Integrating Data Across Misaligned Spatial Units

Zhukov, Yuri M.; Byers, Jason S.; Davidson, Marty A.; Kollman, Ken

doi:10.1017/pan.2023.5

Cited by 2 publications

(1 citation statement)

References 24 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…surveys) may not be georeferenced at all. Different data integration choices may yield different results, raising concerns over generalizability [5]. Differences in sampling, question wording and sequence, primary sources, operational definitions, digital image processing algorithms, and other factors ensure that no two datasets are perfect substitutes for one another, making it difficult to distinguish case-specific idiosyncracies from general patterns, and to ask, "what does country A tell us about country B?"…”

Section: Methodsmentioning

confidence: 99%

Promoting data harmonization to evaluate vaccine hesitancy in LMICs: approach and applications

Rego,

Zhukov,

Reneau

et al. 2023

BMC Med Res Methodol

Self Cite

View full text Add to dashboard Cite

Background Factors influencing the health of populations are subjects of interdisciplinary study. However, datasets relevant to public health often lack interdisciplinary breath. It is difficult to combine data on health outcomes with datasets on potentially important contextual factors, like political violence or development, due to incompatible levels of geographic support; differing data formats and structures; differences in sampling procedures and wording; and the stability of temporal trends. We present a computational package to combine spatially misaligned datasets, and provide an illustrative analysis of multi-dimensional factors in health outcomes. Methods We rely on a new software toolkit, Sub-National Geospatial Data Archive (SUNGEO), to combine data across disciplinary domains and demonstrate a use case on vaccine hesitancy in Low and Middle-Income Countries (LMICs). We use data from the World Bank’s High Frequency Phone Surveys (HFPS) from Kenya, Indonesia, and Malawi. We curate and combine these surveys with data on political violence, elections, economic development, and other contextual factors, using SUNGEO. We then develop a stochastic model to analyze the integrated data and evaluate 1) the stability of vaccination preferences in all three countries over time, and 2) the association between local contextual factors and vaccination preferences. Results In all three countries, vaccine-acceptance is more persistent than vaccine-hesitancy from round to round: the long-run probability of staying vaccine-acceptant (hesitant) was 0.96 (0.65) in Indonesia, 0.89 (0.21) in Kenya, and 0.76 (0.40) in Malawi. However, vaccine acceptance was significantly less durable in areas exposed to political violence, with percentage point differences (ppd) in vaccine acceptance of -10 (Indonesia), -5 (Kenya), and -64 (Malawi). In Indonesia and Kenya, although not Malawi, vaccine acceptance was also significantly less durable in locations without competitive elections (-19 and -6 ppd, respectively) and in locations with more limited transportation infrastructure (-11 and -8 ppd). Conclusion With SUNGEO, researchers can combine spatially misaligned and incompatible datasets. As an illustrative example, we find that vaccination hesitancy is correlated with political violence, electoral uncompetitiveness and limited access to public goods, consistent with past results that vaccination hesitancy is associated with government distrust.

show abstract

Section: Methodsmentioning

confidence: 99%

Promoting data harmonization to evaluate vaccine hesitancy in LMICs: approach and applications

Rego,

Zhukov,

Reneau

et al. 2023

BMC Med Res Methodol

Self Cite

View full text Add to dashboard Cite

show abstract

Integrating Data Across Misaligned Spatial Units

Zhukov¹,

Byers²,

Davidson³

et al. 2023

Polit. Anal.

Self Cite

View full text Add to dashboard Cite

Theoretical units of interest often do not align with the spatial units at which data are available. This problem is pervasive in political science, particularly in subnational empirical research that requires integrating data across incompatible geographic units (e.g., administrative areas, electoral constituencies, and grid cells). Overcoming this challenge requires researchers not only to align the scale of empirical and theoretical units, but also to understand the consequences of this change of support for measurement error and statistical inference. We show how the accuracy of transformed values and the estimation of regression coefficients depend on the degree of nesting (i.e., whether units fall completely and neatly inside each other) and on the relative scale of source and destination units (i.e., aggregation, disaggregation, and hybrid). We introduce simple, nonparametric measures of relative nesting and scale, as ex ante indicators of spatial transformation complexity and error susceptibility. Using election data and Monte Carlo simulations, we show that these measures are strongly predictive of transformation quality across multiple change-of-support methods. We propose several validation procedures and provide open-source software to make transformation options more accessible, customizable, and intuitive.

show abstract

Integrating Data Across Misaligned Spatial Units

Cited by 2 publications

References 24 publications

Promoting data harmonization to evaluate vaccine hesitancy in LMICs: approach and applications

Promoting data harmonization to evaluate vaccine hesitancy in LMICs: approach and applications

Integrating Data Across Misaligned Spatial Units

Contact Info

Product

Resources

About