Riverside Comments on the Friends for Education Report

Abstract

The authors, both affiliated with Riverside Publishing Company, discuss the factors they think explain the Lake Wobegon phenomenon and call for more appropriate use of normative comparisons and more complete reporting of test results.

Educators' Perceptions of NRT Misuse

“…Despite warnings from measurement experts about the limitations of standardized tests and the possible outcomes of their misuse (Drahozal & Frisbie, 1988;Lenke & Keene, 1988;Loyd, 1986), the public has demanded and governing bodies have mandated high test scores as "proof' of educational quality. Members of these same groups have been quick to denounce educators when predictable outcomes, such as teaching to the test or more blatant cheating, have occurred (e.g., Cannell, 1988;CBS News, 60 Minutes, 1990).…”

“…Some responsible measurement specialists (Linn et al 1989;Shepherd, 1990) agree that scores often do appear spuriously high, and they point out some legitimate reasons, e.g., old norms. Others have documented occurrences of inappropriate testing practices (Gonzalez, 1985).…”

The Lake Wobegon Effect Revisited

“…The second explanation-that some states are manipulating the scores-is made convincingly in the Riverside response (Drahozal, 1988). They discuss test exclusion policies asking, "How many fourth graders are there in state X who were not tested.…”

Developing a national frame of reference on student achievement by weighting student records from a state assessment

“…The recency of test norms can also affect the utility of norm-referenced interpretations (Drahozal & Frisbie, 1988;Williams, 1988). As test norms become dated their reflection of achievement in the population of students may become less accurate.…”

“…The primary responses from test publishers focused on educating test users about the life expectancy of norms and developing annual norms (Shepard, 1990). Drahozal and Frisbie (1988) emphasized that the recency of norms is an important consideration when student achievement has changed. Linn, Graue, and Sanders (1990) also believed that Cannell's results were in part due to the use of old norms; however, they did not think that annual norms was the solution to the problem of inflated test scores.…”