WPC 2^BJ  Z|xTimes New Roman (TT)Times New Roman (TT)Symbol (TT)Arial (TT)CourierArial (TT)TimesHelveticaCourier New (TT)Line Printer 16.67cpiWingdings (TT)Harrington (TT)Arial Narrow (TT)Book Antiqua (TT)Bookman Old Style (Light) (TT)Century Gothic (TT)Century Schoolbook (TT)Monotype Corsiva (Italic) (TT)Monotype Sorts (TT)MS LineDrawAlgerian (TT)Arial Rounded MT Bold (Bold) (TT)Braggadocio (TT)Britannic Bold (Bold) (TT)Brush Script MT (Italic) (TT)Colonna MT (TT)Desdemona (TT)Footlight MT Light (Light) (TT)Impact (TT)Kino MT (TT)Wide Latin (TT)Matura MT Script Capitals (DemiBold) (TT)Playbill (TT)ModernScriptRomanXC\  P6QC\  P6Q,a\  P[A9J2PQEd6X@CMJ2PQYC\  PU_`2PkCid6X@DQHH{H H@;r P?pQ:PQ<xzP"@QJz PQB  0*QQ^P-0Q M P:+Q&;6B>Hxx'QF}'' PV/Q[SL7 @QgHUP73QvAopL1 QXM6PhQ^^4pQ+ @xćQ^8. 0Q=fP7bQ7O 0Q(7KtPF Q5-E 9PX6QCnJx PQT9x @P%Q!uP:QCt0n U9 @HxUK\  P@Q2<Ephoenix C\  P6Q  2EDefault Paragraph FoDefault Paragraph Font.. XP\  P6Q  C\  P6Q   ` hp x (#4 <DL!Title: Front Matter Type: S&E Research Facilities at Universities & Colleges: 1994 vol. I Exposure: Public NSF Org: SRS Date: 02/28/95 Replaces: none Delete: none Scientific and Engineering Research Facilities at Universities and Colleges Highlights... * In 1994, the 565 researchperforming universities and colleges in the United States had 127 million net assignable square feet (NASF) of science and engineering (S&E) research space. The top 100 institutions had 82 percent of total academic research and development (R&D) expenditures and accounted for 72 percent of this S&E research space. Other doctorategranting institutions accounted for 24 percent, and the nondoctorategranting institutions for 4 percent of S&E research space. * S&E research space has increased since 1988 at an annual average rate of about 2 percent, from 112 million NASF to 127 million NASF in 1994. Other National Science Foundation (NSF) surveys show that research spending grew by just under 9 percent per year and that graduate enrollment increased by nearly 4 percent per year during this same time period. * More than 40 percent of all researchperforming universities and colleges indicated inadequate amounts of S&E research space in engineering, the physical sciences, the biological sciences outside of medical schools, and the medical sciences in medical schools. * Twentysix percent of all S&E research space was judged to be "suitable for use in most scientifically sophisticated research," while 17 percent was rated as needing either major repair/renovation or replacement. * Overall, projects to construct S&E research space totaled $2,812 million in fiscal years 1992 1993. This amount represented a decline of $290 million in constant dollars (dollars adjusted for inflation) from fiscal years 19901991, the first decline in construction spending since NSF began collecting data on S&E research facilities. * Aggregate spending on repair/renovation of S&E research space declined from $861 million in constant dollars in fiscal years 19901991 to $837 million in fiscal years 19921993. This overall decline resulted from decreased spending by the top 100 institutions; spending for repair/renovation by other doctorategranting and nondoctorategranting institutions increased. * Public universities and colleges accounted for 70 percent of total fiscal years 19921993 spending for S&E research facilities construction and repair/renovation. Almost half (46 percent) of their funds came from state or local sources, another 18 percent from tax exempt bonds, and 14 percent each from the Federal government and institutions' own funds. * In 1994, the estimated cost of unfunded and deferred capital projects that were included in an institutional plan was $5,744 million. These estimates directly reflected the needs of the 40 percent of universities and colleges that had identified these deferred needs in an approved institutional plan. * For the panel of 29 Historically Black Colleges and Universities (HBCUs) included since the 1988 survey, expenditures for research space construction declined from $83.2 million (constant dollars) in fiscal years 19861987 to $8.6 million in fiscal years 19921993. Information about This Publication Suggested Citation National Science Foundation, Scientific and Engineering Research Facilities at Universities and Colleges: 1994, NSF 94315, (Washington, D.C., 1994). Availability of Publications Single copies are available at no cost from the Division of Science Resources Studies, National Science Foundation, Arlington, Virginia 22230. If you are a user of electronic mail and have access to either BITNET or Internet, you may order publications electronically. BITNET users should address requests to pubs@nsf. Internet users should send requests to pubs@nsf.gov. In your request, include the NSF publication number and title, number of copies, your name, and a complete mailing address. Printed publications may also be ordered by FAX (703/6444278). Publications should be received 3 weeks after receipt of your request. See the back of this publication for the availability of S&T data on the Science and Technology Information System (STIS). Telephone Device for the Deaf The National Science Foundation has Telephone Device for the Deaf (TDD) capability, which enables individuals with hearing impairments to communicate with the Division of Personnel Management about NSF programs, employment, or general information. The phone number is (703) 3060090. Foreword to Volume I Since World War II the Federal government has recognized the key contribution made by academic research to the knowledge base for U.S. technological innovation, as well as to advanced training of the U.S. science and engineering workforce. Over the decades Federal agencies have provided a generous portion of the total support for academic research. This investment in the nation's scientific and engineering enterprise has paid off over the years by contributing to our competitiveness in global hightechnology markets and has improved the health, welfare, and quality of life of our citizens. It is obvious that stateoftheart academic research facilities are a necessary element in this successful enterprise. Over the past decades the resources to construct and renovate academic research facilities have been provided by loose partnerships among state, private, and Federal agencies. The relative roles of these sources have fluctuated considerably over time, and there is continuing debate about the appropriate Federal contribution to this function. Under these circumstances, the need for accurate, reliable, and comprehensive information on academic research facilities is clear. The National Science Foundation was directed to collect the necessary data by the U.S. Congress in section 108 (42 U.S.C. 1886). A pilot study published in 1986 provided the initial materials for a comprehensive report. Subsequent biennial surveys have included modifications and improvements, and Volumes I and II of this year's report contain the results of the fifth survey in the series. This volume of the report, Volume I, provides a broad quantitative picture of existing research facilities, current construction and renovation efforts, funding sources, plans for future projects, and deferred projects. This report of survey findings does not address the policy issues at hand. Nevertheless, the data presented here can support a useful policy dialogue among all who strive for a healthy and productive U.S. science and engineering academic research enterprise. Neal Lane Director National Science Foundation Acknowledgments The 1994 Survey of Scientific and Engineering Research Facilities at Universities and Colleges was developed and guided by Ann T. Lanier, Senior Science Resources Analyst, Education and Human Resources Program (EDU), Division of Science Resources Studies (SRS), National Science Foundation (NSF), under the overall direction of Mary J. Golladay, EDU Program Director. Guidance and review were provided by Kenneth M. Brown, Director, SRS, and Cora B. Marrett, Assistant Director for Social, Behavioral, and Economic Sciences (SBE), NSF. Review and comments were provided by Carolyn B. Arena, Project Director for the Academic Instrumentation Survey, SRS, and Nathaniel G. Pitts, Director, Office of Science and Technology Infrastructure, NSF. The National Institutes of Health (NIH) cosponsored the project and provided significant financial support as well as guidance and review under the direction of Paul Seder, NIH Office of Science Policy and Technology Transfer. An Expert Advisory Panel contributed to the survey design, the analysis plan, and the review of this report. Members include the following: * Dennis W. Barnes, President, Southeastern Universities Research Association * Richard Blatchly, Professor of Chemistry, Keene State College * Joanne Cate, Principal Administrative Analyst, Resource Management, University of California System * Fred Jones, Dean of the School of Graduate Studies and Research, Meharry Medical College * Robert H. McGhee, Director of Research Facilities Planning, Howard Hughes Medical Institutes * Julie Norris, Assistant Vice President and Director of Sponsored Programs, University of Houston * Stanley Stark, Haines Lundberg Waehler, New York, NY In addition, NSF sought advice on the development of the survey from higher education association and university representatives, who graciously provided information of considerable importance to the success of the project. Administrators at the higher education associations and societies who reviewed and commented on the report included * Marvin E. Ebel, Council on Governmental Relations (COGR) * Howard Gobstein, Association of American Universities (AAU) * Wayne Leroy, Association of Higher Education Facilities Officers (APPA) * Jeanne Narum, Independent Colleges Offices (ICO) * Jerold Roschwalb, National Association of State Universities and LandGrant Colleges (NASULGC) * John G. Stevens, Council on Undergraduate Research (CUR) * William Tibbs, Society for College and University Planning (SCUP) The 1994 survey was conducted by The Gallup Organization of Rockville, Maryland, under contract to NSF (Contract Number SRS9317363). Gregory Gaertner served as overall Project Director from Gallup; Jennifer Spielvogel led the field operation; Manas Chattopadhyay was the project statistician; and Renuka Mahalingam directed the data processing for the survey. Subcontractors for the project were Pelavin Associates and The American Institutes for Research (AIR). Rita Kirshstein headed the Pelavin team, including Jon Cohen who collaborated on the data analysis. Susan Kleimann directed the AIR contributors, including Andrew Rose, Sue Teigen, Renee Ater, and Susan Padrino. Finally, NSF and the project team acknowledge the indispensable contribution of the many officials and staff members at the sample institutions who completed the survey questionnaires. Volume I Analysis Table of Contents Executive Summary xiii Introduction xxv Chapter 1 Existing Research Space: Quantity 11 Highlights 11 Context 12 The Survey Questions 12 Data Considerations 12 Findings 13 Chapter 2 Adequacy and Condition of Research Space 21 Highlights 21 Context 22 The Survey Questions 22 Data Considerations 23 Findings 24 Chapter 3 New Construction and Repair/Renovation 31 Highlights 31 Context 32 The Survey Questions 32 Data Considerations 32 Findings 33 Chapter 4 Funding of Research Facilities Projects 41 Highlights 41 Context 42 The Survey Question 42 Data Considerations 42 Findings 43 Chapter 5 Deferred Construction and Repair/Renovation 51 Highlights 51 Context 52 The Survey Question 52 Data Considerations 53 Findings 53 Chapter 6 Historically Black Colleges and Universities 61 Highlights 61 Context 62 The Survey Question 62 Data Considerations 62 Findings 63 Chapter 7 Predominantly Undergraduate Institutions 71 Highlights 71 Context 72 The Survey Question 72 Data Considerations 73 Findings 74 Appendix A Technical Notes A2 Appendix B List of Sampled Institutions B2 Appendix C Survey Questionnaire C2 Appendix D Reference List D2 List of Text Tables Table 11. Amount of science and engineering research space by institution type: 1994 14 Table 12. Science and engineering research space utilization: 1994 15 Table 13. Trends in the amount of science and engineering research space by institution type: 1988!1994 16 Table 14. Trends in the average amount of science and engineering research space per institution by institution type: 1988!1994 17 Table 15. Trends in percentage of leased science and engineering research space by institution type: 1988!1994 18 Table 16. Trends in the amount of science and engineering research space by field: 1988!1994 19 Table 17. Percentage of institutions with science and engineering research space by institution type and field: 1994 110 Table 18. Amount and distribution of space for laboratory animal facilities by institution type: 1994 111 Table 21. Percentage of institutions reporting inadequate amounts of science and engineering research space in existing fields by institution type and field: 1994 24 Table 22. Institutional assessment of quality/condition of science and engineering research facilities by institution type: 1994 25 Table 23. Trends in the percentage of total stock of science and engineering research space requiring repair/renovation or replacement by field: 1988!1994 27 Table 24. Percentage of animal care research space meeting government regulations by institutional type: 1994 28 Table 31. Trends in expenditures to construct science and engineering research facilities by institution type: 1986!1995 (constant dollars) 34 Table 32. Trends in expenditures to construct science and engineering research facilities by institution type: 1986!1995 (current dollars) 34 Table 33. Trends in expenditures for capital projects costing over $100,000 to repair/renovate science and engineering research facilities by institution type: 1986!1995 (constant dollars) 36 Table 34. Trends in expenditures for capital projects costing over $100,000 to repair/renovate science and engineering research facilities by institution type: 1986!1995 (current dollars) 36 Table 35. Trends in expenditures for science and engineering research facilities repair/renovation projects costing less than $100,000 by institution type: 1990!1993 37 Table 36. Trends in expenditures for capital projects to construct science and engineering research facilities by field: 1986!1995 38 Table 37. Trends in expenditures for capital projects to repair/renovate science and engineering research facilities by field: 1986!1995 39 Table 38. Percentage of institutions doing construction or repair/renovation to science and engineering research space by institution type: 1992!1993 310 Table 39. Trends in percentage of institutions starting capital projects to construct science and engineering research facilities by institution type: 1986!1995 311 Table 310. Trends in percentage of institutions starting capital projects to repair/ renovate science and engineering research facilities by institution type: 1986!1995 312 Table 311. Trends in percentage of institutions starting capital projects to construct science and engineering research facilities by field: 1986!1995 313 Table 312. Trends in percentage of institutions starting capital projects to repair/r renovate science and engineering research facilities by field: 1986!1995 314 Table 313. Cost of planned construction and repair/renovation for laboratory animal facilities by institution type: 1994!1995 315 Table 41. Trends in the sources of funding for capital projects to construct and repair/renovate science and engineering research facilities by institution type: 1986!1993 43 Table 42. Trends in the sources of funding for capital projects to construct and repair/renovate science and engineering research facilities by institution control: 1986!1993 46 Table 43. Trends in public institution's sources of funding for construction of science and enginneering research facilities: 1986!1993 47 Table 44. Trends in private institution's sources of funding for construction of science and engineering research facilities: 1986!1993 48 Table 45. Trends in public institutions' sources of funding for repair/renovation of science and engineering research facilities: 1986!1993 410 Table 46. Trends in private institutions' sources of funding for repair/renovation of science and engineering research facilities: 1986!1993 411 Table 51. Percentage of institutions with needed but deferred capital projects to construct and/or to repair/renovate science and engineering research facilities by institution type and project type: 1994 54 Table 52. Expenditures for deferred capital projects to construct or repair/renovate science and engineering research facilities by institution type and type of project: 1994 55 Table 53. Percentage of institutions with deferred capital projects to construct and/or to repair/renovate science and engineering research facilities by field and project type: 1994 57 Table 61. Trends in the amount of space assigned to science and engineering fields at Historically Black Colleges and Universities: 1988!1994 63 Table 62. Trends in the distribution of science and engineering research space at Historically Black Colleges and Universities by field: 1988!1994 64 Table 63. Assessment of quality/condition of science and engineering research facilities at Historically Black Colleges and Universities: 1988!1994 65 Table 64. Trends in construction of science and engineering research facilities at Historically Black Colleges and Universities: 1986!1995 66 Table 65. Trends in repair/renovation of science and engineering research facilities at Historically Black Colleges and Universities: 1986!1995 66 Table 66. Trends in the sources of funding for capital projects to construct science and engineering research facilities at Historically Black Colleges and Universities: 1986!1993 67 Table 67. Trends in the sources of funding for capital projects to repair/renovate science and engineering research facilities at Historically Black Colleges and Universities: 1986!1993 68 Table 68. Historically Black Colleges and Universities with need for capital projects to construct or repair/renovate science and engineering research facilities: 1994 69 Table 71. Science and engineering research space at predominantly undergraduate institutions: 1994 75 Table 72. Assessment by predominantly undergraduate institutions of quality/condition of science and