produced by each country in each scientific area. 

 The general rationale for such an index is the 

 expectation that the most significant literature 

 will be most frequently cited, whereas relatively 

 unimportant research articles will attract few, if 

 any, citations. Support for the validity of this 

 indicator is the high correlation, found in a 

 number of studies, between the significance of 

 papers as judged by researchers in the field and 

 as measured by the number of citations. How- 

 ever, articles may fail to be noticed because of the 

 language or journal in which they are published, 

 whereas others may be heavily cited because of 

 the criticisms they provoked or because they 

 describe a minor improvement in methodology. 

 These and similar limitations of the indicator, 

 however, are minimized by the extremely large 

 number of citations involved in the present case. 



The data source for the "significance" indi- 

 cator was the Science Citation Intiex, which includes 

 all publications cited in the 500 journals used in 

 this study. The total number of citations 

 received by the literature from each country in 

 each area was divided by the number of research 

 articles produced by the particular country in 

 that scientific area. The resulting "cita- 

 tion/publication" ratios were determined for the 

 years 1965, 1967, 1P69, and 1971. Since the 

 ratios did not change significantly over the 

 period, the mean annual value was used. These 

 data are presented in figure 5, 



Overall, the U.S. -produced literature had the 

 highest "significance" ratio in five of the seven 

 fields, with systematic biology and mathematics 

 the two exceptions. ("Economics" was omitted 

 because of the lack of reliable citation data.) 

 Literature of the United Kingdom received the 

 next highest rankings, placing either first or 

 second in each field. Ranking after the United 

 States and the United Kingdom were West 

 Germany, Japan, U.S.S.R., and France, in that 

 order. ■• 



The "Patent Balance" 



Data on patent applications and awards are 

 also measures of inventive output. Inventions of 



new processes and products, of sufficient 

 originality to be patented, represent potential 

 technological advances. Patents, however, vary 

 greatly in their technical and economic 

 importance and the basis for their award differs 

 in important ways from country tocountry. Not 

 only does the rigor of the tests for originality 

 vary, but there are also considerable differences 

 in the relative success of litigation involving 

 patent rights; this determines the relative ease 

 and value of obtaining patents in different coun- 

 tries. 



The absolute number of patent applications or 

 awards in individual countries is not an ade- 

 quate indicator for the purposes of inter- 

 national comparisons. It is more meaningful to 

 compare the number of patents awarded to 

 nationals with those awarded to foreigners in 

 each country. 5 This yields an index which 

 reflects the relative success of countries in 

 developing products and processes of sufficient 

 potential significance to warrant international 

 patent protection. 



Figure 6 presents the total number of patents 

 awarded to U.S. nationals by five countries 

 (United Kingdom, U.S.S.R., West Germany, 

 Japan, and France), those awarded by the United 

 States to nationals of these countries, and the 

 resulting U.S. balance. This shows that the 

 "patent balance" of the United States fell by 

 some 40 percent between 1966-70. The decline is 

 due principally to the reduced number of patents 

 awarded to U.S. nationals by foreign countries. 



Since patent applications in a foreign country 

 are usually the result of simultaneous patent 

 application in the applicant's own country, these 

 data indicate that the rate of growth of patent- 

 able ideas of international merit have been 

 expanding at a greater rate in other countries 

 than in the United States. 



The patent balance of the United States rela- 

 tive to each of the other countries (except the 

 U.S.S.R. which accounts for less than one per- 

 cent of the total patent transactions considered) 

 is presented in figure 7. Overall, the U.S. balance 



•■ The particular sample of journals used in estimating the 

 national origins of literature and the relative significance of 

 the literature may have resulted in some bias favoring 

 English-language publications. 



5 Patents awarded to U.S. nationals by foreign countries 

 minus patents avi^arded to foreign nationals by the United 

 States. 



10 



