The urban-rural divide in policy priorities across time and space
The urban-rural divide in policy priorities across time and space (DOI: https://doi.org/10.1017/psrm.2025.33) was written by Tevfik Murat Yildirim and Knut M. Solvig in 2025. It was published in Political Science Research and Methods.
The authors study policy priorities with respect to geography and partisanship, to establish if urban residents differ from rural residents and if those differences overlap with partisanship.
The authors use the MIPD. Responses are coded into 110 issue categories. The authors recode these into 12 focus issues:
- budget deficit
- agriculture
- moral values
- immigration
- economy
- tax
- civil rights
- crime
- education
- foreign policy
- health
- drugs
The authors then estimate prioritization among them.
Importantly, this dataset includes a standardized urbanicity variable that is reduced to binary classification: 1=urban.
Also importantly, this dataset lacks measures of income before 1960. The authors therefore restrict the primary analysis to after 1960 but provide some analysis in the appendix that excludes the income variable but covers the full time range.
The authors attached annual state-level statistics as controls, e.g. income inequality.
Survey weights are attached to the dataset, but the authors designed new weights against Census targets. "[T]hese weights are not comparable across time and space, and we are therefore unable to use them in our empirical models." They fit both weighted and unweighted models to establish robustness.
The authors estimate main effects with logistic regression. They find that "residents of rural and other peripheral communities are more likely than their urban counterparts to prioritize health care, moral values, agriculture, budget deficit, and issues related to the economy" and "urban residents seem to be more concerned about such issues as education, drugs, foreign policy, immigration, crime, and civil rights". Effect sizes are modest.
The authors then estimate marginal effects over time, establishing that most of these effects are stable in the period 1960-2020. Prioritization of agriculture, drugs, and moral values has risen among rural residents relative to urban voters; prioritization of the economy has risen among urban voters relative to rural voters.
They then construct multilevel models with nesting at the state level to estimate place of residence and state level effects. They find that the gap in prioritization of moral values disappears in states with low income inequality, while the effect grows given high income inequality. They also find that "prioritization of the economy, agriculture, moral values, crime, foreign policy, and drugs diminishes as state-level policy liberalism increases."
They then incorporate partisanship, and find that "place-based gaps are overshadowed by partisan gaps." One exception is moral values; rural Republicans are more likely to prioritize them relative to urban Republicans.
Notably, "[b]oth rural Democrats and Republicans were significantly more likely than their urban counterparts to prioritize issues related to agriculture until around the 1980s, after which the urban–rural gap disappeared."
The authors focus on drugs and moral values issues with qualitative analysis. They find substantive evidence that drugs are coded language for moral values, and then run more focused modeling on subcategories of the moral values category:
- morals and values
- family issues
- lack of religion
- abortion
In terms of these issues, rural Democrats started very similar to rural Republicans in 1960, but have moved towards urban Democrats since. Rural Republicans' prioritization of these issues has grown over time relative to urban Republicans.