of the huge mass of information available on the Web requires the use of special tools. The Web browser is the most common navigation tool. One can create his or her own home page and post links to various remote home pages. WWW has become the fastest growing part of the Internet because it makes the Internet much easier to use and brings together many disparate, and often arcane, services on the Internet in a user-friendly environment.

It is necessary to know the World Wide Web (WWW) address, referred to as the Universal Resource Locator (URL), in order to access a WWW site. The URL address identifies a specific site or home page. For example, the home page address or URL for contacting the authors of this paper is: http://www.staff.uiuc.edu/~royw/. Thanks to the configuration of WWW browsers, it is rarely necessary to type in such long addresses in order to reach a specific site. To go straight to the desired location it is a simple matter of using a mouse to click on an icon, or a menu option. Favorite and frequently used sites can be stored in a list of bookmarks which can be easily used at a later time without retyping the address.

Connection services to the World Wide Web (WWW) are getting better despite the fact that the connection from the home or office computer must be fast and of a different type than that needed for email. Complete Internet WWW access is now being offered by commercial providers, including the software necessary for use and templates for developing one's own home page. Commercial providers can usually solve difficult technical problems involved in accessing the Internet. The two most widely used access programs are the Netscape Navigator and the Microsoft Explorer.

Currently, significant advances in technology have set the stage for making the Internet available to almost anyone world-wide who has a television (TV) set. There are 95.9 million homes with at least 1 TV set, which is 98% of all U.S. households (World Almanac, 1997), and 855 million TV sets in the world (UNESCO, 1995). People will be using their remote control to see the Internet on their regular TVs just as users can now watch TV channels on minimally upgraded personal computers (Maney, 1997). In this age of television (TV), one notion is to dramatically increase the number of users of the Internet with over 40 million users by interfacing with TV with hundreds of millions of viewers world-wide. This fusion is now known in the media as Web TV.

In the typical instructor-based law enforcement program, the media of instruction (texts, lectures, audio-visual aids) tend to promote a degree of passiveness on the part of the learner because he/she need not participate actively during much of the instruction. Some discussion work is included, as are sessions of role playing, but they do not continuously involve students on an individual basis as would be found in the most effective instructional settings. Among the evaluation modes implied, the most efficient and, consequently, most heavily used is the written test. The press of time and expense of observational evaluations severely limits the level of true performance assessment that may be done on an individual basis. Most of this practice and

evaluation must be relegated to early professional experience under the supervision of a field training officer/instructor.

By introducing the Internet and World Wide Web (WWW) into the picture, we can involve the benefits of branching, tailored feedback, required responses, repetition, and simulation that would be present in the best tutoring environment to complement the traditional classroom instruction. We can also extend the instruction to any number of locations simultaneously, or even replace the physical classroom. Students can interact with the instructor outside of class hours, view materials from libraries and other virtual resources around the world, run interactive software, complete homework assignments, consult subject matter experts, collaborate with other students, and communicate in other ways consistent with active learning (Ball, 1998).

For one of the authors of this paper, Dr. Walker has been inviting his students to use the Internet and World Wide Web (WWW) since May 1996 to contact him through his aforementioned Internet home page outside of class hours, and any time after graduation throughout their careers. He is willing to respond to questions which might be answered based on his experience in training over 20,000 sworn police officers by computer since 1978. To facilitate an ongoing relationship with his students and others in the field, Dr. Walker maintains the currency of over 300 links on his Internet home page. These links to criminal justice resources are selected based on his experience in teaching information sources at the police academy level for over 20 years. Thus, past, present and future students of Dr. Walker and others in the field can return to a "virtual classroom" for updating of their skills and knowledge throughout their careers, from any location in the world and at any hour of the day.

Our experience with the Internet, World Wide Web (WWW) demonstrates that

(1) techniques proven in past applications of distance learning continue to be valid for this medium, (2) instructional design and production know-how are vital components of successful application of the medium, and (3) available production tools do permit dedicated workers to generate effective distance learning materials with minimal technical support. All of the materials produced by the authors were generated entirely outside of our normal working hours and without institutional support. We do not recommend such a "cottage industry" approach, but do present the results of its application as evidence that use of the WWW for distance learning of criminal justice topics is a current reality that is within the reach of all institutions.

The authors can be contacted at the Internet address Ahttp://www.staff.uiuc.edu/~royw/@ for reviews, an e-mail box for asking questions including how to get started in developing your own courseware, and running computer based police training demos live via the Internet.

LESSONS Hour (ECH)¹

Pretest and Posttest (Variations by ALessons@ and Modules) . . 2

Drill and Practice² . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-30