CHAPTER II
Composition of Total Stock of Academic Research Instrumentation: Supersystems and Non-Supersystems
Data Considerations for this Chapter[21]
This chapter offers a profile of the composition of the supersystem items included in the total inventory discussion in Chapter I.[22] It gives an overview of the types of instruments that are being called supersystems and a comparison of these supersystems as they relate to the entire inventory of academic research instruments. Supersystems are excluded in the instrument status analyses in Chapter III. (See the Technical Notes for a more detailed explanation of supersystems.)
Context for this Chapter
In 1993, there were an estimated 150 supersystems, comprising less than 1 percent of the total number of instruments costing at least $20,000 (61,700) located in the 318 institutions. These same supersystems totaled $1.2 billion, which was 19 percent of the total aggregate cost of all instruments ($6.3 billion). The bulk of the supersystems were in four very specialized instrument types. The majority (approximately 120), were in computers. Three other categories (telescopes, nuclear reactors/nuclear science instruments, and research vessels/planes) contained fewer than 15 supersystems each. In addition, there were a total of 5 other supersystems found within electronic instruments, wind/wave/water/shock tunnels, and NMR spectrometers.
Where to find the data: Figure 3 depicts supersystems as a proportion of the total number and total cost of all instruments, by type of supersystem. Figure 4 depicts the same information by field. Table 4 presents the distribution of the numbers and aggregate purchase price of supersystems and non-supersystems by the types of instruments, and Table 5 presents the same information by ownership distribution among the various disciplines.
Relative Importance of Supersystems by Cost
Many (but not all) of the supersystems were extremely expensive for their type of instrument. For instance, only 21 percent of the approximately 40 research vessels/planes/helicopters represented in the survey were administered as supersystems; however, supersystems represented 81 percent of the total aggregate cost of all research vessels/planes/helicopters. Similarly, only 4 percent of the approximately 300 telescopes in the survey fit the definition of supersystems, but they comprised 24 percent of the total aggregate cost of all telescopes costing at least $20,000 (see Figure 3).
Relative Importance of Supersystems by Field of Science
As shown in Table 5, all supersystems were located in four academic fields: computer science (computers); physics/astronomy (NMR spectrometers, nuclear reactors/nuclear science instruments, research vessels/planes, and telescopes); environmental sciences (research vessels/planes); and engineering (electronics instruments, wind/wave/shock tunnels, and nuclear reactors/nuclear science instruments).
For each of the four fields which contained any supersystems, the supersystems comprised less than 10 percent of the total number of their instruments (see Figure 4). The discipline with the highest percentage of numbers of instruments classified as supersystems was computer science, with 6 percent of the computers fitting the definition of supersystem. All 118 of the supersystem computers were inventoried to computer science.[23]
In terms of total cost, the supersystems comprised a higher percentage of the aggregate purchase price of instruments in the inventory of the same four affected fields: 19 percent of the cost of all the instruments. The percentage varied widely among the four fields owning supersystems. Once again, computer science had the highest proportion; supersystems comprised 64 percent of the total aggregate cost of all computers inventoried to computer science (see Figure 4, Table 5).
Footnotes
[21] For the reader's convenience, the text discussion in this chapter generally rounds dollars to the nearest million and numbers to the nearest hundred. The accompanying tables, however, contain unrounded data.
[22] As explained in the background section and in the Technical Notes, many of these surveyed institutions own and administer a number of large, specialized research instruments that are built around or consist of a single integrated system (such as a telescope, computer center, nuclear reactor, or oceanographic research vessel). On some campuses, these are administered as being virtually identical to the academic unit itself, rather than as a separately inventoried part of a unit. To gather survey data for these special cases, the Supersystem Data Sheet questionnaire was developed, which was less detailed than the survey's regular Instrument Data Sheet. It is important to note that in many instances there was little difference between a supersystem instrument and another large integrated instrument system of the same type; only the administrative process was different, resulting in less data collected about the supersystem.
[23] It is understood that large computers and supercomputers are used by researchers from a variety of disciplines, and many are heavily used for large multidisciplinary research projects. However, in this report, all computers that fit the supersystem definition were automatically tabulated in the discipline of computer science.
