An Introduction to Scientific Research Methods in Geography


Daniel R. Montello & Paul C. Sutton

  • Citations
  • Add to My List
  • Text Size

  • Chapters
  • Front Matter
  • Back Matter
  • Subject Index
  • Dedication

    To Violet and Élan, for their love, support, and relentless pursuit of our improvement.


    View Copyright Page


    In this text, we provide a broad and integrative introduction to the conduct and interpretation of scientific research in geography. We cover both conceptual and technical aspects of research as they apply to all topical areas within geography, including human geography, physical geography, and geographic information science. We have attempted to produce a comprehensive text that discusses all parts of the research process, including scientific philosophy; basic research concepts; generating research ideas; quantitative and qualitative data collection; sampling and research design; data analysis, display, and interpretation; reliability and validity; using geographic information techniques in research; communicating research and using library resources; and ethical conduct in research. The text is intended for English-language undergraduate and graduate courses on research methods in geography and related disciplines, such as environmental studies. In addition, we hope it will have some value as a primer or reference work for students, faculty, and other research professionals who want an integrated yet concise introduction to scientific research methods in geography.

    In this text, we apply the research philosophy and methods of the social and natural sciences to research in geography. At the same time, we recognize and respect the diversity and heterogeneity of geography, and avoid simplistic conceptions of scientific geography as narrowly “positivistic,” “objective,” or “quantitative.” In this way, the text attempts to promote rigor and progressiveness in geography, helping to build bridges among the various subfields of geography and the other social and natural sciences, while avoiding some of the limiting meta-theoretical conflicts that have characterized geography in recent decades.

    Geography is a very heterogeneous discipline. Not every subdiscipline of geography accepts the intellectual preferences characteristic of scientists (discussed in Chapter 1) or practices systematic empirical research to the same degree. In particular, geography also includes humanities, engineering, and craft approaches that we make no attempt to cover thoroughly in this text. But make no mistake—our focus on a scientific approach still leaves quite a bit of room for conceptual and methodological pluralism in geography. For example, we make it clear that “subjective” mental constructs, such as beliefs and attitudes, can be validly measured and studied scientifically; so can many of the theoretical ideas and concepts that stem from various critiques of strict positivism in geography, such as unequal power relationships and constructed worldviews. At the same time, we do not naively treat phenomena such as symbolic meaning, so important to human experience and activity, as if they were no different whatsoever from temperature or chemical composition. However, we believe the differences can at least partially be conceptualized within a common scientific framework, thereby allowing the study of diverse geographic subject matters to be approached with a largely common set of intellectual and analytic tools.

    Overview of Chapters

    We have organized the chapters of this text to correspond roughly to the order of tasks one would encounter while actually carrying out research. However, we have written the chapters to be largely self-contained, so they can to a degree be read in different orders. However, we do recommend that, if you are a novice to research methodology, you read the entire book before embarking on research. Each chapter starts with a short list of learning objectives, a set of questions that serves to briefly summarize and preview the chapter's major topics and to stimulate your thought processes by getting you to begin actively pondering what you are about to read. Similarly, each chapter ends with a set of review questions designed to remind you of important general ideas within the chapter and help you consolidate what you have read. Important terms—we call them key terms—are in boldface throughout the text; each chapter's terms are listed and defined in glossaries at the end of the chapters. The end of each chapter also contains a bibliography that lists books, articles, and other resources that have shaped our thinking about research methods and deserve to be recommended as further sources for the interested reader.

    Chapter 1 is entitled Introduction: A Scientific Approach to Geography. It introduces scientific research and methodology with a discussion of an example research project. We discuss problems with the conduct and interpretation of this project; we intend this to provide a concrete example of the scope of issues that are part of scientific research methodology. We introduce the logic and philosophy of science, both by defining its approach and goals, and by considering “characteristic beliefs” held by most scientists as scientists. We also consider the very definite limits of science. Science is not just a search for truth or the application of rational thought, but an appreciation of the value of empirical observation coupled with rational thought as a way to increase our understanding of the world. As such, many of the most important questions that the humanities address cannot or should not be answered scientifically, although they may be informed by scientific results. Furthermore, scientists are humans, of course, and scientific institutions are human institutions. Human shortcomings are certainly a part of science. We finish this chapter by reviewing the history and conceptual systems of the discipline of geography. We note the distinction between regional and systematic approaches (our concern in this book is mostly with systematic geography), and go over the basic areas of human and physical geography, as well as research on geographic information techniques.

