Skip to main content

Using Longitudinal Multilevel Models to Investigate the Relationship Between Urbanization and Protest Mobilization in Africa

By: & Published: 2019 | Product: SAGE Research Methods Cases Part 2
+- LessMore information
Search form
No results
Not Found
Download Case PDF


Urbanization has long been seen by scholars and policymakers as a disruptive process that can contribute to social and political unrest. But there has been very little systematic quantitative research on the topic. We set out to address this gap in the empirical literature by investigating the relationship between “urbanization” and the frequency of protests in African countries. Our paper makes two key methodological contributions. First, we empirically disaggregated the phenomenon of “urbanization” in our modeling strategy. We did this by incorporating variables capturing scale effects (i.e., absolute number of people in cities), ratio effects (i.e., relative share of a population living in cities), rate of change effects (i.e., how fast urban populations are changing), and population distribution effects. No previous study has investigated all these dynamics simultaneously. This allowed us to capture relationships in the data that have been overlooked. Second, we used a multilevel model (or random effects model) that allowed analysis of effects at two levels—within a country over time, and between countries. This modeling strategy differs from the fixed-effects approach that dominates research in this field by political scientists and economists. Both innovations yielded novel insight. We found that levels of urbanization within and between countries to be negatively associated the frequency of protests. But we also found the absolute size of a country’s urban populations to be positively correlated with protest frequency between countries. These results provide a nuanced picture of the links between spatial-demographic change and protests that challenges conventional wisdom.

Looks like you do not have access to this content.

Methods Map

Multilevel analysis

Copy and paste the following HTML into your website