Appropriate Research Methods

8. The Essentials

Essentials of Methodological Approaches

This digital anthology discusses ways to improve both the quantitative and the qualitative approaches to behavioral and social science research. Whereas natural scientists (such as physiologists or chemists) and clinicians measure with varying degrees of precision seemingly more tangible phenomena (e.g. blood pressure, urodynamic flow rates, or cholesterol levels) social and behavioral scientists have a more difficult task---to capture phenomena which are intrinsically elusory (some philosophers would term these “incorporeal”).  They may include the measurement of largely emotional states, like anxiety, depression, quality of life, happiness, and grief.

Measuring these is akin to measuring clouds---we can observe them and know that they can have important consequences. How to actually operationalize and measure these elusory phenomena is discussed by Keith Widaman (Professor, University of California, Davis) in the Psychosocial Variables* chapter. *Note: this chapter is still in development and will be available soon.

In the chapter on Sample Surveys Sarah Nusser (Director of Survey Research at Iowa State University) discusses the key ingredients of a well designed and rigorously conducted social survey. Survey research is the bread and butter of much social science, but much of this work falls far behind minimal standards required for quality science. To take a few examples: the training and monitoring of field interviewers is vitally important to the production of valid and reliable research results. Indeed, interviewer variability, especially when gathering social and behavioral data, is clearly the Achilles heel of much survey research: variability in measured outcomes may be a function of variability between the data gatherers.

Based on extensive field experience with national surveys in the UK, Steve Woodland (Office for National Statistics, UK) discusses the “dos and don’ts” of interviewer training and some well-tested techniques that are likely to yield quality data in the chapter Social Survey Data Collection. This is a deliberately practical contribution because the devil besetting much social and behavioral research often lies in the practical details of implementation.  It is sometimes possible to address emerging issues in health care not by collecting new studies, but by using and creatively integrating existing data sets (often collected for other purposes). In the Administrative Data Systems chapter Vince Mor (Chair, Department of Community Health and Professor of Medical Research at Brown University) discusses and illustrates the potential (and limitations) of existing data for health care research.