What is I-O?

In this column, I want to take this opportunity to inform you about two things on the Washington, DC front that I believe will be of interest to I-O psychologists—the soon-to-be released O*NET (the Occupational Information Network), and the revised Standard Occupational Classification System (SOC). Following up on my last column, I will also report some statistics on the SIOP Program composition.

In 1993, the U.S. Department of Labor initiated a project to provide a comprehensive occupational information system that would be accessible to government, industry, and workers and would replace the Dictionary of Occupational Titles (DOT) as the nation’s primary source of occupational information. Primary goals of this initiative were to develop a new occupational information system that would (a) promote the effective training, counseling, and employment of the American workforce; (b) provide a database that identifies, defines, classifies, and describes occupations in the economy in an accessible and flexible manner; and (c) provide a common language that serves as a national benchmark for all users of occupational information. A prototype of the new Occupational Information Network (O*NET) database, which runs on a Windows-based personal computer, is due to be released this fall.

At the heart of O*NET is a content model that specifies various types of occupational information and serves as the framework for the system. Unlike the original DOT, which was based on descriptions of the tasks workers perform, the O*NET incorporates more comprehensive descriptions of worker and job attributes; it also allows for describing jobs at both cross-job and job-specific levels. With respect to across-job descriptors, the O*NET system contains the following types of information:

• Person Requirements (skills and knowledge areas),
• Person Characteristics (abilities, interests, and values),
• Experience Requirements (training, licensure, experience),
• Job Requirements (generalized work activities, physical and social work context factors, organizational context factors), and
• Labor Market Characteristics (occupational outlook, wages, and pay).

The across-job variables were systematically taxonomized, based on the psychological and job analytical literature. In addition, the across-job variables can be used to organize more specific job descriptive information such as occupation, specific tasks, skills, knowledges, and so forth, and thereby link job-specific information to the broader, across-job common language framework.

The U.S. Department of Labor Employment and Training Administration would like to encourage the I-O community to start using the O*NET tool and provide them with feedback on its application to our needs. Also, they will soon be implementing an extensive data collection plan to complete the O*NET database with information gathered from on-the-job workers. Members of SIOP who hold decision-influencing positions within organizations could directly impact O*NET development by encouraging participation in the data collection effort. More information will be forthcoming on this initiative.

A complete description of development of the O*NET model and system can be found in technical reports available from the National Center for O*NET Development, North Carolina Employment Security Commission, P.O. Box 27625, Raleigh, NC 27611; e-mail: O*NET@esc.state.nc.us. Also, APA is publishing a book describing the O*NET which will be available in January (Peterson, Mumford, Borman, Jeanneret, & Fleishman, in press). Additional and current information about O*NET can be obtained by contacting O*NET Project, U.S. Department of Labor, ETA/Office of Policy and Research, 200 Constitution Avenue NW, Room N5636, Washington, DC 20210; (202) 219-7161; or at www.doleta.gov/programs/onet.

At present, there are a numerous occupational classification systems in use within the Federal Government, states, and throughout the country in private organizations. This large number of parallel, but incompatible, systems makes comparing jobs and occupations across organizations very difficult. To address this issue, several federal agencies that use occupational classification systems (e.g., the Office of Personnel Management, Department of Labor, Department of Defense Manpower Data Center, Bureau of Labor Statistics, Bureau of the Census, the EEOC, among others) have cooperatively developed a standardized system for classifying jobs—the revised Standard Occupational Classification System (SOC). This system can be used to classify all occupations in the economy, including public, private, and military occupations.

The SOC proposes 810 occupational codes and uses four levels of aggregation: major group, minor group, broad occupation, and detailed occupation. The SOC differs from O*NET in that it contains job titles and brief definitions of the occupations within each code, whereas O*NET contains more extensive job descriptive information, as described above.

The SOC will be used by all federal agencies collecting occupational data to provide a means to compare occupational data across agencies. For example, O*NET will use the SOC set of occupational codes as its occupational structure for sampling and data collection purposes. Similarly, all state and local government agencies are being strongly encouraged to use this national system to promote a common language for categorizing occupations in the United States.

The 1998 SOC is published in two volumes. The first volume lists all occupations and definitions within the structure and provides illustrative examples of specific jobs associated with each title in each occupation. The second volume contains a comprehensive list of job titles in each occupation and the industries in which they are located. Volume II will be updated and maintained in electronic format to ensure that currency and comparability is maintained. Federal Register notices and related documents describing the SOC can be obtained from the Bureau of Labor Statistics web site at status.bls.gov/soc/soc home.htm. Inquiries about the SOC or requests for electronic copies of the SOC structure can be directed to: Laurie Salmon, Standard Occupational Revision Policy Committee, Bureau of Labor Statistics, Room 4840, Washington, DC 20212; (202) 606-6511.

In my last column, I mentioned that several SIOP conference attendees commented that they would like to see more "O" sessions on the SIOP program and this led me to wonder about the composition of the program overall. To explore this issue, Mike Burke, the 1999 SIOP Program Chair, compiled some information on "I" versus "O" program sessions that I wanted to report back to you.

While there are a number of different ways one could examine the balance of "I" versus "O" program sessions, Mike chose to provide information on content areas grouped according to "Industrial," "Organizational," "Methods," and "Other." The "Industrial" category included content areas such as selection, testing, training, job analysis, job performance, performance appraisal, compensation, HR management, and job attitudes. The "Organizational" category included motivation, culture/climate, leadership, organizational change, turnover, decision making, groups/teams, intergroup conflict, participation, socialization/careers, work and family, gender/diversity, aging, ethics, and stress. The "Methods" category included statistical techniques and research methodology content areas. A few other sessions did not fit well into these general content categories and were classified as "Other."

Counts and Percentages for Types of Sessions (Industrial, Organizational, Methods, and Other) on the 1998 SIOP Program were as follows:

Counts and Percentages of Acceptance Rates (Acceptances/Submissions) by Types of Sessions were:

The overall acceptance rate for all submissions was 62%.

In general, the 1998 program appeared to be reasonably well balanced between "I" versus "O" sessions, and there did not appear to be much of a difference in acceptance rates for "I" versus "O" submissions. Of course, depending on how one classifies the submissions into the content areas, I versus O representation could be interpreted differently. If, for example, Methods sessions are considered "I" type sessions, the conclusion would be that there were fewer "O" sessions overall. On the other hand, there were 25 acceptances (i.e., symposia, panels, posters) for the "job attitudes" content area. If these acceptances had been included in the "O" as opposed to the "I" category, the above numbers and percentages for I versus O would be almost identically reversed.

While these data suggest a reasonably balanced program, I also again encourage anyone who would like to see different content or more of a particular content area represented to develop and submit program proposals in those areas.

Peterson, N. G., Mumford, M. D., Borman, W. C., Jeanneret, P. R., & Fleishman, E. A. (Eds.). (in press). An occupational information system for the 21st century: The development of O*NET. Washington, DC: American Psychological Association.