Technical Note
This news release presents statistics from two major surveys, the
Current Population Survey (household survey) and the Current Employ-
ment Statistics survey (establishment survey). The household survey
provides information on the labor force, employment, and unemploy-
ment that appears in the "A" tables, marked HOUSEHOLD DATA. It is a
sample survey of about 60,000 households conducted by the U.S. Cen-
sus Bureau for the U.S. Bureau of Labor Statistics (BLS).
The establishment survey provides information on employment, hours,
and earnings of employees on nonfarm payrolls; the data appear in the
"B" tables, marked ESTABLISHMENT DATA. BLS collects these data each
month from the payroll records of a sample of nonagricultural busi-
ness establishments. The sample includes about 140,000 businesses and
government agencies representing approximately 410,000 worksites and is
drawn from a sampling frame of roughly 8.9 million unemployment in-
surance tax accounts. The active sample includes approximately one-
third of all nonfarm payroll employees.
For both surveys, the data for a given month relate to a particular
week or pay period. In the household survey, the reference period is
generally the calendar week that contains the 12th day of the month.
In the establishment survey, the reference period is the pay period
including the 12th, which may or may not correspond directly to the
calendar week.
Coverage, definitions, and differences between surveys
Household survey. The sample is selected to reflect the entire
civilian noninstitutional population. Based on responses to a series
of questions on work and job search activities, each person 16 years
and over in a sample household is classified as employed, unemployed,
or not in the labor force.
People are classified as employed if they did any work at all as
paid employees during the reference week; worked in their own business,
profession, or on their own farm; or worked without pay at least 15
hours in a family business or farm. People are also counted as employ-
ed if they were temporarily absent from their jobs because of illness,
bad weather, vacation, labor-management disputes, or personal reasons.
People are classified as unemployed if they meet all of the follow-
ing criteria: they had no employment during the reference week; they
were available for work at that time; and they made specific efforts
to find employment sometime during the 4-week period ending with the
reference week. Persons laid off from a job and expecting recall need
not be looking for work to be counted as unemployed. The unemployment
data derived from the household survey in no way depend upon the eli-
gibility for or receipt of unemployment insurance benefits.
The civilian labor force is the sum of employed and unemployed per-
sons. Those not classified as employed or unemployed are not in the
labor force. The unemployment rate is the number unemployed as a per-
cent of the labor force. The labor force participation rate is the
labor force as a percent of the population, and the employment-popula-
tion ratio is the employed as a percent of the population. Additional
information about the household survey can be found at www.bls.gov/
cps/documentation.htm.
Establishment survey. The sample establishments are drawn from pri-
vate nonfarm businesses such as factories, offices, and stores, as
well as from federal, state, and local government entities. Employees
on nonfarm payrolls are those who received pay for any part of the
reference pay period, including persons on paid leave. Persons are
counted in each job they hold. Hours and earnings data are produced
for the private sector for all employees and for production and nonsu-
pervisory employees. Production and nonsupervisory employees are defin-
ed as production and related employees in manufacturing and mining and
logging, construction workers in construction, and nonsupervisory em-
ployees in private service-providing industries.
Industries are classified on the basis of an establishment’s princi-
pal activity in accordance with the 2007 version of the North American
Industry Classification System. Additional information about the estab-
lishment survey can be found at www.bls.gov/ces/#technical.
Differences in employment estimates. The numerous conceptual and
methodological differences between the household and establishment
surveys result in important distinctions in the employment estimates
derived from the surveys. Among these are:
--The household survey includes agricultural workers, the self-
employed, unpaid family workers, and private household workers
among the employed. These groups are excluded from the
establishment survey.
--The household survey includes people on unpaid leave among the
employed. The establishment survey does not.
--The household survey is limited to workers 16 years of age and
older. The establishment survey is not limited by age.
--The household survey has no duplication of individuals, because
individuals are counted only once, even if they hold more than one
job. In the establishment survey, employees working at more than
one job and thus appearing on more than one payroll are counted
separately for each appearance.
Seasonal adjustment
Over the course of a year, the size of the nation's labor force and
the levels of employment and unemployment undergo regularly occurring
fluctuations. These events may result from seasonal changes in weather,
major holidays, and the opening and closing of schools. The effect of
such seasonal variation can be very large.
Because these seasonal events follow a more or less regular pattern
each year, their influence on the level of a series can be tempered by
adjusting for regular seasonal variation. These adjustments make non-
seasonal developments, such as declines in employment or increases
in the participation of women in the labor force, easier to spot. For
example, in the household survey, the large number of youth entering
the labor force each June is likely to obscure any other changes that
have taken place relative to May, making it difficult to determine if
the level of economic activity has risen or declined. Similarly, in
the establishment survey, payroll employment in education declines by
about 20 percent at the end of the spring term and later rises with
the start of the fall term, obscuring the underlying employment trends
in the industry. Because seasonal employment changes at the end and
beginning of the school year can be estimated, the statistics can be
adjusted to make underlying employment patterns more discernable. The
seasonally adjusted figures provide a more useful tool with which to
analyze changes in month-to-month economic activity.
