Types of Research

Overview

Planners commonly engage in three types of research: qualitative, quantitative, or mixed. All research starts with a research question or questions, and these three fundamental approaches to research provide the means by which you answer the question (s). Qualitative research explores the non-numerical traits (qualities) that describe the subject or context under investigation, whereas quantitative research explores the numerical traits (quantities) that are measured or counted, hence their respective names. Mixed methods research combines qualitative and quantitative research, often triangulating to answers based on evidence from both.

The first section of this chapter describes the history of the different research traditions. The second section describes the general concepts and terminology that form the context of research. The third section provides an overview of each of the three types of research and the methods commonly associated with them. Further detailed descriptions of these methods occur later in this book. This chapter concludes with considerations when you choose your research methodology'.

History

Planning research has its roots in other disciplines. Qualitative and quantitative research grew out of the scientific method of observation and analysis. Qualitative research approaches descend directly from methods performed on indigenous populations at the height of European colonization (Denzin & Lincoln, 2005). In the 1920s and 1930s, qualitative methods developed at the University of Chicago along with the contemporaneous work in the anthropology of Boas, Mead, Benedict, Evans-Pritchard, Alfred Radcliffe-Brown, Bronislaw Malinowski, and others formed the foundations for contemporary qualitative research (Denzin & Lincoln, 2005).

Drawing from these earlier methods, such noted planning authors asjacobs (1961), Lynch (1960, 1972), Whyte (1980, 1988), Jackson (1984), and Gehl (1987) published groundbreaking works based on their qualitative observational methods. Indeed, we would argue that the most influential works in planning are based on qualitative, astute observation. Quantitative research tends to come later and test theories grounded in qualitative research.

Quantitative research in planning emerges from methodologies used in economics, political science, public health, business, and other disciplines. The individual chapters in this book, which deal primarily with quantitative research, provide the relevant history of each method.

Research using both qualitative and quantitative approaches has often been overlooked in academia and existed long before the term mixed methods appeared in textbooks and publications (Maxwell, 2016). Early investigators who used mixed methods without explicitly stating so include Charles Booth, Jane Addams, Wilhelm Wundt, and Max Weber. Anthropology, archeology', and linguistics have continuously employed qualitative and quantitative methods for many years, but little has been published on how to integrate the two approaches and types of data (Maxwell, 2016). Even newer approaches like design-based research don’t get categorized as mixed methods even though they integrate “both qualitative and quantitative data [to] inform the conclusions ... and test the interpretation (theory) of what took place” (Maxwell, 2016, p. 19).

General Concepts

As you venture into your research, you need to understand the general concepts that provide the underpinnings of all research. This section presents the concepts of data type, inductive versus deductive logic, timeframe, research design, and triangulation.

Data Type

The types of data that you seek will inform and be informed by the research that you intend to do. Greater reliance on primary data dictates more time and funding to collect it and develop the information that informs the findings. Primary data come directly from the responses to surveys, observations, and other data collected by the researcher. Secondary data come from other researchers (e.g., published reports, journals), existing data archives held by a second party (e.g., databases), archival resources (e.g., libraries), or other sources (e.g., books,journal articles, websites) that were not produced by the researcher. The expected results (e.g., verbal descriptions, mathematical models) of the research are the impetus for data collection. Consideration of the story that you ultimately tell will influence your selection of data collected.

Qualitative data reflect a nuanced understanding of the research context and the subsequent development of findings. Due to the time intensive nature of qualitative data collection, it tends to involve smaller samples that may not equally well represent the larger demographic population commonly modeled in quantitative analysis. The findings are also sensitive to research bias in both the collection and analysis of data (Regents of the University of Minnesota, 2014).

Conversely, quantitative data are numerically specific. When properly analyzed, they are objective and reliable. Due to its numeric orientation, quantitative data can be readily communicated using graphs and charts. Best of all, many large datasets already exist. Quantitative data may not accurately describe complex situations as subtleties are distilled into broader singular categories or go missing entirely. Lastly, the use of some of the more sophisticated quantitative analysis methods requires specific expertise that may not be commonly available (Regents of the University of Minnesota, 2014). The authors of this chapter are strong advocates for the use of statistical consultants, and, indeed, have enlisted Bill Greene, Jim Grace, Mark Stevens, Simon Brewer, John Kircher, and others as co-authors of chapters in this book or its advanced-version companion (Ewing & Park, 2020).

We are providing two quantitative datasets with this book on quantitative methods. Both are examples of secondary data, collected in one case by the federal government and processed by us, and in the other case collected by local agencies and processed by us. Chapter 1 provides a description of and data dictionary for each dataset.