engineering research space: 1994 76 Table 73. Capital project costs for science and engineering research space at predominantly undergraduate institutions: 1992!1993 77 Table 74. Sources of funding for capital projects of science and engineering research space at predominantly undergraduate institutions: 1992!1993 (percentage) 78 Table 75. Sources of funding for capital projects of science and engineering research space at predominantly undergraduate institutions: 1992!1993 (dollars in millions)) 79 Table 76. Deferred capital project costs of science and engineering research space at predominantly undergraduate institutions: 1992!1993 710 Table A1. Number of institutions in the survey universe of researchperforming universities and colleges: weighted estimates, 1988 A3 Table A2. Number of institutions in the 1990, 1992, and 1994 samples of researchperforming universities and colleges A6 Table A3. Academic institutions response rates, by category of institutions: 1994 A8 Table A4. Standard errors for selected estimates A11 Table A5. Composite FixedWeighted Price Index for Construction inflation adjustments A15 List of Executive Summary Figures Figure 1. Top 100 doctorategranting institutions have 72 percent of the total 127 million net assignable square feet of science and engineering research space: 1994 xv Figure 2. Dollars for construction projects for science and engineering research space declined since fiscal years 1990!1991 xviii Figure 3. Public and private institutions have different funding sources of capital projects for construction and repair/renovation of science and engineering research space: 1994 xix Figure 4. Unfunded science and engineering capital needs total $5,744 million xxi Figure 5. Funding for construction at Historically Black Colleges and Universities has declined xxiii List of Chapter Figures Figure 11. Top 100 doctorategranting institutions have 72 percent of the total 127 million net assignable square feet of science and engineering research space: 1994 14 Figure 12. Total net assignable square feet of science and engineering research space remains proportional by institution type 16 Figure 31. Dollars for construction projects for science and engineering research space declined since fiscal years 1990!1991 35 Figure 41. Public and private institutions have different funding sources of capital projects for construction and repair/renovation of science and engineering research space: 1994 45 Figure 51. Unfunded science and engineering capital needs total $5,744 million 55 Figure 71. Predominantly undergraduate institutions have relatively little science and engineering research spaces 74 Executive Summary Universities and colleges are a critical component of the nation's research system. The availability and condition of research facilities at these institutions influence the ability of scientists and engineers to conduct research and train the future science and engineering workforce. Numerous Congressional committees have expressed concerns about the quality of these facilities and the costs of maintaining them. Hearings held in both House and Senate committees on science and technology in the mid1980s led to the conclusion that the condition of these facilities posed a "serious and ongoing problem . . . ." However, insufficient information existed to assess the extent of the problem. Recognizing the need for information on the amount and quality of scientific and engineering (S&E) research space, Congress mandated that the National Science Foundation (NSF) gather this information and report it to Congress: The National Science Foundation is authorized to design, establish, and maintain a data collection and analysis capability in the Foundation for the purpose of identifying and assessing the research facilities needs of universities and colleges. The needs of universities by major field of science and engineering, for construction and modernization of research laboratories, including fixed equipment and major research equipment, shall be documented. University expenditures for the construction and modernization of research facilities, the sources of funds, and other appropriate data shall be collected and analyzed. The Foundation, in conjunction with other appropriate Federal agencies, shall report the results to the Congress. The first report shall be submitted to the Congress by September 1, 1986 (42 U.S.C. 1886). Since 1986, NSF has collected data on a biennial basis to address these concerns of Congress. The first study, a "quick response" survey, provided limited data regarding S&E facilities issues. In 1988, 1990, 1992, and 1994, fullscale surveys have provided considerable information about the nations' academic research facilities. This report describes the findings from the 1994 survey and places them in historical context by comparing results with those from earlier surveys. Following a brief discussion of the study methods, the remainder of this executive summary addresses several key questions regarding S&E research space in researchperforming universities and colleges: * How much space is available for S&E research, and how has this changed over time? * How much more space is needed? * How good is the existing space, and what improvements are needed? * What is being done to address these needs? * Where does the money come from? * What more remains to be done? The final section of the summary profiles two distinct groups of institutions that play important roles in the training of future scientists and engineers: Historically Black Colleges and Universities (HBCUs) and a select group of academic institutions that are oriented primarily to undergraduate education. These predominantly undergraduate institutions consist of comprehensive universities and liberal arts colleges. What Methods Did This Study Use? The 1994 Survey of Scientific and Engineering Research Facilities at Universities and Colleges collected data from a universe of 565 institutions, which included all those with research and development (R&D) expenditures of $50,000 or more and HBCUs with any R&D expenditures. The 1994 survey was mailed to all sampled institutions in the fall of 1993. Extensive telephone followup was used to elicit a 93 percent response rate and to resolve questions regarding incomplete or inconsistent responses. Sampled institutions that had participated in the 1992 survey were also sent a computer generated "facsimile" of their previous responses. (See Appendix A, Technical Notes, for a detailed description of the sampling procedures and data collection methods.) How Much Space Is Available for S&E Research, and How Has This Changed Over Time? In 1994, universities and colleges devoted about 282 million net assignable square feet (NASF) of space to S&E fields. Of this space, about 127 million NASF was devoted to research.1 The top 100 institutions in R&D expenditures housed the most S&E research space, 91 million NASF, comprising about 72 percent of all S&E research space (Figure 1).2 The top 100 institutions also had 82 percent of total academic R&D expenditures. Since 1988, the amount of S&E research space has increased steadily, growing at an average rate of about 2 percent per year. From 1988 to 1994, the available S&E research space grew from 112 to 127 million NASF.3 Most of this increase occurred in the top 100 institutions; engineering experienced the largest growth of any single S&E field. How Much More S&E Research Space Is Needed? To answer this question, institutions assessed, for each S&E field, how adequate the amount of existing space was for current research programs. Of those institutions with some research space in each field, at least 40 percent reported inadequate amounts of space in four S&E fields: * Engineering * Physical sciences * Biological sciences outside of medical schools; and * Medical sciences in medical schools. The institutions that had the most S&E research space also expressed the greatest need for more space. Over half of the top 100 institutions reported inadequate amounts of research space in all four of the above S&E fields. Fewer than 40 percent of the nondoctorategranting institutions reported inadequate amounts of space in any field. How Good Is Existing S&E Research Space, and What Improvements Are Needed? Of all S&E research space, over a quarter (33 million NASF) was considered suitable for the most sophisticated research, a result driven by the high quality of space at doctorategranting institutions. Both the top 100 institutions and other doctorategranting institutions designated about 27 percent of their overall space as "suitable for use in the most highly developed and scientifically sophisticated research . . . ." The nondoctorategranting institutions classified 16 percent of their space into this category. The second category of space, space that is "suitable for most uses," included another 33 percent of the total S&E research space. This percentage was fairly consistent across doctorategranting institutions. Nondoctorategranting institutions rated 42 percent of their space as belonging to this category. Twentythree percent of existing S&E research space was considered to need limited repair/renovation. This percentage was generally consistent across institution types. Thirteen percent of S&E research space was rated as needing major repair/renovation, and another 4 to 5 percent required replacement. Repair/renovation needs were concentrated in a few S&E fields. The single field requiring the largest proportion of major repair/renovation or replacement was agricultural sciences, in which 22 percent of the existing research space was rated in this category. However, few institutions (20 percent) had agricultural sciences research space. In environmental sciences, about 19 percent of the existing research space needed major repair/renovation or replacement. Other fields in which 15 percent or more of the total S&E research space needed major repair/renovation or replacement included the biological sciences outside of medical schools (19 percent); the physical sciences (18 percent); the medical sciences, both within and outside medical schools (17 percent for both); engineering (15 percent); and the biological sciences in medical schools (15 percent). What Is Being Done to Address S&E Research Facility Needs? In fiscal years 19921993, universities and colleges began over $2,812 million of construction for S&E research space. Most of this construction activity, approximately $2,000 million or 72 percent, occurred at the top 100 institutions. Other doctorategranting institutions spent approximately $691 million on the construction of new S&E research space, and nondoctorategranting institutions spent $92 million. For the first time since NSF began collecting data on S&E research facilities, in fiscal years 19921993, the inflationadjusted amount (the constant dollar amount) spent on construction of S&E research space declined.4 (See Figure 2 on the following page.) Repair/renovation spending for existing S&E research space also declined to $837 million in fiscal years 19921993 from $861 million in fiscal years 19901991, continuing the decline that began in fiscal years 19881989. The decline in repair/renovation was driven by a decline at the top 100 institutions. Some level of capital projects (either construction or repair/renovation) took place at almost half (46 percent) of all institutions during fiscal years 19921993. However, almost all of the top 100 institutions (90 percent) began capital projects, while 25 percent of the nondoctorategranting institutions did so. Fortyeight percent of the other doctorategranting institutions began capital projects in fiscal years 1992 1993. As noted above, agricultural sciences was the S&E field with the largest proportion of space that needed major repair/renovation or replacement. With the exception of S&E fields in medical schools (biological and medical sciences), agricultural sciences was also the field in which the highest percentage of institutions was undertaking construction projects. In fiscal years 19921993, 27 percent of all institutions with research space in the agricultural sciences had construction projects underway in that field. Institutions were more likely to have begun repair/renovation projects than construction projects in most S&E fields. The two exceptions were agricultural sciences and mathematics. In agricultural sciences, 27 percent of all institutions began construction, and 18 percent began repair/renovation projects. In mathematics, 2 percent of all institutions began both construction and repair/renovation projects. Who Funds Capital Projects? In fiscal years 19921993, public and private institutions drew upon substantially different sources to fund the construction and repair/renovation of S&E research space. Public institutions relied primarily on state and local funding, which accounted for 46 percent of their total funding for capital projects. Private institutions relied primarily on institutional contributions (institutional funds, taxexempt bonds, and other debt); these funds accounted for 62 percent of the total funding for their capital projects (Figure 3). At public institutions, capital funding declined about 4.5 percent between fiscal years 19861987 and fiscal years 19921993. Private institutions experienced a large drop in capital funding from state and local governments between fiscal years 19901991 and the following two fiscal years of 19921993. However, the high level of state and local capital funding that private institutions received in fiscal years 19901991 was somewhat atypical; it was considerably higher than at any time since fiscal years 1986 1987. What More Remains to Be Done? Congress is concerned with determining what universities and colleges need with regard to S&E research space. Determining need is a complex matter, because what is needed must be placed within a framework that is realistic from a budgetary perspective. In an effort to measure real (as opposed to speculative) needs, the 1994 survey adopted a conservative approach to this issue. (See Appendix A, Technical Notes, for a discussion of differences from previous surveys.) It combined institutions' assessments of S&E research space needs with deferred plans to repair/renovate or to construct S&E research space. Institutions reported whether an approved institutional plan existed that included "any deferred space that requires new construction or repair/renovation." Four criteria were used to define deferred space: * The space must be necessary to meet the critical needs of current faculty or programs; * Construction must not be scheduled to begin during fiscal years 19941995; * Construction must not currently have funding; and * The space must not be for developing new programs or expanding the number of faculty. Using these standards, respondents were asked to estimate for each S&E field the construction and repair/renovation costs of such deferred projects. The strength of this approach is the fact that institutions must make decisions about the distribution of scarce resources to develop and approve these plans. In short, these plans are not wish lists. Therefore, when approved institutional plans include construction or repair/renovation that is deferred (i.e., not planned for fiscal years 19941995), it is reasonable to see these deferred projects as needed projects yet to be addressed. Forty percent of responding institutions could report deferred space meeting these criteria; sixty percent did not; the information reported here is based on reports from the forty percent. Although a less formal definition might well lead to a different estimate than the one reported here, the needs expressed based on these criteria provide a framework for meaningful interpretation of results and the development of trends over future years. These estimates reflect a thoughtful process of deliberation and compromise at the responding institutions. Deferred capital projects at the 40 percent of institutions with institutional plans amounted to $5,744 million. Of this, $4,047 million reflected needs for construction, and $1,697 million reflected needs for repair/renovation (Figure 4). The fields in which capital projects were most often deferred included engineering, with 18 percent of responding institutions reporting deferred capital projects; the physical sciences, with 16 percent reporting deferred capital projects; the medical sciences in medical schools, with 16 percent; and the biological sciences outside of medical schools, with 14 percent. What Is the State of S&E Research Facilities at Historically Black Colleges and Universities? Historically Black Colleges and Universities (HBCUs) have played an important role in the education of black students at all higher education levels for over 100 years. These universities and colleges consist of both public and private institutions as well as twoyear, fouryear, and professional schools. In 1991, approximately 269,000 students attended the 105 institutions of higher education considered HBCUs by the U.S. Department of Education. Although the HBCUs have considerably less S&E research space than other researchperforming institutions, the HBCUs are an important source of science and engineering degrees for the black students who are currently enrolled in college.5 Researchperforming HBCUs contained 7.9 million NASF of S&E space, of which 2.2 million were devoted to research. Among a panel of 29 institutions that has been sampled consistently since 1988, the amount of S&E research space dropped slightly, from 1.78 million NASF in 1988 to 1.76 million in 1994. HBCUs reported that their S&E research space was in fairly good shape. Over 30 percent of space was reported to be suitable for the most sophisticated research, and 9 percent was in need of major repair/renovation. Construction starts at HBCUs continued a precipitous decline. In fiscal years 19861987, the panel of 29 HBCUs spent $83.2 million in constant dollars on construction of S&E research space. By fiscal years 19921993, this figure had dropped to $8.6 million, about a tenth of its earlier level. (See Figure 5 on the following page.) HBCUs continued to receive most of their funding for S&E research space construction and repair/renovation from the Federal government. What Is the State of S&E Research Facilities at Predominantly Undergraduate Institutions? In the National Science Foundation Authorization