Data Literacy: Navigating Sources for Employment Data

Employment and unemployment are important economic indicators – we use them as part of our set of indicators and include them on our Regional Dashboard – and the unemployment rate is an oft-cited statistic when it comes to a community’s economic well-being. But despite how widely discussed they are, employment and unemployment data can be tricky to interpret, and even the unemployment rate has a pitfall or two for the unwary data user. Also, the number of sources for this data can make it difficult to narrow down which is ideal for your project. So for our fourth data literacy post, we’re exploring these pieces of data, their common sources, and their common problems so you can read, discuss, and interpret them like a pro. Or at least like a savvy amateur.

How is the unemployment rate defined, and what exactly is the “labor force”?

The Bureau of Labor Statistics (BLS) defines the unemployment rate as the percentage of total labor force that is unemployed[1]. To break this down further, it’s important to understand exactly what’s meant by “labor force” and, for that matter, what’s meant by “unemployed.”

“Unemployed persons” also has a BLS definition: “All persons who had no employment during the reference week, were available for work, except for temporary illness, and had made specific efforts to find employment some time during the 4 week-period [sic] ending with the reference week.”[2] To paraphrase this, the unemployed population is made up of individuals who currently have no job, but who want and are looking for a job.

The labor force includes individuals who are employed, and individuals who do not have a job but who are looking for one.[3] Individuals who do not have a job and are not currently looking for a job are not counted as among the labor force. This includes groups like full-time students choosing not to work while pursuing their degrees, stay-at-home parents and other family caregivers, retired persons, and individuals who may have been looking for a job previously, but who have since given up.

Where do different employment and unemployment rate statistics come from?

Part of what makes employment and unemployment statistics challenging to interpret is that they can be found from several sources, which can vary in timeframe, level of geography, and methodology, resulting in data that isn’t always comparable.

The American Community Survey

The U.S. Census Bureau offers employment and unemployment data as part of the American Community Survey (ACS), broken down by a variety of demographic and economic factors: age, race and ethnicity, sex, educational attainment, disability status, household or family type, poverty status, and more. Tables are available in 1-Year and 5-Year Estimates datasets, and geographic availability depends, as it always does with ACS data, on the dataset selected. Tables from the 1-Year Estimates dataset are available for areas with a population of 65,000 or greater; tables from the 5-Year Estimates dataset are available for all areas regardless of population. As we’ve discussed in previous blog posts, ACS data is comprised of rolling estimates: a statistic from the 2015 1-Year Estimates dataset is reflective of all of 2015, and a statistic from the 2011-2015 5-Year Estimates dataset is reflective of all five years between 2011 and 2015.

Employment status data is also available from the U.S. Census Bureau’s Decennial Census up to 2000, as part of the long-form Census questionnaires. It’s not available as part of the 2010 Decennial Census: by 2010, the long-form questionnaires had been replaced by the rolling ACS estimates.

For a more concrete comparison of the types of data offered by the various sources, we’ll take a look at the five most recent unemployment rate figures published via the ACS for Champaign County.

Table: Unemployment Rate of Champaign County: ACSDownload table data for Unemployment Rate of Champaign County: ACS.

Source: U.S. Census Bureau; American Community Survey, 2015 American Community Survey 1-Year Estimates, 2014 American Community Survey 1-Year Estimates, 2013 American Community Survey 1-Year Estimates, 2012 American Community Survey 1-Year Estimates, 2011 American Community Survey 1-Year Estimates, Table S2301; generated by CCRPC staff; using American FactFinder;(4 August 2017).

As we see in the table above, the ACS offers estimates on an annual basis. And it’s a viable source for this example because we’re looking at all of Champaign County. However, if we were interested in an area that falls under the 65,000 population threshold for 1-Year Estimates (like the City of Urbana, for example), or if we wanted data updated even more frequently, we might want to look elsewhere.

The U.S Bureau of Labor Statistics

The U.S. Bureau of Labor Statistics (BLS) offers unemployment data via the Local Area Unemployment Statistics program. As one might assume from the name, this dataset primarily offers unemployment data, including unemployment rates, change in unemployment rates, and overall employment status. Unlike the ACS data, it doesn’t come associated with those other economic or demographic factors: you can find the unemployment rate of a given area at a given time, but you can’t connect it to the same area’s poverty rate or educational attainment, or find the unemployment rate for a certain age bracket.

The major advantages of LAUS data are its publication frequency, its historical range, and its wide variety of geographic areas. LAUS data is published on a monthly basis, and is available online back to 1994. It’s available for states, metropolitan areas, metropolitan divisions, micropolitan areas, combined statistical areas (CSAs), counties and county equivalents, and cities and towns with a population greater than 25,000.

There are also options for geographies that might otherwise be difficult to find data for. If your area of interest is a city (which meets or exceeds the minimum population of 25,000) that falls partly into two different counties, and you want the unemployment rate for only the part of the city that’s in one county, it’s available from the LAUS. If your area of interest is a metropolitan or micropolitan statistical area that falls into multiple states, and you’re interested only in the part of the area that’s in one of the states, that’s available too. For Illinois, the LAUS also offers data for the “Balance of Illinois” area: the remainder of the state without the Chicago area.

