Distance education (DE) refers to the use of digital, networking, and telecommunications technologies to provide education while allowing geographical separation between teachers and students. Any single evaluation of DE is an inquiry into the impact of a specific program on a well defined group of students. In the aggregate, however, different evaluations can be combined to yield an overall sense of DE's impact on the educational process. A model for organizing data from these evaluations is presented in Figure 1. The model consists of four high-level categories of variables, and two perspectives on change.

• characteristics of the student body
• student outcomes
• characteristics of DE programs
• institutional arrangements

• short term (blue arrows)
• long term (red arrows)

• Substitute delivery method: This is a 1:1 replacement of one teaching method for another. For instance, a lecture in a classroom may be delivered via television or through the Web. Class readings may be delivered in electronic form instead of hard copy. The innovation is in the medium by which a traditional method of education is offered.
  
  
• Value add: These are techniques that improve traditional education, but which are impractical or impossible outside of a digital environment. One example may be establishing project teams with members who are separated by geography. Another example may be using various simulation techniques that require computer technology. A third example may be using discussion group technology organize students' questions over time. (Of course these methods can also be part of traditional classroom education.)
  
  
• Efficiency: In a traditional class, adequate attention to students requires a maximum ratio of students to teacher (or qualified teaching assistant) of about 30:1. I believe that ratios much beyond this limit make it difficult to carefully grade papers, answer questions, and otherwise attend to students' individual needs. For many (most?) DE programs, the only efficiency gain is in "lecture" elements, where it does not matter if the number of students is 3, 30, 300, or 3000. As long as efficiency is unchanged for other elements of education, the economics of providing education will not change, no matter what the mode of delivery. DE systems that can shift efficiency can have profound affects on the cost and availability of education.

• Existing Students, New Access Route for Same Amount of Education: In this situation students who would ordinarily use face-to-face educational methods switch to DE. The size and make-up of the student body remains constant.
  
  
• Existing Students, New Access Route Leading to More Education: Here the traditional student body remains unchanged, but because of the availability education through DE, those students seek more education than they normally would.
  
  
• New students: In this case the customer base for education goes up because DE brings in new students who would not otherwise avail themselves of particular educational offerings.
  
  
• Student Clustering : Here what is important is not the number of students or amount of education they receive. Rather, the operative factor is the richness of communication among students. For instance, DE may result in more education for people in a particular company or interest group. Beyond any individual learning that takes place, we may expect additional impact from interaction among students who share common interests.

• Shifting Customer Base: Traditionally, schools compete for students on the basis of quality, cost and geography. Prospective students choose the best school they can within limits of their ability, budgets and freedom to travel or relocate. DE weakens the importance of geography in that competition. This is not only true for the teaching component of education, but it can also be true for the research component. For example, when complex machinery can be manipulated remotely through easily accessible Web-based interfaces, students doing research may have less need to live near their universities.
  
  
• Change Within Educational Establishments: As DE shifts the customer base, schools will deal with the same problem that has bedeviled many traditional companies, i.e. the balancing act required to move one's traditional customer base from a bricks and mortar environment to a digital environment. Executives fear that great expense will go into cannibalizing an existing customer base, without commensurate change in either the amount of business from existing customers, or the development of new customers. To exacerbate the problem, traditional companies fear that if they do not go digital, new competition, without sunk costs in infrastructure, will move in to compete. Companies are experimenting with many different solutions to this challenge, with varying degrees of success. The result is considerable stress, and a lot of reorganization, in many companies. The dynamic is exactly the same in the field of education. As DE exerts its influence, students' choices for receiving educational services will force schools to confront these problems.
  
  
• Inter-organizational Arrangements: DE technology allows new business models for relationships among schools. For instance, schools can develop courses jointly, share a student body, or deploy each others' faculty for specialized training. Schools are already experimenting with these models.

• Skills/Knowledge: One important outcome measure of any educational effort must be the amount of information imparted to students.
  
  
• Critical Thinking, Problem Solving: Depending on circumstances, improving students' problem solving ability may be as important, or more important, than imparting specific information or knowledge. To illustrate, consider two courses, one an introductory course in literature, and one an advanced programming course for experienced professionals. Certainly a key outcome of the literature course should be an increase in students' ability to analyze literature with an eye toward discerning structure, style, and form. If they achieved this goal at the expense of reading fewer pieces than in the original syllabus, nobody would complain. In the programming class however, students are likely to already be excellent critical thinkers about program logic. What they need is factual information about how a particular language works. For them, the emphasis is on factual learning rather than increasing their ability to think critically about programming.
  
  
• Community Building: In many educational settings, community building is as important as imparting knowledge or teaching critical thinking. One good example is graduate education, where one goal of the process is to develop a generation of colleagues who can interact with each other over the course of their careers. Another example might be undergraduate settings where school ties affect life-long career and business choices. Even in shorter term training, community building may be important. For instance, participation in training on an innovative technology may lead, not only to mastery, but to an informal support group which can provide ongoing assistance.

• Short Term changes are immediate effects whose observation does not require an extended time frame. (While a longitudinal methodology may be strengthen assessment of this change, there are likely to be good opportunities to obtain good data without much emphasis on change over an extended time frame.)
  
  
• Long term change requires a methodology that tracks change over time.

Characteristics of DE Program and Student Outcomes: From the point of view of educators and planners, this is the primary feedback loop in the model. This information is critical because it is key to improving program quality. The feedback loop is short because much of the needed data can be obtained during a course, or shortly after its completion.

Characteristics of Student Body and Student Outcomes: Because interaction among students can be so important for learning, this factor must also be included in assessments of student outcomes. As with the content of the program, the necessary data can be obtained during, or soon after, a course is completed.

Institutional Arrangements and Characteristics of DE Program: Institutional arrangements can also have direct impact on the characteristics of DE programs. For example, as a way of capitalizing on DE's reach, a school may choose to begin offering courses to non-traditional students. Doing so, however, may require special teaching elements to accommodate students whose study skills are rusty. I treat this as a short feedback loop because I assume that reasonably good planning for any given course will take this issue into account.

Student Outcomes and Characteristics of Student Body: As particular DE programs gain reputations for quality they will draw more, or different students. Assessing these changes will take a relatively long time because it takes time for students to learn about, and then commit to, alternative sources of education.

Institutional Arrangements and Characteristics of Student Body: Institutional arrangements, of course, may have an impact on the characteristics of a student body. For instance, a school may change its admissions criteria, or begin to market new programs to potential new types of students. While immediate impact may be observed for a particular course or semester, it will require time to determine whether those are enduring changes.