In an important article in the current issue of Educational Evaluation and Policy Analysis, Douglas Harris makes the common-sense argument that, in order to understand the potential value of educational interventions, we need consider not only their effect sizes, but more importantly their “cost-effectiveness ratio” (CER), i.e., how much effect size in desired outcomes is achieved per unit of cost. Though the basic point he makes is so evident to be almost trivial, the article analyzes in depth many of the complications with determining CER, such as how to account for benefits that may occur long in the future, benefits that might extend beyond the bounds of higher test score outcomes, or costs that are born by a wide variety of individuals and groups. Definitely worth a read.
The article reinforced in my mind why the CER of educational technology programs, and, especially, one-to-one laptop programs, is likely high. Here are a few reasons:
(1) We frequently think about the money spent in these programs, but we less frequently think about the money saved. For example, currently, most schools have dedicated space for computer laboratories. If such labs are cut back or eliminated, we can save on building space, supervision, equipment, etc. In addition, as more textbooks and other materials are ported to less-expensive electronic realms, we can also save on the purchase of educational materials.
(2) The cost of one-to-one laptop programs is falling dramatically and can be expected to fall further in coming years. Thus any plans for such programs in the future must take into account how the costs are falling. In particular, I expect the emergence of inexpensive ARM-based laptops in coming years that will sharply cut the price of purchasing or insuring laptops. These laptops will also use free online software (such as Google Docs), thus reducing both software costs and technical support costs.
(3) I believe that the long-term impact of laptops on academic achievement is underestimated. Most studies of laptop use take place only in the initial year or two of implementation. As students and teachers get used to using laptops, the impact on academic achievement should be higher. This is particularly so if laptops can be brought home, since we know that the long-term effects of access to a home computer on academic achievement are substantial.
(4) The benefits of laptop programs extend far beyond effects on standardized test scores and include preparing our youth for the types of employment and civic involvement required in the future. Though these are difficult to measure, their inclusion increases the CER of technology interventions.
(5) Finally, though laptop programs are expensive, other interventions that bring about positive effects, such as universal pre-K, after-school instruction, and smaller class size are even more expensive. This is not to suggest that those interventions are not also needed, but just to point out that the CER of laptop programs might be comparatively high.