Act of 1994, the Committee on Science, Space, and Technology expressed concern "that NSF's biennial survey of academic research facilities needs ... has not focused adequately on the needs of undergraduate institutions." The 1994 facilities survey, in the field at the time, was not designed to collect data regarding the specific needs of undergraduate institutions. Furthermore, the sampling frame for this study did not represent all the types of undergraduate institutions of concern to Congress. Nevertheless, the 1994 survey and sample can provide insights into several issues regarding the S&E research facilities of a select group of undergraduate institutions. Predominantly undergraduate institutions that engaged in separately budgeted S&E research had a total of approximately 25 million NASF of space in S&E disciplines. Analysis divided these institutions into two groups: comprehensive universities (institutions that granted a master's degree as well as a bachelor's degree) and liberal arts colleges. Over 80 percent of the S&E space at predominantly undergraduate institutions was in the comprehensive universities. Of all S&E space at predominantly undergraduate institutions, only 17 percent was devoted to S&E research, as might be expected because of their educational mission. Almost half of this space was characterized as "effective for most purposes," though not generally suitable for the most advanced research. Approximately 3 percent of the S&E research space was classified as in need of replacement, about the same as all other institutions. To address S&E research space needs, these institutions invested a total of about $92.3 million in capital projects in fiscal years 19921993. Of this total, $65.2 million was spent on construction and $27.1 million was spent on repair/renovation. Of the predominantly undergraduate institutions which had an approved institutional plan that included deferred or unfunded capital projects for S&E research space, deferred capital projects totaled over $356.6 million. Introduction Background Universities and colleges are a critical component of the nation's science and engineering (S&E) research system. The availability and condition of research facilities at these institutions influence the ability of scientists and engineers to conduct research and train future scientists. Numerous Congressional committees have expressed concerns about the quality of these facilities and costs of maintaining them. Hearings held in both House and Senate committees on science and technology in the mid1980s led to the conclusion that the condition of these facilities posed a "serious and ongoing problem . . . ." However, insufficient information existed to assess the extent of the problem. Recognizing the need for information on the amount and quality of S&E research space, Congress mandated that the National Science Foundation (NSF) gather this information and report it to Congress: The National Science Foundation is authorized to design, establish, and maintain a data collection and analysis capability in the Foundation for the purpose of identifying and assessing the research facilities needs of universities and colleges. The needs of universities by major field of science and engineering, for construction and modernization of research laboratories, including fixed equipment and major research equipment, shall be documented. University expenditures for the construction and modernization of research facilities, the sources of funds, and other appropriate data shall be collected and analyzed. The Foundation, in conjunction with other appropriate Federal agencies, shall report the results to the Congress. The first report shall be submitted to the Congress by September 1, 1986 (42 U.S.C. 1886). Since 1986, NSF has collected data on a biennial basis to address these concerns of Congress. The first study, a "quick response" survey, provided limited data regarding S&E facilities issues. In 1988, 1990, 1992, and 1994, fullscale surveys have provided considerable information about the nation's academic research facilities. The Survey and its DesignThe Survey and its Design The 1994 Survey of Scientific and Engineering Research Facilities at Universities and Colleges, like earlier efforts, collected data on the amount of S&E research space in the nation's higher education institutions, the adequacy and condition of this space, the extent to which universities and colleges were constructing facilities and repairing/renovating space, and the funding of this activity. In addition, the 1994 survey gathered for the first time information about unfunded and deferred capital projects for S&E research facilities. The sample for the 1994 survey was designed to provide efficient and unbiased estimates of the amount of S&E research space in universities and colleges and to retain, as much as possible, comparability with the 1992 sampling procedures. This sample represented a universe of 565 institutions with more than $50,000 in research and development (R&D) as well as Historically Black Colleges and Universities (HBCUs) with any R&D expenditures. As was the case in previous years, most institutions in the sample were selected with a probability proportional to the square root of their R&D expenditures in thousands. (See Appendix A, Technical Notes, for a more complete discussion of sampling procedures.) The final sample of 309 universities and colleges, which represented the universe of 565, included the following: * All of the top 100 universities and colleges in terms of R&D expenditures (n=100); * Other public doctorategranting universities (n=50); * Other private doctorategranting universities (n=35); * Public nondoctorategranting institutions (n=72); and * Private nondoctorategranting institutions (n=52). The sample of HBCUs was distributed among the 309 universities and colleges. These 309 universities and colleges are described as researchperforming institutions throughout the report. The 1994 survey was mailed to all sampled institutions in the fall of 1993. Extensive telephone followup was used to elicit a high response rate and to resolve questions regarding responses. Sampled institutions that had participated in the 1992 survey were also sent a computergenerated "facsimile" of their previous responses. Overall, 93 percent of all universities and colleges sampled completed the survey, an increase from 89 percent in the 1992 survey. The Report The 1994 report has been reformatted to help readers obtain answers to a number of key policy concerns. Each chapter contains the following sections: * Highlights, a summary of key findings; * Context, the rationale and background for the data presented in the chapter; * The Survey Question(s), a description of the question or questions around which the chapter is focused; * Data Considerations, a presentation of data limitations or interpretations; and * Findings, tables, graphs, and text that address questions frequently posed about S&E research facilities. This report provides information presented in previous reports, particularly data pertaining to trends in the amount, condition, capital activity, and funding of S&E research space, as well as a profile of HBCUs. In addition, the 1994 report includes a chapter on deferred and unfunded construction and repair/renovation projects as well as a profile of institutions that are predominantly undergraduate in orientation. Although information on animal care facilities was presented as a separate chapter in previous reports, this information is incorporated into individual chapters in the current report. In most chapters, differences among types of institutions and S&E fields are presented. Throughout the report, type of institution refers to the following categories: * Doctorategranting, which includes . The top 100 institutions in R&D expenditures . The other doctorategranting institutions not in the top 100 * Nondoctorategranting Fifteen percent of the HBCU institutions are doctorategranting; 85 percent are classified as nondoctorategranting. Throughout this report, HBCUs are included in the data of their appropriate institution type except in Chapter 7, which focuses on predominantly undergraduate institutions. In this chapter, nondoctorate HBCU data are reported separately. For this survey and report, the S&E fields include the following: engineering; physical sciences; environmental sciences; mathematics; computer sciences; agricultural sciences; biological sciences, both in universities and colleges and in medical schools; medical sciences, both in universities and colleges and in medical schools; psychology; social sciences; and other sciences, not elsewhere classified. Chapter 1 presents findings on the amount of research space available in S&E fields at researchperforming institutions, currently and over time. Chapter 2 examines the adequacy of the amount of S&E research space as well as its condition as assessed by the institutions. Chapter 3 provides information on the costs in constant and current dollars of constructing facilities and repairing/renovating S&E research facilities. The sources of funds for these capital projects are presented in Chapter 4. Chapter 5, new to the 1994 report, examines deferred and unfunded construction and repair/renovation projects. Chapter 6 provides a profile of HBCUs, and Chapter 7, also new to this report, profiles institutions that are predominantly undergraduate in their focus. Several appendices provide interested readers with more detailed information. Appendix A, Technical Notes, presents additional material about the study design and methodology. Appendix B includes a list of sampled institutions. Appendix C contains the survey instrument. Appendix D lists references. A second volume, Statistical Tables, contains detailed statistical information. Chapter 1 Existing Research Space: Quantity Highlights . . . * In 1994, the nation's researchperforming academic institutions devoted a total of 282 million net assignable square feet (NASF) to science and engineering (S&E) fields. This total included space used for instruction as well as space used for research. Of the 282 million NASF, research occupied 127 million NASF. * As in prior years, the top 100 institutions in research and development (R&D) expenditures had the largest share of S&E research space. Of the 127 million NASF that was devoted to S&E research, the top 100 institutions had 91 million NASF, or 72 percent. * The amount of S&E research space increased steadily, from 112 million NASF in 1988 to 127 million NASF in 1994. This increase amounts to about 2 percent a year, with most of the growth occurring at the top 100 research institutions. Context To understand the research facility needs of universities and colleges, it is necessary to know how much research space scientists and engineers in U.S. universities and colleges currently use. Has the amount of space increased or decreased over time? What are the trends for particular types of institutions? What are the trends for particular fields? Do different S&E fields require different amounts of space? This chapter addresses these issues. The next chapter will address whether the amount of research space is adequate and whether the condition of the space is sufficient for conducting competitive research. The Survey Questions This chapter discusses the information reported in Item 1a and Item 1b of the survey. (See Appendix C.) For each S&E field individually, and for all S&E fields combined, Item 1a collects data in units of NASF on * Instructional and Research NASF. This is total space; it includes space that is used for instruction and space that is used for research, and * Research NASF. This is space that is used for research; it does not include space that is used for instruction. Item 1a also asks for the total NASF for instruction and research for all nonscience fields combined as well as a total for instruction and research NASF for both S&E and nonscience fields. For all S&E fields combined, Item 1b requests the amount of research NASF that is leased. Data Considerations "How much space do universities and colleges devote to S&E research?" Although this question may appear to be straightforward, several issues complicate the response. * Space may be used for more than one purpose or be shared by more than one field. Examples include a laboratory that is used for research only part of the time or a building that is shared by two or more fields. For multipurpose or shared space, the survey asks respondents to prorate the space. For example, if a laboratory is used for research 30 percent of the time, respondents should count 30 percent of the laboratory's NASF as research space. If mathematics and computer sciences both use the same laboratory, the NASF reported for each field should reflect the amount of space prorated by the amount of time that field uses the space. * Some fields require more space for research than others. For example, agricultural research requires considerably more space than mathematics research. Thus, a larger amount of research space in any field does not necessarily translate into sufficient research space for that field. * In the 1994 survey, research is defined more broadly than in the 1992 survey. However, this change in definition does not reflect a change in how institutions actually report S&E research space. The 1994 definition includes all research and development activities that are budgeted and accounted for. In some cases, it can also include departmental research not separately budgeted. In prior years, institutions were asked to exclude departmental research that was not a separate budget item. Conversations with respondents from earlier surveys revealed that some departmental research had been included; thus, the current definition of research reflects what institutions had been reporting all along. Findings How Much Space Was Available for S&E Research? In 1994, the nation's 565 researchperforming academic institutions had a total of 511 million NASF of instructional and research space in all academic fields. (See Table 11 on the following page.) S&E fields occupied 282 million NASF of this total, and research space within the S&E fields comprised 127 million NASF. Other National Science Foundation (NSF) surveys show that S&E research spending grew by just under 9 percent per year and graduate enrollment increased by nearly 4 percent per year.1 The top 100 institutions contained more space in all academic fields than all other types of institutions. Although the top 100 institutions made up 18 percent of the 565 researchperforming universities and colleges, they accounted for 52 percent of the space in all academic fields (265 million NASF). These top 100 institutions had an even greater percentage of the total S&E research space, containing 72 percent of all S&E research space (Figure 11). The top 100 institutions had 82 percent of the total academic R&D expenditures in 1991. How Was S&E Space Utilized? In 1994, the total space devoted to S&E fields, including both instruction and research, comprised 55 percent of the total academic space at the nation's researchperforming institutions (Table 12). The ratio of S&E space to total academic space varied by the type of institution. The top 100 universities devoted 65 percent of all academic space to S&E. Nondoctorategranting institutions used 35 percent of their total academic space for S&E fields. The percentage of S&E space that was used for research also varied. In the top 100 institutions, 53 percent of their S&E space was devoted to research; in other doctorategranting institutions, 38 percent of the S&E space was devoted to research; and nondoctorategranting institutions devoted 17 percent of the S&E space to research. Has the Amount of S&E Research Space Increased? Since 1988, the amount of S&E research space has increased steadily, from 112 million NASF in 1988 to 127 million in 1994. (See Table 13 on the following page.) These numbers represent a total increase of approximately 13 percent. Most of this increase resulted from steady growth in research space at the top 100 institutions. In 1988, these universities devoted 81 million NASF to research space; by 1994, research space had grown by 10 million NASF in the top 100 universities, to 91 million NASF. Research space in other doctorate granting institutions also increased between 1988 and 1994, from 27 million NASF to 31 million NASF. Research space at nondoctorategranting institutions remained constant (Figure 12). However, the average amount of S&E research space per institution has declined since 1992 (Table 14). This decline resulted solely from a decline in the average amount of research space in doctorategranting institutions that are not in the top 100. At these institutions, the average amount of S&E research space dropped from 154 thousand NASF in 1992 to 141 thousand NASF in 1994. During that same period, the average amount of research space at the top 100 institutions increased from 875 thousand to 910 thousand NASF. At nondoctorategranting institutions, the average space increased slightly, from 20 thousand to 22 thousand NASF. How Much S&E Space Was Leased? Some universities and colleges lease academic and research space. In 1994, researchperforming universities and colleges leased 2 percent of their total S&E research space. (See Table 15 on the following page.) In 1994, the top 100 institutions leased the highest percentage of their S&E research space, 4.9 percent, continuing an upward trend from previous years. Other doctorategranting institutions leased 2 percent of their total S&E research space, and nondoctorategranting institutions leased less than 1 percent of their total S&E research space. How Was Research Space Distributed Across S&E Fields? In 1994, more institutions had research space in the biological sciences (87 percent) and the physical sciences (86 percent) than in any other S&E field. (See Table 16 on the following page.) In contrast, 20 percent of all researchperforming universities and colleges had research space in the agricultural sciences. However, the total amount of NASF of research space in the agricultural sciences (20 million NASF) was greater than the total in either the physical sciences, the biological sciences, or medical sciences in medical schools (17 million NASF each). Conducting agricultural research, thus, appears to require a relatively large amount of