
Punch, Keith F. Introduction to social research : quantitative and qualitative approaches. 2nd.ed. London: Sage, 2005. 320 p. ISBN 0761944176. £45.95Books on research methods are frequent in social science. Still, it is not easy to find the book that would suit your students and still cover most of the issues you would like to include in a course on research methodology. You often find that you would need several books and only can use parts of them. Punch has come close to solving this dilemma in this second edition of his introduction to social research. The book covers both qualitative and quantitative methods as well as preempirical stages, ethical issues and research writing. This comprehensive approach, by necessity, results in many sections are short and some research approaches are mentioned briefly or not at all. However, the author manages to avoid some common problems, like going into technical details with basic statistics and not mentioning the problems with many interconnected variables, or describing the differences between quantitative and qualitative approaches in a crude or prejudiced way. The book is intended for upperlevel undergraduate and beginning graduate students in different areas of social science. Thus, you would probably want to add examples and details from your own field. The book starts by emphasizing the importance of research questions and then discuss the problem of going from research questions to data, quantitative as well as qualitative data. The section on quantitative data is rather comprehensive and includes research design, data collection and measurement issues and analysis in separate chapters. Each chapter ends with a review of concepts and recommended further reading. An impressive number of important issues are dealt with and the reader will be able to understand the basic principles in correlation, analysis of variance and covariance, multiple correlation and factor analysis and statistical inference. However, statistical details and formulae are not included. Basic knowledge of statistics and computation is supposed to be aquired by the students earlier or treated in supplementary readings. Unfortunately in many countries you cannot presuppose that basic education has already solved those problems. The section focusing on qualitative data is structured similarly to the quantitative sections, but is longer than the chapters on quantitative methods, probably because of the diversity within the field, which the author describes as one of his problems in the selection of contents. Grounded theory is here supplemented with sections on narratives, ethnomethodology, discourse analysis and semiotics as well as documentary and textual analysis. New sections in the second edition include action research and discourse analysis. Some of the sections, however, are quite brief and will serve only to introduce the approach. After reading this section you will have a good overview and an understanding facilitating choice of specific methods. There will probably be a need for more indepth material on some of the approaches depending on the scope and level of your course and also here some examples which fit your area of study. In a separate chapter the author discusses the possibilities of combining qualitative and quantitative approaches and how to evaluate the result. Generally, the basic qualities of both approaches are described without obvious prejudice and this is one of the advantages of this book. Scholars heavily focusing on postmodern approaches might not agree with this judgement! Brief guides for the use of the SPSS program for quantitative analysis and NUDIST, a program for analysis of qualitative data, are offered in appendices. It is regrettable that the proofreading or printing process somehow failed with this book. At least in my copy the reference list and index are messed up with some pages on advertising and in the process some of the references were lost. Hoping that this will be corrected soon I can recommend the book to be included in courses where a comprehensive and balanced introduction to qualitative and quantitative approaches is desirable. Prof. Lars Höglund
