A Participant-Centered, GIS-Based Approach to Improving Contextual Measurement

A Participant-Centered, GIS-Based Approach to Improving Contextual Measurement

It’s clear that where people live matters

Research across a number of disciplines including sociology, public health, and political science shows that the type of neighborhood people inhabit has far-reaching consequences, influencing outcomes such as school dropout, exposure to violence, quality of mental health, and levels of political participation (Crane, 1991; Branas, Rubin, & Guo, 2012; Kim, 2010; Schlozman, Verba, & Brady, 2012; Cohen & Dawson, 1993). 

It’s less clear, however, how to measure neighborhoods. Unlike other contexts like cities or counties, neighborhoods rarely have clear, standardized, or officially recognized boundaries. To work around this, researchers most commonly approximate neighborhoods with census tracts, geographic subdivisions created for statistical purposes intended to have roughly equal population sizes and demographic characteristics. What are the risks of measuring neighborhoods this way? In this Blog Post, I highlight three limitations and explain how participant-drawn neighborhoods, collected through GIS tools, offer a more precise way to measure neighborhoods.

The limits of conventional neighborhood measurement

1. Administrative boundaries may fail to meaningfully capture how residents themselves experience their surroundings. 

While frequently treated as merely physical spaces, neighborhoods are also social constructions shaped by people’s lived experiences, identities, and daily routines (Lee et al., 1994). Consequently, individuals occupying the same geographic area may perceive their neighborhoods differently, often in ways that diverge from administrative boundaries (Lee & Campbell, 1997). In the least problematic case, the imposition of artificial boundaries can yield an incomplete understanding of local dynamics. In the most harmful case, such as indigenous territories, the imposition can result in sacred or historically significant lands being mislabeled as “ungoverned” or “unproductive,” effectively erasing local knowledge and enabling exploitative interventions.

2. Administrative boundaries can misrepresent contextual effects 

Contextual effects refer to how characteristics of the surrounding environment influence a person’s attitudes or behaviors. Neighborhoods are just one type of setting in which these effects occur, but they can also arise in other social, institutional, or physical settings, such as a school or workplace. Such effects may stem from the demographic composition, the built environment, or other features of the setting. When estimating neighborhood effects, the assumption is that everyone within a single tract shares the same exposure and spatial understanding of their neighborhood. Yet individuals living in the same tract may define and experience their surroundings differently. This means that residents’ perceptions may shape or mediate behavior, not just context (Wong et al., 2012). Treating exposure as uniform conflates perceived and actual effects and introduces measurement error, since administrative units rarely match lived spaces.

3. Administrative boundaries make it difficult to draw consistent conclusions

Because these boundaries are chosen arbitrarily, analyses are prone to the Modifiable Areal Unit Problem (MAUP), where observed patterns reflect the scale and shape of the boundary rather than actual social processes. (Openshaw & Taylor, 1979; 1981). For example, rates of participation in protests may appear uniform across a city, but tract-level analysis could reveal more concentrated participation that is masked by the aggregation. Such patterns at these different scales could make it difficult to draw consistent conclusions across studies. 

Self-defined neighborhood boundaries as a solution

As others have proposed, letting individuals define their neighborhood boundaries offers a practical way to address the three limitations described above (Coulton et al., 2001; Wong et al., 2012; 2020). First, self-defined neighborhood boundaries ensure that the measure accurately reflects the individual's lived experiences. Second, self-defined boundaries capture the environment as experienced by the individual, allowing researchers to measure contextual effects more accurately. Relatedly, this approach reduces measurement error by aligning the spatial unit of analysis with each individual’s true unit of exposure. Finally, it helps mitigate the MAUP by grounding analyses in units defined by the individual rather than arbitrarily selected administrative boundaries.

Methods for capturing self-defined boundaries

Researchers typically capture self-defined boundaries in three ways: (1) buffers, circular areas centered on an individual’s home that are uniform in size but vary in coverage and may overlap; (2) behavioral activity spaces, tracked via GPS to capture daily movements; and (3) perceived neighborhoods, either drawn by residents using GIS tools or estimated through self-reported size or walking distance. Here, I will focus on the final method for capturing perceived neighborhoods, highlighting the advantages and drawbacks of GIS-based versus non-GIS methods.

What are GIS-based and non-GIS-based methods?

With GIS-based methods, respondents digitally outline their perceived neighborhoods on a map via survey platforms like Qualtrics, Partimap, or Kobo Toolbox. An example of how this can be done in Partimap is presented in Figure 1. Using the purple cursor, I delineated my hypothetical “neighborhood” on an administrative map of Nairobi. While the map in the example is zoomed out, I could have easily zoomed in using the buttons at the bottom right. In addition to completing this task digitally, respondents can also freely draw their neighborhood on physical paper. Such hand-drawn maps are later digitized by tracing the boundaries and converting them into shapefiles. Both digital and hand-drawn boundaries yield continuous, spatially explicit measures of neighborhoods.