    Chapter 2 is Fundamental Research Concepts. In it, we introduce a series of basic concepts that are fundamental to the conduct and interpretation of scientific research. We classify these as “idea concepts” (theory, hypothesis, causality, model, construct) or “empirical concepts” (case, variable, measurement, data, measurement levels, continuity versus discreteness, accuracy and precision). We go over the key concept of scale in geographic research and thought, including phenomenon scale, analysis scale, and cartographic scale. We finish the chapter by going over some ways to generate research ideas, including a concise “plan of action” for generating good research questions and designing studies to address them.

    In Chapter 3, Data Collection in Geography: Overview, we discuss the difference between primary and secondary sources of data. We introduce the various types of data collection in geography that we cover in subsequent chapters, grouped into just five types. Finally, we provide an introduction to the distinction between “quantitative” and “qualitative” methods in research, rooting it in its historical development in different social-science and humanities disciplines. We interpret the distinction in terms of data collection and analysis methods, although we do not favor drawing the distinction in an excessive way. Some of the distinctions people occasionally make between the two are misleading or not useful, however, and we think they contribute to the unnecessary philosophical and methodological chasm one sometimes sees between scientific research in human and physical geography.

    Chapter 4 is entitled Physical Measurement. In it, we discuss a broad type of data collection that is very popular in geography—in fact, the most common way data are collected in most areas of physical geography. Physical measurements are collected by recording physical properties of the earth surface, the ocean, the atmosphere, and the biota. Geographers measure a great variety of physical properties, using a variety of tools and techniques. Physical properties include size and number, temperature, chemical makeup, moisture content, texture and hardness, the reflectance and transmissivity of electromagnetic energy (including optical light), air speed and pressure, and more. Physical measurements are often made via aerial and satellite remote sensing, which we discuss in Chapter 12. Human geographers also make physical measurements, often by observing the “physical traces” left behind by human behavior or activity. For example, geographers observe house design and agricultural patterns.

    In Chapter 5, entitled Behavioral Observations and Archives, we discuss two types of data collection that human geographers often use. We place these two together mostly for convenience; they are not that closely related, but neither requires an entire separate chapter. Behavioral observation is the systematic observation of people's actions. Archives are records of events or characteristics that were generally collected for purposes other than scientific research (financial records or crime statistics, for instance). Behavioral observation and archives do share the fact that as measures of human activities, both are often relatively “nonreactive” as compared to the explicit reports of the next chapter. That is, they often do not require the people being studied to realize that they are being studied, a realization that can change people's actions. Finally, both types of data collection often produce “records” that do not serve directly as data but must be coded (classified) to serve as scientific data. We discuss how to do coding.

    Chapter 6 is entitled Explicit Reports: Surveys, Interviews, and Tests. In this chapter, we cover one of the most flexible ways of collecting data in human geography. Explicit reports are written surveys (questionnaires) and oral interviews that assess beliefs, attitudes, activities, and demographic characteristics. They also include tests, which assess factual knowledge. Explicit reports are usually verbal but can be designed in a number of other formats as well. They may be administered in a variety of ways, via a variety of administration media, including the Internet. We discuss ways to structure report items, especially the distinction between closed-ended and open-ended items. We give specific instructions for designing different types of closed-ended items. We also discuss ways to generate items—how do you create questions for a survey, for instance? We also cover secondary report data, particularly the very important secondary source of U.S. census data. We conclude with some interesting, more conceptual, issues about the limits of explicit-report measures, such as the fallibility of memory, the impenetrability of certain aspects of mental activity, and the imperfect relationship of mind to behavior.

    In Chapter 7, Experimental and Nonexperimental Research Designs, we discuss the topic of research designs—the structure of manipulations/comparisons in a study that determines which questions can be asked of data, and with what degree of validity they can be answered. We extensively consider the distinction between true experiments and nonexperimental studies, as this is such a fundamental distinction with respect to the validity of causal conclusions in research. We discuss three ways of increasing the validity of causal conclusions in research: physical, assignment, and statistical control. Assignment control is the manipulation of variables, which is at the heart of the experimental/nonexperimental distinction. In this light, we introduce the concepts of control groups, random assignment, independent and dependent variables, and quasi-experiments. We explain the distinction between lab and field research. We present some specific research designs, both between-case and within-case. We present developmental, or temporal-change, designs. We also discuss single-case and multiple-case designs. We finish with a section on computational modeling, an approach to research design that is also a type of data collection and method of analysis.

