nsf.gov - NCSES International Investment and R&D Data Link: 2004–07 - US National Science Foundation (NSF)
text-only page produced automatically by LIFT Text
Transcoder Skip all navigation and go to page contentSkip top navigation and go to directorate navigationSkip top navigation and go to page navigation
National Science Foundation National Center for Science and Engineering Statistics
Contents

General Notes

Data Tables

Appendix. Technical Notes

Suggested Citation, Acknowledgments



Francisco Moris,
Project Officer
(703) 292-4678
Research and Development Statistics Program

NCSES Home
International Investment and R&D Data Link: 2004–07

 


Appendix. Technical Notes

 

Scope of Linked Surveys

The three surveys included in this project are the National Science Foundation (NSF)/U.S. Census Bureau's (Census's) Survey of Industrial Research and Development (SIRD) and the Bureau of Economic Analysis's (BEA's) surveys of Foreign Direct Investment in the United States (FDIUS) and U.S. Direct Investment Abroad (USDIA). These surveys cover for-profit companies on a fully consolidated domestic (U.S.) basis and use comparable definitions and accounting guidance for research and development expenditures. Certain measurement differences in other variables do not significantly impair the comparability of the data. For example, R&D employment data from SIRD are based on full-time equivalents (FTEs), whereas R&D employment data from the FDIUS and USDIA surveys are based on counts of full-time and part-time employees who devote the majority of their time to R&D activities (see "Definitions"). Another difference between the surveys is that the NSF/Census data are collected on a calendar-year basis, whereas the BEA data are collected on a (company) fiscal year basis.

Differences in industry classification, however, remain to be investigated. The NSF/Census data classify companies based on industry distribution of payroll, whereas BEA data classify companies based on sales by industry. In this publication, tabulations with industrial sector or industry-specific data present information from a given survey source as indicated in appropriate tables.

SIRD

SIRD is a nationally representative sample of all for-profit companies with five or more employees, publicly or privately held, that performed R&D within the 50 United States and the District of Columbia. The primary focus of the survey is U.S. industry as a performer of R&D rather than as a source of funds. The reporting unit is the company, defined as a business organization of one or more establishments under common ownership or control.

SIRD statistics include total R&D, the portion of the total financed by the federal government, and the portion financed by the companies themselves or by other nonfederal sources, such as state and local governments or other industrial firms under contract or subcontract. Total R&D is also separated into the types of costs, including wages and fringe benefits of R&D staff, materials and supplies, depreciation, and other costs. Other statistics include R&D financed by domestic firms but performed outside the 50 U.S. states and the District of Columbia, R&D performed by organizations outside the firm, R&D performed in collaboration with other organizations, and the funds spent to perform energy-related R&D. It also provides information on R&D-performing firms, including domestic net sales, number of employees, number of R&D-performing scientists and engineers, geographic location where the R&D was performed, and R&D funds spent per R&D-performing scientist and engineer. For SIRD survey forms see "Questionnaire(s)" at http://www.nsf.gov/statistics/srvyindustry/. For SIRD methodology and sample information see the "Technical Notes" section in the 2004, 2005, and 2006–07 reports for Research and Development in Industry (2004, http://www.nsf.gov/statistics/nsf09301/; 2005, http://www.nsf.gov/statistics/nsf10319/; and 2006–07, http://www.nsf.gov/statistics/nsf11301/).

BEA surveys

BEA conducts the FDIUS and USDIA surveys annually and conducts more comprehensive FDIUS and USDIA benchmark surveys every 5 years. For FDIUS, 2004, 2005, and 2006 were annual survey years and 2007 was a benchmark year. For USDIA, 2004 was a benchmark year and 2005–07 were annual survey years. These surveys provide financial and operating data, including total R&D expenditures and R&D employment. The FDIUS survey covers U.S. (i.e., U.S.-located) affiliates of foreign multinational corporations, whereas the USDIA survey covers U.S. parent companies and their foreign affiliates. U.S. affiliates and U.S. parent companies must file on a fully consolidated domestic (U.S.) basis. For the purpose of the BEA international surveys, the United States includes the District of Columbia, Puerto Rico, and other U.S. territories and possessions. For methodology details on BEA surveys see http://www.bea.gov/methodologies/index.htm#international_meth.

