useful for establishing how different variables relate to one another conceptually. Some GIS software programs provide a flowchart tool as a means to develop your analysis approach and, once populated with data and analysis functions, to run your model. Even when the flowchart is simply drawn on paper, it can guide you in developing a systematic or holistic approach to a particular issue under study.
For instance, suppose that you are conducting a study about an individual’s attitudes about environmental issues. You may hypothesize that the nature of the community and its geographic location (e.g., proximity to unspoiled natural features, such as state or national parks or wilderness areas) influence community concern about environmental issues. Your conceptual framework would then incorporate the geographic variable of park and wilderness proximity into your model (figure 3.10). Using a GIS, you could map the locations of these natural features and overlay sociodemographic characteristics and residents’ levels of environmental concern to see if there are differences.
Figure 3.10 An example of a conceptual model or framework for environmental concern. Sociodemographic factors, such as age, race, gender, and income, might have some relationship to where individuals choose to live (dashed arrow) in addition to their environmental concern (solid arrow). Proximity to parks or wilderness is also expected to have an influence on environmental concern (dotted arrow).
Choose research methods
What are the factors to consider in choosing a research method? One should first consider the project goal and the sorts of data most appropriate in meeting it. Can you use existing data, or will you require new data (or both)? Should the data you use in your study be quantitative or qualitative (or both)? What will the boundaries of your study be? Research can be conducted at the local, regional, national, or global level or at any combination of the four levels. Will you be doing a descriptive study of one area or comparing multiple research locations? Are there one or several methods you might use in the collection and analysis of the data?
If time and money allow, it is sometimes beneficial to incorporate multiple research methods in studying your topic because a variety of methods adds a greater empirical angle to your study. One popular approach is triangulation, or cross-examination, which is simply studying the same phenomena using three different research methods. Triangulation gives the researcher greater choice in gathering information on the topic. When data collected using multiple methods all point to a similar result, confidence in your results is strengthened.
So you could approach your problem or issue using a variety of methods, such as surveys, key-informant interviews, external or participant observation, and a review of historic or archival data. You could incorporate a geographic component into some or all of these different methods.
AN ASIDE ON TRIANGULATION
Interestingly, the term triangulation is historically associated with geometry, not the social sciences. If you work primarily in the social sciences, you may be familiar with the term as it applies to using multiple sources or methods to arrive at a result or conclusion. However, if you do an Internet search on the term, you will find that the first several pages of resulting hits relate to mapping and geometry. The properties of triangles were well understood by the ancient Greeks and are the basis for many modern mapping and land surveying techniques. As we were writing this chapter, we realized that an unintentional relationship between the mapping and social sciences may have originated in the translation of this geometric mapping term into the social science lexicon. In mapping, triangulation refers to the process of calculating a distance to a location by knowing the length of one side of a triangle and the related angles. This is accomplished with an instrument known as a theodolite, or its modern equivalents, including the Global Positioning System or surveyor’s total station (figure 3.11).
Figure 3.11 A surveyor making assessments in the field. CandyBox Images/Shutterstock.com.
Operationalization: Measurement
As you consider the selection of your research methods, it is important to refer back to your conceptual framework. You need to decide exactly what variables you will measure, and how. This requires that you define each of the concepts that you are using in your study and its associated level of measurement. Operationalizing a variable requires you to explain how you are going to measure the concepts mentioned in your hypothesis. Typically, the variety of data collection approaches that you employ is documented in the methods section of your study. When your methods relate directly to the creation of a GIS dataset, you need to be sure to record the data creation procedures in a metadata file, which should be included with each individual data layer.
In the example of the relationship between pollution and poverty, you might choose to operationalize pollution as shown in figure 3.9 into three simple categories: low, medium, and high. In doing so, you would need to define exactly what is meant by these terms. Doing this would be an example of translating data from a simple numerical form into a perhaps more palatable and easily understood frame of low, medium, and high. Of course, you would have to explain the data that are used as part of your categorization process for low, medium, and high. For instance, is “low pollution” something that is measured as a number of pounds or gallons released, or by a medical definition such as the odds of getting sick from exposure (e.g., one in 1 million as low; one in one hundred thousand as medium; and one in ten thousand as high)? Of course, you could come up with a variety of definitions, each viewed as appropriate or inappropriate, depending, for example, on legal definitions, regulatory recommendations, expert opinions, or local experience.
Coming up with a true definition is not necessarily as important (or possible) as explaining your definition so others can understand and interpret your results. One advantage of the GIS is that if you choose to change your definition or examine different scenarios, it is relatively easy to rerun your analysis with the adjusted definitions to determine how the results change.
Collect and prepare the data
When using a GIS as a tool for your research, collecting and preparing the data for analysis can be significant to the research process. For any kind of research, data analysis and organization are time consuming; for GIS-based projects, you can expect that 75 to 80 percent of the time and effort you expend will be used in collecting, creating, or converting data to ensure that everything is ready for the analysis. In later chapters, we extend our discussion of additional types of data and their collection and creation in GIS formats. When we say “creating data,” we don’t just mean making it up. Because GIS technology is fairly new to some disciplines, data may exist in hard-copy formats (paper maps, field notes, etc.) but not necessarily in a computerized, GIS-compatible format. For these reasons, the researcher may have to be persistent and a bit creative when seeking out potential sources of data. It is not uncommon to find multiple sources of what appear to be the same data, so it is essential to review and evaluate each prospective dataset before settling on the perfect source of information for your particular study.
Of course, not all data you may need will exist, so it is likely you could be collecting your own data for use in a GIS analysis. For many social scientists, data collection is one of the most enjoyable components of the research process. Why? This is the part of the research process in which researchers actively implement ideas that, until then, have only existed in their minds (and, ideally, during the two previous stages of the research process, have been