Unemployment in the United States is measured through a monthly survey of about 60,000 households conducted by the U.S. Census Bureau on behalf of the Bureau of Labor Statistics. This survey, called the Current Population Survey, asks people a series of questions about their work activity to sort every adult into one of three categories: employed, unemployed, or not in the labor force. The headline unemployment rate you see in the news is the percentage of the labor force that falls into the “unemployed” category.
Who Counts as Unemployed
A person must meet all three of these criteria to be officially classified as unemployed:
- They don’t have a job. Any job counts, including part-time work. If you worked even one hour for pay during the survey’s reference week, you’re counted as employed.
- They actively looked for work in the past four weeks. “Actively” means something concrete: submitting applications, interviewing, contacting employers, or similar steps. Simply reading job listings doesn’t qualify.
- They’re currently available to start work. Someone who wants a job but can’t take one right now (because of a medical issue or school schedule, for example) wouldn’t be counted.
If you don’t have a job and aren’t looking for one, you’re classified as “not in the labor force.” This group includes retirees, full-time students, stay-at-home parents, and people who have stopped searching. The unemployment rate only measures those who are actively trying to find work but haven’t yet, as a share of the total labor force (which is employed people plus unemployed people combined).
How the Data Is Collected
The Current Population Survey contacts roughly 60,000 households each month, covering the civilian noninstitutional population age 16 and older. “Noninstitutional” means it excludes people in prisons, long-term care facilities, and active-duty military. Trained interviewers ask household members about their work activity during a specific reference week, typically the week that includes the 12th of the month. The answers determine each person’s classification.
This is a survey of people, not employers. That distinction matters because it captures groups that employer records miss: self-employed workers, agricultural workers, people in private households, and unpaid family workers. It also counts each person only once, even if they hold multiple jobs.
The Payroll Survey: A Second Data Source
The BLS also runs a separate survey called the Current Employment Statistics survey, sometimes called the payroll or establishment survey. This one goes directly to businesses and government agencies, sampling about 119,000 employers representing roughly 622,000 individual worksites each month. Instead of counting people, it counts nonfarm jobs. Someone who works two jobs shows up twice in this survey but only once in the household survey.
The payroll survey doesn’t measure unemployment at all. Its job is to track how many positions exist across industries, along with hours worked and earnings. When a jobs report says “the economy added 200,000 jobs last month,” that number comes from the payroll survey. When it says “the unemployment rate is 4.2%,” that comes from the household survey. The two surveys sometimes tell slightly different stories about the labor market because they’re measuring different things in different ways.
What the Headline Rate Misses
The official unemployment rate, known technically as U-3, has real blind spots. It doesn’t count someone who gave up looking for work after months of rejection. It doesn’t distinguish between a person happily working 10 hours a week and someone desperate for full-time hours but stuck with a part-time schedule. To fill these gaps, the BLS publishes six different measures of labor underutilization, labeled U-1 through U-6.
- U-1 counts only people who have been unemployed for 15 weeks or longer.
- U-2 counts only job losers and people who completed temporary jobs.
- U-3 is the official unemployment rate: total unemployed as a percentage of the civilian labor force.
- U-4 adds discouraged workers to U-3. Discouraged workers want a job and are available but have stopped looking specifically because they believe no work is available for them.
- U-5 adds all marginally attached workers, a broader group that includes discouraged workers plus anyone who wants a job, is available, and looked for work sometime in the past 12 months but not in the last four weeks.
- U-6 takes U-5 and adds people working part time for economic reasons, meaning they want full-time work but can only find part-time hours. This is often called the “real” unemployment rate because it captures the widest range of labor market pain.
U-6 typically runs several percentage points higher than U-3. When you hear debates about whether the unemployment rate understates the true jobs picture, U-6 is usually what people are pointing to as the more comprehensive alternative.
Seasonal Adjustment
Raw unemployment numbers swing predictably throughout the year. Retailers hire heavily before the holidays and let workers go in January. Construction slows in winter. Schools release teenagers into the job market every summer. If the BLS reported only the raw numbers, a normal January spike in layoffs might look like a recession, and a summer hiring surge might look like a boom.
To strip out these predictable patterns, the BLS applies seasonal adjustment using a statistical program called X-13ARIMA-SEATS. The process builds a model of each data series, identifies the recurring seasonal component, and removes it so that only the underlying trend and genuinely unexpected changes remain. The seasonally adjusted number is what gets reported as the headline figure each month. The BLS reruns this adjustment every month as new data arrive, which means previously published seasonally adjusted figures can be slightly revised.
Why the Measurement Matters for You
Understanding how unemployment is measured helps you interpret the monthly jobs report with more nuance. If the unemployment rate drops but the labor force also shrank, that could mean people stopped looking rather than finding jobs. If U-3 holds steady but U-6 is climbing, more people may be getting stuck in part-time work involuntarily. And because the household survey only requires one hour of paid work to count someone as employed, a falling unemployment rate doesn’t necessarily mean everyone has stable, full-time work.
The monthly jobs report typically comes out on the first Friday of each month. Both the unemployment rate from the household survey and the job-creation number from the payroll survey are released in the same report, published by the BLS. You can find the full data, including all six U-1 through U-6 measures, on the BLS website at bls.gov.