Washback in Language Testing: Research Contexts and Methods
Liying Cheng, Yoshinori Watanabe & Andy Curtis (Eds.) (2004)
Mahwah, N.J.: Laurence Erlbaum & Associates
Pp. xxi + 237
ISBN 0-8058-3987-9 (paper)
$27.50 (paper) $59.95 (cloth)
Washback in Language Testing is an edited volume of noteworthy chapters revolving around the concept of washback, or the influence of testing on teaching and learning language. This book is the latest update which began to attract attention in language testing nearly 20 years ago. It is a germane and timely topic because of increasing use of government- and institution-initiated, standardized testing to influence and control accountability in teaching and curriculum (K-12 merit pay in Colorado and assessment concerns of university accrediting agencies), program success (subsequent new and refunding in K-12 and Adult Education programs in the US), student progress (testing to combat "social promotion" in Georgia's elementary schools), and university admissions (all over the world).
Washback in Language Testing consists of two major sections, Concepts and Methodology of Washback (chapters 1-3) and Washback Studies (chapters 4-11), an interesting Foreword by J. Charles Alderson, a thorough reference section, and author and subject indices. In all but one of the research study chapters, surveys, explanation of test format, instruments of analysis and other rubrics are given after each chapter to enable readers to understand the studies better and continue research strands by adapting them to specific situations.
In the first chapter, editors Liying Cheng and Andy Curtis review the historical background of washback. They then discuss the definition of washback, whether it has positive (beneficial changes in teaching and curriculum brought about by changing or instituting examinations) or negative effects (not recognizing course goals and learning objectives to which the test is supposed to relate), how testing has become a device for change in education, and its place in currents assessment trends. This chapter illustrates well the complexities of testing in relationship to teaching and learning.
In the second chapter, "Methodology in Washback Studies", Editor Yoshinori Watanabe discusses the complexity of washback. He characterizes it according to the dimensions of specificity, intensity, length, intentionality and value, along with what aspects of an examination can influence teachers and learners, and factors influencing washback. After this discussion he states that qualitative research methodology may better characterize washback than quantitative methods and then outlines a general method for investigating the phenomenon.
Stephen Andrews looks at the relationship between curriculum and washback in the third chapter. In it, he outlines the negative influences of tests and how tests are used to promote innovations in curricula, especially language tests. He illustrates these through Sri Lanka's "O" level test revisions, Hong Kong's Certificate of Education Examination (English Language), university entrance examinations in Japan, TOEFL, and general education in England, Trinidad and Uganda. He concludes the chapter with insightful suggestions to be thoughtfully considered when introducing or revising high-stakes tests. [-1-]
In Chapter 4, Brian Stecher, Tammi Chun and Sheila Barron report on their study of the effects of assessment-driven testing (WASL) on writing teaching in Washington State. The testing is administered in grades 4, 7 and 10 in certain subject areas for each grade. The study focuses on grades 4 and 7 because those were the only grades tested at the time of the study. As a result of analyzing statewide surveys to both principals and teachers, the researchers found that although the approach to writing, a process approach, changed little before and after the tests were instituted, curriculum (writing conventions, emphasis on audience, purpose, styles and formats) and instructional methods (greatest emphasis on WASL rubrics for student feedback) did change. The study concludes that the WASL influenced instruction; however, it was difficult to determine the importance of the standards in shaping the instruction.
In Chapter 5, "The IELTS Impact Study: Investigating Washback on Teaching Materials", Nick Saville and Roger Hawkey describe the development of instruments for collection of data to analyze and evaluate relevant textbooks. Thus far, the work has been focused on developing and validating the instrument for the IELTS and done by raters about textbooks; the finalized instrument is presented in an appendix to the chapter. That being the main thrust of the article, some preliminary responses from the pilot raters indicate that micro-skills and communicative skills required for future professional or academic study and authenticity of texts, both written and oral, have a positive influence for test takers.
Belinda Hayes and John Read investigate the "Academic Module of the IELTS in New Zealand" in Chapter 6. In a first phase, they used questionnaires and 23 follow-up interviews with teachers preparing students. In the second phase, a classroom study compared two courses, one with an EAP orientation the other more test-focused. Phase One of the study showed that most IELTS preparation courses are in private, commercial language schools. From Phase Two of the study, they conclude that a commercial, test-focused course may have more negative washback effects, while an EAP course addresses a wider range of needs directly related to academic study and promoting language development.
"Washback in the Certificate in Spoken and Written English (CSWE) in Australia" is studied by Catherine Burrows in the seventh chapter. This test, implemented in 1993, introduced mandatory, formal, competency-based assessment with guidelines which included model assessment tasks for teachers to use in their individual classroom-based assessment. This study was done as doctoral research and used a survey to determine classroom practices, interviews to find out teacher beliefs, and observations to see if the practices and beliefs actually were borne out in the respondents' classrooms. Conclusions made in this study are: test designers should take into consideration teachers' beliefs, assumptions, and knowledge in implementing new tests; the importance to washback of empirical research; and quantitative and qualitative data being used to analyze washback.
In the next chapter, "Teacher Factors Mediating Washback", Yoshinori Watanabe reports on the effect of high-stakes Japanese university entrance examinations on secondary school instruction. Classroom observations were done in high schools from which students had been accepted by prestigious universities regularly and, not coincidently, offered special exam prep classes. His conclusions from this study were that teachers' beliefs and educational background were important for beneficial washback to take place.
In Chapter 9, Liying Cheng researches the question if changes to the high-stakes Hong Kong Certificate Examinations in English (HKCEE) to a more task-based approach had any impact on teachers' perceptions about how teaching should take place in the classroom. In a 3-phase study, teachers from 60 schools were surveyed. The results were that teachers were reacting positively to the change although their daily teaching activities and the teachers' reliance on textbooks, quickly rewritten by publishers, remained relatively unchanged. Therefore, changing the exam may influence the focus and type of exam practice; however, it will not change emphasis on teaching towards the exam. [-1-]
In "Has a High-Stakes Test Produced the Intended Changes?" Luxia Qi reports on a study of China's National Matriculation English Test (NMET). The study was done by interviewing participants, 8 test constructors, 3 English inspectors, and 10 Senior III secondary school teachers from both rural and city schools. Qi's conclusions are that the NMET has had little intended positive washback because linguistic knowledge and teaching only skills tested on the NMET are still emphasized in the classroom.
In the final chapter, Irit Ferman examined the effects of a national oral matriculation test, the English Foreign Language Oral Matriculation Test for 12th grade, in Israel. In this study, questionnaires, interviews (structured and non-structured), and document analyses were done. Ferman concludes that there were mixed results on the intended effects, some positive and some negative.
In conclusion, washback exists and has consequences, often unintended or negative or positive or both. This book is highly recommended for pre- and in-teachers interested in testing, especially pre-university English for Academic Purposes instructors, because it describes educational systems from which their students may have emerged and gives excellent descriptions of these situations. For administrators and test makers, this book would be an excellent read so that they are aware that high-stakes tests are not always representative of language use or enable the successful students to learn content in another language, but represent only that test's skills and language content.
Utah State University
© Copyright rests with authors. Please cite TESL-EJ appropriately.
Editor's Note:Dashed numbers in square brackets indicate the end of each page for purposes of citation..