Defining R&D: Is Research Always Systematic?

1 Defining R&D: Is Research Always systematic ? Beno t Godin 3465, rue Durocher Montr al (QUE). Canada H2X 2C6. Project on the History and Sociology of S&T Statistics Paper No. 7. 2001. Previous papers in the series: 1. B. Godin, Outlines for a History of Science Measurement. 2. B. Godin, The Measure of Science and the Construction of a Statistical Territory: The Case of the National Capital Region (NCR). 3. B. Godin, Measuring Science: Is There Basic Research Without Statistics? 4. B. Godin, Neglected Scientific Activities: The (Non) Measurement of Related Scientific Activities. 5. H. Stead, The Development of S&T Statistics in Canada: An Informal Account. 6. B. Godin, The Disappearance of Statistics on Basic Research in Canada: A Note. 1. Defining Research : Is Research Always systematic ? Defining R&D: Is Research Always systematic ? According to the Oxford Dictionary, the term Research has French origins and appeared in the 16th century.

2 1. It is rooted in the term search invented in the 14th century and defined as examine thoroughly . Research meant an act of searching closely and carefully , or intensive searching . It was first applied to science in 1639 as scientific inquiry , but rarely used in that context before the end of the 19th century. These definitions all included the essential idea of systematicity. Current definitions also focus on this idea. Whether one looks at 20th century dictionaries or international conventions on R&D, definitions of Research Always contain the idea of systematicity. The 1939 edition of the Webster dictionary, for example, defined Research as diligent inquiry or examination in seeking facts or principles , 2 while more recent definitions often specify diligent and systematic . Similarly, the OECD's definition of Research and development (R&D) uses the word systematic explicitly: R&D is creative work undertaken on a systematic basis to increase the stock of scientific and technical knowledge, including knowledge of man, culture and society and the use of this stock of knowledge to devise 3.

3 New applications . Despite these parallels, the current concept of systematic science is radically different from what it originally was. The purpose of this present paper is to discuss a change in the notion of Research that occurred in the twentieth century, as a result of using a specific sense of systematic . There are three parts to the thesis. First, the meaning of systematic in definitions of Research has drifted from an emphasis on the scientific method to an emphasis on institutionalized Research . Second, this drift is closely related to the (modern) instrument 1. Oxford Dictionary of English Etymology, Onions (ed.), Oxford: Clarendon Press, 1966; The Shorter Oxford English Dictionary, W. Little, Fowler, J. Coulson, Oxford: Clarendon Press, 1959. 2. Webster's 20th Century Dictionary of English Language, New York: Guild Inc., 1939. 2. Defining Research : Is Research Always systematic ?

4 Used for measuring Research and to its limitations, namely the survey. Third, the definition had important consequences on the numbers generated, the most important one being the undercounting of R&D. Research Institutionalized The OECD international standards on R&D suggest that governments measure institutional rather than individual Research : it is institutions that should be surveyed, not their individual researchers. 4 The measurement of Research is thus based on the classification of institutions by economic sector as defined in the National Accounting System (NAS). The original sectors included government, industry and households (individuals) but, for R&D statistics, the latter has been eliminated and the university sector added. The OECD Frascati manual (1993) suggests two approaches for surveying R&D: one only surveys a few known R&D performers (or institutions), as most countries do, while the other surveys a random sample of all R&D performers: There are at least two feasible approaches for establishing the survey population of the business enterprise sector.

5 One is to survey a sample drawn from the entire sector, choosing the sample on the basis of the company data available to the methodologists, such as employees and sales, by industry and region. The other is to try and survey only firms supporting R&D (para. 399, p. 107). Until 1993, there was no precise standard for deciding which firms to include in the survey. Member countries therefore interpreted the text differently, making international comparisons difficult. This meant that small and medium sized firms (SME) were, usually, poorly surveyed because R&D was thought to be a statistically rare event in smaller 3. OECD (1993), The Measurement of Scientific and Technical Activities: Proposed Standard Practice for Surveys of Research and Experimental Development, Paris: paragraph 57, p. 29. 4. During the discussions in 2000 about the revision of the Frascati Manual, the OECD began, however, to look at how to measure R&D performed by consultants which amount to 13% of all current costs for R&D in the business sector in Sweden for example.

