Planning Data and Analysis
Urban planning is information-driven decision-making. Cities and regions are “systems of systems"; their characteristics are many, including the dimensions of time and space. There is much information about the history and characteristics of cities and regions as well as information generated by cities and regions. Plan-making, it can be argued, is a data hungry process as suggested by the rational planning model. As seen in the model (Figure 4.1), identifying the relevant goals and objectives is the first step, which itself requires information from a range of sources and stakeholders. Planners are trained to ask questions, matching them with appropriate data. This chapter focuses on planning data as an important element in planning analysis and research.
Planning questions and issues are often framed as problems, including, for instance, traffic congestion, lack of affordable housing, and land use conflicts. Planning questions can also take the form of opportunities. Economic development, open space and parks planning, and community involvement represent situations where alternative actions can lead to increased efficiency, increased satisfaction, and new ideas— while not being specifically rooted in discreet, identifiable problems. Therefore, in addition to identifying deficiencies and challenges, a planner’s role includes drawing upon experience and evidence to create or enhance change. Again, this relies on the ability to match appropriate information with the question at hand, as efficiently as possible. Whether a problem or an opportunity, quantitative and qualitative evidence is used to support decisions and actions as part of the plan-making process. The following discusses planning data types, structures, and formats that are common to planning research and analysis. An understanding of data capacities and limitations is an important part of research design.
Planning Data Types and Structures/Formats
Planners are expected to be fluent in several data types and formats given the wide range of what they observe and measure. These include, but are not limited to, economic, demographic, physical, and environmental conditions over time and space (Ma, Lam, & Leung, 2018). These data may also be aggregated, depending on the type of analysis, the unit of analysis, and the methods of data collection. The data collection step is important not only for acquiring information for analysis, but also because it dictates the kinds of questions that can be answered. This includes dimensions of who, what, where, when, and why. Sufficient time and forethought is needed
Figure 4.1 Rational Planning Model
to clarify how the data will be used and the types of statements and results that will be communicated (Wang & Hofe, 2008). Answering these questions first will better inform the data collection strategy and overall analysis.
Data are collected from either primary or secondary sources. Primary data collection occurs when data are obtained firsthand, directly from a source. Examples include surveys, interviews, field observation, or sensed data. For instance, surveys are commonly used by planners to obtain feedback from residents for community involvement activities to gauge satisfaction and to inventory needs and preferences. Such surveys can result in different data types, structured or unstructured, which involve particular data management and analysis approaches (discussed later). This includes open and closed-ended questions depending on question types.
Primary data collection is often expensive because it involves the effort of identifying, contacting, and directly observing individual sources, along with coding, cleaning, and verifying the data. On the other hand, secondary data have been acquired by another (primary) source, which in many cases, has processed, cleaned, and verified
Planning Data and A nalysis 63 the data. Avery common source of secondary data for urban planners is the U.S. Census. While secondary data can save time and effort, it is essential to understand the methods and purposes of the original data collection to determine appropriateness. This information should be included as part of the meta-data, which documents the origins and reliability of the data.
Sampling and Bias
During either primary or secondary data collection it is important to understand the sampling method employed and how it may introduce bias or error in the data. Sampling usually occurs when it is not feasible or efficient to collect information from every case in a study population. Collecting information about every individual observation in a large population can be expensive and time consuming. In addition, it is often true that the variability of characteristics in a population can be captured with less than a 100 percent sample. At a certain point the increment of new information (or variation) learned by an additional observation begins to decline. This depends on the overall size and variation within a population. Understanding the potential for bias within a dataset has direct implications for the reliability of resulting analyses (see Chapter 6).
Bias in relation to data sampling refers to the misrepresentation of population characteristics. Biased sampling leads to the selection of observations that are not in proportion to the distribution of values in the study population. Analyses using biased data can produce biased or unreliable results. For example, a study of the general public that had only male survey respondents will only represent the characteristics of males, and not that of the whole population as was intended. This can result from faulty sampling methods or poor study design.