The majority of the indicators used in the Digital Advancement Muni Index are from the American Community Survey 1-Year Data.
The American Community Survey (ACS) is an ongoing survey that provides data every year -- giving communities the current information they need to plan investments and services. The ACS covers a broad range of topics about social, economic, demographic, and housing characteristics of the U.S. population. ACS provides different levels of estimates depending on the population size of a region. For example, ACS-1 Data is available for the nation, all 50 states, the District of Columbia, Puerto Rico, every congressional district, every metropolitan area, and all counties and places with populations of 65,000 or more.
More on how to use ACS data: When to Use 1-year or 5-year estimates?
In order to scope our initial dataset, we focused on cities with populations over 100,000. This meant we would likely be able to use ACS-1 data for these urban areas and could start our process with a small dataset of about 300 cities. To determine the cities to use, we used the 2019 population estimate in the US Census Cities Population Dataset. This dataset is from the Pop Estimates Program and only provides estimates for incorporated places so unincorporated cities like Paradise, NV or Arlington, VA are scoped out.
More information about the population estimates program can be found in their ESTIMATES AND PROJECTIONS AREA DOCUMENTATION.
Each of the cities in the dataset has a unique combination of a FIPS state code and FIPS Place code. These are used to identify the city as a Census Designated Place. Note that the Census Designated Place map does not always map perfectly to the actual city boundaries. In some cases, there may not always be a Census Designated Place corresponding to a large city. This may be due to cities not being incorporated, having county level government structures, or other atypical attributes.
After identifying our initial list of cities by filtering the Census pop estimates dataset to only cities with populations larger than 100,000 in 2019, we discovered that the following large (> 100K) cities were missing from our dataset:
Paradise, NV
Arlington, VA
Nashville, TN
Honolulu, HI
Anchorage, AK
Macon, GA
Augusta, GA
Athens, GA
Lexington, KY
Louisville, KY
First we identified that Nashville, TN has a unique government structure, but by using the Nashville-Davidson metropolitan government (balance) with FIPS state 47 & place 52006, we approximate Nashville.
Based on guidance from principal researcher John Horrigan, PhD, we decided to query county-level data where city-level data was not available to use due to lack of incorporation or other atypical governance structure. We chose to query for these cities as the majority of county level population is in those cities:
Anchorage municipality, AK (queried as a county)
Arlington county, VA
Fayette County, KY (Lexington)
Jefferson County, KY (Louisville)
This meant that we intentionally excluded the following cities where county level data was not available or the county level population was not majority made up of the city population:
Paradise, NV
Honolulu, HI
Macon, GA
Augusta, GA
Athens, G
Next, we also discovered that we were missing the Black-White Residential Segregation (Indicator 2) data for the following two cities:
South Fulton, GA (recently incorporated in 2017)
Jurupa Valley, CA (recently incorporated in 2011)
This is due to the fact that the Federal Reserve Bank of Chicago Peer City Identification Tool Dataset does not produce this data for recently incorporated cities. Due to this missing data, we decided to exclude these cities.
We’ve also recently discovered that our initial city identification query only looked for cities and not large towns. This caused us to unintentionally exclude:
Gilbert, AZ (2019 POP: 254,109)
Cary, NC (2019 POP: 170,568)
Davie, FL (2019 POP: 106,486)
We plan to include these in our next update.
All of the cities we used in our final list had both a FIPS state and FIPS Place code. The counties we chose had both a FIPS state code and a FIPS county code. These codes were then used to pull the data we needed from other sources.
To learn more about the data, feel free to download the data set or reach out to us with any questions in a chat.