Department of Labor Logo United States Department of Labor
Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Occupational Employment Statistics
PRINT:Print
OES OES Program Links

Technical Notes and Survey Method and Reliability Statement for 1998 OES Estimates

Technical Notes

Employment Estimates

Employment represents the estimate of total wage and salary employment in an occupation across the industries in which it was reported. The OES survey form sent to an establishment contains between 50 and 225 OES occupations. The number of occupations listed on a form depends on the industry classification and size class of the sampled establishments. To reduce paperwork and respondent burden, no survey form contains every OES occupation.

Wage Estimates

Wages for the OES survey are straight-time, gross pay, exclusive of premium pay. Included are base rate, cost-of-living allowances, guaranteed pay, hazardous-duty pay, incentive pay including commissions and production bonuses, and on-call pay. Excluded are back pay, jury duty pay, overtime pay, severance pay, shift differentials, nonproduction bonuses, and tuition reimbursements.

Annual Wage

Most employees are paid at an hourly rate by their employers and may work less than or more than 40 hours per week. The annual wage estimates on this website are calculated by multiplying the hourly wage estimates by a "year-round, full-time" hours figure of 2,080 hours per year (52 weeks by 40 hours). Thus, the annual wage estimates may not represent the actual annual pay received by the employee. There are a small number of occupations where hourly wages are not published. For these occupations the annual wages have been directly calculated from the reported survey data. The workers in these occupations are paid based on an annual amount, but generally work less than the usual 2,080 hours per year. Since the survey does not collect the actual hours worked, the hourly rate cannot be calculated with a reasonable degree of confidence from the annual wages. Occupations that typically have a work-year of less than 2,080 hours include musical and entertainment occupations, flight attendants and pilots, and teachers.

Mean Hourly Wage

The mean hourly wage is the estimated total wages for an occupation divided by its weighted survey employment.

Median Hourly Wage

The median hourly wage is the estimated 50th percentile of the distribution of wages; 50 percent of workers in an occupation earn wages below, and 50 percent earn wages above the median wage.

Relative Standard Error (RSE)

The particular sample used in this survey is one of a large number of all possible samples of the same size that could have been selected using the same sample design. Estimates derived from different samples would differ from each other. The variance of a survey estimate is a measure of the variation among the estimates from all possible samples. The standard error of a survey estimate is the square root of its variance; the relative standard error is the ratio of the standard error to the estimate itself.

The sample estimate and its standard error allows us to construct an interval estimate with a prescribed level of confidence that the interval will include the mean value of the estimates from all possible samples.

To illustrate, if all possible samples were selected, and if each of these were surveyed under essentially the same conditions, and an estimate and its estimated sampling error were calculated from each sample, then:

  • Approximately 90 percent of the intervals from 1.6 standard errors below to 1.6 standard errors above the derived estimate would include the average value of the estimates from all possible samples. This interval is called a 90-percent confidence interval.


  • Approximately 95 percent of the intervals from two standard errors below to two standard errors above the derived estimate would include the average value of the estimates from all possible samples. This interval is called a 95-percent confidence interval.

For example, suppose that an estimated occupational employment total is 5,000 with an associated relative standard error of two percent. Based on this data, the standard error of the estimate is 100 (= 5,000 X 0.02) and the 95-percent confidence interval for the estimate is (5,000 + 200) or (4,800 to 5,200). This confidence interval is one of many that could be constructed based on the same sample design. Approximately 95 percent of these confidence intervals would encompass the average value of the estimates from all possible samples.




Survey Method and Reliability Statement for the 1998 OES Survey All-Industry Wage Rate Estimates

General

The Occupational Employment Statistics (OES) survey is an annual mail survey measuring occupational employment and wage rates for wage and salary workers in nonfarm establishments, by industry. The survey samples approximately 400,000 establishments per year, taking 3 years to fully collect the sample of 1.2 million establishments. BLS and the Employment and Training Administration (ETA) provide the funding for the survey. BLS provides the procedures and technical support, while the State Employment Security Agencies (SESAs) collect the data. The SESAs produce occupational estimates by detailed industries for local areas and the states. BLS produces similar industry-specific estimates for the nation as well as employment and wage estimates for 750 occupations across all industries for the nation, each of the 50 states plus the District of Columbia, and Metropolitan Statistical Areas (MSAs).

