Survey Research

by

Jorge Delva, Debora S. Tauiliili

• LAST REVIEWED: 30 July 2014
• LAST MODIFIED: 30 July 2014
• DOI: 10.1093/obo/9780195389678-0182

Introduction

In its simplest term, a survey refers to the administration of questions to obtain information about people’s behaviors, attitudes, and beliefs about a topic. A survey may include a single question by asking, for example, if respondents approve of child spanking, or if they ever smoked cigarettes; or a survey may consist of a battery (a compilation) of questions to assess, measure, or determine, for example, whether the respondent has ever experienced depressive symptoms. When more than one question is asked, the set of questions is often referred to as an instrument or questionnaire. Questionnaires may include stand-alone questions and multiple questions about the same topic and constructs. Examples of stand-alone questions include asking someone to disclose his or her age or to describe his or her attitude toward abortion. Questionnaires may also include a measure, (a set of related questions that, when combined, may better assess the intended construct). Thousands of measures exist, and most can be found on the Internet. A survey can be administered once or repeatedly, to a handful of individuals or to hundreds of thousands, and it can be administered by mail, in person, by phone, email, Internet, or through mobile devices. A considerable amount of survey research in social work, and in several social science disciplines and allied fields (e.g., education and public health) is devoted to measuring a construct with a high degree of validity and reliability. Validity refers to ensuring that the question(s) do measure the construct they are purported to measure, and reliability refers to measuring the construct consistently. The authors’ experience is that most education about survey research in social work and the social sciences and allied disciplines focuses on measurement issues; there is less concern with best practices to draw samples that are representative of the larger population. The analytic methods that are needed to properly analyze these data also tend to be neglected. To address these gaps, in this annotated bibliography we review books, manuscripts, reports, and Internet sites on the administration and analyses of surveys that rely on samples drawn to be generalizable to the larger population. From time to time, we also provide examples of surveys conducted with samples that were not drawn to be representative of the larger populations to illustrate some aspect of survey research, data collection, or analytic methods, with some of the most nascent approaches such as those studies using real-time data capture and social media.

Textbooks

Healey 2002, Henry 1990, Kalton 1983, and Rubin and Babbie 2014 provide excellent introductions to the various survey sampling and survey (or questionnaire) administration methods that survey researchers can use without relying on advanced statistical terms. Stone, et al. 2007 and Thyer 2010 include chapters on various sampling methods, but Lavrakas 2008 provides the most comprehensive coverage of survey methods and analytic approaches.

• Healey, J. F. 2002. Statistics: A tool for social research. 2d ed. Belmont, CA: Wadsworth/Thomson Learning.

  This book has a chapter that provides a clear introductory description of the concepts of sampling and sampling distribution for those interested in conducting survey research by drawing samples that can be representative of the general population.

• Henry, G. T. 1990. Practical sampling. Newbury Park, CA: SAGE.

  This book provides an easy-to-read description of various sampling strategies that can help the survey researcher draw representative samples, leading to the most precise estimate. It includes numerous examples of various sampling strategies. This book does not require advanced knowledge of statistics for the reader to get a general overview of representative samples.

• Kalton, G. 1983. Introduction to survey sampling. Newbury, CA: SAGE.

  This book also provides an easy-to-read description of various sampling strategies (proportionate and disproportionate stratification, cluster and multistage sampling, probability proportional to size sampling, sampling frames) as well as a description of nonresponse, sample size calculations, and survey analysis. This book will be of interest to those who desire to have a general knowledge of the ways survey research may be conducted with sampling methods that can generalize the findings of the survey to the larger population.

• Lavrakas, P. J. ed. 2008. Encyclopedia of survey research methods. Thousand Oaks, CA: SAGE.

  This encyclopedia, over one thousand pages with over 320 contributors, provides what is perhaps the most comprehensive and detailed coverage of survey research methods, covering a considerably wide range of methodological and statistical topics. Available online by subscription.

• Rubin, A., and E. R. Babbie. 2014. Research methods for social work. Belmont, CA: Brooks/Cole.

  This book includes chapters, with an excellent introduction to sampling methods with easy-to-follow examples. A chapter called “Survey Research” essentially refers to survey (or questionnaire) administration methods (interviews, self-administered Internet, telephone).

• Stone, A. A., S. Shiffman, A. A. Atienza, and L. Nebeling, eds. 2007. The science of real-time data capture. New York: Oxford Univ. Press.

  This edited book provides a comprehensive discussion and review of real-time data capture methods, focusing on one that is called ecological momentary assessment (EMA). This is an excellent book for anyone interested in detailed methodological and some statistical considerations of EMA.

• Thyer, B. 2010. The handbook of social work research methods. 2d ed. Thousand Oaks, CA: SAGE.

  This edited book includes a chapter, “Probability and Sampling” (pp. 37–50), that provides a clear description of the concept of probability and different types of sampling procedures (random, stratified, systematic), with easy-to-understand examples. Those with introductory knowledge of statistics will be able to more easily understand the basic statistical concepts that are included as part of the examples. However, those without statistical knowledge will still be able to gain an understanding of different sampling procedures.