space. Between 1988 and 1994, most of the growth in S&E research space could be attributed to growth in a few fields. Engineering research space grew the most, from 16 million NASF of research space in 1988 to 21 million NASF in 1994. Medical school space in the biological sciences and medical school space in the medical sciences each grew by 3 million NASF during this period. The distribution of research across all S&E fields shows growth in the NASF devoted to engineering space. In 1988, engineering occupied 14 percent of all research space; by 1994, engineering occupied 16 percent of this space. The computer sciences and mathematics each occupied 1 percent of all research space in all survey years, the least of all S&E fields. The top 100 universities were more likely to have research space in every S&E field than other types of institutions. Among the top 100 institutions, 93 percent contained research space in the biological sciences outside of medical schools, and 91 percent had research space in the physical sciences (Table 1 7).2 How Much Space Was Devoted to Facilities for Laboratory Animals? Scientific research in several fields relies on animals. Federal laws and regulations have been instituted to protect animals used in research and to ensure that the space in which they are kept is adequate (42 U.S.C. 289d and 9 CFR Part 3). Eightyseven percent of all researchperforming universities and colleges had laboratory animal facilities. Doctorategranting institutions were more likely to have such facilities than were nondoctorategranting institutions, 95 percent and 77 percent respectively (Table 18). In 1994, researchperforming universities and colleges devoted approximately 11.3 million NASF to laboratory animal facilities; of this total space, approximately 8.6 million NASF, or 76 percent was used for research. Doctorategranting universities contained 10.6 million NASF, or 94 percent of the total animal laboratory space. These institutions used 78 percent of laboratory animal space for research. Nondoctorategranting institutions had 7 percent of the total laboratory animal space and devoted 50 percent of it to research. Chapter 2 Adequacy and Condition of Research Space Highlights * Forty percent or more of all researchperforming institutions indicated inadequate amounts of science and engineering (S&E) research space in engineering, the physical sciences, the biological sciences outside of medical schools, and the medical sciences in medical schools. * The top 100 institutions in terms of research and development (R&D) expenditures were more likely to report inadequate amounts of S&E research space than other types of researchperforming institutions. * Twentysix percent of the S&E research space at researchperforming institutions was considered to be ". . . suitable for use in the most scientifically sophisticated research." * A combined total of 17 percent of the S&E research space at researchperforming institutions was rated as needing either major repair/renovation or replacement. Context The amount of S&E research space at researchperforming universities and colleges increased steadily between 1988 and 1994, particularly within the top 100 universities. Growth in the amount of S&E research space does not necessarily mean, however, that enough space exists to meet research needs in all S&E fields; nor does it mean that the condition of existing space is suitable for conducting competitive research. This chapter examines assessments, both currently and over time, of the amount of S&E research space and its condition. The Survey Questions Amount of Research Space Respondents were asked to rate the amount of research space in each field by choosing one of the following: A Adequate amount; sufficient to support all the needs of your research in the field; B Generally adequate amount; sufficient to support most of your research needs in the field but may have some limitations; C Inadequate amount; not sufficient to support the needs of your research in the field; D Nonexistent space but needed; or E Not applicable or not needed. (See Item 2 of the survey in Appendix C.) In this report, inadequate space is defined as either category C, inadequate amount, or category D, nonexistent space but needed. Condition of Research Space For each field, institutional respondents reported the percentage of space falling into one of the following categories: A Suitable for use in the most highly developed and scientifically sophisticated research in the field; B Effective for most purposes but not applicable to category A; C Effective for some purposes but in need of limited renovation or repair; D Requires major repair or renovation to be used effectively; E Requires replacement; or NA Not applicable or no research space in this field. (See Item 3 of the survey in Appendix C.) To determine the overall amount and percentage of space that was rated in each of the abovelisted categories, the amount of research space in each field (reported in Item 1a) was multiplied by the percentage of space reported in each of the above categories and totaled across fields. For example, if a university had 1,000 net assignable square feet (NASF) of research space in environmental sciences and 20 percent of this space "requires replacement," 200 NASF (1,000 x .20) was considered to require replacement. These calculations were performed for each field and each institution and summed to provide the total amount of space meeting each condition. The amount of space meeting each condition was then divided by the total research NASF to provide an overall percentage. Data Considerations The survey measures both the adequacy of the amount of S&E research space and the condition of this space in each S&E field. Responses are based on the assessments of a variety of different individuals, including the survey coordinator at the institution, deans, and other administrators. The two questions gathering information about the adequacy of the amount of research space and its condition thus elicit more subjective responses than do other survey items. Findings Was the Amount of S&E Research Space Sufficient for Current Research Programs? Universities and colleges were more likely to rate research space as inadequate in some S&E fields than in others. Forty percent or more of all institutions indicated inadequate amounts of S&E research space in engineering, the physical sciences, the biological sciences outside of medical schools, and the medical sciences in medical schools (Table 21). The top 100 institutions had considerably more S&E research space than any other type of institution (see Table 11, page 14); and they were generally more likely to indicate that the existing amount of S&E research space was inadequate. The two exceptions were in mathematics and psychology, fields in which more nondoctorategranting than doctorategranting institutions rated S&E research space as inadequate. In four fields, over half of the top 100 institutions reported inadequate amounts of S&E research space: engineering (55 percent); the medical sciences in medical schools (55 percent); the physical sciences (51 percent); and the biological sciences outside of medical schools (51 percent). What Was the Condition of S&E Research Space? Of the S&E research space at institutions (see Table 11, page 14), 26 percent (33 million NASF) of the S&E research space at universities and colleges was rated "suitable for use in the most scientifically sophisticated research." Twentyseven percent of the S&E research space at both categories of doctorategranting institutions was rated this way, and 16 percent of the S&E research space at nondoctorategranting institutions was rated this way (Table 22). Universities and colleges classified a total of 17 percent of their S&E research space as needing either major repair/renovation (13 percent) or replacement (4 percent). There was general consistency among institutions in the percentage of S&E research space assessed as needing major repair/renovation or replacement. A total of 18 percent of the S&E research space at the top 100 institutions was rated as needing major repair/renovation or replacement; a total of 14 percent of the S&E research space at other doctorategranting institutions and a total of 16 percent at nondoctorategranting institutions were rated as needing major repair/renovation or replacement. These similar percentages, however, mask large differences in the actual amounts of space rated as needing major repair/renovation or replacement. The 18 percent of all S&E research space that the top 100 institutions rated in this way represented 16.4 million NASF in 1994; the 16 percent of S&E research space that the nondoctorategranting institutions rated this way represented 800 thousand NASF. In total, 21.6 million NASF in researchperforming institutions needed major repair/renovation or replacement. What Percentage of the Total Amount of Research Space in Each Field Required Either Repair/Renovation or Replacement? In 1994, a higher percentage of the total S&E research space in the agricultural sciences, 22 percent, needed major repair/renovation or replacement than any other field. (See Table 23 on the following page.) While 20 percent of all researchperforming colleges and universities had agricultural research space, this field accounted for a large share of S&E research space, 20 million NASF. (See Table 16, page 19.) Thus, this relatively large need was concentrated in a small number of institutions. In addition to agricultural sciences (22 percent), fields in which 15 percent or more of the total S&E research space needed major repair/renovation or replacement included the following: engineering (15 percent); the physical sciences (18 percent); the environmental sciences (19 percent); the biological sciences, both in medical schools (15 percent) and outside of medical schools (19 percent); and the medical sciences, both within and outside of medical schools (17 percent for each). At this point, there are no evident trends in repair/renovation needs across S&E fields. What Was the Condition of Facilities for Laboratory Animals? Across all researchperforming universities and colleges, 84 percent of the research NASF for animal facilities met government regulations. Seven percent needed major repair/renovation or replacement in order to meet these standards. There was very little variation across institutions in the percentage of laboratory animal space that either met or did not meet government regulations. (See Table 24 on the following page.) Chapter 3 New Construction and Repair Renovation Highlights * In fiscal years 1992!1993, expenditures on projects to construct science and engineering (S&E) research space in researchperforming universities and colleges totaled $2,812 million. This amount represents a decline of $290 million in constant dollars (that is, adjusted for inflation) from the previous two fiscal years. This amount also represents the first decline in construction spending since the National Science Foundation (NSF) began collecting data on S&E research facilities in 1986. * Expenditures on projects to repair/renovate S&E research space also declined from fiscal years 1990!1991 levels. In fiscal years 1992!1993, researchperforming universities and colleges spent $837 million to repair/renovate S&E research space. In fiscal years 1990!1991, these institutions spent $861 million (in constant dollars). This overall decline resulted from a drop in the amount spent on such projects by the top 100 institutions in research and development (R&D) expenditures. * In fiscal years 1992!1993, the biological sciences and the medical sciences accounted for over half of all construction dollars as well as repair/renovation dollars spent for S&E research facilities. * Fiftyfive percent of all researchperforming universities and colleges were either constructing S&E research facilities or repairing/renovating their S&E research space during fiscal years 19921993. Context Studies of higher education institutions have noted the declining state of the physical facilities across university and college campuses. (See Appendix D, References, for more detailed information.) Budgetary constraints have forced many institutions to defer construction of facilities and maintenance of older buildings. As a result, many of these institutions now have less than satisfactory academic and research space. Furthermore, changes in technology that have altered the way in which research is conducted and students are trained in S&E fields have put additional pressures on the nation's universities and colleges to update and replace buildings. This chapter examines the extent to which researchperforming universities and colleges are constructing S&E research facilities and repairing/renovating S&E research space. The Survey Questions Institutions were asked to estimate the researchrelated cost and space for construction and repair/renovation projects begun during fiscal years 1992!1993 and to make the same estimates for projects planned for fiscal years 1994!1995. Project start was defined as the institution's fiscal year in which actual construction or repair/renovation work began or was expected to begin. In the case of multiyear projects, total project costs were allocated to the fiscal year in which the construction or repair/renovation actually began. The reported costs, defined as the cost to complete a project, included planning, construction, and fixed equipment. Projects over $100,000 and under $100,000 were reported separately. If a project was to serve both research and nonresearch purposes, the construction costs and space estimates were to be prorated to reflect the researchrelated portion of the cost. (See Item 4a and Item 7 of the survey in Appendix C.) Data Considerations Data reported in this chapter reflect the extent of construction and repair/renovation activity underway in fiscal years 1992!1993. Tables that report expenditures or costs over time are presented in constant and current dollars but discussed only in terms of 1993 constant dollars. Constant dollars are "inflation adjusted" dollars and compensate for variations in the purchasing power of the dollar over time. Constant dollars thus adjust for the fact that what $100 will buy today is not the same as what $100 would purchase ten years ago or even one year ago. The specific deflator used in this chapter is the Bureau of the Census' Composite FixedWeighted Price Index for Construction which more closely tracks inflation within the construction industry than a more general index does. The fixedweighted price index reflects changes in prices and remains unaffected by changes in the mix of construction projects during any given year. (See Appendix A, Technical Notes, for further discussion of the price index.) Previous reports presented trends in current dollars; however, comparisons in current dollars tend to overstate increases in spending over time because more current dollars are needed to buy the same products each year. In this report, trends are reported in constant dollars and provide a more accurate picture of expenditure trends. Throughout this chapter, as well as the rest of the report, the term "capital projects" refers to either construction projects or repair/renovation activities. Construction always refers to building facilities that currently do not exist; repair/renovation implies remodeling or restoring existing facilities. Findings How Much Did Institutions Spend on Construction and Repair/Renovation Projects? In fiscal years 1992!1993, expenditures for S&E research space construction projects in researchperforming institutions totaled $2,812 million. This amount represented a decline of $290 million from the previous two fiscal years. This amount also represented the first decline in construction spending since NSF began collecting data on S&E research facilities. (See Table 31 and Table 32 on the following page.) All types of institutions experienced decreases in construction spending in constant dollars. However, in relative terms, the top 100 experienced the smallest decrease, a 4 percent decline from $2,107 million in fiscal years 19901991 to $2,029 million in fiscal years 19921993, while the nondoctorategranting institutions experienced the largest decrease, 31 percent from $133 million to $92 million during the same period. (See Figure 31 on the following page.) Expenditures for repair/renovation projects costing over $100,000 also declined from fiscal years 1990!1991 levels. In fiscal years 1992!1993, researchperforming universities and colleges spent $837 million to repair/renovate S&E research space. In fiscal years 1990!1991, these institutions spent $861 million. This overall decline resulted from a drop in what the top 100 institutions spent: $660 million in fiscal years 1990!1991 and $623 million in fiscal years 1992!1993. (See Table 33 and Table 34 on the following page.) Other doctorategranting institutions spent approximately $12 million more in fiscal years 1992!1993 than in the two previous fiscal years to repair/renovate S&E research space, while the spending of nondoctorategranting institutions remained stable. Expenditures for S&E research facility repair/renovation projects costing less than $100,000 told a somewhat different story. Expenditures increased by twothirds, from $152 million in fiscal years 1990!1991 to $241 million in fiscal years 1992!1993. Other doctorategranting institutions were the only type of institution that experienced a decline in these types of expenditures. (See Table 35 on the following page.) How Much Did Institutions Spend on Construction and Repair/Renovation of S&E Research Space in Different Fields? The biological sciences and the medical sciences accounted for over half of all construction dollars spent by researchperforming universities and colleges in fiscal years 1992!1993. (See Table 36 on the following page.) Overall, $2,812 million were spent; academic institutions spent $999 million to construct research space in the medical sciences and $633 million to construct research space in the biological sciences. Within each of these fields, the majority of the construction dollars went to construction of medical facilities. Institutions spent the next largest amounts of money to construct research space in the physical sciences, $337 million, followed by engineering, $286 million, and the agricultural sciences, $210 million. The