A new paper on the how to reliably assess the degree of urbanisation using a single-item self-report measure
The impact of rural versus urban residence on mental health remains a controversial topic, and this calls for valid and reliable measurement tools to define, measure and compare degrees of urbanisation across settings. In a recent study conducted within the Centre for Urban Mental Health, Brinkhof et al. looked at the utility of a single-item self-report measure as a tool to classify areas along a rural-urban continuum. They found that simply asking individuals to score their living environment on a 7-point scale ranging from ‘not urban at all/hamlet’ (1) to ‘very urban’ can approximate objective measures of the degree of urbanisation, and could therefore be used as a valid proxy.
But why would you want to rely on a self-report measure if objective measures are readily available? To answer this question, the authors explain that it is important to consider how the degree of urbanisation is typically defined and measured. While there are multiple definitions of what constitutes an urban environment, population/address density, or some derivative thereof, is the most commonly used measure to classify areas along a rural-urban continuum. In the Netherlands, the surrounding address density is used to categorize environments from ‘not urbanized’ (fewer than 500 addresses/km²) to ‘extremely urbanised’ (2500 addresses/km² or more). This objective categorization is widely accepted. Yet, the applicability is somewhat limited by the fact that linking such objective environmental estimates to other research data requires personal, and potentially highly sensitive, information on participants’ location of residence (e.g., postal code). In the article, the authors explain: “Ever since the General Data Protection Regulation (GDPR) has been introduced to protect data and privacy of individuals in the European Union, researchers have been required to adhere to the principle of data minimization. This comprises that processing of personal data is not allowed if one can achieve the same goal without the concerning data.” Therefore, they conclude that: “research into the interplay between mental health and urbanisation would benefit from an alternative measure of the degree of urbanisation that does not require precise information about the location of residence.”
In two independent older adult samples, the authors found a strong coherence between the single-item self-report measure and the surrounding address density, with correlations ranging from .77 to .83 that were not meaningfully affected by age or gender. Moreover, they showed they a self-report score of 6 or higher could accurately distinguish extremely urbanized areas from other areas.
The authors therefore conclude that, when the only goal is to distinguish urbanisation subgroups*, this self-report measure may be a favorable alternative to objective, postal code reliant, measures. They do argue that it is important to validate the scale in younger and/or psychopathological/cognitively impaired samples to establish the generalizability of their results. However, ultimately, this scale may provide a widely applicable tool that is even more comprehensive than objective measures, given that subjective report leaves room for individual perceptions on what confers an urban settlement.
*In case researchers want to gain detailed and precise information about specific environmental characteristics (e.g., availability of green space), they most likely do need information about individuals’ residency.
Brinkhof, L. P., Ridderinkhof, R. K., Krugers, H. J., Murre, J. M. J., & de Wit, S. (2022). Assessing the Degree of Urbanisation using a Single-Item Self-Report Measure: A Validation Study. International Journal of Environmental Health Research. https://doi.org/10.1080/09603123.2022.2036331
Read the publication here.
Read this article here.