Qualitative Methods

The numerous historical approaches to qualitative research are shown in Figure 3.3. There are basically four research traditions: narrative (storytelling), phenomenology (lived experiences), grounded theory (theoretical explanation), and ethnographic (culture-sharing groups) (Groat & Wang, 2002; Lewis-Beck, Bryman, & Fitting Liao, 2004; Rudestam & Newton, 2007; Roberts, 2010; Creswell, 2013; Edmonds & Kennedy, 2013).

Figure 3.3 Qualitative Research Family Tree Source: Miles & Huberman (1994, p. 6)

In narratives, the researcher seeks out individuals to tell the story of their lives. Phenomenology is similar but with an expanded view of a group that has witnessed or attended a phenomenon. Grounded theorists develop a theory based primarily on observation, with multiple stages of data collection and analysis. Ethnographic researchers study cultural groups in their natural setting over long periods.

Specific qualitative methods used by planners for narrative, phenomenological, grounded theory, and ethnographic research include:

• • Case studies (Ewing, August—September 2008)
• • Survey research (Ewing, December 2007)
• • Observational methods (Ewing, February 2014)
• • Interviews and focus groups (Ewing, October 2007)

Draft chapters have already been written on each of the preceding. The next edition of this book will include these chapters.

Quantitative Methods

This is a book about basic quantitative methods used by planners. Individual chapters that follow present all the methods in common use.

• • Descriptive statistics
• • Chi-square
• • Correlation
• • Difference of means
• • Analysis of variance (ANOVA)
• • Linear regression
• • Logistic regression
• • Quasi-experimental methods

The companion book, Advanced Quantitative Research Methods for Urban Planners, covers the following more advanced methods:

• • Poisson and negative binomial regression
• • Factor analysis
• • Cluster analysis
• • Multilevel modeling
• • Structural equation modeling
• • Spatial econometrics
• • Meta-analysis and meta-regression

The construction and application of these methods make it impractical to describe them here, and they will instead be described in detail later in this book and in its companion.

Mixed Methods

In February 2013, the second author of this chapter wrote a Research You Can Use column for Planning magazine on mixed methods research. We borrow from that column now.

The argument for mixed methods goes like this. First, quantitative and qualitative studies answer different, but equally important questions. Qualitative research deals with the why and how of something, whereas quantitative research mostly responds to questions of how much. Second, qualitative research is often the best way to determine causality between variables, not just correlation. Third, quantitative and qualitative studies can provide independent validation of the same phenomenon. Triangulation, or the use of multiple methods to double-check results, was first advocated by Todd Jick of Cornell University, one of the pioneers of mixed method research. In a 1979 article, he wrote, “The convergence or agreement between two methods ... enhances our belief that the results are valid and not a methodological artifact” (p. 602).

Adam Millard-Ball, an assistant professor of environmental studies at the University of California, Santa Cruz, performed a mixed methods study of climate action planning by California cities. The quantitative portion of the study, published in the Journal of Urban Economics, found that cities with climate action plans have more green buildings, spend more on pedestrian and bicycle infrastructure, and implement more programs to divert waste from methane-generating landfills. However, the same study concluded that these actions result not from the climate plans themselves, but rather from the pressure of citizen groups. In short, the correlation between climate action planning and emission-reduction measures does not mean that the former causes the latter, but rather that confounding factors are at work, revealing a classic problem with quantitative studies.

To get at the thorny issue of causation, Millard-Ball conducted a second study, published in the Journal of Planning Education and Research. The second study was qualitative. Based on case studies of municipal climate action planning, he found little evidence of causal impacts. Instead, cities use climate plans to codify policies that were likely to happen anyway. This result is not too surprising since climate action planning by local governments in California is strictly voluntary.

The connection between Millard-Ball’s quantitative and qualitative studies is twofold (Figure 3.4). First, and most obviously, the qualitative study validates the findings of the earlier quantitative study. Second, the quantitative study provides a clever method of case study selection. By selecting cases for the qualitative study that outperform expectations in the quantitative study, Millard-Ball could conduct cross-case comparisons of two cities that perform above expectations on environmental indicators (only one with a climate plan) and two cities with climate plans (only one performing above expectations).