Survey Research: The Basics

Keith F. Punch
Sage Publications: London
0761947051 (pb); 0761947043 (hb)
£15.76 (pb); £42.25 (hb)

Order this book?

Front Cover This textbook is very much a 'how-to-do book'. It accompanies another textbook An Introduction to Social Research: Quantitative and Qualitative Methods also written by Keith Punch, published by Sage in 1998. The book consists of seven chapters and two appendices and is aimed at the student researcher and the new social researchers. The key purpose of the book is to teach small-scale quantitative methods, in other words, the questionnaire survey. "To do that, it takes a simple model of the quantitative-survey-relating- variables, dismantles it to show how it works, and illustrates it with a detailed example" (p.2). The focus of this book is on a particular kind of surveys, namely those which are: (1) small scale; (2) cross sectional; (3) based on individuals as the unit of analysis; and (4) using self- administered questionnaires. The book is a countering "the massive swing towards qualitative methods of the last 25 or so years' (p.13) since Punch feels that there is a risk that such a central tool as survey research can be "overlooked or downgraded in the way new generations of social science researchers are trained' (p.13).

The chapters are laid out clearly and they have reasonably useful headings, such as 'Elements of the Survey: Description' or 'The Survey Report'. There is a section called 'review concepts' at the end of each chapter, with some additional footnotes. Appendix 1 contains useful further readings on technical topics, such as 'sampling and sample size', 'validity' or 'statistical significance'. Appendix 2 is a worked-out example of a two-variable quantitative survey.

The index of a textbook needs to be perfect to help the inexperienced researcher find the necessary terms in the text, unfortunately this one is not. Someone looking for the entry 'open-ended question' will not find this entry under the heading 'open-ended' or 'question', but as a sub-entry under the heading 'questionnaire'. Which is rather odd for a book on survey research focusing mainly on questionnaire studies. Similarly 'pilot testing' will not be found under 'pilot' nor under 'testing', again it is indexed under 'questionnaire'. I could not find an entry in the index for several mentioned in the actual text terms such as 'Likert scale', 'content validity, or 'covering letter' or 'stepwise regression'.

I really do not like the author's name on the front cover in lower case letters only. A name needs to start with a capital, otherwise it does not look quite right. This is, of course, a very personal view.

There are some minor hiccups in this otherwise excellent textbook. For example, the recommendation that "parametric or continuous variable techniques" should not be used in analysing categorical variables, is followed by the throwaway line "Other techniques are required" (p. 56). There is no information for the reader what kind of techniques is appropriate or whether the author is going to return to this issue later on in the book. Also on page 93 there is a sentence about a sample size of 350 being sufficient, but it is not made clear to the reader whether this an educated guess based on years of experience, or whether it is based on a particular sample size calculation.

The book is easy to read and very useful for students conducting their first real research project for a dissertation or thesis. The book is not based on an ideal-type of questionnaire survey, which is well resourced, and generally operating under ideal conditions. Punch identifies the problem that graduate students will be daunted by the perfect flawless research process as presented in some of the research literature (p. 22).

The book is full of little bits of useful advice I find myself giving my students all the time. "'Writing as you go' is a particularly useful strategy when the survey is also a dissertation" (p.69) is something that my students always hear me tell them. Or another one we often argue about in our department is where should the literature review come in a dissertation or thesis. Punch puts it after the Objectives and research question and before the Methods chapter (Table 5.1 p.75), but he recognizes that it is "sometimes recommended that the literature review comes before the research questions" (p.79).

I would recommend this book to final year students and postgraduates doing their research project. I would also recommend teachers of research methods and supervisors of students to recommend it to their fledgling researchers.

Edwin van Teijlingen
University of Aberdeen and visiting professor at Kalamazoo College, Michigan (USA)