largest increase in spending for construction of S&E research space between fiscal years 1990!1991 and fiscal years 1992!1993 occurred within the medical sciences, from $841 million to $999 million. Most S&E fields experienced a decline in construction spending. Spending on engineering construction, for example, dropped from $412 million in fiscal years 1990!1991 to $286 million in fiscal years 1992!1993. Spending to construct research space in the physical sciences declined from $448 million to $337 million. Despite the relatively high level of funding in fiscal years 1992!1993 to construct research space in the biological sciences, spending in this field dropped over $200 million, from $867 million in fiscal years 1990!1991 to $633 million in fiscal years 1992!1993. Repair/renovation spending for research space across S&E fields demonstrated similar patterns. In fiscal years 1992!1993, researchperforming institutions spent more to repair/renovate S&E research space in the medical sciences ($262 million) and the biological sciences ($224 million) than in any other S&E field (Table 37). As was the case with construction, the majority of repair/renovation dollars within these two fields went to research space in medical schools. Researchperforming universities and colleges spent $139 million to repair/renovate engineering research space and $134 million to repair/renovate research space in the physical sciences. In no other individual S&E field did repair/renovation expenditures exceed $40 million. Some S&E fields experienced increases in spending to repair/renovate research space between fiscal years 1990!1991 and fiscal years 1992!1993, while others experienced declines. The largest increase occurred in engineering. Researchperforming universities and colleges spent $85 million to repair/renovate research space in this field in fiscal years 1990!1991 and $139 million the following two fiscal years. Despite a large share of all repair/renovation dollars, the biological sciences experienced the largest decrease, from $270 million in fiscal years 1990!1991 to $224 million in fiscal years 1992!1993. To What Extent Were Universities and Colleges Involved in Capital Projects? During fiscal years 1992!1993, 55 percent of all researchperforming institutions undertook some type of S&E capital project costing over $100,000, either construction or repair/renovation (Table 38). Ninetyfive percent of the top 100 institutions began some type of capital project during this period. Fiftyseven percent of other doctorategranting universities, and 35 percent of nondoctorategranting institutions undertook such projects. Regardless of the type of institution, universities and colleges were more likely to have begun repair/renovation projects than construction projects. Doctorategranting institutions were more likely to have begun both repair/renovation and construction projects than were nondoctorategranting institutions. Did Capital Project Activity Change over Time? Overall, the percentage of researchperforming universities and colleges engaged in either the construction of S&E research space or the repair/renovation of S&E research space fluctuated somewhat over time. The level of construction activity in doctorategranting universities increased between fiscal years 1986!1987 and fiscal years 1990!1991, but dropped considerably in fiscal years 1992!1993 (Table 39). In fiscal years 1990!1991, 57 percent of all doctorategranting universities began some type of construction project, but in fiscal years 1992!1993, 44 percent of these institutions began construction projects. This decrease can be accounted for by the rather sharp drop in the percentage of other doctorategranting institutions (those not in the top 100) that began construction projects. In fiscal years 1990!1991, 45 percent of these institutions began construction projects; in fiscal years 1992!1993, 28 percent did so. For nondoctorategranting institutions, the drop in construction project starts occurred between fiscal years 1988!1989 and fiscal years 1990!1991, a decline from 32 percent to 12 percent. Not only was there a decline in the percentage of institutions undertaking S&E research construction projects between fiscal years 1990!1991 and fiscal years 1992!1993, but no increase was planned for fiscal years 1994!1995. (See Table 39.) In fiscal years 1992!1993, 32 percent of researchperforming institutions constructed S&E research space; for fiscal years 1994!1995, 30 percent planned to undertake construction projects.1 As construction activity dropped at other doctorategranting institutions, repair/renovation activity also dropped sharply at the same institutions. In fiscal years 1988!1989 and fiscal years 1990!1991, 65 percent of other doctorategranting institutions were repairing/renovating some S&E research space (Table 310 ). In fiscal years 1992!1993, 48 percent of the other doctorategranting institutions were beginning to repair/renovate S&E research space. Planned repair/renovation projects to S&E research space for fiscal years 1994!1995 were also down from fiscal years 1992!1993. (See Table 310.) Fortysix percent of all institutions undertook repair/renovation projects to S&E research space in fiscal years 1992!1993; 41 percent planned such projects for fiscal years 1994!1995.1 In What Fields Were Capital Projects Undertaken? Both construction and repair/renovation projects were more likely to have occurred in certain S&E fields than others. This was the case for the most recent fiscal years as well as in prior years. In fiscal years 1992!1993, onethird of the researchperforming institutions with medical schools and with S&E space in the medical sciences began construction projects. In fiscal years 1992!1993, those institutions with research space in the agricultural sciences were also heavily engaged in constructing S&E facilities; 27 percent of these universities and colleges had such projects underway. Other S&E fields in which at least 10 percent of researchperforming institutions began construction projects in fiscal years 1992!1993 were engineering (17 percent); both the biological sciences outside of medical schools (10 percent) and those in medical schools (20 percent); and medical sciences outside of medical schools (11 percent) (Table 311). Repair/renovation projects to S&E research space were also most likely to have been started in the medical sciences within medical schools. Sixtyone percent of all researchperforming universities and colleges that had research space in this field either repaired or renovated some portion of this research space. Medical schools were also active in repairing research space in the biological sciences (39 percent). At least onefifth of researchperforming universities and colleges repaired research space in engineering (30 percent); the physical sciences (22 percent); and the biological sciences outside of medical schools (22 percent) (Table 312). Over time, for construction as well as repair/renovation activities, the level of activity by S&E field fluctuated, indicating that researchperforming universities and colleges may focus efforts and resources on specific S&E fields in certain years. As an example, 28 percent of all academic institutions started projects to construct research space in engineering in fiscal years 1986!1987. In fiscal years 1988!1989, 18 percent started construction projects in this field, and the percentages were similar in fiscal years 1990!1991. (See Table 311.) Similarly, the percentage of institutions that started repair/renovation to agricultural research space declined from 27 percent in fiscal years 1990!1991 to 18 percent in fiscal years 1992!1993. (See Table 312.) The fields in which institutions planned to construct S&E research space or to repair/renovate space in fiscal years 1994!1995 were similar to those in which projects were undertaken in fiscal years 1992!1993. Universities with medical schools still planned to be active in constructing space (29 percent of these institutions had plans for such projects in fiscal years 1994!1995) and in repairing/renovating space (45 percent planned to undertake such projects). What Did Institutions Plan to Spend on Animal Facilities? Researchperforming universities and colleges planned to spend over $294 million on construction and repair/renovation projects for laboratory animal facilities in fiscal years 1994!1995. Ninetyone percent of these planned expenditures, $266.5 million, were accounted for by the top 100 universities. Nondoctorategranting universities planned to spend $1.7 million, or less than 1 percent of the total (Table 313). Chapter 4 Funding of Research Facilities Projects Highlights * For public universities and colleges, state and local governments continued to be the largest single funding source of science and engineering (S&E) research facilities construction projects, contributing $930 million, or 46 percent of all funds for fiscal years 1992!1993. * State and local governments also contributed 46 percent of all dollars for repairing/renovating S&E research facilities in public universities and colleges. Their contributions totaled $237 million in fiscal years 1992!1993. * For private universities and colleges, total institutional contributions (institutional funds, tax exempt bonds, and other debt) provided over half (54 percent) of the funds for construction projects for S&E research space in fiscal years 1992!1993. * Private universities and colleges also relied heavily on institutional contributions to fund S&E research repair/renovation projects in fiscal years 1992!1993. Seventytwo percent of the total repair/renovation funding, $225 million, came from these institutional contributions. Context Although researchperforming universities and colleges were involved in considerable capital project activity in fiscal years 1992!1993, both the construction of facilities and the repair/renovation of S&E research space declined somewhat from the previous two fiscal years. The amounts and relative proportions of money received from different funding sources varied over time, possibly reflecting changes in both the economy and the types of projects undertaken. This chapter examines how higher education institutions financed S&E capital projects between 1986 and 1993, with particular attention to declines in specific funding sources. The Survey Question Institutional respondents were asked to report funding sources for projects to construct S&E research facilities and to repair/renovate S&E research space. Respondents reported only the projects that cost over $100,000. These projects were to have begun in fiscal years 1992!1993. Possible sources included the Federal government, state or local governments, private donations, institutional funds, taxexempt bonds, other debt financing, and other sources. (See Item 5 of the survey in Appendix C.) Data Considerations Institutions reported only on construction and repair/renovation projects that were for S&E research space and that exceeded $100,000. Within the seven funding categories provided on the survey, considerable diversity is possible. For example, Federal funding can include specific facilities support programs administered by the National Science Foundation (NSF) and the National Institutes of Health (NIH). Federal funding might also include nonpeerreviewed projects that are specified individually through Congressional legislation rather than specific agency programs. There may be some overlap in the categories as well. For example, indirect costs included as institutional funds can come from Federal, state, and local governments. No information was gathered in the survey that distinguished indirect cost recovery from other institutional funding, such as the use of operating or endowment funds. In this report, all dollar figures for years prior to 1993 were adjusted using the Bureau of Census' Composite FixedWeighted Price Index for Construction. This adjustment means that dollar figures presented in this report do not match the previous reports' figures, which were in current dollars. Findings How Did Institutions Fund Capital Projects? Type of Institution For doctorategranting institutions, the decline in combined costs for the construction of S&E research space and the repair/renovation of research space between fiscal years 1990!1991 and the following two fiscal years resulted from declines in nearly all funding categories (Table 41). Other debt and other funding sources increased, but the relative contribution of these two sources was small. For nondoctorategranting institutions, combined construction and repair/renovation expenses for projects involving S&E research space decreased steadily, in constant dollars, from fiscal years 1986!1987 to fiscal years 1992!1993. In fiscal years 1992!1993, nondoctorategranting institutions spent $124 million for capital projects. Furthermore, nondoctorategranting institutions spent about 3 percent of what doctorategranting universities did for these same types of projects. Although Federal support to doctorategranting universities for capital projects entailing S&E research space declined between fiscal years 1990!1991 and fiscal years 1992!1993, from $536 million to $499 million, the Federal government still provided these institutions with considerably more money than in either fiscal years 1986!1987 or fiscal years 1988!1989. (These dollar figures have been adjusted for inflation.) State and local government support to doctorategranting institutions for S&E research space capital projects dropped slightly between fiscal years 1990!1991 and fiscal years 1992!1993. For nondoctorategranting institutions, funds from state and local governments increased from $26 million in fiscal years 1990!1991 to $84 million in fiscal years 1992!1993. This amount was, nevertheless, considerably below fiscal years 1986!1987 funding level of $139 million. Institutional contributions to capital projects that involved S&E research space occurred through institutional funds, taxexempt bonds, and other debt. The institutional contribution of doctorategranting institutions to both construction and repair/renovation of S&E research space was considerably larger in both absolute dollars and relative contribution than that of nondoctorategranting institutions. Doctorategranting institutions contributed 41 percent of all construction and repair/renovation dollars; nondoctorategranting institutions contributed 10 percent.1 Control of Institution Because of the support that state governments provide public higher education institutions, the control of institutions is very relevant to discussions of who funds capital projects involving S&E research space. State governments subsidize over 1,600 universities and colleges in the United States, providing support for operating expenses as well as capital projects. Private institutions, although greater in number, enroll fewer students and cannot rely on state and local governments for capital funding as readily as public universities and colleges. (See Figure 41 on the following page.) During fiscal years 1992!1993, public universities and colleges spent a total of $2,537 million on the construction and repair/renovation of S&E research space (Table 42). Since fiscal years 1988!1989, the amount spent on capital projects declined slightly, from $2,618 million to $2,573 million in fiscal years 1990!1991 to $2,537 million in fiscal years 1992!1993. During fiscal years 1992!1993, private universities and colleges spent $1,110 million on construction and repair/renovation projects involving S&E research space. Over the four time periods represented by the surveys, the amount spent on these activities fluctuated slightly from year to year. For public universities and colleges, state and local governments provided the largest share of funding for S&E research capital projects, $1,167 million. Funds from this source increased between fiscal years 1990!1991 and fiscal years 1992!1993, from $1,087 million to $1,167 million. Funds from taxexempt bonds also increased slightly during this period, from $428 million to $446 million. For private universities and colleges in fiscal years 1992!1993, institutional contributions (a total of institutional funds, taxexempt bonds, and other debt) provided the largest share of funding for capital projects entailing S&E research space, $654 million. Relative to both public institutions and to other funding sources, funds from state and local governments in fiscal years 1992!1993 were small, $54 million. Although the contribution of state and local governments in fiscal years 1992!1993 dropped sharply from the two previous fiscal years, fiscal years 1990!1991 contribution of $164 million should be viewed as an anomaly. How Did Institutions Fund Construction Projects? Public Institutions As was the case in prior years, public universities and colleges relied heavily on state and local support for the construction of S&E research space in fiscal years 1992!1993. For this period, state and local governments provided a total of $930 million or 46 percent of all funds (Table 43). Federal funding of construction projects for S&E research space increased from a relative contribution of 3 percent in fiscal years 1986!1987 to 16 percent in fiscal years 1992!1993. This increase occurred at a time when private support for S&E research construction at public universities and colleges declined from 19 percent to 8 percent. For public universities and colleges, the institutional contribution (institutional funds, taxexempt bonds, and other debt) to construction projects for S&E research space represented 30 percent of all construction dollars in fiscal years 1992!1993. Private Institutions For fiscal years 1992!1993, taxexempt bonds provided private universities and colleges with $230 million for the construction of S&E research facilities or 29 percent of all construction funds (Table 44). Although this amount represented a decline from fiscal years 1990!1991 contribution of $343 million, taxexempt bonds provided private institutions with considerably more funds in fiscal years 1992!1993 than in fiscal years 1986!1987. In fiscal years 1986!1987, taxexempt bonds provided $144 million (Table 44). Total institutional contributions (institutional funds, taxexempt bonds, and other debt) provided an increasing percentage of the total funding spent on the construction of S&E research facilities. In fiscal years 