Assessing Language Ability in the Classroom, 2nd ed.
Cohen broadens the meaning of the term testing from its usual restricted sense to include "the collection of any data that can be used to assess the language abilities of respondents" (p. 11). This broader definition, then, allows scope for discussion not only of traditional topics such as different types of validity and item analysis, but also for a review of currently popular trends such as self-assessment and portfolio assessment. The result is a volume that is accessible to novices and much more in the mainstream of the field than the original edition, which, as Cohen himself observes, was written "around the edges of other testing books—, i.e., covering topics not covered in much depth elsewhere" (p. 358).
The volume consists of eight chapters in addition to an introduction and conclusion. In keeping with the focus on the needs of practitioners, chapters are organized around key practical questions or topics, e.g. "What is the purpose of assessment?" and "How are quizzes to be used?" For those whose concerns include research, each chapter contains research notes, separated from the main text, which summarize evidence about the topics discussed and provide sources for further investigation. One valuable feature of the research notes is that Cohen draws upon studies of learners of a variety of target languages and a wide range of educational [-1-] contexts, thus increasing the usefulness of the book as a text in foreign language education as well as TESL. Finally, each chapter concludes with a series of discussion questions. For the most part, the questions are well conceived and suggest a variety of appropriate activities for students and teachers who are novices in test design and implementation.
Space limitations preclude a full discussion of all the topics Cohen covers. However, a brief examination of his treatment of self-assessment and portfolio assessment of writing will serve to illustrate how Cohen has moved into the mainstream with this new edition.
Cohen devotes considerable attention to self-assessment and includes sample instruments for listening, speaking, reading, and writing, as well as an instrument designed to assess the small-group participation of elementary school students. Cohen offers a summary of the major studies that support self-assessment, as well as those that challenge its validity. He concludes cautiously that "careful self-assessment by students in a classroom could be one of the means for multiple assessment, while the use of self-assessment for large-scale assessment might be more problematic…" (p. 199).
In recent years, school districts in many states have adopted portfolio assessment, and proponents have argued its case with considerable zeal (see, e.g., DeFina, 1992). In contrast, Cohen pays careful attention to the limitations of portfolio assessment, while at the same time recognizing the value of portfolio work in building a sense of community among peers and in increasing communication between teacher and student. He notes that portfolios require great amounts of teacher time and that grades for portfolios tend to cluster together, thus failing to provide adequate discrimination. Cohen also cites at length Hamp-Lyons and Condon's (1993) study questioning a number of basic assumptions about portfolios, including the belief that portfolio evaluation necessarily means that variation in quality within each portfolio is taken into account or that multiple genres are always represented in a portfolio. Despite his attention to the problems inherent in portfolios, however, Cohen is by no means wholly negative about their potential, concluding that the portfolio movement is "potentially beneficial...to the field of language assessment since the emphasis is on convergent and repeated measures over time rather than on single measures at one point in time" (p. 361).
To sum up, Cohen offers a clear treatment of most of the topics usually covered in an introductory language testing course in a text that is highly accessible to classroom teachers and students in teacher education programs. Cohen's emphasis on the sociocultural context and on the use of multiple means of assessment, as well as his judicious treatment of currently popular approaches, is particularly valuable. The numerous examples of assessment [-2-] instruments for a variety of age and proficiency levels also increase the usefulness of the book for practitioners. Classroom teachers will most likely come away from the text with a much clearer idea of how to construct measures that reflect and reinforce current communicative approaches to language teaching. The volume's main limitation, one which may make it unsuitable for those whose interests are primarily in research or whose responsibility for assessment extends beyond the individual classroom, lies in its scant treatment of the statistical dimensions of testing. As Cohen himself rightly notes in his very brief discussion of item analysis (pp. 101-104), readers who need to understand the basis of, for example, test reliability formulas, need to consult one of the more quantitatively oriented texts such as Bachman (1990), Hatch and Lazaraton (1991), or Henning (1987).
Bachman, L. (1990). Fundamental considerations in language testing. Oxford: Oxford University Press.
De Fina, A. A. (1992). Portfolio assessment: Getting started. New York: Scholastic Professional Books.
Hamp-Lyons, L., & Condon, W. C. (1993). Questioning assumptions about portfolio-based evaluation. College Composition and Communication, 44, 176-190.
Hatch, E., & Lazaraton, E. (1991). The research manual: Design and statistics for applied linguistics_. New York: Newbury/HarperCollins.
Henning, G. (1987). A guide to language testing. New York: Newbury.
The University of Texas at San Antonio
© Copyright rests with authors. Please cite TESL-EJ appropriately.
Editor's Note: Dashed numbers in square brackets indicate the end of each page in the paginated ASCII version of this article, which is the definitive edition. Please use these page numbers when citing this work.