Journal of the American Society for Information Science (JASIS) 
48(10) p.963, 1997


Validation of Citation Analysis


Although Dr. Gene Garfield and I have been in correspondence for about a year, during which I have tried to explain my work to him, he remains perplexed. I have no complaint with citation indices being used for what they were designed for. From time to time, I use them to help locate literature. I am complaining about using citations for evaluative purposes. I have pointed out that scientific articles cite only a fraction of their influences, and that citing is highly biased (MacRoberts & MacRoberts, 1986. 1987a). I have pointed out that citation analysts do not address the problems raised by their critics ( MacRoberts & MacRoberts. 1989a. 1989b. 1996) .

Gene admits that scientists do not cite all their influences (letter dated 12 Sept. 1996); he also admits that citing is biased: ''I don't think that there is any question that individual authors may be biased in their selection of resources" (letter dated 14 October 1996). He also disagrees with the Coles' (Cole & Cole, 1979) conclusion that only a handful of scientists are responsible for scientific progress and that the vast majority are drones, and instead. agrees with Ortega y Gassett that science is the work of all ( letter dated 26 July 1996) (MacRoberts & MacRoberts. 1987b ) .

There seems to be only one point of disagreement between us. Gene thinks that biased citing is random and I think it is nonrandom (MacRoberts & MacRoberts. 1987a. 1996). If Gene is correct. his conclusion flies in the face of all the evidence regarding citations being used to persuade audiences (Gilbert, 1977) and contradicts even the "Matthew Effect'' hypothesis Merton 1973) .

Otherwise, as far as I can tell,Gene and I basically agree. Surely it would be possible to test whether biased citing is random or nonrandom. As a matter of fact. we tested the matter a decade ago (MacRoberts & MacRoberts 1987a).

As to other points raised in Gene's letter. I not only claim but can assure him that I was not aware of the special issue of Scientometrics (v. 12. n. 5 - 6 Nov. 1987) he mentions until he informed me of it (letter dated 26 July 1996) and later sent me photocopies of the articles (letter and enclosures dated 12 August 1996). But that I was never informed by Scientometrics of the special issue or invited to respond to my critics or by the authors of the various critiques is not surprising; certainly it would not be surprising to modern students of the sociology of science (SSK). Scientific disputes generally are not settled by the "gentlemanly'' rules supposed to exist in the "fairy tale" view of science, but more often are settled by kangaroo courts.

Gene and I are clearly speaking past each other. I have no complaint with citation indices used as aides to locating scientific literature; I do object to the use of citations as data for evaluative purposes. The references given below contain a thorough discussion of my reasons.

Michael MacRoberts 
Bog Research
740 Columbia
Shreveport, LA 71104


Cole, J. R.. & Cole. S. (1972). The Ortega hypothesis. Science. 178. 368-375.

Gilbert. G. N. (1977). Referencing as persuasion. Social Studies of Science. 7, 113-122.

MacRoberts. M. H. & MacRoberts B. R. (1986). Quantitative measures of communication in science. A Study of the formal level. Social Studies of Science. 16, 151-172.

MacRoberts, M.H. and MacRoberts, B.R. (1987a). Another test of the normative theory of citing. Journal of the American Society for Information Science, 38, 305-306.

MacRoberts, M.H. and MacRoberts, B.R. (1987b). Testing the Ortega hypothesis: Facts and artifacts. Scientometrics, 12, 293-295.

MacRoberts, M.H. and MacRoberts, B.R. (1989a). Problems of citation analysis: A critical review. Journal of the American Society for Information Science, 40, 342-349.

MacRoberts, M.H. and MacRoberts, B.R. (1989b). Citation analysis and the science policy arena. Trends in Biochemical Science, 14, 8-10.

MacRoberts, M.H. and MacRoberts, B.R. (1996). Problems of citation analysis. Scientometrics. 36. 435-444.

Merton, R.K. (1973). The Sociology of Science. Chicago: University of Chicago Press.