1986!1987, institutional contributions provided less than 45 percent of construction funding. By fiscal years 1992!1993, they provided over half, 54 percent, of the funds for S&E research facilities projects. In fiscal years 1992!1993, the Federal government provided $133 million, or 17 percent of all construction dollars, for S&E research space to private universities and colleges, a proportion similar to that provided to public institutions. In contrast to public institutions, state and local governments provided $39 million, or 5 percent, of the total S&E research construction costs for private institutions for fiscal years 1992!1993. How Did Institutions Fund Repair/Renovation Projects? Public Institutions In fiscal years 1992!1993, state and local governments provided public universities and colleges with $237 million for the repair/renovation of S&E research facilities. (See Table 45 on the following page.) As a relative contribution, this amount comprised 46 percent of all repair/renovation funding. The constant dollar contribution from state and local governments decreased between fiscal years 1990!1991 and fiscal years 1992!1993, from $244 million to $237 million. The relative contribution from state and local governments also decreased from 52 percent of all funds for repair/renovation projects to 46 percent. In fiscal years 1992!1993, institutional funds provided $154 million or 30 percent of all repair/renovation dollars for S&E research space. When taxexempt bonds and other debts are considered, public institutions made an institutional contribution of over 40 percent of all funds for the repair/renovation of S&E research space. The Federal government provided 5 percent of all funding for the repair/renovation of S&E research facilities. Private Institutions In all years covered by the surveys, private universities and colleges funded a substantial portion of their S&E repair/renovation projects with relatively little Federal, state, or local government support (12 percent). In fiscal years 1992!1993, institutional contributions (institutional funds, taxexempt bonds, and other debt) provided $225 million, or 72 percent, of all repair/renovation costs. (See Table 46 on the following page.) In fiscal years 1992!1993, private donations totaled $48 million or 15 percent of the total amount spent to repair/renovate S&E research space. The relative contribution of the Federal government was similar for private (7 percent) as for public institutions (5 percent). However, private institutions received considerably less in constant dollars ($22 million) than public institutions ($34 million). In fiscal years 1992!1993, state and local governments provided 5 percent of all S&E research repair/renovation dollars to private universities and colleges. Chapter 5 Deferred Construction and Repair/Renovation Highlights * Of all researchperforming universities and colleges, 40 percent reported an approved institutional plan that included deferred or unfunded construction or repair/renovation projects for science and engineering (S&E) research space. (See Appendix A, Technical Notes.) * Based on these reports from institutions, the estimated cost for deferred construction projects for S&E research space was $4,047 million; the estimated cost for deferred repair/renovation projects for S&E research space was $1,697 million. These estimates directly reflected the needs of the 40 percent of universities and colleges that had identified these deferred needs in an approved institutional plan. * Five fields were mentioned by at least 10 percent of the researchperforming universities and colleges that reported deferred needs in S&E research space: agricultural sciences, engineering, the physical sciences, medical sciences in medical schools, and biological sciences outside of medical schools. Context Previous chapters have provided information about the amount of S&E research space available in researchperforming universities and colleges (see Chapter 1) and its adequacy and condition (see Chapter 2). Of central concern to policymakers, however, is the level of funding researchperforming institutions need for both the construction of S&E research facilities and the repair/renovation of S&E research space. Institutions reported that 13 percent of all S&E research space required major repair/renovation and another 4 percent required replacement in 1994. (See Table 22, page 25.) This does not mean, however, that institutions have the resources to repair/renovate or replace this space. This chapter provides insights into the amount of funding that is required for current S&E research space needs that cannot be funded with available resources. The Survey Question To address the issue of need, the 1994 survey introduced a new item. In order to obtain an estimate of needed funding for capital projects involving S&E research space, institutions were asked to report whether an approved institutional plan existed that included "any deferred space that requires repair/renovation or new construction." Four criteria were used to define deferred space: * The space must be necessary to meet the critical needs of current faculty or programs; * Construction must not be scheduled to begin during fiscal years 1994!1995; * Construction must not currently have funding; and * The space must not be for developing new programs or expanding the number of faculty. Using these standards, respondents were asked to estimate for each S&E field the construction costs and the repair/renovation costs of such projects. (See Item 9 of the survey in Appendix C.) Data Considerations Beyond issues of sheer survival, the concept of "need" becomes subject to interpretation from different perspectives. The specific criteria used in the 1994 facilities survey were developed to place a defined boundary around the concept of "need." The criteria cannot preclude the possibility that, by some different definition, additional facilities needs might be identified. For example, estimates of deferred construction and repair/renovation needs in the 1988 and 1990 surveys, derived from several survey items, were considerably higher than the 1994 estimate of $5,744 million. On the other hand, use of the present multiple criteria cannot guarantee that all facilities included in approved plans reflect the same level of construction urgency. The chief benefit of using these multiple, explicit criteria is that they provide a specific description of the facilities needs to be included and excluded from consideration, thus allowing meaningful interpretation of individual data points and the development of trends across future surveys. For this survey, 40 percent of all institutions indicated that they had an approved institutional plan that included deferred space; 60 percent had no such approved plans. Thus, the deferred space needs reported here directly reflect the needs of these 40 percent of institutions who met the definitional criteria. No attempt is made to estimate "facilities needs" under some other, less formal definition. (See Appendix A, Technical Notes, for further discussion of estimates.) Findings What Was the Extent of Needed, but Deferred, Capital Projects? A total of 40 percent of all researchperforming universities and colleges had an approved institutional plan that included either construction or repair/renovation projects that were deferred and unfunded. (See Table 51 on the following page.) The top 100 institutions were most likely to have had such a plan (60 percent) and the nondoctorategranting institutions were least likely (26 percent). Universities and colleges were, in general, more likely to have had plans for deferred repair/renovation projects than for the deferred construction of facilities. The top 100 institutions, however, were as likely to have had plans for construction projects as for repair/renovation. Fiftytwo percent of the top 100 institutions that responded to this item indicated deferred construction projects in their plans; 48 percent of that same group had deferred projects for repair/renovation of S&E facilities. What Was the Estimated Cost of Deferred Capital Projects? Deferred construction costs were estimated at $4,047 million across all researchperforming institutions that responded to this item, while deferred repair/renovation costs totaled $1,697 million. (See Table 52 on the following page.) The top 100 institutions in research expenditures that responded to this item accounted for over twothirds of the estimated construction costs, $2,823 million, and 61 percent of repair/renovation costs, $1,052 million. In nondoctorategranting institutions with an approved institutional plan, $199 million were estimated for deferred construction and $239 million for deferred repair/renovation (Figure 51). All of these estimates of deferred need must be interpreted cautiously. These figures are based on universities and colleges that reported approved institutional plans that included deferred space for either construction or repair/renovation that was unfunded. As a result, these dollar figures do not represent total need for S&E facility improvements; at best, they provide a lower limit estimate of unmet need. How Were Deferred Capital Projects Distributed across S&E Fields? Deferred and unfunded need existed in all S&E fields. For researchperforming universities and colleges with approved institutional plans for S&E research space, unfunded need for construction projects in the agricultural sciences was indicated more frequently than in any other field. Twentyone percent, or slightly over onefifth, of all responding institutions with research space in the agricultural sciences reported unfunded need for new facilities in this field. Four other fields were mentioned by at least 10 percent of the responding group. The fields were engineering, named by 18 percent of these institutions; the physical sciences and the medical sciences in medical schools, each named by 16 percent; and the biological sciences outside of medical schools, named by 14 percent of these institutions with space in this field. (See Table 53 on the following page.) Institutions most frequently indicated the physical sciences as the S&E field with unfunded need for repair/renovation of their research facilities. Onefourth of responding universities and colleges indicated that their institutional plans included unfunded repair/renovation projects in this field. Over 20 percent of the responding researchperforming universities and colleges reported unfunded need for repair/renovation in the following three fields: engineering (22 percent); the biological sciences outside of medical schools (22 percent); and agricultural sciences (21 percent). Chapter6 Historically Black Colleges and Universities Highlights * The panel of 29 Historically Black Colleges and Universities (HBCUs) that has been sampled since 1988 contained 6.1 million net assignable square feet (NASF) of science and engineering (S&E) space; 29 percent of this space was designated for research. * Expenditures for S&E research space construction in the panel of 29 HBCUs sampled since 1988 declined from $83.2 million (constant dollars) in fiscal years 1986!1987 to $8.6 million in fiscal years 1992!1993. * The Federal government has consistently provided HBCUs with the majority of their funds for both construction and repair/renovation projects. Context HBCUs have played an important role in the education of black students at all higher education levels for over 100 years. These universities and colleges consist of both public and private institutions as well as twoyear, fouryear, and professional schools. In 1991, approximately 269,000 students attended the 105 institutions of higher education considered HBCUs by the U.S. Department of Education. Although HBCUs have considerably less S&E research space than other researchperforming institutions, they are an important source of science and engineering degrees for black students who attend college. A recent study of science and engineering doctorates revealed that almost 30 percent of black science and engineering doctorate degree recipients between 1985 and 1990 received their bachelors' degrees from HBCUs. (See Appendix D, Undergraduate Origins of Recent Science and Engineering Doctorate Recipients, 1992.) This chapter profiles the state of S&E research facilities at the researchperforming HBCUs. It examines all of the topics covered in previous chapters, including the amount of S&E space, its adequacy and condition, construction and repair/renovation activities, funding sources for these projects, and the need for additional or renovated space. The Survey Question The profile of HBCUs in this chapter is based on all of the survey questions considered in previous chapters. Data Considerations The National Advisory Committee on Black Higher Education and Black Colleges and Universities identifies 107 HBCUs.1 Of this group, 29 reported separately budgeted research expenditures in 1988, the year in which the first fullscale facilities survey was conducted by the National Science Foundation (NSF). All of these institutions were included in the 1988 and subsequent samples. In 1992, NSF identified an additional 41 HBCUs that had separately budgeted research and development (R&D) expenditures. In both 1992 and 1994, the survey sample included the original panel of 29 institutions and the additional 41 for a total of 70 researchperforming HBCUs. As a result, two sets of estimates for HBCUs can be presented. This chapter presents an overall space estimate for all 70 researchperforming HBCUs. All remaining analyses focus on the 29 HBCUs included in the sample since 1988. These institutions are referred to as the panel of 29 HBCUs. (See Volume 2, Statistical Tables, for more detailed information regarding all 70 HBCUs.) Findings How Much Research Space did HBCUs Have? In 1994, the 70 researchperforming HBCUs contained 7.9 million NASF of S&E space; 28 percent of this S&E space, 2.2 million NASF, was designated as research space (Table 61). In 1994, the panel of 29 HBCUs contained 6.1 million NASF of S&E space; 29 percent of this space was designated for research. Eighty percent of all S&E research space in HBCUs was concentrated in the panel of 29 HBCUs. Thus, the panel of 29 universities and colleges that has participated since the first survey represented the bulk of all research space in HBCUs. To facilitate analyses of trends in HBCUs' research space and funding, the remainder of this chapter focuses on this panel of 29 institutions. The total amount of S&E space in the panel of 29 HBCUs decreased somewhat between 1992 and 1994, from 6.6 million to 6.1 million NASF. S&E research space in the panel of 29 HBCUs was basically unchanged at 1.8 million NASF in 1992 and 1994; thus, the decline occurred primarily in nonresearch space. S&E research space as a percentage of total S&E space has steadily increased among the panel of 29 HBCUs since 1988, when the amount was 18 percent. In 1994, two fields occupied half of all the research space in the panel of 29 HBCUs: the agricultural sciences and the biological sciences. (These two fields accounted for 37 percent of the research space in the total 565 researchperforming institutions in this study.) (See Table 16, page 19.) The amount of engineering research space was also relatively high in the 29 HBCUs, occupying 18 percent of all research space. The physical sciences and the medical sciences each comprised 12 percent of the total (Table 62). What Was the Condition of Research Space in HBCUs? In 1994, 31 percent of the S&E research space in the panel of 29 HBCUs was rated as "suitable for the most scientifically sophisticated research;" this represented a slight drop from the 34 percent of space rated this way in 1992. Nine percent of all space needed major repair/renovation or replacement (Table 63). What Was the Extent of Capital Projects at HBCUs? Since the inception of the survey, the panel of 29 HBCUs experienced large decreases in expenditures for research space construction. In constant dollars, there was a decline of almost $75 million in construction expenditures between fiscal years 1986!1987 and fiscal years 1992!1993. In fiscal years 1986!1987, the 29 HBCUs spent $83.2 million (constant dollars); in fiscal years 1992!1993, these institutions spent $8.6 million on construction projects. (See Table 64 on the following page.) Repair/renovation expenditures also dropped, but not quite as dramatically as construction expenditures. In fiscal years 1986!1987, the panel of 29 HBCUs spent $16.3 million in constant dollars; in fiscal years 1992!1993, repair/renovation to S&E research space at these institutions totaled $8.7 million. Repair/renovation projects that were planned for fiscal years 1994!1995 totaled $13.9 million (Table 65). For both types of activities, construction and repair/renovation, the panel of 29 HBCUs indicated that they would spend more during fiscal years 1994!1995 than they did in fiscal years 1990!1991 and fiscal years 1992!1993. What Was the Source of Funding for Capital Projects? The Federal government provided the majority of funds for both construction and repair/renovation projects to the panel of 29 HBCUs. For fiscal years 1992!1993, the Federal government contributed $6.6 million, or 77 percent of all S&E research construction funds (Table 66), and $5 million, or 55 percent of all S&E research repair/renovation funds. (See Table 67 on the following page.) Relatively speaking, the Federal government provided a much larger share of total capital project funds to HBCUs than to either public or private higher education institutions in general. The Federal government provided less than 20 percent of all construction funds and less than 10 percent of all repair/renovation funds to researchperforming institutions overall. (See Tables 43 through 46.) In fiscal years 1992!1993, state and local governments were the only other source of funding for S&E research construction projects for the panel of 29 HBCUs and were the second highest contributors to repair/renovation funds. What Was the Extent of Unfunded Capital Projects? Thirtysix percent, or slightly over a third of the panel of 29 HBCUs, reported deferred and unfunded S&E research construction projects that were included in an approved institutional plan. These projects totaled $93.8 million. It should be noted that three HBCUs had particularly large S&E research construction