Skip all navigation and go to page content

Chapter 5. Academic Research and Development

Infrastructure for Academic R&D

Physical infrastructure is an essential resource for the conduct of R&D. Not long ago, the capital infrastructure for R&D consisted primarily of research space (such as laboratories and computer rooms) and instrumentation. Accordingly, the square footage of a designated research space and counts of instruments are principal indicators of the status of research infrastructure.

Over the last 20 years, however, advances in information technology have brought significant changes to both the methods of scientific research and the infrastructure needed to conduct R&D. The technologies, human interfaces, and associated processing capabilities resulting from these innovations are often included in the term cyberinfrastructure. Cyberinfrastructure may involve mainly one resource, such as a network used to transfer data, or it may involve a complex interaction of many resources resulting in sophisticated capabilities, such as high-performance computation or remote use of scientific instrumentation. No matter how simple or complex these technologies and their human interfaces may be, cyberinfrastructure has become an essential resource for science.

Indicators for research facilities, research equipment, and cyberinfrastructure capacity are discussed in this section. (For an overview of the sources of data used see the sidebar, "Data on the Financial and Infrastructure Resources for Academic R&D," earlier in this chapter.)

Research Facilities

Research Space

At the close of academic FY 2009, research-performing colleges and universities had 196.1 million net assignable square feet (NASF) of research space available (appendix table 5-13).[12] This was 2.2% above the assignable square footage at the end of FY 2007.

This increase represented continuity in a now two-decade long trend of academic institutions investing to expand their research space. Even so, the pace of growth has slowed in the last few years. The 2.2% expansion over the FY 2007–09 period was the slowest since the 1988–90 period. (The rate of increase peaked in 2001–03 at 11%, and has gradually declined since then.)

The S&E field of biological/biomedical sciences currently accounts for the largest portion of research space, or 50.3 million NASF in FY 2009 and 26% of the academic total (figure 5-11 and appendix table 5-13).[13] The related field of health/clinical sciences was the second largest, accounting for 36.3 million NASF and 19% of the total. Still sizable are engineering (30.2 million NASF, 15%), agricultural/natural resources (29.5 million NASF, 15%), and physical sciences (28.5 million NASF, 15%). The other fields are substantially smaller: social sciences (5.5 million NASF, 3%), computer/information sciences (5.2 million NASF, 3%), psychology (5.2 million NASF, 3%), mathematics/statistics (1.5 million NASF, 1%), and all other sciences (3.9 million NASF, 2%).

The aforementioned slowing pace of growth in overall academic research space since FY 2005 has played out in a variety of ways across the S&E fields (appendix table 5-13). The large amount of space for biological/biomedical sciences continued to expand at a substantial rate in both the FY 2005–07 and 2007–09 periods. The agricultural/natural resources field also increased its research NASF in both periods. Engineering expanded in FY 2005–07, but had no growth in the 2007–09 period.

Even so, the amount of research space available to a sizable number of S&E fields has experienced no growth or a decline since FY 2005. Health/clinical sciences and physical sciences, both fields with large amounts of research space, experienced declines in each of the FY 2005–07 and 2007–09 periods.[14] The decline in the health/clinical sciences is particularly notable, because this field exhibited some of the largest increases in research space in any S&E field in the first half of the 2000 decade. While much smaller in NASF size, the social sciences exhibited a research space decline in both FY 2005–07 and 2007–09. And, also small, the mathematics/statistics field exhibited a decline in 2007–09.

Compared with other fields, the computer sciences exhibited among the largest rates of increase in research space from FY 2001 to 2007 (appendix table 5-13). Nonetheless, its total research space, currently at 5.2 million NASF, is less than most fields.

New Construction

Concomitant with the slowing expansion of overall academic research space, new construction also slowed in the second half of the 2000 decade (table 5-2). The 16.2 million NASF of new construction in FY 2002–03 dropped to about 8.8 million in FY 2006–07, even if up somewhat, to 9.9 million in FY 2008–09. Similarly, within the broad decline of total research space, the amount and direction of change in new construction varied significantly across the S&E fields.