Top of page. Back to Top

Link Methodology

General Observations

The linked data resulted from matching SIRD separately to the two BEA surveys (FDIUS and USDIA surveys) for a given survey year. (No linking was performed across years.) SIRD companies were matched to majority-owned U.S. affiliates from the FDIUS survey (tables 1–13 [year 2004], 26–38 [2005], 51–63 [2006], and 76–88 [2007]) and were then separately matched to parent companies of U.S. multinational companies (MNCs) from the USDIA survey (tables 14–25 [2004], 39–50 [2005], 64–75 [2006], and 89–100 [2007]). BEA matched data cover majority-owned, nonbank U.S. affiliates and parent companies of U.S. MNCs. The link between any two surveys was performed in two basic steps.

The first step was a computer match of the BEA records to Census's Business Register (BR), a database covering all U.S. businesses with paid employees. The BR includes companies and their establishments, and it contains records of company names, addresses, and other identifying information, as well as key economic data obtained from Census surveys and administrative records. The computer match was made using Employer Identification Numbers (EINs) that are reported in BEA surveys and included in the BR. EINs are used by companies and their establishments when they file federal and state payroll and income taxes.[1]

For companies that did not link in the computer match of EINs, other identification information, such as names and addresses, was used to link U.S. affiliates and U.S. parent companies to BR companies.

The second step was to link BEA identification codes (IDs) that were matched to the BR and link them to the SIRD using Census IDs obtained from the BR and used in SIRD processing. Additional steps were required to verify the matches, evaluate the accuracy of the matched data, and tabulate the matched data. The link also allowed identifying corrections or updates based on cross-survey information. Table A1 provides a summary of company counts.

TABLE A1. Summary of company counts
Company counts 2004 2005 2006 2007
Published SIRD (U.S.-located for-profit companies)
Companies in frame 1.9 million 1.9 million 1.8 million 1.9 million
Consolidated companies in sample 31,916 31,847 31,913 31,766
Unweighted number of company respondents with non-zero R&D performance 8,000 8,980 9,379 9,491
FDIUS survey (U.S. affiliates)
Consolidated U.S. affiliates 15,776 16,281 15,855 17,613
U.S. affiliates matched to Business Register 7,265 7,409 7,192 8,275
U.S. affiliates matched to SIRD 1,073 1,087 1,128 1,115
Unique SIRD companies matched to FDIUS 994 1,010 1,041 1,028
With non-zero R&D performance 907 962 977 967
USDIA survey (parent companies of U.S. MNCs)
Consolidated U.S. parent companies 3,483 3,331 3,317 3,350
U.S. parent companies matched to Business Register 3,199 3,133 3,103 3,091
U.S. parent companies matched to SIRD 1,527 1,557 1,648 1,695
Unique SIRD companies matched to USDIA 1,505 1,531 1,596 1,586
With non-zero R&D performance 1,497 1,499 1,546 1,512

FDIUS = Foreign Direct Investment in the United States; SIRD = Survey of Industrial Research and Development; MNC = multinational company; USDIA = U.S. Direct Investment Abroad.

NOTES: Link consisted of matching data across surveys for a given year. No linking was performed across years. Year-to-year variation in this table may reflect a combination of factors: changes in the multinational status of companies (e.g., acquisition or divestiture by a parent company), changes in R&D-active status (e.g., a company may be an R&D performer in one year but not the next), or changes in coverage (e.g., a company not captured by a given survey in one year because of size thresholds or other reasons [hence not matched], but sampled and matched in another year). See coverage and related information for each survey in "Scope of Linked Surveys." Data in this table were extracted from flowcharts developed as part of this project. These flowcharts are available upon request from the author.

SOURCE: National Science Foundation/National Center for Science and Engineering Statistics, U.S. Bureau of Economic Analysis, and U.S. Census Bureau. R&D Data Link Project, 2004–07.

2004

SIRD-FDIUS match

FDIUS-BR. The 2004 BEA FDIUS file contained information on 15,776 consolidated U.S. affiliates. Editing resulted in 12,942 U.S. affiliates, which were then matched to the BR using EINs. This resulted in 7,204 U.S. affiliates that were initially matched to the BR and 5,738 that were unmatched. U.S. affiliates that did not match initially to the BR and that had more than $1 billion in sales or had non-zero R&D expenditures were manually searched for in the BR by using the name and address information for the BEA affiliate. This process resulted in an increase to 7,265 U.S. affiliates matched to the BR. The Census ID was merged onto the FDIUS file for these matched IDs.