projects included in their institutional plans. Eighteen percent of the panel of 29 HBCUs indicated the presence of approved plans with unfunded and deferred repair/renovation projects for S&E research space. These projects totaled $9.2 million (Table 68). The estimated cost of unfunded and deferred capital projects, both construction and repair/renovation, represents a conservative estimate of the total need for S&E research capital projects. The restrictive nature of the survey question limited the number of institutions that could respond.2 Chapter 7 Predominately Undergraduate Institutions Highlights * Predominantly undergraduate institutions, defined in this chapter as comprehensive universities and liberal arts colleges with research and development (R&D) expenditures of $50,000 or more, had approximately 25 million net assignable square feet (NASF) of space in the science and engineering (S&E) disciplines. Seventeen percent of this space, 4.1 million NASF, was devoted to research. * Predominantly undergraduate institutions spent $65 million to construct S&E research space in fiscal years 1992!1993, and over $27 million to renovate their S&E research space during this same period. The overwhelming majority of construction, 91 percent, took place at comprehensive universities. * Comprehensive universities financed capital projects primarily through state funds; liberal arts colleges depended mostly on private and Federal funds. Context Previous chapters have examined differences in S&E research facilities across two types of institutions: doctorategranting and nondoctorategranting. In this chapter, a subset of nondoctorategranting institutions is examined more closely!predominantly undergraduate institutions. These predominantly undergraduate institutions exclude the HBCUs which were dealt with in the previous chapter and consist of the remaining nondoctorategranting institutions, that is the comprehensive universities and liberal arts colleges. Predominantly undergraduate institutions have considerably less S&E research space than doctorategranting universities. However, their contributions to the scientific enterprise are typically noted through their role in training future scientists and engineers. A National Science Foundation (NSF) study, Undergraduate Origins of Recent Science and Engineering Doctorate Recipients, reports that 34 percent of the individuals who were awarded science and engineering doctorates between 1985 and 1990 received their undergraduate degrees from either comprehensive universities (20 percent) or liberal arts colleges (14 percent). (See Appendix D, References.) In April 1994, the Committee on Science, Space, and Technology of the House of Representatives expressed concern "that NSF's biennial survey of academic research facilities needs . . . has not focused adequately on the needs of undergraduate institutions."1 Although the 1994 survey and sample could not be adjusted to address this concern, its results can provide insight into several issues regarding the S&E research facilities of a select group of undergraduate institutions. These findings are presented in this chapter. The Survey Question The profile of predominantly undergraduate institutions presented in this chapter is based on all of the survey questions considered in previous chapters. Data Considerations Predominantly undergraduate institutions contribute to research primarily through educating students and training them to become researchers. Although considerable research activity does occur at these institutions, direct research is not their primary contribution to the scientific enterprise. The current NSF facilities survey, designed to collect information on the size, condition, and needs of the nation's researchperforming universities and colleges, collects data from a sample of higher education institutions that report R&D expenditures of at least $50,000 in S&E fields.2 Many universities and colleges that focus on undergraduate education do not report such expenditures and therefore are not included in this survey; yet, these institutions do teach large numbers of students and award degrees in S&E fields to individuals who conduct S&E research. Results from analyses reported in this chapter, however, cannot be generalized to undergraduate institutions that did not report R&D expenditures of at least $50,000. The Carnegie Classification of Institutions of Higher Education is used to distinguish between two different groups of predominantly undergraduate institutions: comprehensive universities, colleges that offer a liberal arts program along with other programs such as engineering, business administration, or nursing; and liberal arts colleges, institutions that primarily award bachelor's degrees and that grant more than half their degrees in the liberal arts.3 The NSF facilities sample includes 54 comprehensive universities that represent 136 institutions, and 26 liberal arts colleges that represent 52 such institutions. The 5 tables presented in this chapter also include results from the 54 nondoctorategranting HBCUs,4 so that the resulting totals of all nondoctorates match the data presented in all previous totals for nondoctorates in Chapters 1 through 5. Discussion, however, is limited to the predominantly undergraduate institutions. (See Chapter 6 for more information on and discussion of HBCUs.) Findings How Much Research Space Did Predominantly Undergraduate Institutions Have? The amount of S&E research space at predominantly undergraduate institutions (that is, comprehensive universities and liberal arts colleges) was a relatively small portion of the total reported by all institutions included in the study. Comprehensive universities had 7 percent of the total S&E research space, and liberal arts colleges had a modest 1.8 percent. The balance of S&E research space (91.2 percent) was in the other types of institutions (Figure 71). All nondoctorate institutions held a total of 29.5 million NASF of S&E space. Predominantly undergraduate institutions had about 25 million NASF of that space (Table 71). Of the predominantly undergraduate institutions, comprehensive universities contained over 19 million NASF of S&E space, or almost 80 percent of the total S&E space in predominantly undergraduate institutions. Predominantly undergraduate institutions dedicated 4.1 million NASF to S&E research, about 17 percent of their total S&E space. Of that 4.1 million NASF, comprehensive universities dedicated 3.1 million NASF to S&E research, and liberal arts colleges dedicated 1 million NASF to S&E research. What Was the Condition of the Research Space at Predominantly Undergraduate Institutions? Almost half of the S&E research space in both comprehensive universities (44 percent) and liberal arts colleges (46 percent) was rated as "effective for most uses," though not suitable for the most sophisticated research in the field (See Table 72 on the following page.) Combined with the amount of space rated as suitable for the most sophisticated research, the predominantly undergraduate institutions had a total of 54 percent for comprehensive universities and 63 percent for liberal arts colleges of their space in the top two categories. These amounts are comparable to the doctorategranting institutions which rated 59 percent of their space in these two categories. (See Table 22, page 25.) A somewhat higher percentage of the S&E research space was rated as suitable for sophisticated research in liberal arts colleges (17 percent), than in comprehensive universities (10 percent). In both types of predominantly undergraduate institutions, 3 percent or less of the S&E research space was rated in need of replacement. Overall, in all researchperforming institutions, 4 percent of the S&E research space was evaluated as needing replacement. (See Table 22, page 25.) In predominantly undergraduate institutions, 12 percent of all S&E research space was rated as needing major repair/renovation. How Extensive Was Construction and Repair/Renovation Activity at Predominantly Undergraduate Institutions? All nondoctorate institutions spent $92.3 million to construct S&E research space and $34.5 million to repair/renovate S&E research space. Combined, the predominantly undergraduate institutions spent $65.2 million to construct S&E research space in fiscal years 1992!1993 and $27.2 million to repair/renovate their S&E research space (Table 73). Comprehensive universities spent over ten times as much on construction projects and almost one and onehalf times as much on repair/renovation projects as did liberal arts colleges. It should be noted, however, that comprehensive universities are generally larger than liberal arts colleges and that more comprehensive universities are actively engaged in research than liberal arts colleges. In fiscal years 1992!1993, the amount spent for S&E research construction projects at predominantly undergraduate institutions represented about 2 percent of all money spent for S&E research construction at all researchperforming universities and colleges. (See Table 31, page 34.) Repair/renovation to S&E research facilities at predominantly undergraduate institutions represented approximately 3 percent of all repair/renovation dollars in fiscal years 1992!1993. (See Table 33, page 35.) How Did Predominantly Undergraduate Institutions Fund Capital Projects? Comprehensive universities relied upon vastly different sources than liberal arts colleges to fund capital projects. During fiscal years 1992!1993, 80 percent of all capital projects at comprehensive institutions were funded through state revenues, while liberal arts colleges received no funding from states (Table 74). The 80 percent represented $60 million. This reflects the fact that comprehensive universities are more likely to be public institutions than are liberal arts colleges. Liberal arts colleges relied about equally on Federal and private funds, each accounting for slightly over onequarter of the capital funds or $4.1 million each. Institutional contribution (institutional funds, taxexempt bonds, and other debt) accounted for 47 percent or 6.9 million of the capital funds at liberal arts colleges. (See Table 75 on the following page.) What Was the Extent of Unfunded Capital Projects? Of all researchperforming institutions in this study, 40 percent had an approved institutional plan that included deferred and unfunded construction or repair/renovation projects for S&E research space. Based on their plans, about 10 percent of the comprehensive institutions and 23 percent of the liberal arts colleges indicated unfunded and deferred construction projects for S&E research space. Twenty percent of the comprehensive universities and 29 percent of the liberal arts colleges reported unfunded and deferred repair/renovation. The deferred needs of the institutions who did not respond to this item remain unknown. Although the liberal arts colleges that were engaged in funded S&E research had considerably less S&E research space than did comprehensive institutions, deferred construction costs at liberal arts colleges were notably larger. In the liberal arts colleges, deferred costs for construction were $97.1 million; in comprehensive universities, these costs were $26.4 million. However, the comprehensive universities reported much larger unfunded needs in deferred repair/renovation than did the liberal arts colleges, $178.3 million and $54.8 million respectively. Combined, the deferred capital projects, both construction and repair/renovation, at predominantly undergraduate institutions totaled $356.6 million (Table 76). Appendix A Technical Notes This appendix discusses the study methodology as well as various other technical aspects that the reader should consider when interpreting the data presented in this report. In addition to the current 1994 survey, the discussion includes the original 1988 survey, the 1990 survey, and the 1992 survey. The following topics are covered: * Universe and sample * The surveys * Data collection and response rates * Item nonresponse * Weighting * Reliability of survey estimates * Data considerations, definitions, and limitations Universe and Sample 1988 Survey The 1988 survey was designed to provide estimates for all researchperforming academic institutions, as defined in the National Science Foundation's (NSF) Fiscal Year (FY) 1983 Survey of Scientific and Engineering Expenditures at Universities and Colleges. The universe datafile for the 1983 expenditures survey included all universities and colleges that offered a master's or doctorate degree in science and engineering (S&E), all others that reported separately budgeted S&E research and development (R&D) expenditures of $50,000 or more, and all Historically Black Colleges and Universities (HBCUs) that reported any R&D expenditures. This datafile represented the most recent available universe survey of R&D expenditures at academic institutions. The datafile contained a total of 566 institutions. All HBCUs in the frame were included in the sample with certainty (N = 30), and a stratified probability sample of 223 institutions was selected from among the remaining institutions in the frame. These institutions were first stratified by control (public versus private) and highest degree awarded in S&E (doctorategranting versus nondoctorategranting). A minimum sample size of 25 was set for each of the four resulting strata, and the remaining sample size was allocated to strata in proportion to the "size" of each stratum. Stratum size was defined as the square root of the aggregate R&D expenditures in S&E of the institutions in the stratum. Academically administered Federally Funded Research and Development Centers were excluded from this survey. Within strata, institutions were sampled with probability proportionate to size. Again, size was defined as the square root of the institution's fiscal year 1983 R&D expenditures. Following the selection of an initial sample of 253 institutions, NSF determined that several of the sampled institutions were out of the scope of the survey. Outofscope institutions included those in outlying territories, military academies, and three highly specialized institutions considered inappropriate, given the nature of their programs. Elimination of these outofscope cases reduced the final sample to 247 institutions, of which 29 were HBCUs and 99 had (or were) medical schools. Institutions in the sample accounted for more than 75 percent of all academic R&D expenditures in fiscal year 1983 and encompassed at least 70 percent of the spending in each major S&E discipline. The sample represented a weighted national total of 525 institutions. The composition of this survey universe, by type of institution, is shown in Table A1. 1990 Survey The institution sample for the 1990 survey was the same as for the 1988 survey, except for these two changes: * The sample was updated to reflect recent R&D patterns as shown in NSF's fiscal year 1988 R&D expenditures survey, which collected expenditures data for all institutions in the survey frame for the first time since 1983. Schoolbyschool comparisons of these two databases resulted in the identification of 12 institutions whose 1988 R&D expenditures would have given them substantially higher probabilities of selection than they had using 1983 expenditures. These 12 institutions were made certainty selections for the 1990 survey. Five were already in the sample, having been noncertainty selections in the 1988 study; the other 7 were added to the sample for the 1990 survey. * One institution from the 1988 sample became out of scope when it distributed its assets among other institutions in the same state system. Therefore, this institution was eliminated from the sample. The sample changes noted above produced a net increase of 6 institutions, increasing the sample size to 253 in 1990. The universe represented by the sample, however, did not change. The sample design for the 1990 survey is summarized in Table A2. 1992 Survey The institution universe and sample for the 1992 survey were the same as for the 1990 survey, except for three changes: * Shortly after the sample for the 1990 facilities survey was selected, NSF conducted a universe survey of all HBCUs and identified an expanded group of 70 that reported separately budgeted R&D expenditures in S&E disciplines. A sample of 46 of these 70 institutions was selected for the 1992 facilities survey, with probability proportionate to size. Size was measured as the square root of the institution's reported 1989 R&D expenditures (a minimum size measure of $10,000 was used to afford the smallest institutions some possibility of selection). * The sample was expanded to include all institutions in the top 100 in 1988 R&D expenditures. Only two institutions from this analytically important category were not already in the sample, and they were made certainty selections in 1992. * To improve the precision of estimates for nondoctorategranting institutions, an expanded sample of 91 institutions in this category was selected (excluding HBCUs, which were sampled separately). The sample included all (10) public institutions with1988 R&D expenditures of $2 million or more, and all (11) private institutions with 1988 expenditures of $1 million or more. Institutions with R&D expenditures below these cutoffs were sampled with equal selection probabilities. Of the 91 sampled nondoctorategranting institutions, 9 were later determined to be out of scope, since they reported in the 1992 facilities survey that they had no S&E research space and also reported in the1988 R&D expenditures survey (which provided the basis for the sampling frame) that they had less than $50,000 in separately budgeted R&D expenditures. The exclusion of these outofscope institutions reduced the sample of nondoctorategranting institutions to 82. The sample design for the 1992 survey is summarized in Table A2. 