The American Community Survey (ACS) conducted by the United States Census Bureau provides estimates of the characteristics of the population over a specific time period. The ACS collects data from the 50 states, Washington, DC, and Puerto Rico, where it is called the Puerto Rico Community Survey. It is a continuous survey, in which each month a sample of housing unit addresses receives a questionnaire, with approximately three million addresses surveyed each year. Each year the survey produces data pooled to produce 1-year, 3-year, and 5-year estimates for geographic areas in the United States and Puerto Rico, ranging from neighborhoods to congressional districts to the entire nation. Data for each release of the 5-year estimates were collected over a 5-year period ending December 31 of the reference year (e.g., data in the 2014-2018 5-year estimates were collected January 1, 2014 - December 31, 2018). The statistics reported represent the characteristics of the population for the entire period vs. a specific year within that period.
The 5-year estimates are published for areas with populations of all sizes and are the most reliable and precise of the ACS period estimates as well as the most comprehensive, albeit the least current. (The 1-year and 3-year estimates provide data on areas with populations of 65,000+ and 20,000+, respectively. Note that the ACS 3-year estimates were discontinued with the 2011-2013 release.) The ACS estimates provide information about the social and economic needs of communities and are used to help determine how more than $400 billion in federal and state funds are distributed each year. It is conducted under the authority of Title 13, United States Code, Sections 141 and 193. Note that counts of the population are provided by the Census of Population and Housing conducted by the U.S. Census Bureau every 10 years; and official estimates of the population are derived from the previous census and from the Census Bureau’s Population Estimates Program.
Source: U.S. Census Bureau. American Community Survey. Accessed May 25, 2020, https://www.census.gov/programs-surveys/acs/technical-documentation/summary-file-documentation.html.
What Do the ACS 5-Year Estimates Tell Us?
The 600+ individual datasets that make up the American Community Survey contain over 21,000 sociodemographic indicators. Estimates are available for nations, states, counties, census tracts, metropolitan and micropolitan statistical areas, place, and block groups, resulting in hundreds of thousands of data points. Data Planet allows you to view estimates separately by indicator and geography and to compare and contrast estimates across geographies. You can also compare the indicators with indicators in other datasets that you find in the repository.
The Data Planet repository holds the 5-year estimates for releases since the 2010 U.S. Census of Population and Housing. Currently available is the 2023 release, covering the period January 1, 2018 - December 31, 2022, as well as the seven prior releases.
Remember This! The estimates represent data collected over the entire 5-year period covered by the survey, not to a specific point of time within that period.
American Community Survey Sample Embed
Locating ACS Indicators in Sage Data
The Data Planet platform provides options to both browse and search for indicators by category and subject.
To browse the available American Community Survey, 5-year estimates, select the dataset in the "Subject" → "Population and Income" listing, or select it in the "United States Census Bureau" entry in the "Source" tab as shown in this screen grab:
Open up a listing of interest in the indicator tree and review the indicators. The titles below in the "Education" subject category give a sense of the granularity of the data:
Those familiar with the ACS may prefer to use the search box and do a precise search by ACS table number.
Alternatively, do a search on "American Community Survey" to retrieve a listing of the ACS indicators. Select "Show All Results" and filter by ACS release, geography, and subjects to narrow your results set.
From the results set, you can link directly from a record back into the repository to view the data and chart selected.
Viewing and Exporting ACS Indicators in Sage Data
Create maps, trends, rankings of the indicators using the options in the tool bar. The chart below compares the number of native-born, foreign-born naturalized, and noncitizen Hispanic or Latino males over age 18 in three states:
Create maps of the data by selecting the Map icon in the tool bar above the chart. Below find a map of Mississippi counties by numbers of persons ages 19-64 living with a disability that do not have health insurance coverage:
Below the chart on the right, you will see either a DOI or an option to "Create DOI link." The DOI (Digital Object Identifier) ensures you'll be able to retrieve an exact view of the data at the time it was created. For more on DOIs, visit the page on it.
From the chart, you can download the download the data, export a DataSheet with a statistical abstract describing the indicator, export to a reference manager, or further manipulate the data.
Manipulating ACS Indicators
Data Planet makes it possible to compare and contrast multiple estimates and geographies.For more information on using Data Planet Statistical Datasets, see http://data-planet.libguides.com/statisticaldatasets.
To select multiple indicators, you can simply tick off the checkboxes alongside the indicators. The chart below compares four indicators: the numbers of males and females ages 25-39 and 40-64 whose bachelor's degree major was Science and Engineering. The criteria selection shows two of the four indicators selected to create the chart:
Below is the resulting chart:
You can also create pie charts comparing geographies. For example, the chart featured in the embed below compares the aggregate number of hours worked by workers ages 16-64, by counties within the state of Colorado:
You can also create charts that show rankings. Below, the chart ranks the top 10 counties in Oklahoma with at least one computing device:
Try it yourself with other indicators—the possibilities are myriad!
Remember This! Keep in mind that the graphs you create do not necessarily imply causality: the results may suggest a potential relationship between the variables you select, which may be an interesting line of inquiry for your own research.