FDIUS-SIRD. The 7,265 U.S. affiliates matched to the BR were in turn matched with 32,065 individual SIRD records using the Census ID available in the two files.[2] This process resulted in three subsets: 1,073 affiliates matched to the SIRD file, 6,208 affiliates not in the SIRD file (all but 370 affiliates had no R&D expenditures), and 31,071 SIRD records not in the FDIUS file. Information on the 370 U.S. affiliates that performed R&D and were not included in the SIRD file was used to improve SIRD sampling, but these affiliates were otherwise excluded from the data tables of the link project. The 1,073 matched affiliates included 994 unique companies in the SIRD file and 1,057 unique companies in the FDIUS file as a result of differences in consolidation patterns. Of the 994 companies in the SIRD file, 907 had non-zero R&D performance in 2004.

SIRD-USDIA match

USDIA-BR. There were 3,483 consolidated U.S. parent companies from the BEA USDIA survey. This count was reduced to 3,395 after editing, and these companies were then matched to the BR. This resulted in 3,114 U.S. parent companies that were initially matched to the BR and 281 that were unmatched. U.S. parent companies that did not match initially to the BR and that had more than $1 billion in sales or had non-zero R&D expenditures were manually searched for in the BR by using name and address information for the BEA parent company. This process resulted in an increase to 3,199 BEA parent companies matched to the BR. The Census ID was merged onto the USDIA file for these matched parent companies.

USDIA-SIRD. The 3,199 U.S. parent companies matched to the BR were in turn matched with 32,065 individual SIRD records using the Census ID available in the two files. This process resulted in three subsets: 1,527 U.S. parent companies matched to the SIRD file, 1,672 U.S. parent companies not included in the SIRD file (all but 116 had no R&D expenditures), and 30,560 SIRD companies not in the USDIA survey file. Information on the 116 U.S. parent companies that were R&D performers not included in the SIRD file was used to improve SIRD sampling, but these companies were otherwise excluded from the data tables of the link project. The 1,527 matched U.S. parent companies included 1,505 SIRD companies due to differences in consolidation patterns. Of the 1,505 companies in the SIRD file, 1,497 had non-zero R&D performance in 2004.

2005

SIRD-FDIUS match

FDIUS-BR. The 2005 BEA FDIUS file contained information on 16,281 consolidated U.S. affiliates. Editing resulted in 13,039 U.S. affiliates, which were then matched to the BR using EINs. This resulted in 7,263 U.S. affiliates that were initially matched to the BR and 5,776 that were unmatched. U.S. affiliates that did not match initially to the BR and that had more than $1 billion in sales or had non-zero R&D expenditures were manually searched for in the BR using the name and address information for the BEA affiliate. This process resulted in an increase to 7,409 U.S. affiliates matched to the BR. The Census ID was then merged onto the FDIUS file for these matched IDs.

FDIUS-SIRD. The 7,409 U.S. affiliates matched to the BR were in turn matched with 32,057 individual SIRD records using the Census ID available in the two files. This process resulted in three subsets: 1,087 affiliates matched to the SIRD file, 6,339 affiliates not in the SIRD file (all but 405 affiliates had no R&D expenditures), and 31,047 SIRD records not in the FDIUS file. Information on the 405 U.S. affiliates that performed R&D and were not included in the SIRD files was used to improve SIRD sampling, but these affiliates were otherwise excluded from the data tables of the link project. The 1,087 matched affiliates included 1,010 unique SIRD companies and 1,070 unique companies in the FDIUS file as a result of differences in consolidation patterns. Of the 1,010 companies in the SIRD file, 962 had non-zero R&D performance in 2005.

SIRD-USDIA match

USDIA-BR. There were 3,331 consolidated U.S. parent companies from the BEA USDIA survey. This count was reduced to 3,266 after editing, and these companies were then matched to the BR. This resulted in 3,005 U.S. parent companies that were matched to the BR and 261 that were unmatched. U.S. parent companies that did not match initially to the BR and that had more than $1 billion in sales or had non-zero R&D expenditures were manually searched for in the BR by using name and address information for the BEA parent company. This process resulted in an increase to 3,133 BEA parent companies matched to the BR. The Census ID was merged onto the USDIA file for these matched parent companies.