But for right now we’re interested, again, in Champaign County, for the purposes of comparing what’s available. Remember, we’re not trying to compare the published unemployment rates in the following table with those in the above table – for one thing, they’re not comparable, as we’ll discuss more later. We’re just looking at the different types of data offered by the different sources.

Table: Unemployment Rate of Champaign County: BLSDownload table data for Unemployment Rate of Champaign County: BLS.

Source: U.S. Department of Labor; Bureau of Labor Statistics; Local Area Unemployment Statistics, “Databases, Tables & Calculators by Subject,” Series ID LAUCN170190000000003; retrieved by CCRPC staff; from <>; (4 August 2017).

Here, we again have the five most recent data points for the Champaign County unemployment rate from the source of interest, but this time going back five releases doesn’t even get us out of 2017. The level of granularity offered by monthly publications can be great for some projects, but may be more detail than is needed for others. Annual data – from other BLS tables, or from other sources – may be as much as you need.

The Illinois Department of Employment Security

The Illinois Department of Employment Security (IDES) offers a wide variety of data products. Some are available nationally from the BLS and Census Bureau, while others are exclusive to Illinois and IDES. Many of these data products and programs are organized into the Illinois Virtual Labor Market Information dashboard. They offer data in a variety of timeframes (monthly, quarterly, and annually), on a variety of topics (employment, unemployment, payroll, wages, commuting, and more), and at a variety of levels of geography (state, county, MSA, Economic Development Region, Local Workforce Investment Area, municipality, and legislative district). Not every program offers its data at all geographic levels or publishes at all frequencies, and most specialize by topic.

Below are the five most recent unemployment figures for Champaign County as available from IDES.

Table: Unemployment Rate of Champaign County: IDESDownload table data for Unemployment Rate of Champaign County: IDES.

Source: Illinois Department of Employment Security; Economic Information and Analysis Division; U.S. Department of Labor, Local Area Unemployment Statistics, “Monthly LAUS Reports”; “Illinois Monthly Labor Force Report”; retrieved by CCRPC staff; (4 August 2017).

Notice anything about these numbers? Like how they’re completely identical to those shown in the table for the Bureau of Labor Statistics? There’s a good reason for that – the IDES data is also pulled from the LAUS. This means that even though it comes from a different source agency and webpage, it is exactly the same data, redistributed by IDES through a different channel and in a different report format.

It’s important to pay attention to where data comes from – not just the direct source, but any indirect ones, whether there’s a common program like LAUS or a citation that indicates that the person or agency doing the analysis is not the person or agency that collected the data. Knowing more about the origin of the data you’re discussing, presenting, or writing about makes you a more credible source as you pass the data along.

How are they comparable and not comparable?

The range of data products and sources available present comparability challenges in both timeframe and level of geography.

When you’re comparing employment or any other data, it’s important that the timeframes do not overlap or nest. Just as it’s not appropriate to compare overlapping American Community Survey 5-Year Estimates datasets (e.g., the 2010-2014 unemployment rate with the 2011-2015 unemployment rate), it’s not appropriate to compare a statistic from a shorter timeframe (e.g., April 2015) to a statistic from a longer timeframe that it falls within (e.g. full year 2015), even if they’re from the same program and reporting agency.

Geographic comparability is just as important. As we discussed above, there are many levels of geography with associated data, and not all of them may be familiar to a beginning data user. Obviously, we all know that states are made up of counties, and counties are not shared across state lines. However, MSAs can fall into multiple counties, and can include multiple municipalities, some or all of which might meet the BLS minimum population of 25,000 and have their own, single-municipality data available. The other less familiar geographies, Economic Development Regions (EDRs) and Local Workforce Investment Areas (LWIAs), are both multi-county areas within Illinois, and, unlike with MSAs, counties are not shared between multiple EDRs or LWIAs. Although they share those characteristics, we can’t compare, for example, occupational wages from an EDR to occupational wages from an LWIA: they’re two separate geographic classifications, and in many cases they overlap with each other. Illinois has 10 EDRs, but 26 LWIAs, so the EDR that our county of interest falls within is highly unlikely to include exactly the same group of counties as its LWIA.

The important thing to remember when starting to work with employment data is to make sure you’re looking for the right dataset, time, and geography for your specific project, whether you know what that is at the outset or you have to do a little trial and error along the way. Not every data source will have every statistic that you want in your area of interest and for your period of study, but with the variety that’s available, there will almost certainly be a source or program that meets the needs of your project or research question. So we advise spending some time at the beginning of the process, winnowing down the above list of sources until you find the one that best fits your project, rather than cherry-picking data points from several sources and trying to force them to be comparable later on. If you start your data gathering process knowing what you’re looking for, and you can avoid being overwhelmed by the number of options available, your research process can be easier and, dare we say it, fun.

[1] United States Department of Labor. Bureau of Labor Statistics. 2016. “BLS Information: Frequently Asked Questions (FAQs). (Accessed 4 August 2017).

[2] Ibid.

[3] Ibid.

Related Documents

Leave a Reply

Please be respectful. All fields are required, and all comments are subject to moderation. URLs are not allowed.