Many seasonally adjusted series are independently adjusted in both
the household and establishment surveys. However, the adjusted series
for many major estimates, such as total payroll employment, employment
in most major sectors, total employment, and unemployment are computed
by aggregating independently adjusted component series. For example,
total unemployment is derived by summing the adjusted series for four
major age-sex components; this differs from the unemployment estimate
that would be obtained by directly adjusting the total or by combining
the duration, reasons, or more detailed age categories.
For both the household and establishment surveys, a concurrent
seasonal adjustment methodology is used in which new seasonal factors
are calculated each month using all relevant data, up to and including
the data for the current month. In the household survey, new seasonal
factors are used to adjust only the current month's data. In the es-
tablishment survey, however, new seasonal factors are used each month
to adjust the three most recent monthly estimates. The prior 2 months
are routinely revised to incorporate additional sample reports and re-
calculated seasonal adjustment factors. In both surveys, 5-year revi-
sions to historical data are made once a year.
Reliability of the estimates
Statistics based on the household and establishment surveys are
subject to both sampling and nonsampling error. When a sample rather
than the entire population is surveyed, there is a chance that the
sample estimates may differ from the "true" population values they
represent. The exact difference, or sampling error, varies depending
on the particular sample selected, and this variability is measured by
the standard error of the estimate. There is about a 90-percent chance,
or level of confidence, that an estimate based on a sample will differ
by no more than 1.6 standard errors from the "true" population value
because of sampling error. BLS analyses are generally conducted at the
90-percent level of confidence.
For example, the confidence interval for the monthly change in
total nonfarm employment from the establishment survey is on the order
of plus or minus 100,0001. Suppose the estimate of nonfarm employment
increases by 50,000 from one month to the next. The 90-percent confi-
dence interval on the monthly change would range from -50,000 to
+150,000 (50,000 +/- 100,0002). These figures do not mean that the
sample results are off by these magnitudes, but rather that there is
about a 90-percent chance that the "true" over-the-month change lies
within this interval. Since this range includes values of less than
zero, we could not say with confidence that nonfarm employment had, in
fact, increased that month. If, however, the reported nonfarm employ-
ment rise was 250,000, then all of the values within the 90-percent
confidence interval would be greater than zero. In this case, it is
likely (at least a 90-percent chance) that nonfarm employment had, in
fact, risen that month. At an unemployment rate of around 5.5 percent,
the 90-percent confidence interval for the monthly change in unemploy-
ment as measured by the household survey is about +/- 280,000, and for
the monthly change in the unemployment rate it is about +/-0.19 per-
centage point.
In general, estimates involving many individuals or establishments
have lower standard errors (relative to the size of the estimate) than
estimates which are based on a small number of observations. The pre-
cision of estimates also is improved when the data are cumulated over
time, such as for quarterly and annual averages.
The household and establishment surveys are also affected by
nonsampling error, which can occur for many reasons, including the
failure to sample a segment of the population, inability to obtain
information for all respondents in the sample, inability or unwill-
ingness of respondents to provide correct information on a timely
basis, mistakes made by respondents, and errors made in the collec-
tion or processing of the data.
For example, in the establishment survey, estimates for the most
recent 2 months are based on incomplete returns; for this reason,
these estimates are labeled preliminary in the tables. It is only
after two successive revisions to a monthly estimate, when nearly
all sample reports have been received, that the estimate is consi-
dered final.
Another major source of nonsampling error in the establishment
survey is the inability to capture, on a timely basis, employment
generated by new firms. To correct for this systematic underestima-
tion of employment growth, an estimation procedure with two compo-
nents is used to account for business births. The first component
excludes employment losses from business deaths from sample-based
estimation in order to offset the missing employment gains from
business births. This is incorporated into the sample-based esti-
mation procedure by simply not reflecting sample units going out
of business, but imputing to them the same employment trend as the
other firms in the sample. This procedure accounts for most of the
net birth/death employment.
The second component is an ARIMA time series model designed to
estimate the residual net birth/death employment not accounted for
by the imputation. The historical time series used to create and
test the ARIMA model was derived from the unemployment insurance
universe micro-level database, and reflects the actual residual net
of births and deaths over the past 5 years.
The sample-based estimates from the establishment survey are ad-
justed once a year (on a lagged basis) to universe counts of payroll
employment obtained from administrative records of the unemployment
insurance program. The difference between the March sample-based em-
ployment estimates and the March universe counts is known as a bench-
mark revision, and serves as a rough proxy for total survey error.
The new benchmarks also incorporate changes in the classification of
industries. Over the past decade, absolute benchmark revisions for
total nonfarm employment have averaged 0.3 percent, with a range from
-0.7 to 0.6 percent.
Other information
Information in this release will be made available to sensory im-
paired individuals upon request. Voice phone: (202) 691-5200; Federal
Relay Service: (800) 877-8339.
