Tom Mathews
Book Review 

The Research Manaual: Design and Statistics for Applied Linguistics. Evelyn Hatch and Anne Lazaraton. New York: Newbury House. 1991. 

Review published in Language Learning, 42, 429-432 (1992). 

        Hatch and Lazaration's Research Manual is a very useful and needed resource that will benefit students as well as professionals in applied linguistics. In many ways this book represents an improvement over Hatch and Farhady's Research Design and Statistics for Applied Linguistics. Indeed, many sections of the present work appear to have been taken almost verbatim from Hatch and Farhady. Nevertheless, this manual is considerably longer and more thorough than the other. The text is presented in three parts. Happily, the first part gives exclusive attention to defining research questions, describing data, and constructing experimental designs. These important areas are discussed in depth before any discussion of statistics. The second and third parts describe univariate statistics and statistics used in comparing groups, respectively. 
        The text of each chapter alternates several times between instruction and exercises. The placement of exercises interspersed with the text is confusing. Due to their constant interruption, reading the text, and more importantly, rereading it, is frustratingly difficult. Sometimes the exercises themselves are confusing. The answers may be overly obvious, so obvious that readers may overlook the answer and continue to search for another. The inclusion, in an appendix, of solutions to these exercises is very helpful. 
        Each chapter also ends with a series of activities which involve questions relating to actual studies done in the field of language learning. These studies are well documented and generally well described (albeit often very briefly), and they supply the reader with valuable background in applied linguistics along with their primary purpose of giving examples of statistical or experimental problems. 
        The Research Manual is written in a friendly and chatty style. This "reader freindly" tone is helpful and should to keep the novice in statistics reader from feeling intimidated. However, at times the style becomes verbose and is sometimes trite and even condescending. All of the topics presented are illustrated with examples which are, appropriately, taken from the field of language study and language learning. Students will find many of these examples understandable and meaningful. However, some of the examples of research from the literature are too complicated, given the point they are trying to illustrate. That is, if the authors had described properly the particulars of the research involved, they would have used more space doing so than in describing the use of the statistical procedure being illustrated. The result is that many complex and interesting examples are glossed over very quickly, oversimplified, or incorrectly represented. An example of this is the discussion of Keenan and Comrie's (1977) noun phrase accessibility hierarchy in Chpater 14. The authors, in an attempt to illustrate the Chi-square procedure, define the noun phrase accessibility hierarchy as categorizing relative clauses that follow subjects, objects, or the objects of a preposition, etc. They state that "you categorize the relative clauses by position" (p. 394). This gross oversimplification of the hierarchy is a disservice to students who may not be acquainted with it, and is totally unacceptable to those who are. 
        Another illustration tells when it might be appropriate to share a z score with a student (p. 199). The argument is unnecessary. The student would be just as satisfied if she were told that grades were based on percentages, and she would undoubtedly understand that explanation better. 
        There are a number of bothersome errors than can most likely be blamed on the copy editor, but which are, nevertheless, frustrating to the reader. One such error is referring the reader to appendix D, when it should be appendix E (p. 11). Others include sending the reader to the wrong page for an example in a set of exercises (p. 197), in another exercise, asking the reader to compute a T score and then labeling the column z score (p. 203), and retaining the words "align = center" at the end of a paragraph before a formula (p. 492). 
The authors seem to confuse random samples with representative samples. Although good experimental design would require both, they should not be equated. 
        This reviewer's major criticism is that the Research Manual is neither an introductory textbook nor a reference manual. If the book is intended to be used as a reference manual, then the intercalation of exercises into the text several times in each chapter is a hindrance to its use. Exercises would be better placed at the end of each chapter, or even at the end of the entire text, with solutions following. Nevertheless, the book does include a very complete content index, as well as an index of authors and several useful appendices, including a list of formulas and a list of important journals in applied linguistics. If, on the other hand, the Research Manual is intended as a coursebook for a student population, the assumptions regarding those students experience in the field of applied linguistics are certainly exaggerated. A great number of the examples given will be of little meaning to a beginning graduate student, and the brief way in which most are presented diminishes their illustrative value. 
        Overall the book is useful and represents an improvement over Hatch and Farhady work. It explains more procedures in more depth and makes no assumptions about the amount of experience readers may have with statistical concepts and terminology. 
        Lazaraton has written an excellent 75-page computer supplement that illustrates many of the statistical procedures described in the Research Manual. This supplement parallels the main text in its organization, and carefully describes how statistical procedures can be performed on mainframe computers using readily available software packages. 

 Thomas J. Mathews 
 Brigham Young University 


Hatch, E., & Farhady, H. (1982). Research design and statistics for applied linguists. Rowley, Massachusetts: Newbury House. 
Keenan, E. L. & Comrie, B. (1977). Noun phrase accessibility and universal grammar. Linguistic Inquiry, 8, 63-99. 
Lazaraton, A. (1991). A computer supplement to accompany Hatch and Lazaraton: The research manual: Design and statistics for applied linguistics. New York: Newbury House.