Overview of Geographic Information
IPUMS MICS provides geographic information to facilitate disaggregation at subnational levels of geography as well as cross-temporal comparative analyses. Multiple sample years are available for most countries in the database, and geography variables are spatially harmonized when there are changes in administrative boundaries over time.
IPUMS MICS provides two kinds of geography variables: spatially harmonized and sample-specific
- Spatially harmonized geography variables. Provide consistent geographic units for a country across sample years. When there are changes in boundaries over time, IPUMS MICS creates spatially consistent footprints across time by aggregating units as necessary. Some geographic detail is lost in the construction of spatially consistent geography.
- Sample-specific geography variables. Where geographical boundaries have changed over time for a country, MICS provides sample-specific variables in addition to the harmonized version. These variables retain all of the original detail from each sample, but they are usually not fully consistent over time.
We provide a list of geography variables along with GIS files across all IPUMS MICS samples.
Other information about IPUMS MICS geography variables
- Most MICS samples do NOT have any changes in geography across the samples. In such cases, users should use the "primary geography variable" listed on this page for all purposes.
- When there are changes in geography across the samples, IPUMS creates consistent boundaries across time. These spatially consistent variables provide identical geographic units for a country across samples years, facilitating comparisons across time.
- Use "sample-specific geography" variables when only one sample is required for analysis.
- All MICS geography variables correspond to a GIS shapefile that is available for download.
- All geography variables begin with GEO followed by the number 1, 2, or 3 (e.g. GEO1_PK, GEO2_PK). GEO1 indicates the highest level of geography, GEO2 will give more detail, and GEO3 (when available) is the most detailed.
- If a geography variable ends with a 4-digit year (e.g. GEO1_PK2014, GEO3_PK2011), the variable corresponds to the geography for that specific sample year, like 2014 or 2011. Variables without a year are harmonized or "primary".
The basic set of potential geography variables for any MICS country includes the following (using Mali and Malawi as examples):
| Variable | Country (Sample years) | Type of geography | What variable to use? |
|---|---|---|---|
| GEO1_ML | Mali (2009, 2015) | Consistent boundaries 2009-2015 [GIS] | No changes in geography, use GEO1_ML for all survey years |
| GEO1_MW | Malawi (2006, 2013, 2019) | Consistent boundaries 2006-2019 [GIS] | Geography is different between the survey years. For analysis across time use GEO1_MW |
| GEO1_MW2006 | Malawi (2006) | Sample-specific boundaries 2006 [GIS] | Geography is different between the survey years. For analysis of 2006 only, use GEO1_MW2006 |
| GEO1_MW2013 | Malawi (2013) | Sample-specific boundaries 2013 [GIS] | Geography is different between the survey years. For analysis of 2013 only, use GEO1_MW2013 |
| GEO1_MW2019 | Malawi (2019) | Sample-specific boundaries 2019 [GIS] | Geography is different between the survey years. For analysis of 2019 only, use GEO1_MW2019 |
"ALT" in the name of a variable indicates that there is an alternative geography available at that level (e.g., GEO1ALT_GY).
Supported By