USDIA-SIRD. The 3,133 U.S. parent companies matched to the BR were in turn matched with 32,057 individual SIRD records using the Census ID available in the two files. This process resulted in three subsets: 1,557 U.S. parent companies matched to SIRD; 1,604 U.S. parent companies not included in the SIRD file (all but 114 had no R&D expenditures), and 30,526 SIRD companies not in the USDIA file. Information on the 114 U.S. parent companies that were R&D performers that not included in the SIRD file was used to improve SIRD sampling, but these companies were otherwise excluded from the data tables of the link project. The 1,557 matched U.S. parent companies included 1,531 SIRD companies due to differences in consolidation patterns. Of the 1,531 companies in the SIRD file, 1,499 had non-zero R&D performance in 2005.

2006

SIRD-FDIUS match

FDIUS-BR. The 2006 BEA FDIUS file contained information on 15,855 consolidated U.S. affiliates. Editing resulted in 12,218 U.S. affiliates, which were then matched to the BR using EINs. This resulted in 7,161 U.S. affiliates that were initially matched to the BR and 5,057 that were unmatched. U.S. affiliates that did not match initially to the BR and that had more than $1 billion in sales or had non-zero R&D expenditures were manually searched for in the BR using the name and address information for the BEA affiliate. This process resulted in an increase to 7,192 U.S. affiliates matched to the BR. The Census ID was then merged onto the FDIUS file for these matched IDs.

FDIUS-SIRD. The 7,192 affiliates matched to the BR were in turn matched with 31,992 individual SIRD records using the Census ID available in the two files. This process resulted in three subsets: 1,128 affiliates matched to the SIRD file, 6,067 affiliates not in the SIRD file (all but 319 affiliates had no R&D expenditures), and 30,951 SIRD records not in the FDIUS file. Information on the 319 U.S. affiliates that performed R&D and were not included in the SIRD file was used to improve SIRD sampling, but these affiliates were otherwise excluded from the data tables of the link project. The 1,128 matched affiliates included 1,041 unique companies in the SIRD file and 1,125 unique companies in the FDIUS file as a result of differences in consolidation patterns. Of the 1,041 companies in the SIRD file, 977 had non-zero R&D performance in 2006.

SIRD-USDIA match

USDIA-BR. There were 3,317 consolidated U.S. parent companies from the BEA USDIA survey. This count was reduced to 3,250 after editing, and these companies were then matched to the BR. This resulted in 2,988 U.S. parent companies that were initially matched to the BR and 262 that were unmatched. U.S. parent companies that did not match initially to the BR and that had more than $1 billion in sales or had non-zero R&D expenditures were manually searched for in the BR by using name and address information for the BEA parent company. This process resulted in an increase to 3,103 BEA parent companies matched to the BR. The Census ID was merged onto the BEA file for these matched parent companies.

USDIA-SIRD. The 3,103 U.S. parent companies matched to the BR were in turn matched with 31,992 individual SIRD records using the Census ID available in the two files. This process resulted in three subsets: 1,648 U.S. parent companies matched to the SIRD file, 1,504 U.S. parent companies not included in the SIRD file (all but 58 had no R&D expenditures), and 30,396 SIRD companies not in the USDIA survey file. Information on the 58 U.S. parent companies that were R&D performers and were not included in the SIRD file was used to improve SIRD sampling, but these companies were otherwise excluded from the data tables of the link project. The 1,648 matched U.S. parent companies included 1,596 SIRD companies due to differences in consolidation patterns. Of the 1,596 companies in the SIRD file, 1,546 had non-zero R&D performance in 2006.

2007

SIRD-FDIUS match

FDIUS-BR. The 2007 BEA FDIUS file contained information on 17,613 consolidated U.S. affiliates. Editing resulted in 14,195 records, which were then matched to the BR using EINs. This resulted in 8,220 U.S. affiliates that were initially matched to the BR and 5,975 that were unmatched. U.S. affiliates that did not match initially to the BR and that had more than $1 billion in sales or had non-zero R&D expenditures were manually searched for in the BR by using name and address information for the BEA affiliate. This process resulted in an increase to 8,275 U.S. affiliates matched to the BR. The Census ID was then merged onto the BEA file for these matched IDs.