Survey Definitions and Concepts

Many of the concepts and definitions used in the OES Survey are comparable to those in the Current Employment Statistics survey, a monthly BLS payroll survey of nonagricultural establishments. Many others, however, are unique to this survey. Key definitions are as follows:

An establishment is an economic unit, such as a factory, mine, or store, which produces goods or services. It is generally at a single location and engaged predominantly in one economic activity.

The OES survey defines employment as the number of workers who can be classified as full-time or part-time employees; workers on paid vacations or other types of leave; workers on unpaid short-term absences; salaried officers, executives, and staff members of incorporated firms; employees temporarily assigned to other units; and employees for whom the reporting unit is their permanent duty station regardless of whether that unit prepares their paycheck. The survey excludes the self-employed, owners/partners of unincorporated firms, and unpaid family workers. Employees are reported in the occupation in which they are working, not necessarily for which they were trained.

Employment represents the estimate of total wage and salary employment in an occupation across the industries in which it was reported. The OES survey form sent to an establishment contains between 50 and 225 OES occupations. The number of occupations listed on a form depends on the industry classification and size class of the sampled establishments. To reduce paperwork and respondent burden, no survey form contains every OES occupation.

The OES classification system uses seven occupational divisions to categorize workers in one of 750 detailed occupations. The seven divisions are as follows:

1. Managerial and administrative occupations;
2. Professional, paraprofessional, and technical occupations;
3. Sales and related occupations;
4. Clerical and administrative support occupations;
5. Service occupations;
6. Agriculture, forestry, fishing, and related occupations;
7. Production, construction, operating, maintenance, and material handling occupations.

Wages for the OES survey are straight-time, gross pay, exclusive of premium pay. Included are base rate, cost-of-living allowances, guaranteed pay, hazardous-duty pay, incentive pay including commissions and production bonuses, and on-call pay. Excluded are back pay, jury duty pay, overtime pay, severance pay, shift differentials, nonproduction bonuses, and tuition reimbursements.

The OES survey collects wage data in 11 intervals. Employers report the number of employees in an occupation per each wage range. The wage intervals used for the 1998 survey are as follows:

Interval Hourly Wages Annual Wages
Range A Under $6.75 Under $14,040
Range B $6.75 to $8.49 $14,040 to $17,659
Range C $8.50 to $9.99 $17,660 to $20,779
Range D $10.00 to $11.24 $20,780 to $23,399
Range E $11.25 to $13.24 $23,400 to $27,559
Range F $13.25 to $15.74 $27,560 to $32,759
Range G $15.75 to $19.24 $32,760 to $40,039
Range H $19.25 to $24.24 $40,040 to $50,439
Range I $24.25 to $43.24 $50,440 to $89,959
Range J $43.25 to $60.00 $89,960 to $124,820
Range K $60.01 and over $124,821 and over

Hourly versus annual wage reporting: For each occupation, respondents are asked to report the number of employees paid within specific wage intervals. The intervals are defined both as hourly rates and the corresponding annual rates, where the annual rates are constructed by multiplying the hourly wage rate for the interval by the typical work year of 2,080 hours. In reporting, the respondent can reference either the hourly or the annual rate, but is instructed to report the hourly rate for part-time workers.

Annual wage: Most of the annual mean wage estimates in this release are calculated by multiplying the mean wage by a "year-round, full-time" hours figure of 2,080 hours per year (52 weeks by 40 hours). Most employees are paid at an hourly rate by their employers and may work less than or more than 40 hours per week. Thus, the annual wage estimates may not represent the actual annual pay received by the employee. There are a small number of occupations where only an annual wage figure is provided. The workers in these occupations generally work less than the usual 2,080 hours per year. Since the survey does not collect the actual hours worked, the hourly rate cannot be calculated with a reasonable degree of confidence from the annual wages. For these occupations, therefore, only the annual salary is reported, which has been calculated directly from the data (rather than by multiplying an hourly figure by 2,080 hours). Occupations that typically have a work-year of less than 2,080 hours include musical and entertainment occupations, pilots and flight attendants, and teachers.

The Unemployment Insurance (UI) Address File is a micro-level employer file prepared quarterly by each State's Employment Security Agency and submitted to the Bureau of Labor Statistics. For 1998, the file from the second quarter of 1997 is used as a sampling frame while the fourth quarter of 1998 is used as a source of population values for employment (the second quarter of 1998 is used as a source of population employment values for New Jersey).

