Higher Education Research and Development: Fiscal Year 2010
Appendix D. HERD Survey Short Form Data Collection
During the FY 2010 collection cycle, the National Science Foundation (NSF) also conducted an abbreviated version of the Higher Education Research and Development (HERD) Survey. Known as the Higher Education Research and Development Survey Short Form, it was administered to all academic institutions in the United States and outlying areas that granted a 4-year degree or higher and were not already included in the HERD Survey population.
The goal of the HERD Short Form was to measure the amount of academic research and development not being captured by the HERD Survey and to assess possible methodologies for developing future frames for the HERD Survey.
Scope of the Survey
The HERD Short Form was sent to 1,715 higher education institutions in the United States, Guam, Puerto Rico, and the U.S. Virgin Islands that granted bachelor's degrees or higher, in any field, and were not part of the FY 2010 HERD Survey population. The HERD population included all institutions that reported $150,000 or more in R&D expenditures in science and engineering (S&E) fields on the FY 2009 HERD Survey.
The initial HERD Short Form population included both for-profit and nonprofit higher education institutions. At the conclusion of data collection, the response rate for for-profit institutions was significantly lower than that of nonprofit higher education institutions. Many of these institutions were reluctant to share their proprietary financial data with NSF and the public. Also, the amount of R&D expenditures reported by those for-profit institutions that did respond was extremely minimal (only 0.5% of the R&D expenditures reported to the FY 2010 HERD Short Form). NSF decided to remove the 139 for-profit institutions from the survey population frame and the final counts. The final population for the HERD Short Form included 1,576 institutions.
FY 2010 HERD Short Form Frame Design
Together, the HERD Survey and HERD Short Form are a census of all eligible academic institutions. The survey frame was developed using data from the FY 2007 NSF Survey of Federal S&E Support to Universities, Colleges, and Nonprofit Institutions (Federal S&E Support Survey); FY 2008 data from the National Center for Education Statistics Integrated Postsecondary Education Data System (IPEDS); and the 2011 Higher Education Directory, published annually by Higher Education Publications. The Office of Postsecondary Education accreditation database, the Council for Higher Education Accreditation database, and institution websites were used as secondary sources to resolve inconsistencies and help merge the three sources into a unified frame. All institutions identified as offering a bachelor's degree or higher from any of the three main sources were considered for inclusion in the frame.
Information about geographic location and senior administration was also reviewed to determine whether institutions met NSF's new reporting criteria of being geographically separate from other institutions in their university system and being headed by a president or chancellor (see appendix C for more information about changes in reporting levels).
The HERD Short Form was administered as a Web-based survey consisting of four questions. The survey was also available as an Adobe Portable Document Format file (see appendix B for both the HERD Survey and HERD Short Form instruments), but to facilitate a timely collection, respondents were encouraged to use the website. The information needed to complete this survey could generally be found in the institution's year-end accounting records.
Question 1 was a screener asking whether the institution had any R&D in FY 2010. Institutions that indicated they did not have R&D were instructed to skip questions 2 and 3.
Question 2 was a request for total current expenditures for separately budgeted R&D by source of funds. Institutions were asked to identify expenditures from the following sources: U.S. federal government, state and local government, business, nonprofit organizations, institutional funds, and all other sources.
Question 3 requested the month in which the institution's FY 2010 fiscal year ended.
Question 4 requested contact information for future communications regarding the survey response.
The FY 2010 HERD Short Form was conducted in two stages. The first stage focused on soliciting senior administrators to commit to completing the survey and to identifying the proper point of contact. The second stage was the administration of the survey to the designated contact.
Identifying the Point of Contact
In February 2011, letters were mailed to each institution's senior administrator (typically, an individual with the title of chancellor or president). The letter explained the purpose of the survey and notified administrators that they would be receiving an e-mail asking them to designate a respondent for the survey. Eight days later, e-mails were sent to administrators containing a link to a Web-based form that could be used to designate a point of contact. Each link provided respondents with access to the form specific to their institution without having to enter login information. Institutions were asked to name their designated contact by 7 March 2011. Reminders were sent via e-mail before and after the announced due date. Beginning in May 2011, administrators who had not responded were contacted by phone and asked to provide a point of contact. Efforts to identify a point of contact ended in early June 2011. Institutions that responded to requests to name a contact by saying they did not have R&D were considered respondents to the survey.
Administering the Survey
Designated contacts were e-mailed a link to their institution's survey as soon as they were appointed. Each institution was given 2–3 weeks to complete the survey. Reminders were sent via e-mail before and after the announced due date. After two nonresponse reminder e-mails, attempts were made to reach the designated contact by phone. Final nonresponse communications were conducted via phone with the senior administrator, who was asked to appoint a new contact.
Regular data collection efforts ended in July 2011. Special collection efforts continued until early August 2011 for 10 institutions that reported $1 million or more in research expenditures in FY 2008 to the IPEDS Finance Survey.
A total of 1,015 out of 1,576 (64.4%) institutions responded to the FY 2010 HERD Short Form. Responses were received from 80.8% of public nonprofit academic institutions and 61.9% of private nonprofit academic institutions. Table D-1 displays a detailed breakdown of unit response rates, by highest degree granted and institutional control.
Of the 1,015 institutions responding to the Short Form, 187 (18.4%) met the $150,000 inclusion threshold for the HERD Survey population. Of those, 38 (20.3%) were doctoral-granting institutions, 108 (57.8%) were master's-granting institutions, 36 (19.3%) were bachelor's-granting institutions, and 5 (2.7%) granted first professional degrees only. These institutions will be asked to participate in the full HERD Survey for FY 2011.
The combined results of the HERD Survey and HERD Short Form are the best available account of all R&D expenditures at the nation's colleges and universities in FY 2010. Analysis of the HERD Short Form results shows that the HERD Survey captures almost all academic R&D expenditures. Of the more than $61.49 billion in R&D reported to both the HERD Survey and HERD Short Form for FY 2010, roughly $258 million (0.42%) was from the HERD Short Form population. Table D-2 shows R&D expenditures by survey population and source of funds. For institutions in the HERD Survey population, 61.2% of expenditures were from federal sources. For institutions in the HERD Short Form population, 51.3% of expenditures were from federal sources.
Table D-3 shows R&D expenditures, by highest degree offered, institutional control, and source of funds.
Table D-4 shows expenditures—by state, institutional control, and source of funds—for all institutions in the HERD Short Form population reporting total R&D expenditures greater than zero. For more information about the HERD Short Form instrument, data collection procedures, or data, contact the NSF project manager.
 E-mail addresses for 103 institutions were not available. Requests to designate points of contact were sent to these institutions by mail.
 Initial due date for the survey was 3 weeks from the time that the point of contact was appointed. This was changed to 2 weeks beginning in April 2011.