Chapter 3.
Occupational Employment Statistics

Estimating Procedure

Occupational employment estimates
During the sample selection process, each sampled establishment is assigned a sampling weight that is equal to the reciprocal of its probability of selection. For example, if an establishment on the sampling frame had a 1 in 10 chance of being selected into the sample, then its sampling weight is 10.

For establishments that did not respond to the survey, a nonresponse adjustment factor (NRAF) is calculated and applied against the sampling weights of the responding establishments within each State (Nation)/3-digit industry/size class cell. Multiplying NRAFs by sampling weights increases the weight of the responding establishments so they can account for the missing employment data of the nonresponding establishments.

A ratio estimator is used to calculate estimates of occupational employment at the State (Nation)/3-digit industry/size class cell level. For each size class within an industry and State (Nation), the occupational distribution is estimated by calculating a ratio for each occupation. This ratio is the sum of the total weighted employment of an occupation to the sum of the total weighted employment of all responding establishments. These ratios are then multiplied by a known total employment figure (i.e., a benchmark value) for that size class. Higher level estimates of occupational employment are obtained by summing these lower level employment estimates. For example, occupational employment estimates at the three-digit industry level are obtained by summing up size class employment estimates.

Variance estimates of the occupational employment estimates
The OES survey uses a subsample replication technique called the "jackknife random group" to estimate the variance of occupational employment at the 3-digit industry/size class level. In this technique, R subsamples are formed from the parent sample. Next, R estimates of total employment are calculated for each occupation, one employment estimate per subsample. The variability of the R employment estimates for each occupation is calculated and used as an estimate of the variance for each occupation.

Higher level variance estimates of occupational employment are obtained by summing these lower level variance estimates.

Quality control measures
A major goal of a cooperative program like the OES survey is to accommodate State-specific publication needs with limited resources yet standardize the survey procedures across all 50 States and the District of Columbia, while at the same time producing quality employment estimates. Controlling sources of nonsampling error in this decentralized environment can be particularly difficult. Two important quality control tools employed by the OES survey are the Survey Processing and Management (SPAM) system and the Estimates Delivery System (EDS). Both systems were developed to provide a consistent and automated framework for survey processing and to reduce the workload at the State, regional, and national levels.

By standardizing data processing activities such as validating the sample frame, allocating and selecting the sample, refining mailing addresses, addressing envelopes and mailers, editing and updating questionnaires, producing management reports, and calculating employment estimates, across all States, the SPAM system and the EDS have also standardized survey methodology. This has reduced the number of errors on the data files as well as the time needed to review them.

Next: Presentation