    Chapter 8 is entitled Sampling. We define a population as the entire set of entities you wish to generalize to; a sample is any incomplete subset of that. What constitutes a population and what constitutes a sample therefore depends on your research goal—what you intend to generalize to in a particular situation. We make it clear that obtaining entire populations would always lead to more certain, and therefore better, research conclusions. In most basic-science research, however, the population is a hypothetical entity that cannot be obtained in entirety, if for no other reason than that it includes entities that no longer exist and entities that do not exist yet. We also explain that researchers do not just sample cases (people, rivers, cities, soil regions, and so on), but virtually all other components of research (such as times and places of measurement, survey questions, sensors) as well. We introduce the important concept of the sampling frame, which is the set of entities from which samples are actually drawn. The sampling frame is usually a subset of the population to which one wishes to generalize. We describe basic sampling procedures, the way sampling frames are chosen from populations, and then the way individual entities are chosen from sampling frames. We organize our description of procedures in terms of nonprobability and probability sampling. Although nonprobability sampling technically does not allow the derivation of probabilities to be used in inferential statistics, it is nonetheless quite commonly done in scientific research. We go over various specific types of probability sampling, including simple random, stratified, and cluster sampling. We discuss the implications of particular sampling frames and procedures. We also consider the possible effects of nonparticipation and volunteer bias. Then we go over some of the unique implications of sampling entities distributed over space and time. Finally, we consider the issue of sample size, including recommendations for determining adequate sample size and an introduction to formal power analysis.

    Chapter 9 covers the topic of Statistical Data Analysis. We explain why data analysis in geography is typically statistical (probabilistic) in nature—primarily because it is conceptualized as resulting from a sampling procedure and measured with error, but also because of the complex multivariate nature of most geographic phenomena. We describe the two components of most data analysis: (a) describing properties of data sets and (b) inferring properties of population data from properties of sample data. We introduce basic statistical concepts and focus on giving a clear explication of basic statistical logic. In particular, we carefully present the conceptual basis for hypothesis (significance) testing. However, we do not present a detailed tutorial on statistical analysis in this chapter; other courses exist to cover this important and rich topic, and readers should take at least a couple of them. We finish by reviewing some of the special properties of geographic data that arise from their spatiality. These “spatial” properties often lead to some interesting consequences for data analysis in geography, either because the properties are the subject of geographic theories or because the properties strongly influence the validity with which otherwise nonspatial data are interpreted.

    Chapter 10 covers Data Display: Tables, Graphs, Maps, Visualizations. We discuss the purposes of data displays, including the initial examination of data, the interpretation of their meaning, and the communication of data and their meaning to other people. All of these purposes boil down to effective communication as the guiding principle in designing and using data displays. We go over several types of displays, including tables, graphs, maps, and computer visualizations. We discuss maps relatively briefly; mapping is so important that geographers should take at least one separate course on it. We provide advice on choosing the right display type and designing it so it is effective at its purpose. We finish by briefly discussing emerging trends and technologies in visualization, including animations, sonifications, spatializations, and virtual reality.

    Chapter 11 is entitled Reliability and Validity. In it, we discuss the two fundamental concepts of reliability and validity that are germane to all scientific research. Reliability is the “repeatability” of scores or measured values of variables. High reliability occurs when you measure something twice and get the same value each time, assuming the thing you are measuring is actually the same at the time of the two measurements. Low reliability is caused by random errors of measurement and has a variety of detrimental effects on research outcomes. We go over techniques for assessing and increasing reliability. Validity is the “truth-value” of research results and interpretation. Given that research is largely an attempt to increase the truthfulness of our understanding of the world, validity is a core concern for researchers. Following an influential typology, we discuss four classes of validity: internal, external, construct, and statistical conclusion. We finish by discussing certain special classes of validity threats that arise because researchers are human, and in human geography, their research subjects are human as well. These “participant and researcher artifacts” can produce biased and distorted data that reflect various expectancies or beliefs people have about research situations, or about classes of people being tested or doing the testing.

    Chapter 12 concerns Geographic Information Techniques in Research. We describe geographic information and define geographic information techniques to include geospatial analysis, cartography, remote sensing, and computer geographic information systems (GISs). We cover the first two in Chapters 9 and 10, so this chapter focuses more on the fundamental nature of geographic information and research applications of the last two techniques, remote sensing and GIS. The data collected via remote sensing qualify as physical measurement, covered in Chapter 4; in this chapter, we consider aspects of physical measurement especially relevant to aerial and satellite sensing. We also summarize some characteristics of our planet earth and the representation of its surface in globes and digital databases. We explain some basics of coordinate systems and Global Positioning Systems (GPSs). We finish by overviewing basic GIS operations.

