This chapter outlines the underlying cost structure of the basic components of school technology plans and examines the investment choices available. It also presents a framework for analyzing a technology plan's life-cycle costs. Chapter 3 describes a revenue model that will help assess the financial feasibility of the technology plan when used in conjunction with the cost model. The approaches to analyzing the costs and revenue sources we outline in these two chapters will provide the basis for formulating specific funding strategies at the school district level. Those strategies are discussed in Chapter 3.
In developing the plan that will put technology into schools, there are at least six stages of analysis required to properly inform the school district's funding strategy. Exhibit 2.1 illustrates an approach to developing and analyzing a district-based technology plan as a precursor to developing the funding strategy. Any technology plan should be formulated around the educational goals of the school district (stage 1). Conducted properly, this step will clearly define the technical and functional requirements needed to facilitate the district's educational goals (stage 2). Potential technology solutions will be shaped by these specifications (stage 3). Once the plan has taken shape, its technical and operational feasibility should be assessed. Usually, there is more than one alternative technical solution capable of meeting the objectives for a system (stage 4). To ensure a prudent public policy, the cost of at least one other feasible technical alternative should be evaluated.
Once the most appropriate technology has been selected (stage 5), the costs of the entire technology plan should be analyzed by developing a model of the life-cycle costs of the plan (stage 6). The technology plan should be reviewed and possibly refined in light of this analysis. A model of life-cycle revenues should be developed by assessing the funds that are available from existing sources and mapping them back to the technology's life-cycle costs to identify gaps in the available funding streams. Now, we have reached the starting point for developing a creative funding strategy. Part of this strategy may involve amending the plan's goals, technology configuration, or implementation to better accommodate funding availability.
At least five basic technology models provide schools and classrooms with access to the information superhighway. The five are based on a combination of the connectivity models described by Rothstein and McKnight 8 and the infrastructure deployment models described by McKinsey. 9 The models are listed below and presented in a stylized form in Exhibit 2.2.
Adaptations of these models will differ from one site to another, depending on the nature of the facility, the needs of staff and students, and what technology currently exists at the site. However, we present the models in their basic form to provide a framework for considering the economics of the systems that underpin them. These considerations can be used to support the funding strategy.
Model 1 is a basic, low-cost design. Using a modem, it provides only a single line connection such as via "plain old telephone service" (POTS) to the district office server that limits access to one user at a time. Because of the narrow bandwidth, it is suited best to text-based applications on the Internet (i.e., limited bandwidth means that slow response times will not adequately support video or graphical applications).
Model 2 enhances Model 1 by adding a local area network (LAN). The LAN gives every machine connected to the network the ability to access the Internet in the manner described in Model 1. The number of users able to access the Internet at any one time is limited by the number of outside access lines the school has. Apart from an increase in the number of simultaneous users, the benefits of this system are the same as those in Model 1. The technology environment sustained by Model 2 is equivalent to the Laboratory model described by McKinsey.10 It provides very limited access to technology, and access to computers and the information superhighway must be scheduled. For these reasons, this model is unlikely to lead to full integration of technology into the curriculum.
Model 3 differs from Model 2 by using a router instead of a modem. The advantage of a router is that several users at a school can access the Internet at the same time. With a router in place, the LAN can be expanded so that it becomes possible in Model 3 for the system to simultaneously support one or more PC in each classroom. Model 3 is the equivalent of McKinsey's Lab-Plus model.11 It can be a stepping stone toward deployment of Model 4 because it supports additional networked computers in the classroom for student use. It also gives teachers an opportunity to develop technology skills and competencies, as well as the time to adapt their courses and teaching methods to the expanded technology prior to deploying Model 4. In Model 3, the larger network increases initial start-up costs, compared to Model 2, and the larger number of users (involving all the teaching staff) increases training expenditures.
Model 4 is similar to McKinsey's Classroom Model.12 In this model, connectivity to the information superhighway becomes widespread throughout the school. Several features of Model 4 are different from the previous models. First, there is a file server at the school that improves the performance of the network because information can be accessed locally. The increased traffic in this system will require at least an ISDN connection from the school to the district office or local telephone company central office. For the same reason, there is a dedicated high-speed T1 connection from the district office to the Internet. In contrast to the earlier models, this enables the school's network to provide some limited support for video and graphical applications. (A T3 connection would provide an even higher-speed link and support applications such as teleconferencing.) The largest cost in this model is the cost of acquiring many computers, resulting in high start-up costs. Because of the larger number of users, equipment inventory, and network complexity, training and technical support will form most of the recurrent or ongoing costs of this system. Also, the high-speed Internet connection is a significant cost item. (The "E-rate" component of the Universal Service Fund and Telecommunications Act of 1996 can reduce this and related telecommunications costs through discounts.)
