Home » Briefings & Reports » Estimating the Costs and Benefits of Educational Testing Programs

Estimating the Costs and Benefits of Educational Testing Programs

posted in: Briefings & Reports

(click here for full article)

(click here to download a PDF of this article)

By Richard Phelps, Ph.D.
Education Consumers Consultants Network



The Bush administration’s No Child Left Behind Act requires annual testing in grades 3-8, so now there is a huge controversy about how much all of that testing will cost. Some studies estimate costs of nearly $1800 per student and others are in the $10 to $25 range.

Talk about differences of opinion!

Why do the experts differ by such enormous amounts? The short answer is that some proponents of testing count only the cash expenditures added as a result of testing and some opponents count any expense that can be linked to testing regardless of whether it is incurred as a result of testing.


Estimating Costs

For example, a 1998 estimate by The Center for Research on Education Standards and Student Testing (CRESST) counted teacher salary paid for regular instruction covering the subjects on which the students would be tested. Its estimates ranged from $848 to $1,792 per student. A Boston College study counted students’ time as well, i.e., both the time taken for instruction and that taken for testing. Counting teacher and student time as costs makes sense only if one adopts the controversial assumption that these uses of teacher and student time detract from more important uses of the school day. The Boston College study also apportioned building maintenance, capital costs, and other forms of fixed overhead to testing. Plainly, these costs do not increase when testing is implemented.

In contrast to estimates based on cost-allocation are the prices that students and schools actually pay for exams such as the ACT, SAT, and AP. They range from $20 to $70 a student-prices that cover all development and scoring costs, wages for the test administrators, and a profit for the publisher.

In 1993, the U.S. General Accounting Office (GAO) asked state and local testing directors to report cost details on their statewide and districtwide tests. It found that state-mandated, mostly multiple-choice tests cost approximately $15 per student. Performance-based tests cost around $33 per student. These estimates included all recurring costs including the teacher time used in administering tests. The creation of new tests cost an additional $2 per student for multiple-choice tests and $10 per student for performance tests. See http://www.gao.gov search for report PEMD-93-8.

The GAO estimate included all costs that would be incurred if a single test were given districtwide and administered independently–say during the summer months and by hired personnel. If, instead, a test replaced an existing test and school personnel administered it during the regular school year, added costs dropped to $2 per student for a multiple-choice test and $11 per student for a performance test.


Estimating Benefits

Testing incurs costs but it provides benefits as well. As with cost estimates, benefit estimates tend to vary with the leanings of the estimator. Testing proponents tend to count numerous benefits while testing opponents tend to ignore them.

The primary benefit of testing is that it provides information about whether students are learning. Education has a variety of stakeholders: children are the beneficiaries, parents are the purchasers, taxpayers and policymakers are the sponsors, and educators are the suppliers. All of these parties need accurate information about results.

Independent and objective testing is especially important to public education’s lay stakeholders. Educators working directly with children have the opportunity and expertise for assessing student learning. Children, parents, and policymakers lack both assessment skills and knowledge of relevant standards. Given that schools have substantial incentives to maintain a favorable public impression, an absence of independent and objective testing can invite unduly upbeat reports. Without external standardized testing, the public schools are local monopolies accountable only to themselves.

Testing yields other proven benefits. For example, it requires teachers and students to try harder. Studies of high-stakes testing programs estimate motivational benefits to be around $13,000 per tested subject area over the lifetime of students. The educational achievement it produces affords tangible career benefits such as higher wages and greater job security.

Testing also benefits employers and higher education institutions. Good tests used in conjunction with other indicators can predict occupational success and match graduates to appropriate career opportunities. Thousands of studies have shown that achievement and aptitude tests can better predict success than other indicators normally used in hiring decisions Predictive validity is the term used for measuring the degree to which test scores predict future success in matters such as job or college performance. Allocative efficiency is the matter of whether workers and students are well matched to their job or their studies. The better the match, the greater the benefit for both the organization and the individual. Even critics of these studies concede benefits on the order of $5,000 to $8,000 per lifetime.

Although estimates of cost and benefit are debatable, testing benefits are generally conceded to substantially outweigh costs. In the area of accountability, however, testing seems indispensable. Without an equally independent and objective alternative, high stakes standardized tests are likely to remain the public’s principal means of monitoring its investment in public education.

The Education Consumers Consultants Network is an alliance of experienced and credentialed educators dedicated to serving the needs of parents, policymakers, and taxpayers for independent and consumer-friendly consulting. For more information, contact J. E. Stone, Ed.D., at (423) 282-6832, or write: professor@education-consumers.com