1994 Survey The institution universe and sample for the 1994 survey closely matched the 1992 survey, with the following exceptions: * The 1991 R&D expenditures survey information was used to generate the top 100 stratum. Three institutions were added to the top 100 list, and three institutions were moved out. The expenditures data also were used to calculate the measure of size for the doctorategranting institutions. The 1988 expenditures survey data were used to calculate size measures for the nondoctorategranting institutions, since subsequent surveys did not yield complete information for the nondoctorategranting institutions. * Institutions expending less than $50,000 in R&D in S&E fields were removed from the frame prior to sampling. In 1992, they were selected with probability proportionate to size and then excluded after contact. * FICE codes were updated for 50 institutions.1 * Six institutions were misclassified with the 1992 sampling list as nondoctorategranting, when in fact they did award S&E doctorates. These misclassifications were corrected. * Random (rather than systematic) draws from the strata were employed. * The HBCUs selected with certainty were redefined to include 28 from the 1990 list,2 plus all of the new institutions selected with certainty in 1992. This meant that a total of 33 HBCUs was selected with certainty and 12 others were selected with probability proportionate to size. Of the 314 sampled institutions, 5 nondoctorategranting institutions were later determined to be out of scope, since they reported no S&E research space. The exclusion of these outofscope institutions reduced the sample to 309. The sample design for the 1994 survey is summarized in Table A2. (See Appendix B for a list of sampled institutions.) The Survey Questionnaire The 1994 survey questionnaire, which is reproduced in Appendix C, updated information collected during earlier (1988, 1990, and 1992) surveys regarding several topics: * The total net assignable square feet (NASF) of space in S&E fields, and the NASF used for organized research; * The total amount of space in all nonscience fields, and an overall space total across all academic fields; * The amount of research space that is leased by the institution; * The condition of research facilities in each S&E field; * The adequacy of the current amount of research space, by S&E field; * The project costs, NASF, and sources of funds for major construction and repair/renovation (over $100,000) activities initiated in fiscal years 1992!1993, and planned for fiscal years 1994!1995; * Expenditures for research facility repair/renovation projects in the $5,000 to $99,999 range; * Planned expenditures in fiscal years 1994!1995 for construction and repair/renovation of research laboratory animal facilities; * The status of the institutions relative to the cap on taxexempt bonds (this item is applicable to private universities and colleges only). In addition to collecting updated information on the above topics, the 1994 questionnaire also requested information on two topics that had not been addressed in previous surveys. Specifically, in response to questions about unmet construction and repair/renovation needs for S&E research space, the 1994 questionnaire added items asking about the following issues: * The existence of an approved institutional plan that included deferred space requiring new construction or repair/renovation; * The number of years included in the plan; * The estimated costs, by S&E discipline, for needed new construction and repair/renovation that the institution was not scheduled to begin during fiscal years 1994!1995. In addition, to provide preliminary information on the effects of the requirements of the 1990 Americans with Disabilities Act (ADA), institutions were asked to estimate what portion of their repair/renovation costs from fiscal years 1992!1993 was spent to bring S&E research space into compliance with the ADA. Results from this item are not presented in the 1994 report. Data Collection and Response Rates In September 1993, a letter from Frederick M. Bernthal, then Acting Director of NSF, was sent to the president or chancellor of each sampled institution, asking that the institution participate in the study and that a coordinator be named for the survey. A letter of endorsement of the project signed by the heads of eight higher education associations was also enclosed. After the 2week deadline for returning the coordinator identification card, telephone followup was conducted with all sampled institutions that had not yet identified a survey coordinator. Survey materials were mailed to the coordinators in midOctober by certified mail, and the return receipt cards served as a control log. For cards that were not returned, receipt of the survey materials was confirmed by telephone in November. The questionnaire and cover letter requested return of the completed survey by December 1, 1993. Nonresponse followup began in midDecember and continued through March 1994. After the questionnaires were edited, a series of logic and arithmetic checks was run and additional followup was conducted to resolve data inconsistencies within the questionnaire or disparities between the 1992 and 1994 survey responses. After data collection, site visits were conducted, during which NSF and project staff members met with survey respondents to discuss the questionnaire, interpretation and reliability of the data provided, and the survey procedures. The purposes of these visits were to (1) obtain information about the data provided to assist in the analysis of the findings and (2) obtain information that could be used in planning for the 1996 survey. The overall response rate for the survey was 93 percent. As Table A3 indicates, response rates were high for all institution categories. Item Nonresponse After machine editing of questionnaire responses for completeness, internal consistency, and consistency with data from previous questionnaires, extensive telephone data retrieval was conducted to minimize the amount of missing or otherwise problematic responses to individual questionnaire items. One exception was the item (1a) on total academic space in all disciplines outside S&E fields. This item was difficult for some institutions to answer; and although data retrieval was attempted, the item had an unusually high nonresponse rate (17 percent). As a result of these followup activities, most of the individual items had very low item nonresponse rates. The item with the highest nonresponse rate (other than item 1a) was the new item on costs to comply with the 1990 Americans with Disabilities Act (Item 4b). This item had 24 missing values (8 percent). Next highest was the item in 4a asking about the prorated total research space involved in all 1992 and 1993 repair/renovation projects costing $100,000 or more. It had 9 missing values (3 percent). Item 3, the current condition of research space by field, also had 9 missing values (3 percent) for one field: medical sciences, outside of medical school. All other data items had fewer than 9 missing values; that is, all had item response rates over 97 percent. Missing values were imputed for questionnaire items that were involved in the data analysis. Wherever possible, missing values for items 1, 2, and 3 (amount, condition, and adequacy of existing space) were imputed on the basis of information in the institution's 1992 questionnaire. In questions 4 and 8 (on recent and planned capital projects), most missing values involved either missing costs or missing NASF, but not both. In these cases, the missing data element was imputed from the reported element, using 1992 data on average cost per NASF to estimate one from the other. Missing values that could not be imputed using the above methods (for example, a missing value on the amount of research space at an institution that had not provided this information in the 1992 survey) were imputed using a "hot deck" approach. This approach involved imputing the missing value from a "donor" institution that did provide the needed information and that was as closely matched as possible to the institution with the missing information in terms of control, type (doctorategranting or not), and 1988 research expenditures. Weighting After data collection, sampling weights were created for use in preparing national estimates from the data. First, within each weight class, a base weight was created for each institution in the sample. The base weight is the inverse of the probability of selecting the institution for the sample. Second, because some institutions in the sample did not respond to the survey, the base weights were adjusted in each weight class to account for this unit nonresponse. Finally, the weights were adjusted again to bring the number of estimated institutions in accordance with the known number of institutions in various categories. For this final "poststratification" adjustment, the institutions were classified by type (top 100 in research expenditures, other doctorategranting, nondoctorategranting), control, and HBCU status. The poststratified weights were used to produce the estimates shown in this report. The weighting procedures used were very similar to those used in the 1988, 1990, and 1992 studies. Reliability of Survey Estimates The findings presented in this report are based on a sample and are therefore subject to sampling variability. Sampling variability arises because not all institutions are included in the study. If a different sample of institutions had been selected, then the results might have been somewhat different. The standard error of an estimate can be used to measure the extent of sampling variability for that particular estimate. One of the ways that the standard error can be used is in the construction of confidence intervals. If all possible samples were selected and surveyed under similar conditions, then the intervals of two standard errors below the estimates to two standard errors above the estimates would include the average result of these samples in about 95 percent of the cases. Since only one sample is actually selected and surveyed, the standard error must be estimated from the sample itself. The interval constructed using the estimated standard error from the sample is called a 95 percent confidence interval. Estimated standard errors for selected statistics are shown in Table A4 on the following two pages. The standard errors for this study were estimated using a replication method called the jackknife repeated replication method. Using this method, the sample is divided into 15 replicates, and estimates are produced for each replicate. The variability among these replicate estimates is then used to estimate the standard error. Because the 1994 sample was independently drawn, the standard error of the difference between 1994 and 1992 estimates was computed under the assumption of independence. Data Considerations, Definitions, and Limitations In addition to sampling errors, survey estimates can be adversely affected by nonsampling errors. Errors of this type include those resulting from reporting and processing of data. In this survey, extensive followup with respondents was used to ensure that the data were as accurate as possible. This followup included crossyear review that verified inconsistencies between the current and previous questionnaires. Research Square Footage In the 1994 survey, research was defined more broadly than in previous years. However, this change in definition had little effect on how institutions actually reported S&E research space. Like the definition used in previous years, the 1994 definition included all R&D activities that are separately budgeted and accounted for. Unlike the previous definition, the 1994 definition also included departmental research that was not separately budgeted. Conversations with respondents from earlier surveys revealed that some departmental research had been included; thus, the current definition of research reflects what many institutions had been reporting all along. Previous cycles of this survey used the definition of organized research that is specified in OMB Circular A21 (the form used for calculation of indirect costs). That definition is as follows: "Organized research means all research and development activities of an institution that are separately budgeted and accounted for. It includes: (1) Sponsored research means all research and development activities that are sponsored by Federal and nonFederal agencies and organizations . . . (2) University research means all research and development activities that are separately budgeted by the institution under an internal application of institutional funds." Institutions' facility recordkeeping systems vary considerably. In general, most of the larger institutions have central computerized facility inventory systems, often based on space surveys conducted specifically for OMB Circular A21. Many institutions with smaller research programs are not required to calculate square footage for OMB Circular A21, and do not maintain databases that can provide such information. These institutions had to calculate or estimate square footage information specifically for this study. Capital Projects Involving Research Facilities Relatively few institutions maintain information on construction and repair/renovation projects specific to research facilities. Many capital projects involve both research and nonresearch space. When a project was not exclusively for research, institutions had to estimate the proportion of the project that was related to research facilities. For this purpose, the following guideline was included in the questionnaire instructions: For multipurpose facilities, prorate the costs to reflect the proportion of R&D space involved in the projects (e.g., if 20 percent of the space involved is used for organized research, report 20 percent of the total project completion costs). Some projects, such as construction or wholebuilding renovation may take more than one year to complete, and other projects may overlap fiscal years. Projects were allocated to the fiscal year in which actual construction activity began or will begin. Because institutions use different dollar values to identify "major projects," this survey established a guideline to ensure consistency of reporting. As in previous cycles of the survey, projects with costs of $100,000 or more associated with research facilities were included. In 1992 and 1994, the surveys also had a separate question about costs of repair/renovation projects in the $5,000 to $99,999 range. Dollar Amounts: Current Versus Constant Dollars In this report, capital project dollar amounts are presented in both constant and current dollars but discussed only in terms of 1993 constant dollars. Constant dollars are "inflation adjusted" dollars that adjust for variations in the purchasing power of the dollar over time. Dollar amounts were adjusted using the Bureau of the Census' Composite FixedWeighted Price Index for Construction. Unlike a more general index, this construction index closely tracks inflation within the construction industry. This index reflects only changes in prices and is unaffected by changes in the mix of construction projects during any given year. Previous reports used current, not constant dollars to present trends in capital project expenditures. Comparisons in current dollars tend to overstate increases in spending over time because more current dollars are needed to buy the same products each year. Comparisons in constant dollars provide a more accurate picture of expenditure trends. The specific adjustments used for each of the fiscal years is shown in Table A5. Condition and Adequacy of Research Facilities A number of respondents stated that reports of the condition and the adequacy of facilities are, by their very nature, subjective. Two persons may make different assessments of the same facility or have different opinions of what is required in order for a facility to be suitable for a particular type of research. Despite the subjectivity involved, these items do capture an overall picture of the current status of facilities. Discussions with respondents at a number of institutions indicated that, for the most part, deans in consultation with department heads reported on the condition and adequacy of facilities. A few institutions indicated that they have detailed condition data in a central database. In those cases, the facilities office was able to respond to these items. A few institutions indicated that it is conceptually difficult to assess the condition of a research facility without including instrumentation in that assessment. Most respondents, however, indicated that they had no such problem and were able to report on the condition of the "bricks and mortar." Cost per Square Foot Data The study did not collect unit cost data for individual construction or repair/renovation projects. It collected only the aggregate researchrelated costs and the aggregate research space involved in all projects begun during specified periods. These aggregates can be combined into indices of average cost per square foot, which are useful in tracking broad cost trends over time. However, they are of little practical value as guidelines for project planning. By all accounts, unit costs for both construction and repair/renovation projects are highly variable, depending on the specific requirements of the particular project and on many other factors as well (e.g., geographic region of the country). Such differences, which are of crucial importance in project planning, are obscured in the kinds of multiproject averages that can be constructed from this study's data. Deferred Capital Needs The study asked institutions to report on deferred construction and repair/renovation costs that were included in an approved institutional plan. For definition purposes, the survey stated that deferred space must satisfy the following four criteria: the space must be necessary to meet the critical needs of current faculty or programs; construction must not be scheduled to begin in FYs 1994 or 1995; the construction must not currently have funding; and the space must not be for developing new programs or for expanding the number of faculty. Although such a question prevents respondents from being too speculative, the item fails to include needs that may, in fact, exist but not be part of an institutional plan. Given the fiscal realities of the 1990s, many universities and colleges may need new S&E facilities but competing priorities, coupled with decreased budgets, may result in institutions not incorporating such needs into official planning documents. Since 40 percent of all institutions indicated that they had an institutional plan that included deferred capital projects, the estimate of need derived from responses to this question must be interpreted as a conservative estimate of overall S&E facility needs.