With non-GIS-based methods, respondents define their neighborhoods without maps, either as continuous measures or ordinal categories, for example, by reporting the number of blocks their perceived neighborhood covers or choosing from options like: (1) their block, (2) several surrounding blocks, (3) the area within a 15-minute walk, or (4) a larger area.

Figure 1 Drawing Self-Defined Neighborhood Boundaries in Partimap

Each method has distinct advantages and drawbacks

GIS-based methods offer interactive, flexible, and explicit ways for participants to convey their experiences. For example, Indigenous and Afro-descendant communities have used participatory cartography to document and defend their territories (Pineda, 2020), while formerly incarcerated individuals in Chicago have used participatory mapping to highlight areas they felt they could find housing (Hamlin, 2022). These methods, however, can be technically demanding, time-intensive, and require careful ethical safeguards. 

Non-GIS methods are simpler to administer and well-suited for large-N online surveys, but they are less sensitive to variations in neighborhood shape and size and may oversimplify complex spatial perceptions. Table 1 below summarizes the key advantages and drawbacks of each approach. Researchers should note that using either method may yield systematically different results, with map-drawn neighborhoods generally tending to be larger (Coulton et al., 2001). To ensure consistency and comparability, researchers should stick to a single method throughout a study. 

Both methods are great, but GIS methods may offer the strongest potential

As we have seen, both GIS- and non-GIS-based methods have distinct advantages and limitations for measuring neighborhoods. Many of these limitations naturally stem from asking individuals to represent their surrounding space, which can be especially difficult for those with limited cartographic or digital literacy. Despite these challenges, both methods address key shortcomings of traditional measures. GIS-based methods are particularly effective for producing explicit, comparable, valid, and reliable spatial boundaries, while non-GIS methods are well-suited for approximating neighborhood size. Together, they enable researchers to more accurately capture context as experienced by residents, measure its effects robustly, and generate consistent, reliable results. Such advances allow us to better design interventions that improve neighborhoods and, in turn, life outcomes.

References

1. Branas, C. C., Rubin, D., & Guo, W. (2012). Vacant properties and violence in neighborhoods. International Scholarly Research Notices, 2012(1), 246142

2. Coulton, Claudia J., Jill Korbin, Tsui Chan, and Marilyn Su. 2001. “Mapping Residents’ Perceptions of Neighborhood Boundaries: A Methodological Note.” American Journal of Community Psychology 29 (2): 371–83.

3. Coulton, C. J., Jennings, M. Z., & Chan, T. (2013). How big is my neighborhood? Individual and contextual effects on perceptions of neighborhood scale. American journal of community psychology, 51(1-2), 140–150. https://doi.org/10.1007/s10464-012-9550-6

4. Madeleine Hamlin (2022) Participatory Sketch Mapping for Policy: A Case Study of Reentry Housing from Chicago, The Professional Geographer, 74:1, 52-66, DOI: 10.1080/00330124.2021.1952883

5. Kim, J. (2010). Neighborhood disadvantage and mental health: The role of neighborhood disorder and social relationships. Social science research, 39(2), 260-271.

6. Lee, B. A., Oropesa, R. S., & Kanan, J. W. (1994). Neighborhood context and residential mobility. Demography, 31(2), 249–270.

7. Lee, B. A., & Campbell, K. E. (1997). Common ground? Urban neighborhoods as survey respondents see them. Social Science Quarterly, 78, 922–936.

8. Openshaw, S., & Taylor, P. J. (1979). A million or so correlation coefficients: Three experiments on the modifiable areal unit problem. In N. Wrigley (Ed.), Statistical applications in the spatial sciences (pp. 127–144). London: Pion.

9. Openshaw, S., & Taylor, P. J. (1981) The modifiable areal unit problem. In N. Wrigley & R. J. Bennett (Eds.), Quantitative geography: A British view (pp. 60–70). London: Routledge and Kegan Paul.

10. PINEDA, W. (2020). Revealing Territorial Illusions and Political Fictions through Participatory Cartography. In B. SLETTO, J. BRYAN, A. WAGNER, & C. HALE (Eds.), Radical Cartographies: Participatory Mapmaking from Latin America (pp. 65–80). University of Texas Press. http://www.jstor.org/stable/10.7560/320884.7

11. Wong, C., Bowers, J., Williams, T., & Simmons, K. D. (2012). Bringing the person back in: Boundaries, perceptions, and the measurement of racial context. The Journal of Politics, 74(4), 1153-1170.

12. Wong, C., Bowers, J., Rubenson, D., Fredrickson, M., & Rundlett, A. (2020). Maps in People’s Heads: Assessing a New Measure of Context. Political Science Research and Methods, 8(1), 160–168. doi:10.1017/psrm.2018.51