• Overview
• Concepts
• Data sources
• Design
  • Selecting sample establishments (stage 1)
  • Probability sampling of occupations within sampled establishments (stage 2)
• Calculation
• Presentation
• History
• More Info

• Methods Home
• About
• Programs
• Glossary
search

Handbook of Methods Occupational Requirements Survey Design

Occupational Requirements Survey: Design

Occupational Requirements Survey (ORS) data are collected from a national probability sample selected in two stages:  (1) a probability sample of establishments and (2) a probability sample of occupations (PSO)within sampled establishments. Probability samples are subject to sampling and nonsampling errors, which are discussed in the Calculation section.

Selecting sample establishments (stage 1)

In stage 1, the ORS uses a probability proportional to size (PPS) technique to select a sample of private industry and state and local government establishments from across the nation. The larger the establishment, the greater its chance of being selected. Establishments from all 50 states and the District of Columbia are eligible for selection. ORS stratifies by 23 major industry groups and ownership (private industry and state and local government) and are implicitly stratified within each sampling cell for 24 geographic areas. The 24 geographic areas represent the 15 largest metropolitan areas and the 9 census divisions. More detailed information on ORS sample design can be found in the sample design portion of the research section on the public ORS website.

Each establishment is assigned a six-digit industry code using the North American Industry Classification System (NAICS). When a single physical location encompasses two or more distinct economic activities, the industry code assigned is based on the establishment's principal product, or group of products, produced or distributed, or services rendered. When determining the principle product or service rendered, revenue generated or employment are used to determine the primary business activity and assign an industry code. 

The sampling frame, or universe, is the list of establishments from which the survey sample is selected. The ORS establishment sample is drawn from the Quarterly Census of Employment and Wages (QCEW) and units reporting to the Railroad Retirement Board.

Industry classification of establishments. All federal statistical agencies currently use NAICS to classify survey establishments by industry. NAICS is revised every 5 years to stay current with industrial organization in North America. In selecting new establishment samples, the ORS currently uses NAICS 2012 as one of the stratification variables.

Sample groups. The 2017 estimates are from two samples of data collected from the Occupational Requirements Survey. The ORS is an establishment-based survey and uses a national sample design. To maximize the amount of publishable information, the BLS is combining data across three annual ORS samples to produce the 2018 estimates. The number of publishable occupations and the level of occupational detail is expected to increase with the addition of each subsequent year’s sample until the full three-year ORS sample size of up to 26,500 sampled establishments is reached in 2018. Because the ORS combines data across three annual samples selected independently, there is a possibility that an establishment will be reselected in a subsequent sample. However, ORS data are not collected from the same establishment in more than one sample group. In this case, the data from the initial collection are used.

Probability sampling of occupations within sampled establishments (stage 2)

The ORS collects data about requirements of jobs from sampled establishments. In this stage (stage 2), field economists use a four-step process to select and classify jobs for which data are to be collected from the sampled establishment.

Step 1: Field economists receive the establishment’s complete list of employees and their job titles and perform the probability selection of occupations (PSO) technique. The field economists use the PSO technique to randomly select the jobs for which data are to be collected. This process ensures that the probability of selecting a given job is proportional to the number of workers in the job at the establishment. The number of jobs selected for data collection is based on the establishment’s employment size, according to the following criteria:

Exceptions include state and local government units, for which up to 20 jobs may be selected.

Step 2: Field economists classify the sampled jobs into occupations based on the workers’ actual job duties and responsibilities, not on their job titles or specific education. For example, an employee trained as an engineer, but working as a drafter, is reported as a drafter. An employee who performs the duties of two or more distinct occupations is reported as working in the occupation that requires the highest level of skill or in the occupation in which the employee spends the most time if there is no measurable difference in skill requirements. Each sampled job is classified by the Standard Occupational Classification (SOC) system to the six-digit level, and further designated by an eight-digit code in the Occupational Information Network’s (O*NET) detailed occupational taxonomy, when available. These are referred to as O*NET-SOC 2010 occupations. This code is part of a hierarchical structure as shown in the following exhibit.

O*NET-SOC 2010 occupations are grouped under and include the 840 Standard Occupational Classification (SOC) detailed occupations. SOC detailed occupations are grouped under broad occupations; broad occupations are part of a minor group, and minor groups are part of a major group. The example above shows the hierarchy of ‘architectural drafters’ and ‘civil drafters’ O*NET-SOC 2010 occupations.

The SOC designates 23 major groups and there are 1,110 O*NET-SOC 2010 occupations within these 23 groups. For the purposes of the ORS, occupations can fall into 22 major groups and 1,090 occupations; only the major group designating military-specific occupations is excluded (code 55-0000.00 and detailed occupations within this major group).

Step 3: Identification of occupational attributes of the worker in the sampled job, such as full-time or part-time status, union or nonunion status, and whether the work is paid on a time or incentive basis. The field economist records specific occupational attributes of the worker in the sampled job. For definitions of occupational attributes see the Concepts section.

Step 4: Field economists evaluate the job to determine the work level of its duties and responsibilities using a point-factor system of points based on the following factors:

• Knowledge
• Job controls and complexity
• Contacts
• Physical environment

Each factor consists of several points and a description. The duties and responsibilities of the job, along with consideration given to work performed and the skills, education, and training required for the job are evaluated. Points are then totaled to determine the overall work level for the job. Generally, the greater the impact, complexity, or difficulty of the factor, the higher the number of points assigned, and the higher the work level. As the following exhibit shows, there are some occupations that cannot be “leveled,” because for the following jobs points cannot be determined for all four factors, thus points are not assigned and a level cannot be determined.