FDIUS-SIRD. The 8,275 U.S. affiliates matched to the BR were in turn matched with 31,881 individual SIRD records using the Census ID available in the two files. This process resulted in three subsets: 1,115 affiliates matched to the SIRD file, 7,167 affiliates not in the SIRD file (all but 425 affiliates had no R&D expenditures), and 30,853 SIRD records not in the FDIUS file. Information on the 425 U.S. affiliates that performed R&D and were not included in the SIRD file was used to improve SIRD sampling, but these affiliates were otherwise excluded from the data tables of the link project. The 1,115 matched affiliates included 1,028 unique companies in the SIRD file and 1,108 unique companies in the FDIUS file as a result of differences in consolidation patterns. Of the 1,028 companies in the SIRD file, 967 had non-zero R&D performance in 2007.

SIRD-USDIA match

USDIA-BR. There were 3,350 consolidated U.S. parent companies from the BEA USDIA survey. This count was reduced to 3,285 after editing, and these companies were then matched to the BR. This resulted in 3,026 U.S. parent companies that were initially matched to the BR and 259 that were unmatched. U.S. parent companies that did not match initially to the BR and that had more than $1 billion in sales or had non-zero R&D expenditures were manually searched for in the BR by using name and address information for the BEA parent company. This process resulted in an increase to 3,091 BEA parent companies matched to the BR. The Census ID was merged onto the USDIA file for these matched parent companies.

USDIA-SIRD. The 3,091 U.S. parent companies matched to the BR were in turn matched with 31,881 individual SIRD records using the Census ID available in the two files. This process resulted in three subsets: 1,695 U.S. parent companies matched to the SIRD file, 1,551 U.S. parent companies not included in the SIRD file (all but 68 had no R&D expenditures), and 30,295 SIRD companies not included in the USDIA file. Information on the 68 U.S. parent companies that were R&D performers not included in the SIRD file was used to improve SIRD sampling, but these companies were otherwise excluded from the data tables of the link project. The 1,695 matched U.S. parent companies included 1,586 SIRD companies due to differences in consolidation patterns. Of the 1,586 companies in the SIRD file, 1,512 had non-zero R&D performance in 2007.

Top of page. Back to Top

Validity Checks and Data Adjustments

At various points in the link process, Census, with input from NSF and BEA, reviewed or validated outputs prior to proceeding to the next step. Checks were made to confirm matches to the Census BR at the EIN level. For example, the BR match was examined in terms of names associated with matched EINs and the degree to which total employment from the BR and the BEA file agreed (e.g., within 5%, 10%, or 25%). The SIRD match was assessed by examining reported (or imputed) sales, R&D expenditures, total employment, and R&D employment for matched IDs.

Data discrepancies across surveys for matched companies were sorted by size to select cases to investigate and were discussed by analysts from the respective surveys. These cases were then corrected or determined to be unsolvable. Discrepancies with differences of $100 million in R&D expenditures, $5 billion in sales, 15,000 in total employment, and/or 2,000 in R&D employment were examined. Records that fell outside of the discrepancy review criteria were not investigated and were left as is. The criteria levels were set based on an examination of all discrepancies. A determination was made by NSF, BEA, and Census that these levels would have the largest impact on improving the quality of the matched data. Pre-correction and post-correction summary files, along with flowcharts documenting status of matched and unmatched records, were examined by project staff and assessed at interagency meetings.

Among the reasons for data discrepancies in matched companies across surveys were reporting errors and differences in company consolidation or composition. Some discrepancies occurred among matched records that had imputed data items in at least one of the linked surveys. Where available, reported data were used in lieu of imputed data for purposes of the link.

Top of page. Back to Top

Data Comparability

General Observations

Matched R&D data from SIRD differ from total U.S. industry R&D data from SIRD, as published elsewhere by NSF, because linked data are subsets based on ownership categories obtained through links to the BEA surveys. Further, estimates for total U.S. industrial R&D activity are computed by weighting SIRD sample data to universe totals. Data from the link project are not weighted. Thus matched data presented here are obtained from simple aggregations of sample microdata, cover only matched companies, and do not represent the universe of all R&D-performing parent companies or U.S. affiliates. Lastly, matched data reflect updates and adjustments performed only for the purposes of the link project.

