A Demographic Comparison of Unemployment in Lane CountyFebruary 1, 2018 As the result of a long and sustained economic recovery, unemployment rates are near all-time lows for much of the state. Underneath the surface of low unemployment rates, however, there’s still quite a bit of variability. To learn more, I took a closer look at how unemployment varies by demographic group in Lane County.
This information is useful for several reasons. Perhaps the economic recovery isn’t affecting everyone equally or employment programs aren’t reaching everyone effectively. It may also be helpful for employers to understand where there is still a labor pool to draw from in a low-unemployment economy.
For statistical purposes, unemployment is a clearly defined category: one must be out of work, available for work, and have made specific efforts in the last four weeks to find a job. Some examples of people who are not counted as unemployed: a full-time student who doesn’t have time in their schedule for a job, a stay-at-home dad who intends to get a job at some point in the future, or a retiree who receives Social Security benefits.
In this graph, I used American Community Survey (ACS) data to look at unemployment rates by demographic group in Lane County. These ACS estimates are based on surveys drawn from five-year samples, and often lead to slightly different results from the Local Area Unemployment Statistics (LAUS) program data. LAUS data are what most people think of as the current unemployment rate. If you want the most recent and accurate regional unemployment estimates, look at our LAUS data releases on Quality Info.
So why use ACS data? Because with the ACS you can look more closely at differences by group. In this graph I compared the most recent data (2012-2016) with the most recent non-overlapping years (2007-2011). This isn’t ideal, because 2007-2011 contains the period before, during and immediately after the Great Recession and 2012-2016 encompasses a steady period of recovery, but it’s the most statistically sound comparison that can be made with these data.
In the chart, you can compare unemployment rates of different groups or by looking at the change in one demographic group over time.
Margins of error are important to interpreting this data; without them, we can’t say if the data support the conclusion of a statistical difference between two categories. The categories with yellow stars by them show a statistically significant change from 2007-11 to 2012-16. Gray stars indicate a statistical difference with the Lane County labor force as a whole in 2012-16. (A note on data: Demographics within a given age bracket were compared with the unemployment rate for that age bracket; i.e., males aged 20 to 64 were compared with all people aged 20 to 64).
So for example, even though the estimated unemployment rate for 16 to 19 year olds increased from 2007-11 to 2012-16, we can’t say if that’s a statistically significant change or just noise in the data. Data do support the finding of a decrease in unemployment over time for those with some college or an associate’s degree.
These data don’t support the conclusion that Hispanic or Latino Lane County residents were more likely in 2012-16 to be unemployed than the overall population, but do support the conclusion that people with a disability and people age 16 to 24 were more likely to be unemployed.
People with a disability and people with poverty level income were not asked about in 2007-11 so a comparison over time isn’t possible.
These data can’t tell us everything, but they can reinforce some familiar findings about unemployment:
Unemployment declined significantly from 2007-11 to 2012-16, both overall and for certain demographic groups. No group saw a statistically significant increase.
Young adults in Lane County were more likely to be unemployed (remember, this if for those actively looking and available for a job; college students who are anxious about their future career prospects don’t count as unemployed).
Having a disability or having income below poverty level were associated with statistically higher unemployment rates.
Education levels matter for employment: those with no more than a high school education were more likely and those with a bachelor’s degree or higher were less likely to be unemployed.