Choosing between the ACS 1-Year, 3-Year & 5-Year for Head Start programs

By Vincent Rinehart

Data from the American Community Survey

The Census Bureau is releasing 2013 1-Year American Community Survey estimates today.

What is the American Community Survey?

From the Census website:

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. Information from the survey generates data that help determine how more than $400 billion in federal and state funds are distributed each year.

The ACS is critical to Head Start programs. It provides specific data regarding age, sex and income that can be used to estimate the number of Head Start eligible children in a service area. This data is provided at the country, state, county, community and even Zip Code level. With the release of 2013 1-Year ACS data, it’s a good time for programs to start updating their community assessments. However, there are two other 2013 ACS datasets yet to be released: the 2013 3-Year ACS in October and the 2013 5-Year ACS in December.

The difference between the 1-Year, 3-Year and 5-Year ACS

The Census Bureau explains the difference thoroughly. The basics are that the 1-Year ACS is more timely, but less accurate because it uses one year worth of collected data. The 5-Year ACS is less timely, but more accurate because it uses 5 years worth of data. Lastly, the 3-Year ACS is in between; more accurate than the 1-Year ACS and more timely than the 5-Year ACS. When it’s released, the 2013 5-Year survey will include data collected from 2009-2013. Going back 5 years makes it more accurate because it uses a larger sample size. The 2013 3-Year ACS will include data collected from 2011-2013, making it less reliable but more timely.

Which one should you use?

The 1-Year ACS and 3-Year ACS are not available for smaller communities. Even if a Head Start program has all three data sets available for their service area, I almost always recommend the 5-Year ACS be used for consistency and accuracy. It can be tempting to use whichever data set best supports your hypothesis or plan; so sticking with the most reliable data set every time is the safe and conservative thing to do so you don’t misrepresent the data. What I mean by “conservative” is that the 5-Year ACS will tend to understate any recent changes in a service area. But that tendency is also why a program may want to look at the 3-Year ACS or 1-Year ACS along with the 5-Year ACS.

Instances where this is useful are when a Head Start program’s service area is experiencing rapid change. The poverty rate for families with children under the age of 5 could change quickly in the case of a sudden, localized economic boom or depression. These same economic forces could cause the service area to experience a sudden net inflow or outflow of young families; perhaps coming to your service area in search of new jobs, or leaving it in search of jobs elsewhere.

An Example: The United States as a whole

Data from the American Community Survey

In the example at the top of this article, for the entire US, we have the poverty rate for families with children under the age of 5 for the 1, 3 and 5-Year ACS from 2009 to 2012. The 5-Year ACS shows a steady increase in the poverty rate, while the 3-Year ACS shows a slightly more drastic increase and the 1-Year shows a sudden jump from 2009 to 2010 followed by a convergence with the 3-Year ACS. This would indicate to me that the 2008 US Recession was registering very clearly in the 1-Year ACS. Meanwhile, the 2009 & 2010 3-Year and 5-Year ACS were both still reflecting data from prior to the recession. This causes an understatement of the poverty rate for those years. After a few years, the sudden bump in the poverty rate for 2009-2010 is captured and better represented in the 3 and 5-Year ACS. A steadier year to year poverty rate causes the 1-Year ACS data to level out. This is a perfect example of how sudden changes from one year to the next may show up in the different data sets released by the ACS.

Localized Examples

In my 5 years of working on Community Assessments and economic studies for Head Start programs I have seen instances where the 3 or even 1-Year ACS ought to be considered for planning purposes. The most extreme example of this are in regions experiencing economic booms. Communities in region of the Eagle Ford Shale and North Dakota oil region are notable examples of localized economic booms that have drastically effected the poverty rate, the number of families and thus the number of Head Start eligible children in a service area. I will detail these examples in a follow up article.

I am available to assist in your Head Start program’s Community Assessment and economic and demographic studies, planning and anlaysis. Please feel free to contact me here!

Service Area Demographics: Head Start Eligible Children

By Vincent Rinehart

Digging into the Census Bureau’s website for data can be difficult, especially for programs with a geographically large service area or a large funded enrollment. The best way to do this is to present the data for your service area as a whole, and then break it down into smaller chunks.

An example of how to do this is below for southeast Alaska’s American Indian/Alaska Native Program: Tlingit & Haida Head Start. The raw census data from the 5 year American Community Survey can be found here. The program covers hundreds of miles from north to south with 10 distinct, Census designated geographies. The census designated geographies, counties or boroughs in this case, can be combined together to create an overall estimate of the number of age eligible children for 2012, the latest year we have data for.

2012 5-Year ACS
Combined Tlingt & Haida Service Area
Total Population: 72,210
Population under the age of 5: 6.28%
Poverty rate for families with children under the age of 5: 15.02% (Nation at 17.7%)
Estimated Head Start Eligible: 273
Estimated eligible infants & toddlers: 409


A table is created to show the detailed break down of Census data for each county/borough.


I can then use this data to create a custom map. Each geographic area can be clicked to display basic Census data. Additionally, Head Start centers are plotted on the map.

This sort of geographic breakdown is also helpful for urban programs operating in densely populated areas. When working with data from these sort of programs it’s better to look at the service area by Zip Code or Census Tract instead of County/Borough. (Stay tuned for an example.)

These are examples of the basic tables and data that should be included in the demographics portion of your program’s Community Assessment. Combined with internal data and data from a program’s PIR, a Head Start program would have what it needs to do in depth analysis and plan for full enrollment.

I am available to gather and organize this data and guide your Head Start program through analysis, planning and writing of your Community Assessment. I have 6 years of experience working with chronically under enrolled programs. Please see my about page for more information on how to contact me.