Data for matched parent companies of U.S. MNCs and matched U.S. affiliates (U.S.-located foreign owned companies) should not be summed to attempt a U.S. R&D aggregate related to MNCs because some companies both are parent companies of overseas affiliates and are themselves owned by a foreign parent company.

Data on majority-owned foreign affiliates (MOFAs) are limited to companies owned by matched R&D-performing parent companies. These data thus exclude MOFAs of unmatched U.S. parent companies and MOFAs that perform R&D but whose parent companies do not perform R&D in the United States. In 2004 a total of $0.5 billion of R&D was performed by MOFAs whose parent companies did not perform R&D in the United States.

The earlier pilot study of the R&D link project explored methodological issues. It also resulted in limited data for survey years 1997 and 1999 released in 2005 (see "Data Availability"). These matched data are generally comparable in terms of methodology with the 2004 match.

Comparisons with Full Survey Data

2004

SIRD-reported R&D performed by matched parent companies in 2004 accounted for 96% of U.S. parent company R&D published by BEA in the 2004 USDIA survey. R&D performed overseas by MOFAs of matched parent companies accounted for 98% of MOFA R&D published by BEA in the latter survey. SIRD-reported R&D by matched U.S. affiliates accounted for 96% of R&D published by BEA in the 2004 FDIUS survey.

2005

SIRD-reported R&D performed by matched parent companies in 2005 accounted for 93% of U.S. parent company R&D published by BEA in the 2005 USDIA survey. R&D performed overseas by MOFAs of matched parent companies accounted for 97% of MOFA R&D published by BEA in the latter survey. SIRD-reported R&D by matched U.S. affiliates accounted for 96% of R&D published by BEA in the 2005 FDIUS survey.

2006

SIRD-reported R&D performed by matched parent companies in 2006 accounted for 97% of U.S. parent company R&D published by BEA in the 2006 USDIA survey. R&D performed overseas by MOFAs of matched parent companies accounted for 97% of MOFA R&D published by BEA in the latter survey. SIRD-reported R&D by matched U.S. affiliates accounted for 95% of R&D published by BEA in the 2006 FDIUS survey.

2007

SIRD-reported R&D performed by matched parent companies in 2007 accounted for 95% of U.S. parent company R&D published by BEA in the 2007 USDIA survey. R&D performed overseas by MOFAs of matched parent companies accounted for 95% of MOFA R&D published by BEA in the latter survey. SIRD-reported R&D by matched U.S. affiliates accounted for 91% of R&D published by BEA in the 2007 FDIUS survey.

Top of page. Back to Top

Definitions

General

Affiliate: A company or business enterprise located in one country but owned or controlled (in terms of 10% or more of voting securities or equivalent) by a person (in the broad legal sense, including a company) in another country; may be either incorporated or unincorporated.

Foreign direct investment (FDI): Ownership or control of 10% or more of the voting securities (or equivalent) of a business located outside the home country.

Foreign parent company: The foreign entity, or the first entity outside the United States in a foreign chain of ownership, which has direct investment in a U.S. business enterprise.

Investing country: Country of ultimate beneficial owner, entity that ultimately owns or controls U.S. affiliate.

Majority-owned affiliate: Company owned or controlled to the extent of more than 50% of the voting securities (or equivalent) by its parent company.

Multinational company (MNC): A parent company and its foreign affiliates.

U.S. affiliate: Company located in the United States but owned by a foreign parent company.

U.S. parent company: An entity (individual, branch, partnership, or corporation) that is located in the United States and that owns or controls at least 10% of the voting securities, or equivalent, of a foreign business enterprise.

SIRD Survey

Employment, full-time-equivalent (FTE) R&D scientists and engineers: Number of people who were employed in the 50 U.S. states and the District of Columbia by R&D-performing companies and who were engaged in scientific or engineering work at a level that required knowledge, gained either formally or by experience, of engineering or of the physical, biological, mathematical, statistical, or computer sciences equivalent to at least that acquired through completion of a 4-year college program with a major in one of those fields. FTE employees refer to those who were assigned full time to R&D, plus a prorated number of employees who worked on R&D only part of the time. FTEs are measured with reference to January following the survey year.

Employment, total: Number of people employed in the 50 U.S. states and the District of Columbia by R&D-performing companies in all activities during the pay period that included the 12th of March of the study year (March 12 is the date most employers use when paying first-quarter employment taxes to the Internal Revenue Service).

