- Shopping Bag ( 0 items )
"We all know that the actual process of empirical research is a messy, complicated business that at best only approximates the models we impart to students. Research Confidential pulls back the curtain on this process, laying bare the sordid details of the research process, but doing so in a way that respects the ideals of social research and that provides useful lessons for young scholars. It should be required reading for our research methods courses."
---Michael X. Delli Carpini, Dean, Annenberg School for Communication, University of Pennsylvania
"In this impressive volume, some of the brightest young lights in social research have taken us backstage to share what they learned from their innovative projects. Besides providing a wealth of help with methodological concerns, the book includes theoretical and career issues to consider when doing research. Anyone doing research should benefit from reading it."
---Caroline Hodges Persell, Professor of Sociology, New York University
"Research Confidential complements existing methods literature by providing refreshingly honest accounts of key challenges and decision forks-in-the-research-road. Each chapter enlightens and entertains."
---Kirsten Foot, Associate Professor of Communication, University of Washington
"A must-read for researchers embarking on new projects. Rather than the abstract descriptions of most methods textbooks, this volume provides rich accounts of the firsthand experiences of actual researchers. An invaluable resource of practical advice. Critically, it will make new researchers aware of the actual challenges that they are likely to face in their work."
---Christopher Winship, editor of Sociological Methods and Research and Professor of Sociology, Harvard University
This collection of essays aims to fill a notable gap in the existing literature on research methods in the social sciences. While the methods literature is extensive, rarely do authors discuss the practical issues and challenges they routinely confront in the course of their research projects. As a result, editor Eszter Hargittai argues, each new cohort is forced to reinvent the wheel, making mistakes that previous generations have already confronted and resolved. Research Confidential seeks to address this failing by supplying new researchers with the kind of detailed practical information that can make or break a given project. Written in an informal, accessible, and engaging manner by a group of prominent young scholars, many of whom are involved in groundbreaking research in online contexts, this collection promises to be a valuable tool for graduate students and educators across the social sciences.
Eszter Hargittai is Associate Professor of Communication Studies at Northwestern University and Fellow at the Berkman Center for Internet & Society at Harvard University.
Cover art courtesy of Dustin Gerard
EMMA K. ADAM, LEAH D. DOANE, AND KATHRYN MENDELSOHN
Reality is the leading cause of stress amongst those in touch with it. -Lily Tomlin
One of the symptoms of an approaching nervous breakdown is the belief that one's work is terribly important. -Bertrand Russell
Few social scientists concern themselves with the viscosity of saliva on a daily basis. Our research team, however, has become expert on such matters. Even small samples of saliva offer social scientists insights into the minds and bodies of individuals that surveys alone cannot provide. With just a few drops, we can measure a variety of hormones and other biological markers that provide clues to the functioning of physiological systems, including those serving functions related to the allocation of energy and attention, growth, sexual behavior, immune functioning, and responses to stress. A new generation of social scientists (in part through pressure from funding agencies) is realizing the importance of becoming truly multi-method: in addition to combining qualitative andquantitative methods, they are now also incorporating biological measurement-often referred to as "biomarkers"-into their research.
Salivary biomarkers are just one of a wide range of biomarker measurement approaches, but the nonintrusive nature of salivary sampling makes it a particularly popular point of entry into this new genre of research. The apparent ease with which saliva can be collected has, however, also meant that many investigators adopt this approach without having a realistic sense of how complicated measuring and interpreting salivary biomarkers can be. This chapter describes some of the challenges and benefits of incorporating salivary sampling into social science research. Although it focuses on salivary cortisol in particular, it contains cautionary tales and lessons that can easily be generalized to other approaches to biological measurement.
Setting aside the not unimportant fact that many funding agencies are now encouraging social scientists to do so, why should social scientists want to incorporate biomarkers into their research? Perhaps because the constructs that social scientists already care about-social and policy contexts and the interpersonal interactions and experiences that take place within them-can have profound effects on biological processes. Alterations in biological processes, in turn, can have important influences on emotional and behavioral functioning as well as physical health (see fig. 1). Thus, physiological processes that used to be confined to the realm of biology and medicine are now understood to be embedded in social contexts, interacting with them dynamically, bidirectionally, and continually. A number of theoretical and empirical traditions incorporate this insight, including biopsychosocial (Engel 1980), biosocial (Booth, Carver, and Granger 2000), bioecological (Bronfenbrenner and Ceci 1994), life history (Worthman and Kuzara 2005; Ellis 2004) and early/fetal programming approaches (Barker 2004; Barker et al. 2002). Despite distinct theoretical and disciplinary orientations, the basic message is the same and is being heard now louder than ever: Biological and social/environmental processes are tightly and planfully intertwined-biological processes are in fact designed to change in response to changing experiences, with the aim of promoting flexibility, survival, and flourishing in the face of an ever-changing social-environmental landscape. As a result, examining the transactions between biology and social experience will enable us to better understand human behavior, development, and health.
