KIRIS: A Volkswagon test for a Cadillac price?
Are we breaking new ground - or just breaking the bank?
From Kentucky Citizen Digest, July, 1997

When the state’s education reforms were enacted in 1990, lawmakers spurned nationally standardized tests because they allegedly relied on multiple choice questions to assess students. Multiple choice questions, it was argued, did little more than teach children how to fill in little bubbles on an answer page.

Kentucky, it was decided, would take a new road. Instead of relying on a testing system that used multiple choice questions, it would take up the cause of innovation and develop a performance-based assessment and accountability system.

The rest is history and, as some have pointed out, history is not always pretty.

The most recent problems with the test might be a good reason to take another look at standardized tests. But if we decided to go back to standardized tests, would we be abandoning so-called “performance-based assessment”?  Hardly.

In recent years, national test publishers have overhauled their products to meet the demands of school systems across the country that are looking for more performance-oriented tests.

The alternatives to a state test in 1990 may have been limited to multiple choice tests, but the alternatives are much different today.

Today, states can find a wide variety of performance-oriented tests on the market, and they can purchase them without impoverishing taxpayers.

While the estimate of the per-student cost of taking KIRIS ranges as high as $1,791.96, the cost of administering any of the most commonly used nationally standardized tests, such as the CTBS, the CAT-5 or the Iowa Test of Basic Skills (ITBS), can be as low as $40.

The cost figures for KIRIS are based on a study by Lawrence Picus of the University of Southern California which includes development, management, test administration and scoring, and evaluation and reporting costs. The cost figures for KIRIS in the bar graph on this page include portfolios, while the figures for the national tests do not.

The Office for Education Accountability’s panel of testing experts recommended that portfolios be taken out of the accountability index because of the unreliability of their scores. If policymakers chose to include them, however, a portfolio component could be purchased from any of the major companies for a very reasonable price.

The Picus report studied development, training, management, test administration, scoring, evaluation and reporting expenditures.

Most of these indirect costs are included in the total per student cost estimates of nationally standardized tests included on the front page. The costs not included (such as test administration) are, according to several testing companies, minimal.

Development costs, for example, are borne by all states and schools who take the tests. Little training is involved since testing companies place a high priority on user-friendly, easy-to-read instructions.

Nationally standardized tests, unlike KIRIS, would be individually reliable and would include national comparisons - and would probably involve a lot less controversy in terms of reliability and validity.

The cost estimates shown on page 1 are taken from the catalogue prices listed by the publishers of the tests and so do not include any discounts that might be expected to result in a bidding process that would involve a large number of students.

As the 1998 session of the Kentucky General Assembly approaches, many state education policy makers will be weighing the relative merits of Kentucky’s current assessment program against the merits of these other, more widely used national tests - and the benefits that go along with them.
 

Key Family Foundation Contacts: Kent Ostrander, Executive Director Martin Cothran, Senior Associate Policy Analyst