Industry classifications are based on the 1987 Standard Industrial Classification Manual, Office of Management and Budget, 1987. Industry is classified on the basis of the major product or activity of the establishment, as determined by total sales or receipts of the calendar year prior to classification.

Scope of Survey

The survey included private establishments in SIC codes 07, 10, 12-17, 20-42, 44-65, 67, 70, 72, 73, 75, 76, 78-84, 86, 87, and 89 covering agricultural services; mining; construction; manufacturing; transportation and public utilities; wholesale and retail trade; finance, insurance, and real estate; and services. The survey also covered private and government establishments in SIC codes 806, 821, 822, 824, and 829, the Postal Service (SIC 43), as well as all remaining state and local government establishments. Data for the Postal Service are universe counts obtained from the United States Postal Service (USPS). Federal government data are obtained from the Office of Personnel Management (OPM); these data exclude information from selected agencies.

The reference date of the 1998 survey was the week that included October 12, November 12, or December 12 of 1998. The reference date for a particular establishment in this survey is dependent on its two-digit SIC code. See the table below.

Reference Date Industries Surveyed
October 12 07, 15-17, 41, 46, 50-62, 67, 70, 73, 79, 84
November 12 26-28, 30, 35, 36, 40, 42, 45, 47, 48, 63-65, 75, 76, 78, 80, 81, 83, 86, 87, 89
December 12 10, 12-14, 20-25, 29, 31-34, 37-39, 44, 49, 72, 82, and state and local governments

Sampling Procedures

The sampling frame for this survey was the list of establishments which reported to the state Unemployment Insurance (UI) files for the two-digit SICs listed above. For the 1998 survey, the frame's reference date was the second quarter of 1997. This frame was supplemented with a list supplying establishment information on Railroads (SIC 401).

Establishments in the universe were stratified by Metropolitan Statistical Area (MSA), three-digit SIC, and size of firm (i.e., size class). Size classes were defined as follows:

Size class Number of employees
1 1 to 4
2 5 to 9
3 10 to 19
4 20 to 49
5 50 to 99
6 100 to 249
7 250 to 499
8 500 to 999
9 1,000 or more

In 1996 and 1997, establishments in size classes 2 to 6 were selected based on a probability sample. The sampling weights in size class 2 were adjusted to account for the employment in size class 1. In 1998, the OES Survey began sampling establishments in size class 1; thus, establishments in all size classes are now represented in the probability sample. UI reporting units with 250 or more employees are sampled with certainty across the three year cycle of the survey. Approximately one third of these units are selected within each MSA/SIC/Size class each year. The above allocation resulted in a total initial sample size of 409,347, 408,805, and 400,405 UI reporting units or establishments for 1996, 1997, and 1998. The combined initial sample size for 1996, 1997, and 1998 is 1,206,964 UI reporting units or establishments. (Note that the combined sample size is not a simple sum of the three year's samples. Some State government establishments are included in the survey each year. In the tabulations for the combined survey these establishments are only included once, from the most recent year. Federal government units are also included in the combined tabulation.)

Method of Collection

Survey schedules were initially mailed to virtually all sampled establishments. Personal visits, however, were made to some of the larger establishments.

Two additional mailings were sent to nonresponding establishments at approximately three week intervals. Telephone follow-ups and, in some cases, personal visits were made to nonrespondents considered critical to the survey because of their size.

Response

Subsequent to the close-out date for National estimates, additional data were collected by the states and used to prepare their own estimates. Consequently, the response rates in most states are higher than the response rate used to develop estimates of all-industry wage rates for each MSA.

Estimation Methodology

The OES survey samples approximately 400,000 establishments each year and, over a 3-year period, contacts approximately 1.2 million establishments. Each single-year sample represents one-third of both the certainty and non-certainty strata for the full 3-year sample plan. While estimates can be made from a single year of data, the OES survey has been designed to produce estimates using the full 3 years of data. The full 3-year sample allows the production of estimates at fine levels of geography, industry, and occupational detail, while estimates using any one year of data would be subject to a higher sampling error (due to the smaller sample size) and the limitations associated with having only 1/3 of the certainty units. Producing estimates using the 3 years of sample data provides significant sampling error reductions (particularly for small geographic areas and occupations); however, it also has some quality limitations in that it requires the adjustment of earlier years' data to the current reference period—a procedure referred to as "wage updating."