In an effort to observe the ways in which experience and biology interact over moments, days, and years, our team now collects thousands of saliva samples per year from research participants, analyzing them in conjunction with questionnaires and diary reports of social experience gathered in naturalistic settings. This strategy allows us to identify how events and emotions occurring in the everyday lives of children, adolescents, and adults "get under the skin" to influence physiological processes. More specifically, we have spent countless hours preparing, collecting, and processing tiny vials of spit (see fig. 2) in order to identify how everyday life factors influence the stress hormone cortisol, and how differences in cortisol levels between individuals, and changes in cortisol levels within individuals over time, can in turn impact emotional and physical well-being. Although we have collected saliva from many different age groups, much of our recent research (and thus this chapter) focuses on youth aged 13 through 25 years of age-a population that has presented us with some unique challenges. By describing the trials and tribulations of implementing biological measures in our own research, we hope to provide tips and advice that will allow future researchers to avoid (or at least anticipate) the pitfalls of this type of research and to maximize its benefits. Before we can begin to convey these logistical and theoretical challenges, however, we need first to provide some background information about our hormone of interest, cortisol.
CORTISOL: BACKGROUND AND SIGNIFICANCE
Those readers with insomnia, or a bad case of television addiction, will be aware that late-night television ads refer to cortisol as "that nasty little hormone" and offer expensive (and non-FDA approved) "remedies" that promise to reduce your waistline and resolve multiple other "ills" by reducing your cortisol levels. It turns out, however, that the late-night television ads are largely misplaced in giving cortisol a bad rep. Cortisol is in fact essential for everyday survival, not only during times of stress, but also in low-stress situations. Cortisol is present in the bloodstream at all times and is referred to in this context as basal cortisol. Basal levels of cortisol play a central ongoing role in regulating metabolism and a variety of other physiological systems, including digestion, immune and inflammatory processes, and growth (Johnson et al. 1992). Basal cortisol levels change over the course of the day in what is called a diurnal rhythm-levels are typically highest in the morning upon awakening, increase approximately 50-60 percent in the first 30-40 minutes after waking (called the cortisol awakening response or CAR), then decline across the day to near-zero levels by bedtime (Kirschbaum and Hellhammer 1989, 1994; Pruessner et al. 1997). Cortisol levels also rise above these expected basal or baseline values when individuals encounter stressful events-a phenomenon called cortisol reactivity. This stress-related reactivity is what cortisol is best known for (and why it is often dubbed a "stress hormone," despite the fact that its basal regulatory role is equally essential). Stress-induced increases in cortisol have numerous effects on the body, including causing blood levels of glucose to increase in order to provide the individual with the energy needed to face the stressor at hand (Johnson et al. 1992). Thus, for the most part, cortisol is a necessary, functional hormone-our friend rather than our foe. It is only when cortisol is chronically or frequently elevated, or when it fails to elevate in situations when it is needed, that it is thought to be problematic and to contribute to disease states (McEwen 1998). Enter the social scientist, bent on discovering what types of situations or experiences cause this hormone to elevate and hoping to observe when, for whom, and under what circumstances abnormal or problematic functioning might emerge.
The first of these tasks-looking at what elevates cortisol-was originally (and is still frequently) undertaken in the lab context, often according to the following instructions: Round up some unsuspecting undergraduates looking for course credit, expose them to all sorts of stressful tasks, then make them spit a couple of times before and after the stressor. Next, find out what conditions are most effective at causing cortisol levels to spike, and which types of individuals are most likely to show these increases, et voilà: stressors and stress-reactive individuals will have been found.
While important, these lab-based studies nevertheless have several limitations. First of all, undergraduates are not necessarily like everyone else-they may, for example, be more competitive than average; also, racial and economic diversity is limited on some college campuses. Some researchers have attempted to address this limitation by bringing in individuals from the broader community to participate in these lab-based paradigms. But a second problem also emerged-many researchers attempting lab-based paradigms were baffled by the fact that participants' cortisol levels at the start of the task often tended to be high, with cortisol levels declining over the course of a session, rather than increasing as one would expect them to do in response to the stressors. A comparison of levels at the same time of day at home revealed the answer to the mystery-participants were showing "anticipatory" cortisol reactions. That is, they tended to be more frightened by the prospect of many lab tasks than by the tasks themselves. Put another way, their bodies were preparing them for a potential stressor in advance rather than waiting for them to encounter the actual challenge-and this of course made it hard to look at how cortisol levels responded to experiencing the lab tasks.
