General Notes

This report provides statistics on research and development (R&D;) funding for the years 1982-92 and on R&D; personnel for the period from January 1983 to January 1994. The statistics are produced from results of the annual Survey of Industrial Research and Development and provide national estimates of the total expenditures on R&D; performed within the United States by industrial firms, whether U.S. or foreign owned. It is a sample survey that intends to include or represent all R&D-performing; companies, either publicly or privately held. Every year a survey questionnaire is sent to all companies known, through previous surveys or through outside information sources, to spend more than $1 million annually on R&D; in the United States or to have 1,000 or more employees. Remaining firms are subjected to probability sampling and may or may not receive a questionnaire for a given survey year.

Industry statistics are developed from data collected from individual companies or enterprises. Since the survey is enterprise based rather than establishment based, all data collected for the various subparts of each enterprise (plants, divisions, or subdivisions) are tabulated in the major standard industrial classification (SIC) of the company. The resulting industry estimates are reported using the SIC of the companies within each industry. National totals are estimated by summing the industry estimates.

Several changes have been made to the survey recently that are of special importance to users of this report. Prior to the 1992 survey, statistics were based on samples selected at irregular intervals (i.e., 1967, 1971, 1976, 1981, 1987). In intervening years a subset of the last sample (called a "panel") was used. The most recent sample prior to the 1992 survey was for survey year 1987. Estimates for 1988 through 1991 published in previous reports were based on surveys of approximately 1,700 panel companies that reported R&D; activity in the 1987 survey. Beginning with the 1992 survey, statistics are based on samples selected annually. Also, beginning with the 1992 survey, the sample size was increased from approximately 14,000 to nearly 23,400 firms to better account for births of R&D-performing; establishments in the survey universe, to survey more fully and accurately R&D; performed by nonmanufacturing firms especially in the service sector and small firms in all industries, and to gather more current information about potential R&D; performers. In this report, tables containing historical statistics are presented two ways. For the tables in section A, estimates from the 1992 survey are linked with estimates from the 1987 survey. The linking was accomplished using an algorithm that preserved to the greatest extent possible year-to-year trends for each industry. A full explanation of the linking process is included in section B, under "Comparability of Statistics." Also in section B is a series of tables for which no attempt was made to link the estimates derived from data collected in the 1992 survey to estimates derived from data collected in previous surveys.

The Bureau of the Census, Department of Commerce, has conducted the annual Survey of Industrial Research and Development for the National Science Foundation (NSF) since 1957. Census staff conduct the survey under Title 13 of the United States Code, which prohibits publication or release of data or statistics that may reveal information about individual companies. Therefore, in some tables of this report the symbol "(D)" is used as a footnote reference to indicate that estimates are being withheld to avoid possible disclosure of information about operations of individual companies.

The tables containing statistics developed from the survey data are in section A. Detailed information about the history of the survey, survey methodology, comparability of the statistics, and survey definitions are in the technical notes in section B, and survey questionnaires, instructions, and other survey documents are reproduced in section C. Specific questions regarding the survey may be directed to Raymond Wolfe at (703) 306-1772, via e-mail at rwolfe@nsf.gov (Internet), or at the following mailing address:

R&D; Statistics Program
Division of Science Resources Studies
National Science Foundation
4201 Wilson Boulevard, Suite 965
Arlington, VA 22230