Model 5, which assumes there is a computer at every student's desk, will require a high-speed link (T1) between the school and the district office. The costs of this model are high. Hardware costs (for a computer on each desk) are large, and the training and technical support function will be extensive. The large number of computers in the school will require greater modification of the existing facilities and reinforcement of several other elements in the model. For example, the large number of users will require strong system management and maintenance skills, which may inflate technical support costs. Compared to Model 4, the marginal benefits this model may produce may not be commensurate with its higher cost.
The five models of accessing the information superhighway described above can be regarded as a family of models in which one model is a natural precursor to the next. Higher overall implementation costs at the time of full deployment of the classroom or desktop model is traded for lower costs in the early years of deployment. Exhibit 2.3 presents a schematic of this deployment pattern for a given school.
There are other compelling reasons why a school or school district may wish to adopt this sort of developmental approach to technology deployment. At relatively low cost, the introduction of the Laboratory model in a school will enable the school to develop technical leaders from among the staff who may subsequently help develop the technical competencies of other teachers in preparation for the next phase of deployment. The next phase, the Lab-Plus model, provides a networked workstation for each teacher, which will enable teachers to gain experience with their newly acquired technical skills and will give them the opportunity to begin integrating technology into the curriculum. Achieving these goals can smooth the way for the next stage of deployment, the Classroom model. Movement to each stage can be based on an evaluation of the success of the current stage.
So far we have discussed technology deployment and costs at the school level. Usually, though, technology planning occurs at the school district level. The developmental approach of these models can be applied similarly at the school district level. For example, although local educational goals will dictate the ultimate technology configuration the school district wants, the movement toward these goals does not have to occur uniformly in each school in the district. In practice, the implementation timetable is unlikely to be the same in each school. Different schools may require different models. For example, the method of deployment may be different for newly built schools compared to older schools, elementary schools compared to high schools, or schools with falling enrollments compared to schools where enrollments are expanding. Considerations such as these reinforce the need for a cost analysis framework to distinguish costs at the level of individual schools.
Exhibit 2.4 illustrates technology deployment patterns for five different schools in the same school district. In this example, the goal of the school district is to obtain a relatively high level of computer penetration in all its classrooms by the year 2000. Each school follows a different trajectory toward this goal, based on its individual circumstances and achievements. The technology model that ought to be deployed initially in a school depends on several external factors (such as whether the school building is old or new or whether there are teachers already on staff who have technical competencies). How long a school remains in any one stage of the technology transition may depend not only on the school district's goals but also on the school's success in that stage of development, as indicated by student outcomes, staff competencies, changes in school practices, ability to innovate, ability to raise matching funds, and other factors.
To minimize pressure on funding, it is important to identify system constraints in the planning process. If they are not recognized at this stage, the system will not perform as planned and is unlikely to produce the results expected of it. System constraints are most likely to occur in three areas: equipment, software applications, and staff development. In each area, the following factors will impede system performance:
If any of these obstacles cannot be removed or mitigated, an action plan should be developed to facilitate overcoming them.
Technology costs should be analyzed, not simply budgeted, to ensure that the design of the plan is optimal and that it can be financially sustained over its projected life. Provided an appropriate cost analysis framework is used, the flexibility of the technology plan can be assessed, trade-offs can be considered explicitly, and other options that preserve the plan's goals can be evaluated. Using an analytical approach will increase the likelihood that technology will find its way into the district's schools and that its deployment will be successful.
Numerous estimates have been made to evaluate the cost of implementing technology plans at the national level, and school districts commonly use budgeting methods to cost plans at the local level. Whatever the method used, they all seem to have some methodological weaknesses in terms of developing the financing or funding strategies needed to implement technology.
The methods used by RAND, McKinsey & Company, and other organizations to estimate costs at the national level cannot be applied to develop cost estimates for district-based technology plans for several reasons. They typically assume only one technology model is deployed nationwide, and that the technology is deployed in uniform waves over a finite period of time, such as five or 10 years. They may use only one profile of the "average" school, and they typically assume there are no funding constraints.
Even the methods used by school districts themselves to cost their technology plans are not conducive to analyzing the financing implications of these plans because their methods tend to be borrowed from budget analysts. From the standpoint of devising a budget, this approach is fine, but in terms of devising a funding strategy for the technology plan, it is too limiting. For example, if there is no existing budget account to fund a particular element in the technology plan, that element may be excluded from the plan regardless of its potential contribution to education goals. Costs derived for budgeting purposes are rooted in concrete objectives, which limits the scope for exploring strategies, timeframes, or evaluations to gauge the effectiveness or success of the plan.
At the school district level, the methods used to cost technology should provide answers to questions other than just "How much will the plan cost?" Such questions might include: Which technology model best allows the attainment of our goals? Should all schools in the district implement technology at the same time? If not, which schools should go first? What factors can jeopardize the success of the plan? Is a technology leap required or can a series of intermediate steps be taken? Can the technology plan be funded? If not, what can be done to bridge the gap between resources in hand and the total needed?
9 McKinsey & Company (1995).
10 McKinsey & Company (1995).