To solve this problem, researchers have identified certain standardized, highly stressful lab-based tasks that consistently manage to elevate stress hormone levels (see Dickerson and Kemeny 2004 for a review). In addition, modern-day laboratory-based cortisol researchers have outsmarted the problem through a variety of strategies, induding: implementing long baseline periods (multiple samples before the stressor), allowing time for anticipatory stress to wane, and making sure that home comparison samples are collected at the same time of the day so as to make it easier to quantify the degree of anticipatory stress reactivity present at the start of the session. Nonetheless, there are some things that lab-based paradigms cannot accomplish. First, they cannot identify the stressors that actually occur in people's everyday lives that manage to activate cortisol levels in naturalistic environments. Second, they cannot examine people's basal cortisol levels over the course of an entire day (unless you plan to lock participants up in a hotel or hospital room for a 24-hour period, which gets costly and may serve as a stressor in and of itself, thereby making it difficult to observe basal cortisol).
The lab-based tasks that are known to be effective in activating cortisol (making a speech and then performing backward arithmetic in front of a very stern-looking panel of judges) are not necessarily representative of situations that people encounter every day. Certainly, these tasks are designed to capture aspects of experience (e.g., fear of negative social evaluation) that do occur in our daily lives. But why approximate real-life experience when you can get out there in the world and measure it directly? One reason is that measuring cortisol in everyday environments is exceptionally complicated: it not only involves amazing feats of creative problem solving but also an array of gadgets and hours of painstaking work in order to be done right. Examining cortisol in relation to everyday life events in naturalistic settings can also lead to many unexpected (and sometimes amusing) turns of events, no matter how careful the planning. These efforts and amusing stories are rarely apparent in the final written research article, which is why we choose to share them here. Gathering spit, while at first distasteful to the squeamish social scientist, opens many new scientific doors, and the resulting information is worth all the investigator sweat and tears involved in implementing this novel "beyond self-report" method. Our discussion addresses a number of questions relating to the why, who, how, when, and what of salivary cortisol. That is, we examine why one would measure salivary cortisol, who should measure it and how, when it should be measured, and what to do with the data when they are obtained, in terms of analysis and interpretation.
THE WHY'S, WHO'S, HOW'S, WHEN'S, AND WHAT'S OF SALIVARY CORTISOL
Why Measure Salivary Cortisol?
Reading about the challenges of adding a biological measure such as cortisol to a social science study may prompt readers to wonder, why bother? We have already noted that federal funding agencies have recently begun strongly encouraging researchers to incorporate "biomarkers" into social science research. Pressure from funders is of course by no means sufficient reason to do so and should be resisted unless there is a compelling research question that adding biomarker(s) is likely to answer. Gathering biomarkers in the hopes of coming up with such a question eventually is not a compelling rationale for imposing an additional burden on research participants, or spending money on biomarker analysis. Furthermore, a post hoc (after-the-fact) analysis is rarely sensible, particularly in the case of cortisol, as the specific research question needs to guide the specific data collection strategy.
Our own research was guided by our interest in whether individual differences in basal profiles of cortisol are altered over time by the experience of life events or by the accumulation of life events over the years, with individuals who experienced a larger number of such events, or more severe events, over the years showing more dramatic changes in cortisol profiles. We were also interested in whether acute increases in negative mood or perceived stress predict acute increases in cortisol levels in naturalistic settings. Finally, we wanted to understand whether individual differences in basal profiles, or differences in how cortisol-reactive individuals are to everyday stressors, help to predict the likelihood of depression in subsequent months or years. In one study, we examine these questions longitudinally, over an eight-year period, in a diverse sample of approximately 200 youth transitioning from late adolescence (high school) to young adulthood.
The first of these questions-whether changes in basal cortisol correspond to changing life events-required us to choose a measurement protocol that would accurately define the diurnal cortisol rhythm. The second required us to measure multiple cortisol samples over the course of an entire day, along with simultaneously recorded diary reports of events and emotions, in the hope of capturing momentary cortisol increases associated with stressful situations in everyday life.
In general, research has moved beyond the stage where cortisol is considered simply a handy "indicator" of subjective levels of stress. The studies that are most compelling and most likely to receive funding are those that consider the way in which its interaction with social and contextual stressors may influence later developmental or health outcomes. Above all, the reason for adding a biomarker should be clearly known in advance, to help ensure that the data gathered are appropriate to answer the specific question at hand. This may seem an obvious point but it bears repeating since we know of a number of studies in which the investigators decided to simply "gather stress hormone (cortisol) data," only to be later disappointed when they learned that the measure of cortisol they obtained was not sufficient for the question they eventually decided to ask of it. Having a sensible question in advance obviously requires considerable familiarity with the theoretical literature about the ways in which environments and biological processes interact as well as with past research and theory on the particular biomarker of interest (in our case cortisol). Such familiarity requires either extensive reading or collaboration with someone who already has theoretical expertise and research experience in the relevant areas.
Excerpted from RESEARCH CONFIDENTIAL by ESZTER HARGITTAI Copyright © 2009 by University of Michigan . Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.