Net sales and receipts: Dollar values for goods sold or services rendered by R&D-performing companies to customers outside the company, including the federal government, less such items as returns, allowances, freight, charges, and excise taxes. Domestic intracompany transfers and sales by foreign subsidiaries were excluded, but transfers to foreign subsidiaries and export sales to foreign companies were included.

R&D: R&D is the planned, systematic pursuit of new knowledge or understanding toward general application (basic research); the acquisition of knowledge or understanding to meet a specific, recognized need (applied research); or the application of knowledge or understanding toward the production or improvement of a product, service, process, or method (development). Specifically excluded from R&D are quality control, routine product testing, market research, sales promotion, sales service, and other nontechnological activities; routine technical services; and research in the social sciences or psychology.

Funds for R&D, total: The cost of R&D performed within the company in its own laboratories or in other company-owned or company-operated facilities in the 50 U.S. states and the District of Columbia, including expenses for wages and salaries, fringe benefits for R&D personnel, materials and supplies, property and other taxes, maintenance and repairs, depreciation, and an appropriate share of overhead. Capital expenditures and the cost of R&D contracted to outside organizations are excluded.

FDIUS and USDIA Surveys

R&D: R&D includes activities carried on by persons trained, either formally or by experience, in engineering; the physical sciences, such as chemistry and physics; the biological sciences, such as medicine; the mathematical and statistical sciences; and computer science. R&D includes these activities if the purpose is to do one or more of the following: the planned systematic pursuit of new knowledge or understanding toward general application (basic research); the acquisition of knowledge or understanding to meet a specific, recognized need (applied research); and the application of knowledge or understanding toward the production or improvement of a product, service, process, or method (development).

R&D expenditures: These expenditures refer to all costs incurred to support R&D performed whether assigned to separate R&D organizational units of the company or carried out by company laboratories and technical groups not a part of an R&D organization. Costs include wages, salaries, and related costs; materials and supplies consumed; depreciation on R&D property and equipment; cost of computer software used in R&D activities; utilities; travel costs and professional dues; property taxes and other taxes (except income taxes) incurred on account of the R&D organization or the facilities they use; insurance expenses; maintenance and repair, including maintenance of buildings and grounds; company overhead, including personnel, accounting, procurement and inventory, and salaries of research executives not on the payroll of the R&D organization. R&D expenditures exclude capital expenditures, expenditures for tests and evaluations once a prototype becomes a production model, patent expenses, and income taxes and interest. They also exclude expenditures for quality control; routine product testing; market research; sales promotion, sales service, and other nontechnological activities; routine technical services; research in the social sciences or psychology; geological and geophysical exploration activities; and advertising programs to promote or demonstrate new products or processes.

R&D employees: Scientists, engineers, and other professional and technical employees, including managers, engaged in scientific or engineering R&D work at a level that requires knowledge of physical or life sciences, engineering, mathematics, statistics, or computer science at least equivalent to that acquired through completion of a 4-year college course with a major in one of these fields (i.e., training may be either formal or by experience).

Top of page. Back to Top

Data Availability

Linked data are available at http://www.nsf.gov/statistics/rdlink/. Selected data from this project covering 1997 and 1999 data are available from Research and Development Data Link Project: Final Report (2005) http://www.census.gov/mcd/RDD/rddatalink.pdf and from the following NSF reports on this project: InfoBrief 07-310 at http://www.nsf.gov/statistics/infbrief/nsf07310/, and InfoBrief 12-332 at http://www.nsf.gov/statistics/infbrief/nsf12332/.

Total U.S. industry R&D data for SIRD are available at http://www.nsf.gov/statistics/industry/. Full data from BEA international surveys are available at http://www.bea.gov/international/index.htm#omc.

Top of page. Back to Top

Notes

[1]  The Census Bureau also maintains a separate non-employer database. Non-employers also have EINs since, for example, proprietors and other self-employed individuals obtain an EIN for income tax reporting purposes. Some companies may switch between employer and non-employer status over time.

[2]  Several consolidated companies in the SIRD sample were associated with more than one record. Therefore, the number of individual SIRD records is greater than the number of consolidated companies reported in table A1.


 
International Investment and R&D Data Link: 2004–07
Detailed Statistical Tables | NSF 12-327 | September 2012