The construction starts for new research space in the biological/biomedical sciences was the largest among all the fields in FY 2006–07 and 2008–09, or 2.9 million NASF and 3.5 million NASF, respectively. Further back were the health/clinical sciences (1.7 million NASF in FY 2006–07, 1.9 million in 2008–09) and engineering (respectively, 1.3 million and 2.1 million). All the other fields reported some new construction starts in both FY 2006–07 and 2008–09, but at levels well below the top three fields.

Academic institutions draw on various sources to fund their capital projects, including the institutions' own funds, state or local governments, and the federal government. For the construction of new research space initiated in FY 2008–09, about 62% of the funding came from institutions' internal sources, 36% from state/local government, and the remaining 3% from the federal government. This was similar to the new construction initiated in FY 2006–07, where the funding shares were, 62%, 32%, and 6%, respectively. In recent years, the federal portion of funding has been under 10% and declining, with the FY 2009 level the lowest for several decades.

Research Equipment

In FY 2009, about $2.0 billion in current funds was spent for academic research equipment (i.e., moveable items, such as computers or microscopes) necessary for the conduct of organized research projects (appendix table 5-14).[15] The corresponding totals in earlier years were $1.9 billion in FY 2008, $1.9 billion in FY 2004, and $1.3 billion in FY 1999. Adjusted for inflation, the change in this spending from 2008 to 2009 was a 2% increase, which was an increase of 16% over the 1999 spending level, but a 9% decline from the 2004 level.

The $2.0 billion of equipment spending in FY 2009 was just under 4% of the $54.9 billion of total academic R&D expenditures that year. In FY 2004, the share was somewhat above 4% of the academic R&D total. In FY 1999, the fraction was closer to 5%.

This equipment spending continues to be concentrated in just a few S&E fields. In FY 2009, three fields accounted for 82% of the annual total: life sciences (41%), engineering (24%), and the physical sciences (17%). The shares for these three fields have remained similarly predominant for many years (appendix table 5-14). Even so, when adjusted for inflation, the annual level of equipment spending in all three fields has declined since 2005—reversing a trend of steady growth from FY 2001 to 2004 (figure 5-12).

Some of the funding for academic research equipment comes from the federal government. These federal funds are generally received as part of research grants or as separate equipment grants. In FY 2009, the federal government supported 55% of total academic research equipment funding—a figure that has fluctuated between 55% and 63% over the last 20 years (appendix table 5-15). Nevertheless, the federal share of funding varies significantly by S&E field, ranging from 26% to 77% in 2009. In FY 2009, computer sciences had the largest proportion of federally funded R&D equipment (77%), with atmospheric sciences a close second (76%).


Networking is an essential component of cyberinfrastructure. It facilitates research-related activities such as communication, data transfer, high-performance computation, and remote use of instrumentation.[16] In FY 2009, networking infrastructure on many academic campuses was pervasive and still rapidly expanding in capability and coverage. Research-performing institutions had more connections, bandwidth, and campus coverage than they did earlier in the decade.[17] (Network "bandwidth" is the amount of data that can be transmitted in a given amount of time, typically measured in bits per second.) Colleges and universities reported external network connections with greater bandwidth, faster internal network distribution speeds, more connections to high-speed networks, and greater wireless coverage on campus.

Some academic cyberinfrastructure is dedicated primarily to research activities. For example, universities may have high-performance networks (such as the National LambdaRail or networks to government agencies) available almost exclusively for research activities, and this bandwidth capacity is only for these activities. Nonetheless, universities may have other networks that are available to the entire campus community for both research and non-research activities, and this bandwidth capacity is not an indicator solely of research capacity.

Bandwidth to External Connections

Academic institutions can have multiple networking resources, at varying connection speeds. Internet1—the public multiuse, commodity network often called the "Internet"—is one such component. Many institutions also have direct or indirect connections to high-performance networks that support the development and use of advanced applications and technologies. In the academic community, these high-performance networks are chiefly Internet2 (a high performance backbone network providing leading-edge network services to member colleges, universities, and research laboratories across the country), the National LambdaRail (an advanced optical network for research and education, organized by a consortium of universities, private companies, and federal labs), and connections to federal research networks.

