Adapting Educational and Psychological Tests for Cross-Cultural Assessment
reviewed by William L. Brown - October 03, 2006
Title: Adapting Educational and Psychological Tests for Cross-Cultural Assessment
Author(s): Ronald K. Hambleton, Peter F. Merenda, and Charles D. Spielberger (Eds.)
Publisher: Lawrence Erlbaum Associates, Inc., Mahwah, NJ
ISBN: 0805861769, Pages: 367, Year: 2005
Search for book at Amazon.com
With the advent of the No Child Left Behind Act (NCLB), signed into law on January 8, 2002, the market for educational and psychological tests mushroomed. There has been a sharp increase in the number of subjects (mathematics, English language arts, and science) and grades (three through eight and 11) where assessment is now required by federal law. In addition, all students must be testedincluding students with disabilities and English language learners.
Beyond NCLB is the growing interest in international comparisons of student achievement in reading, mathematics, and science. For example, the Trends in International Mathematics and Science Study (TIMSS), formerly known as the Third International Mathematics and Science Study, asserts that they provide reliable and timely data on the mathematics and science achievement of U.S. students compared to that of students in other countries. (See the NCES TIMSS website at http://nces.ed.gov/timss/) The validity of such assessments relies heavily on the accuracy of various translations of the instruments used for making the comparisons.
This book will be beneficial to test developers faced with the necessity of creating equivalent test forms in several languages. As the writers point out repeatedly, simply translating a test directly from English into another language is inadequate. It is not safe to assume that the translated version will measure the same construct as its English-language parent.
The editors and authors are recognized experts in psychometrics and the techniques and ethics of assessment. All have been involved in adapting tests for international use. A committee of the International Test Commission, under the leadership of Dr. Hambleton (with Dr. Spielberger as a member), worked over a period of several years to develop a set of Guidelines for Test Adaptation, which forms the basis for much of the material in this book.
Hambleton, et al., have provided a useful compendium that details the information needed to adapt a test to a different language, both scientifically and ethically. Although much of the content is repetitive, the reader is presented with a clear explanation and practical examples of the issues, designs, and technical guidelines for adapting tests not only for multiple language groups, but also for multiple cultures. The authors cite several factors that have a strong influence on test scores. If these are ignored, they can lead to invalid comparisons of the achievement levels of the examinees. Some of the critical factors are:
Construct Equivalence: Is it sensible to compare the two cultures on the construct? Does the construct have similar meaning in all cultures being compared?
Test Administration: Is training provided to test administrators to ensure that instructions are strictly followed?
Test Format: Does the test require a balance of essays, short answers, and multiple-choice items so that examinee experience with a single item format is not unduly helped or hindered by the format being used?
Speededness: Does test speededness place those lacking experience with speeded tests at a disadvantage?
The authors make it clear that one cannot simply translate a test from one language to another and expect the two instruments to be equivalent (or parallel). Minimally, a translated test should be back-translated to determine if the original meaning has been preserved. For example, if Out of sight, out of mind is translated and then back-translated, the result might be Blind, insane. The authors give several detailed examples of the methodological and analytical steps required to ensure that the translation measures a construct equivalent to the original version.
The authors also caution against making the assumption that bilingual speakers are representative of native speakers of one of the languages or the other: A major drawback of bilingual linking designs is that the bilingual group may not represent either of the monolingual groups that are the groups of interest in a comparative study (p. 148). Typically, bilinguals will not represent the range of abilities a test is designed to assess, since they usually have extensive experience in both cultures that members of either monolingual group have not had.
Anyone with the responsibility of adapting tests for cross-cultural assessment would be well advised to add this book to their library, and refer to it often. It deals with the legal and ethical issues in adapting tests to different languages and/or for diverse cultures. In addition, it offers much practical information on methods for ensuring that the adaptations lead to fair assessment of the underlying constructsa critical issue in ensuring valid assessment.