The 1996 OES survey estimates, which were published in December 1997, were from the first year of the new OES wage survey and were developed using only a single year (i.e., 400,000 sample units) of data. The initial estimation methodology used a weighting-class adjustment procedure for nonrespondents and an employment benchmark at the state/industry level. Since multiple years of data were not available for the 1996 estimates, the estimation procedure did not involve "wage updating."

The 1997 OES survey estimates represent the second year of OES estimates and have been developed using both the 1996 and 1997 surveys. The 1997 estimates also represent the first year of using a "wage-updating" methodology in developing the OES survey estimates. In addition to the wage-updating procedure, the 1997 estimates used an improved estimation methodology, utilizing a "nearest neighbor" imputation approach for nonrespondents and applying employment benchmarks at a detailed MSA by 3-digit industry and broad size-class level. A variant of the imputation procedure is also used to account for item nonresponse. Note: Because of the difference in estimation methods for these first 2 years of OES estimates, the data from 1997 are not strictly comparable with those published from 1996.

The 1998 OES survey estimates are developed from the full three years of the OES sample. The combined 1996, 1997, and 1998 data cover approximately 1.2 million sample units. The 1998 estimates use the wage-updating methodology introduced in 1997, which uses the over-the-year fourth-quarter wage changes from the Bureau's Employment Cost Index to adjust prior years' data before combining them with data from the current year. In addition, the 1998 estimates use the estimation methodology introduced in 1997, which uses a "nearest neighbor" imputation approach for nonrespondents and applies employment benchmarks at a detailed MSA by 3-digit industry and broad size class level.

The wage-updating procedure is used to adjust prior year wages to reflect increases between the previous data and current year data. For wage-updating purposes, the Bureau has used the national over-the-year wage changes from the fourth quarter of 1996 to the fourth quarter of 1997 and from the fourth quarter of 1997 to the fourth quarter of 1998 for the nine occupational divisions for which ECI estimates are available. These factors are applied to both the 1996 and 1997 survey data to update them to the fourth-quarter 1998 level before combining them with the 1998 survey data. Such a procedure assumes that each occupation's wage, as measured in the earlier years, moves according to the average movement of its occupational division and that there are no major geographic or detailed occupational differences—and this may not be the case. Research is being conducted to develop procedures that may account for differences in the rate of change at more detailed levels, than the nine ECI occupational divisions.

The hot deck (nearest neighbor) imputation procedure imputes for unit nonresponse. This type of nonresponse occurs when a unit reports no employment data. In hot decking, units in the sample are stratified into 'year/State/4-digit industry/size class' cells. Within each cell, a donor (i.e., responding unit) is selected to represent each nonrespondent under the proviso that a donor can not be selected twice. The sampling frame employment is used to match donors with nonrespondents. Once a donor and nonrespondent are matched, the occupational employment totals from the donor are copied over to the nonrespondent. In the event that a donor is not available at the 'year/State/4-digit industry/size class' cell level, the procedure advances to succeeding higher level cells until a donor is found.

Occasionally a responding establishment may provide employment information, but omit wage distribution information for selected occupations. The OES survey currently uses a variation of the mean imputation procedure to impute for item nonresponse. This type of nonresponse occurs when a unit reports the total-employment for its occupations but not the corresponding employment by wage intervals. In this procedure, units in the sample are stratified into 'year/MSA/3-digit industry/size class' cells. A wage-employment distribution is then calculated for those occupations with missing wage-employment based on the usable data in the cell. Missing wage-employment is imputed using the just calculated wage-employment distribution to prorate the total-employment of those occupations with missing wage-employment.

A separate ratio estimator is used to develop estimates of occupational employment in each wage interval. The auxiliary variable is the population value of total employment obtained from the refined Unemployment Insurance files for the 1998 reference month. Within each MSA, the estimated employment for an occupation at the reported three-digit SIC/wage interval level was calculated by multiplying the weighted employment by its ratio factor. The estimated employment for an occupation at the all-industry level was obtained by summing the occupational interval employment estimate across all industries within an MSA reporting that occupation. A further adjustment to each occupational employment total was made as described in the Reliability of the Estimates section. This adjustment did not affect the mean or median wage rates. The employment and wage data for federal government workers in each occupation were added to the survey derived data.