ACS Margin of Error Statistics
Certain metrics are published with the American Community Survey (ACS) results to help the U.S. Census Bureau and data users assess the accuracy and reliability of the estimates released in the survey. For example, a statistic called the “margin of error (MOE)” is published with each estimate. The MOE indicates the likelihood that the ACS sample estimate is within a certain range (the MOE) of a true population value. For the ACS, MOEs are provided at a 90 percent confidence level, which means that the estimate is expected to contain the true or population value within a range defined by the associated MOE 90 percent of the time.
Sage Data has published MOE values with ACS estimates for American Community Survey, 5-year results for 2013-2017 to current.
Presentation of ACS Margin of Error Statistics in Sage Data
As you explore the American Community Survey statistics in Data Planet, you will notice that the data values present an estimated statistic and a margin of error (MOE) statistic for most indicators. The MOE indicates the likelihood that the ACS sample estimate and the actual population value differ by no more than the value of the MOE within a specific confidence interval (more on the confidence interval below). Let's take a closer look at the MOE in Data Planet.
The default view of a chart using ACS data in Data Planet is a ranking view. For example in the chart below, two indicators have been selected (counts of males under age 6 with vs. without health insurance) for South Carolina counties. With two indicators selected, two items per county (represented by horizontal bars) appear in the chart—both of which are showing the number of persons, which is the number of male children under 6 years of age with and without health insurance coverage:
If you're only interested in the estimated count of population, this default view works for you. However, if you are interested in seeing the MOE data displayed, look at the far-left side of the criteria panel featuring the "Plot Options" and select "Margin of Error +/-" instead:
Pro Tip! Click on Table to open up a view of data in tabular format—the pop-out window that appears not only provides the table view, but also presents options to download and export the data as Excel, CSV, JSON, and XML!
Calculating and Confidence Interval
As mentioned above, the MOE indicates the likelihood that the ACS sample estimate and the actual population value differ by no more than the value of the MOE, at a certain confidence level. For the ACS, the Census Bureau sets this level at 90%, meaning the the estimate is expected to contain the true or population value within a range defined by the associated MOE 90 percent of the time.
Let's look at an example: We find in B19019 that the ACS estimate for Median Household Income in 3-person households reported in B19019 for the state of Delaware is $84,232 with an MOE of +/-$1,752:
By adding and subtracting the MOE from the estimate, we can calculate the 90 percent confidence interval for that estimate—meaning we expect the true population value to be within this range 90 percent of the time:
Pro Tip! In Data Planet, we can use the Calculator Tool to calculate the range defined by the MOE for an estimate of interest, using a simple addition function:
The chart below shows the calculated Upper Limit of the Confidence Interval, the Lower Limit of the Confidence Level, and the estimate:
For more on MOEs and the confidence levels of ACS statistics, see the Census Bureau’s presentation on MOEs.
This dataset provides annual estimates of personal income and employment for states and counties in the United States. These estimates are developed as part of the Bureau of Economic Analysis Regional Economic Accounts program. Estimates of compensation and earning by industry and place of work indicate the economic activity of establishments within the local area. Estimates of personal income by place of residence provide a measure of fiscal capacity and an indicator of the economic well-being of the residents of an area. The county estimates of personal income are designed to be conceptually and statistically consistent with the national estimates of personal income in the BEA National Income and Product Accounts (NIPA) dataset. Differences between the NIPA estimates and the regional accounts estimates are due to differences in coverage and timing of the availability of source data; eg, the NIPA measure of personal income is broader than county personal income. The state and county personal income and employment estimates are based primarily on administrative records data, and also use some survey and census data.
For more information, visit the BEA website.
Clink on the links below to view DataSheets comprising statistical abstracts complete with infographics of metrics associated with personal income and employment in US states and counties. Enter Data Planet Statistical Datasets to further manipulate the data and explore relationships between these statistics and other indicators in the vast Data Planet repository. These examples are just a sampling to give you an idea of the granularity of the available data. There are many, many more indicators on this topic in the repository - use additional search terms or contact your library staff for help in formulating your search strategy.
Per Capita Personal Income - This chart ranks the top 10 states with respect to per capita personal income in 2022.
Average Wage and Salary - This graph ranks North Carolina counties with respect to average wage per job. Durham leads.
State and Local Personal Income and Employment - This chart shows the trend in annual personal income for Arkansas County, Arkansas.
Personal Income and Earnings by Industry - Presents annual estimates of farm proprietors' income and farm earnings in Kansas.
Compensation of Employees by Industry - Presents annual estimates of compensation of employees in the retail trade industry in Utah.
Value Added by Outdoor Recreation - Maps states by estimated total value added by outdoor recreation in 2022.
Personal consumption expenditures by state
Using BEA Local Areas Statistics in Sage Data
In Sage Data, the indicators included in the BEA Regional Personal Income and Employment dataset can be viewed individually to examine the many data points that comprise the time series. Statistics can be examined by state, county, and income and employment indicator. Below shows the trend in compensation of employees in education in Kentucky over time.
You can also compare statistics for states, counties and economic indicator. The example below shows the geography selections made to generate a sample graph below showing the trend over time in average wage per job in Montgomery and Baltimore counties, Maryland, vs Washington, DC.
Below is the resulting graph:
Indicators can also be compared to other indicators in the repository, such as Gross Domestic Product by State, etc. Keep in mind that the graphs you create do not necessarily imply causality: the results may suggest a potential relationship between the variables you select, which may be an interesting line of inquiry for your own research.