Seven Rules for Social Research

Glenn Firebaugh is Distinguished Professor of Sociology and Demography at Pennsylvania State University. He is the author of "The New Geography of Global Income Inequality".

"Firebaugh has produced a lively and insightful contribution. He discusses thought-provoking examples and has created some truly excellent and innovative end-of-chapter exercises. Firebaugh maintains a refreshingly reassuring conversational style with the reader and communicates difficult concepts in a straightforward manner. [T]his is a book to be studied, rather than just read."--David Shemmings, Times Higher Education "This book is one of a kind and so excellent it will probably go into second and third editions... Seven Rules of Social Research brings important issues of research design to life: this is not a dry or abstract book... I think his book could be used to great advantage by three sorts of human geographers: those doing a thesis for their degree; those teaching introductory and more advanced courses on 'doing research'; and those seeking to refresh their own approach to research as full-time academics, research assistants or post-docs... The book is supremely clear (even when discussing some knotty issues) and, even at almost 300 pages, does not feel long or turgid."--Noel Castree, Progress in Human Geography "Seven Rules for Social Research is an excellent choice as a supplement for an undergraduate methods course as well as a complete text for the basic graduate methods course. Firebaugh's book would also serve as an invaluable supplement for undergraduate senior seminar students and graduate students alike who are looking for a straightforward reference for conducting both course based and independent research projects. This will be a book that students keep on their shelf throughout their academic career. I intend to keep it on mine."--Allison L. Vetter, Teaching Sociology "This ... book provides an interesting set of rules and thoughtful reflections on issues relevant to conducting social research. It also has useful reviews of newer statistical developments. Overall, this is an insightful book that should be of interest to graduate students and researchers in the social sciences."--Debra L. Oswald, PsycCRITIQUES "[T]his book is insightful and clearly written... Simply put, this is an excellent book that I highly recommend for graduate courses I methods and methods comprehensive exam reading lists. All of our graduate students would benefit from reading it."--Robert Andersen, Canadian Journal of Sociology "The book, and especially its chapter exercises, is most suitable for graduate students with an intermediate quantitative background. The author states that the book is 'to serve as a second methods textbook' the social sciences. This is a modest statement. As long as students pursue the types of research that Firebaugh discusses, this should be among the first books introduced in the course of research design."--Yasuyuki Motoyama, Journal of Sociology and Social Welfare "Seven Rules for Social Research, teaches social scientists how to get the most out of their technical skills and tools, providing a resource that fully describes the strategies and concepts no researcher or student of human behavior can do without... Seven Rules for Social Research is ideal for students and researchers who want to take their technical skills to new levels of precision and insight, and for instructors who want a textbook for a second methods course."--World Book Industry

• Preface xi Chapter 1: The First Rule There Should Be the Possibility of Surprise in Social Research 1 Selecting a Research Question 2 Researchable Questions 2 Interesting Questions 4 Selecting a Sample 18 Samples in Qualitative Studies 23 Is Meaningful Social Research Possible? 26 Summary 29 Student Exercises on Rule 1 31 Chapter 2: The Second Rule Look for Differences That Make a Difference, and Report Them 36 You Can't Explain a Variable with a Constant 37 Maximizing Variance to Find the Effect of a Cause 39 Size versus Statistical Significance 41 Comparing Effects Where There Is a Common Metric 42 Calibration: Converting Explanatory Variables to a Common Metric 44 Substantive Profiling: The Use of Telling Comparisons 46 Visual Presentation of Results 51 Policy Importance 53 Importance for Theory 54 Conclusion 56 Student Exercises on Rule 2 58 Chapter 3: The Third Rule Build Reality Checks into Your Research 64 Internal Reality Checks 65 Reality Checks on Data--Dubious Values and Incomplete Data 65 Reality Checks on Measures--Aim for Consistency in Conceptualization and Measurement 69 Reality Checks on Models--The Formal Equivalence Check 71 External Reality Checks: Validation with Other Data and Methods 76 Using Causal-Process Observations to Test Plausibility of Results 77 Using Ethnographic Data to Help Interpret Survey Results 79 Other Examples of Multiple-Method Research 81 Concluding Remark 82 Student Exercises on Rule 3 84 Chapter 4: The Fourth Rule Replicate Where Possible 90 Sources of Uncertainty in Social Research 91 Overview: From Population to Sample and Back to Population 93 Measurement Error as a Source of Uncertainty 100 Illustration: Two Methods for Estimating Global Poverty 101 Toward a Solution: Identical Analyses of Parallel Data Sets 105 Meta-analysis: Synthesizing Results Formally across Studies 106 Summary: Your Confidence Intervals Are Too Narrow 109 Student Exercises on Rule 4 111 Chapter 5: The Fifth Rule Compare Like with Like 120 Correlation and Causality 121 Types of Strategies for Comparing Like with Like 129 Matching versus Looking for Differences 130 The Standard Regression Method for Comparing Like with Like 131 Critique of the Standard Linear Regression Strategy 132 Comparing Like with Like Through Fixed-Effects Methods 134 First-Difference Models: Subtracting Out the Effects of Confounding Variables 134 Special Case: Growth-Rate Models 138 Sibling Models 140 Comparing Like with Like through Matching on Measured Variables 146 Exact Matching 146 Propensity-Score Method 147 Matching as a Preprocessing Strategy for Reducing Model Dependence 151 Comparing Like with Like through Naturally Occurring Random Assignment 152 Instrumental Variables: Matching through Partial Random Assignment 153 Matching Through Naturally Occurring Random Assignment to the Treatment Group 158 Comparison of Strategies for Comparing Like with Like 159 Conclusion 162 Student Exercises on Rule 5 165 Chapter 6: The Sixth Rule Use Panel Data to Study Individual Change and Repeated Cross-section Data to Study Social Change 172 Analytic Differences between Panel and Repeated Cross-section Data 173 Three General Questions about Change 175 Changing-Effect Models, Part 1: Two Points in Time 176 Changing-Effect Models, Part 2: Multilevel Models with Time as the Context 182 What We Want to Know 183 The General Multilevel Model 183 Convergence Models 185 The Sign Test for Convergence: Comparing Your fs and ds 186 Convergence Model versus Changing-Effect Model 191 Bridging Individual and Social Change: Estimating Cohort Replacement Effects 195 An Accounting Scheme for Social Change 197 Linear Decomposition Method 198 Summary 201 Student Exercises on Rule 6 203 Chapter 7: The Seventh Rule Let Method Be the Servant, Not the Master 207 Obsession with Regression 209 Naturally Occurring Random Assignment, Again 209 Decomposition Work in the Social Sciences 218 Decomposition of Variance and Inequality 220 Decomposition of Segregation Indexes 222 The Effects of Social Context 226 Context Effects as Objects of Study 227 Context Effects as Nuisance 230 Critical Tests in Social Research 231 Conclusion 235 Student Exercises on Rule 7 236 References 241 Index 253