6 See: DAS/EAS/STP/NESTI(2000)17; A. Sundstrom (2001), Improve the Quality of R&D Personnel Data, Especially in Respect to Consultants, DSTI/EAS/STP/NESTI. (2001) 14/PART15. 3. Defining Research : Is Research Always systematic ? 5. units , not systematic . In fact, the OECD distinguishes R&D according to whether it is continuous or ad hoc: R&D by business enterprises may be organized in a number of ways. Core R&D may be carried out in units attached to establishments or in central units serving several establishments of an enterprise. In some cases, separate legal entities may be established to provide R&D services for one or more related legal entities. Ad hoc R&D, on the other hand, is usually carried out in an operational department of a business such as the design, quality or production department (para. 153, p. 51). The manual recommends concentrating on continuous R&D only: R&D has two elements.

7 : R&D carried out in formal R&D departments and R&D of an informal nature carried out in units for which it is not the main activity. In theory, surveys should identify and measure all financial and personnel resources devoted to all R&D. activities. It is recognised that in practice it may not be possible to survey all R&D. activities and that it may be necessary to make a distinction between significant R&D. activities which are surveyed regularly and marginal ones which are too small and/or dispersed to be included in R&D surveys. (para. 392, p. 105) This is mainly a problem in the business enterprise sector where it may be difficult or costly to break out all the ad hoc R&D of small companies (para. 394, p. 106). From these citations, one may conclude that systematic R&D is to be understood as Research performed on an institutionalized and, above all, continuous basis. The main requirements to that end were put forward in the 1993 edition of the Frascati Manual, although they clearly reflected past and current practices of statisticians.

8 The Semantics of systematic . The definition of Research as an organized, formal and continuous activity is an important drift in the concept of Research . One historical use of systematic in relation to Research was associated with positivism that defined science as a search for comprehensive regularity 6. and general law. Inductivism was closely associated with this definition. This is, for example, the understanding given by the Canadian Department of Reconstruction and 5. OECD (1981), The Measurement of Scientific and Technical Activities: Proposed Standard Practice for Surveys of Research and Experimental Development, Paris: para. 243, p. 72. 6. C. Hempel and P. Oppenheim (1948), Studies in the Logic of Confirmation, Philosophy of Science, 15. (135): 135-175. 4. Defining Research : Is Research Always systematic ? Supply in its survey of government R&D in 1947: ( ) with the growth of modern scientific methods ( ) which proceeds by observation and experiment , and by the systematizing of the resulting facts and relationships into truth or laws, the search for new knowledge, especially in the scientific and technical fields has become more and more 7.

9 Institutionalized and professionalized . This meaning gave rise to and has been incorporated in the institutional definition of pure Research as the general knowledge of nature and its laws: science as an activity that begins 8. with observations and ends in truth and general laws. This meaning of systematic is closely related to a second one, that of scientific method. One can find it stated explicitly in UNESCO documents, for example. The first edition of the Guide to the Collection of Statistics on Science and Technology defined scientific Research with four elements, among them was the use of scientific methods, or work in a systematic way . 9 Elsewhere, we also read: An activity can be said to be scientific, in fact, when it is based on a network of logical relationships which make it possible to obtain reproducible and measurable results. The methods used to obtain these results may be considered as techniques when the skills they employ are also systematic , when these skills are based on numerical measurements, and when the results which these measurements give are 10.

10 Reliable . The model behind this understanding of Research is, of course, the natural sciences that 11. proceed by way of (laboratory) experimentation. The model was so pervasive that E . 7. Department of Reconstruction and Supply (1947), Research and Scientific Activity: Canadian Federal Expenditures 1938-1946, Government of Canada: Ottawa: p. 5. 8. V. Bush (1945), Science: the Endless Frontier, North Stratford: Ayer Company Publishers Inc., 1995: 81. 9. UNESCO (1977), Guide to the Collection of Statistics in Science and Technology, Paris, p. 18; see also K. Messman (1977), A Study of Key Concepts and Norms for the International Collection and Presentation of Science Statistics, UNESCO: Paris, p. 20. 10. Bochet (1974), The Quantitative Measurement of Scientific and Technological Activities Related to R&D Development, CSR-S-2, Paris: UNESCO: 11. Since the 1970 edition of the Frascati Manual, the OECD adds the adjective experimental to development to avoid a confusion between development, a phase of R&D, and the same term in economics, and to use the same term as Eastern European countries and UNESCO.