    Chapter 13 is about Scientific Communication in Geography. In it, we discuss both formal and informal scientific communication in geographic research, both to other geographers and to lay audiences. We provide a list of forms of scientific communication, including journal articles, books, grant applications, and oral presentations. We describe the “peer review system” for academic publishing and grant applications. We give specific instructions for how to put together an empirical research paper, including its parts, general structuring, and so on. We cover some aspects of writing style, although we recognize that geography has neither a single accepted style nor a universally accepted style manual. We also discuss aspects of giving effective oral presentations. We then turn to a discussion of the most important storehouse of scientific communications—the library, whether “brick-and-mortar” or electronic. We include detailed advice on how to perform scientific literature searches.

    Our final chapter, Chapter 14, is a discussion of Ethics in Scientific Research. Ethics is the study of moral or proper action. Research ethics involves a variety of rights and responsibilities, including those of the researcher, the people who work with the researcher (including human research subjects), and society at large. In fact, many people now accept that ethical concerns extend to the animals, plants, places, and cultures that play a role in the geographer's research. We go over guidelines for the ethical treatment of places encountered on research field trips. We also cover specific rules and guidelines for the ethical treatment of human and animal research subjects, including rules that geographers who work with human or animal subjects must follow, typically as part of the requirements of “Institutional Review Boards.”


    We would like to thank everyone who gave us input and advice on this book. Melanie V. Gray went above and beyond the call of collegiality and friendship in this respect. She read all the chapters, providing many suggestions as to content, grammar, and writing style. Most of the faculty members in the Geography Department at UCSB provided some direct input, and the rest provided indirect input. Oliver Chadwick, Sara Fabrikant, Catherine Gautier, Michael Goodchild, Phaedon Kyriakidis, Joel Michaelsen, Dar Roberts, Dave Siegel, Stuart Sweeney, Waldo Tobler, and Libe Washburn deserve specific mention. We thank several of our students over the years for the sounding board they provided and feedback that helped shape our ideas; this feedback often came in the form of questions like “What in the world are you talking about?” We also thank the many teachers over the years who inspired and informed us, especially about logical, statistical, and methodological issues. We appreciate the people who reviewed earlier versions of portions of this text and provided helpful comments, including Donald Friend, Kathy Graham, Juliana Maantay, Rod McCrae, and several anonymous reviewers. We are very grateful to Sage Publications for the opportunity they have provided to write this book and have it published. Special thanks go to Robert Rojek at Sage, who solicited the proposal for the book and championed it through to fruition. Finally, we wish to express our admiration and appreciation for Reg Golledge. Not only does he continue to teach us much about geographic research, but it was he who “admitted” both of us to UCSB back in 1992, eventually making this book and our friendship possible.

  • About the Authors

    Daniel R. Montello is professor of Geography and affiliated professor of Psychology at the University of California, Santa Barbara (UCSB), where he has been since 1992. Before that, he was a visiting assistant professor at North Dakota State University in Fargo, and a postdoctoral fellow at the University of Minnesota in Minneapolis. Dan received a Ph.D. in Psychology (Environmental Psychology area) in 1988 from Arizona State University in Tempe and a B.A. in Psychology in 1981 from the Johns Hopkins University in Baltimore. He is a member of the Association of American Geographers, the Psychonomic Society, and Sigma Xi Scientific Honor Society. He has published widely in the areas of spatial and geographic perception, cognition, and behavior; cognitive issues in cartography and GIS; and environmental psychology and behavioral geography. Dan has taught courses in a great variety of topics, including research methods, introductory human geography, statistical data analysis, behavioral geography, environmental perception and cognition, cognitive issues in geographic information science, regional geography of the United States, introductory psychology, child and lifespan developmental psychology, cognitive development, perception, environmental psychology, and cognitive science. He lives in Santa Barbara, California.

    Paul C. Sutton is associate professor of Geography at the University of Denver. He received his Ph.D. in Geography from the University of California, Santa Barbara, in 1999. His teaching and research interests are primarily in the domain of the human-environment problematic. Much of his work focuses on developing methods for improving our ability to estimate and map the human population using nighttime satellite imagery. Future work is focusing on using remotely sensed imagery and ground truthed sampling to map and estimate other socioeconomic variables such as impervious surface area, ecosystem services, income, energy consumption, and CO2 emissions. These areas of research are an amalgam of work in remote sensing, geographic information science, and statistics. Paul has published in many journals, including Nature; Remote Sensing of Environment; Photogrammetric Engineering and Remote Sensing; Population and Environment; Geocarto International; and Computers, Environment, and Urban Systems. Prior to his life in academia, Paul worked as an engineer at the Santa Barbara Research Center and as a high school science teacher at the Anacapa School in Santa Barbara. He lives with his wife, Élan, and son, Paris, in Conifer, Colorado.

    • Loading...
Back to Top

Copy and paste the following HTML into your website