A mean wage and a median wage are calculated using wage data from establishments in the industries that reported employment for an occupation.

Mean wage is the estimated total wages for an occupation divided by its weighted survey employment. For the upper open-ended wage interval, a Winsorized mean procedure is used to estimate the mean wage. That is, the mean wage value for the upper open-ended wage interval is set at its lower bound ($60.01). For the other intervals, a mean wage value was calculated based on occupational wage data collected by the Office of Compensation and Working Conditions. These interval mean wage values are then attributed to all workers reported in the interval. For each occupation, total weighted wages in each interval (i.e., mean wages times weighted employment) are summed across all intervals and divided by the occupation's weighted survey employment to obtain a mean wage.

Median wage is the estimated 50th percentile of the distribution of wages; 50 percent of workers in an occupation earn wages below, and 50 percent earn wages above the median wage. The wage interval containing the median wage is located using a cumulative frequency count of employment across wage intervals. After the targeted wage interval is identified, the median wage rate is then estimated by a linear interpolation procedure

Reliability of the Estimates

The occupational wage rates in this report are estimates derived from a sample survey. Two types of errors are possible in an estimate based on a sample survey - sampling error and nonsampling error. Sampling error occurs because the observations are based on a sample, not on the entire population. Nonsampling error is due to response, nonresponse, and operational errors.

Nonsampling Errors—Estimates are subject to various response, nonresponse, and operational errors during the survey process. Sources of possible errors are data collection, response, coding, transcription, data editing, nonresponse adjustment, and estimation. These errors would also occur if a complete census was conducted under the same conditions as the sample survey. Explicit measures of their effects are not available. However, it is believed that the important response and operational errors were detected and corrected during the review and validation process.

The employment total and wage data for the occupation reflects only those industries that reported the occupation. This occurs primarily in those industries where the occupation appeared on the survey form. Since every occupation does not appear on every industry-specific form, there may be a bias in the employment and wage data for some occupations. The extent of this bias is unknown.

Another source of potential bias is the limitations placed on the size of the benchmark factors. A benchmark factor is the ratio of a known employment value to a sample-derived employment estimate. This factor is used to make a post-stratification adjustment that makes the total weighted employment estimate at the state / three-digit SIC industry / Metropolitan Statistical Area (MSA) / employment size class level match the population employment at that level. The source of the population employment data is the states' Quarterly Unemployment Insurance files for the reference period of the survey. In cases where a small sample was taken, the ratio factor can become large or small. In order to prevent an establishment from contributing either too much or not enough to an MSA's wage rate estimates, the benchmark factor was not allowed to exceed a predetermined value. The total employment count for those MSAs where the benchmark factor was limited by this ceiling will be biased to a small degree in those strata. The employment not assigned to those strata because of this ceiling was then distributed across the other MSAs in the state / three-digit industry, so that the estimated employment of the State / three-digit industry would match the known employment totals at that level.

Sampling Errors—The particular sample used in this survey is one of a large number of possible samples of the same size that could have been selected using the same sample design. For example, occupational wage rate estimates derived from the different samples will differ from one another. The deviation of a sample estimate from the average of all possible sample estimates is called the sampling error. The standard error of an estimate is a measure of the variation of estimates across all possible samples and thus is a measure of the precision with which an estimate from a particular sample approximates the average result of all possible samples.

Quality Control Measures

Quality control measures implemented in the OES survey include:

  • review of the specific occupations to be collected for each industry, and those to be collected in residual categories
  • creating and validating the sample frame for all states at BLS-Washington
  • allocating and selecting the sample for all states at BLS-Washington
  • follow up solicitations of nonrespondents (especially critical nonrespondents)
  • review of survey schedules to verify the accuracy and reasonableness of the reported data
  • adjustments of atypical reporting units on the data file
  • validation of the nonresponse adjustment factors
  • validation of the population employment and ratio factors
  • standardized data processing programs and activities

1998 Occupational Employment and Wage Estimates

1998 National Occupational Employment and Wage Estimates

1998 State Occupational Employment and Wage Estimates

1998 Metropolitan Area Occupational Employment and Wage Estimates

Last modified: October 16, 2001