Early in the 2000 decade, some academic institutions reported no Internet1 connections of any kind. By mid-decade, all institutions had Internet1 connections and bandwidth speeds were increasing. Between FY 2005 and FY 2009, the fraction of institutions with total Internet1 and Internet2 bandwidth of more than 100 megabits per second increased from 52% to 80% (table 5-3). Furthermore, the share of institutions with total Internet1 and Internet2 bandwidths of 1 gigabit per second or faster rose from 22% in FY 2005 to 45% in FY 2009. (If current institutional estimates are realized, the percent of institutions with total bandwidths of 1 gigabit or faster will reach 52% in FY 2010.)

Bandwidth has increased broadly across all types of academic institutions. Nevertheless, a greater fraction of doctorate-granting institutions have the faster bandwidths. In FY 2009, 87% of the institutions that granted doctorates had total Internet1 and Internet2 bandwidth of at least 1 gigabit per second, and 32% had bandwidth greater than 2.4 gigabits. In contrast, 71% of nondoctorate granting institutions had total bandwidth at 1 gigabit per second or above and 8% above 2.4 gigabits.

Part of the increase in institutions' bandwidth can be attributed to an increase in the number of connections to high-performance networks (table 5-4). The number of connections to Internet2 has grown gradually over the current decade; by the end of FY 2009, a large majority (75%) of institutions had Internet2 connections. Between FY 2007 and 2009, the percentage of institutions with connections to the National LambdaRail increased, from 25% to 34% of all institutions. The number of institutions anticipating connections to federal government high-performance networks such as the Department of Energy's ESnet or NASA's NREN further increased in FY 2009. Institutions have also begun connecting to more than one high-performance network—for example, in FY 2009, 34% had connections to both Internet2 and the National LambdaRail.

Internal Institutional Networks

The bandwidth speeds of academic institutions' internal networks have also increased considerably. Since early in the present decade, the percentage of institutions with slower bandwidth has rapidly decreased while the percentage with faster bandwidths has rapidly increased. In FY 2003, 66% of institutions had bandwidth less than 1 gigabit per second, but by the end of FY 2009, only 19% did (table 5-5). In FY 2003, no institutions had bandwidth speeds faster than 2.5 gigabits per second, but by FY 2009, 24% of academic institutions did. By FY 2009, 82% of institutions had speeds of 1 gigabit per second or faster.

In FY 2009, all academic institutions had at least some wireless coverage in their campus buildings. In FY 2003, only 14% of these institutions had more than half of their building infrastructure covered by wireless; by FY 2009, the comparable figure was 74%.


[12] Research space here is defined as the space used for sponsored R&D activities at academic institutions that is separately budgeted and accounted for. Research space is measured in net assignable square feet (NASF). This is the sum of all areas on all floors of a building assigned to, or available to be assigned to, an occupant for a specific use, such as research or instruction. NASF is measured from the inside faces of walls. Multipurpose space that is partially used for research is prorated to reflect the proportion of time and use devoted to research.
[13] The S&E fields used in the NSF Survey of Science and Engineering Research Facilities are based on the National Center for Education Statistics (NCES) Classification of Instructional Programs (CIP)—which is updated every 10 years (the current version is dated 2000). The S&E fields used in both the FY 2007 and FY 2009 Survey of Science and Engineering Research Facilities reflect the 2000 CIP update. For a comparison of the subfields in the FY 2005 and FY 2007 surveys, see the detailed statistical tables for S&E Research Facilities: FY 2007.
[14] The S&T field and subfield definitions were updated to the 2000 CIP starting with the FY 2007 facilities survey. Therefore, some of the observed declines in research space for health/clinical sciences and physical sciences between FY 2005 and FY 2007 could reflect definition changes.
[15] Because of rising capitalization thresholds, the dollar threshold for inclusion in the equipment category has changed over time. Generally, university equipment that costs less than $5,000 would be classified under the cost category of "supplies."
[16] The "bricks and mortar" section of the Survey of Science and Engineering Research Facilities asks institutions to report their research space only. Therefore, the reported figures do not include space used for other purposes such as instruction or administration. In the cyberinfrastructure section of the survey, however, respondents are asked to identify all of their cyberinfrastructure resources, regardless of whether these resources were used for research or other functions.
[17] Research-performing academic institutions are defined as colleges and universities that grant degrees in S&E and expend at least $1 million in R&D funds. Each institution's R&D expenditures are determined through the NSF Survey of Research and Development Expenditures at Universities and Colleges.