Joshua Aizenman, Kenneth Kletzer 30 April 2008
Networking, citation of academic research, and premature death
the impact and diffusion of academic research is by studying the
quantity and pattern of citations to published research findings. The
literature on the diffusion of technological innovations frequently uses
patent citation data to study the spread of technological knowledge
(for example, the volume by Jaffe and Trajtenberg (2002)). Citations are
a data source for investigating linkages from results to subsequent
research more generally in the study of the dissemination and creation
of ideas and knowledge.The study of citation patterns may be useful for
addressing a variety of policy concerns about organisation, funding and
incentives in academic research.
Citations to scholarly publications are often used to evaluate the
impact of research findings and the contributions of individual
researchers. In the natural and social sciences, citation counts have
been used to evaluate the productivity of individuals for appointments,
salaries, and research awards for years. Empirical studies by
Hammermesh, Johnson and Weisbrod (1982), Diamond (1986), and others find
that the compensation of economists rises significantly with cumulative
citations. Within economics, ranking university departments, journals
and economists using raw citation counts or recursively weighted
citation counts is increasingly popular.
The origins of citationsThe use of citation data to allocate research resources and reward
individual effort has generated a substantial literature on the validity
of using citations to measure the value of research. Much of the
analysis of the use and usefulness of citations is associated with the
sociology of science following Robert K. Merton. Merton (1973) observes
that citations to the publications of researchers by other researchers
may be influenced by considerations beyond the strict linkage between
new findings and old findings. Citation practices depend on academic
culture and traditions, and the incentives to cite may not accord with
the notion that citations provide an approximate record of the origin of
Richard Posner (2000) summarises the insights of the literature on
the incentives to cite. He distinguishes between informational and
strategic citations. Informational citations serve an important
expositional role by identifying the context of the article, allowing
the incorporation of supporting evidence without repetition,
establishing linkages to authoritative works, and acknowledging the
priority of others. Each corresponds to the notion that citations
measure the importance of a cited work. Strategic citations are made
simply to benefit the author and are due to the rewards to being cited
and the role of others in the peer reviewing process. The possibilities
of strategic citation behaviour and editorial bias have received
attention in the economics literature (Posner (2000) and Laband and
Piette (1994), respectively).
Information costs are also important for understanding citations.
Because scholarship is costly, the rates of citation to specific
articles can also depend on the promotion by the authors or more general
personal networking by authors. Networking increases the familiarity
with a researcher’s work lowering the information cost to others of
citing that work. In the economics profession, networking has been
studied in the context of citation circles based on graduate education
(Stigler and Friedland (1975), gender differences in citation frequency
(Ferber (1988)) and the research gains associated with co-authorship
(Sauer (1988) and Laband and Tollison (2000)).
Death and citationsIn a recent working paper, “The Life Cycle of Scholars and Papers in Economics - the ‘Citation Death Tax’,”
we consider whether citations depend only on the intrinsic contribution
of a publication or if they are influenced by the author’s professional
presence by estimating the impact of premature death of productive
economists on their citations. Premature death terminates activities
that help enhance the prominence of scholar’s publications, such as
presenting papers, pursuing follow-up research, encouraging related
research by others, and supervising Ph.D. students. Some of these costs
may be mitigated in circumstances where the research was done jointly
with active and productive scholars. The loss of some citations may
result from the termination of the incentives for strategic citation. A
researcher’s death can also have the direct effect of reducing research
activity on a particular topic.
We consider the potential importance of professional presence and
networking for citations by first constructing a sample of publications
and citations to highly-cited economists who died well before retirement
age during the period from 1975 to 1997. We identified sixteen such
economists and used the Web of Science to assemble a panel of the annual
citations of 428 papers written by these 16 economists over the years
1957 to 2006. The members of the sample vary in terms of prominence.
Five have average citations per year between 2 and 10, five receive
average citations/year between 10 and 20, 4 have average citations/year
between 20 and 100, and two, with average citations/year exceeding 100,
died before co-authors received the Nobel prize (Fischer Black and Amos
Tversky). We removed citations by the authors to their own work.
The dynamics of the citations generally correspond to the expected
life cycle pattern of an article in which its citation rate first rises
sharply as awareness of the article spreads. Then citations gradually
decline over time. This pattern is violated for publications by the two
otherwise Nobel laureates in the sample. We constructed the
intertemporal path of missing citations of a paper in the sample
relative to the hypothetical citations had the author been alive, and
calculated for each paper the cumulative missing citations. The break in
citations following the economist’s death yields an estimate of the
Our results are mixed. For half of the economists in the sample, we
identify a large and significant “citation death tax” for the average
paper written by these scholars. For these authors, the estimated
average missing citations per paper attributed to premature death ranges
from 40% to 140% (the overall average is about 90%), and the annual
costs of lost citations per paper are in the range 3% and 14%. Hence, a
paper written ten years before the author’s death avoids a loss of
citations that ranges between 30% and 140%. For the other half of the
sample, there is no citation death tax; and for two extraordinary
scholars in this second group, Black and Tversky, citations took off
over time, reflecting the growing recognition of their contributions.
The value of presenceOur sample of prominent economists who died early suggests that being
there is important for generating citations. It is natural to expect
that an author’s premature death leads to a sudden drop in strategic
citations to his publications. Informational citations may also fall
following a researcher’s death because he can no longer promote his own
research or continue training students. Researchers play a role in
promoting their own research simply by being visible to the research
community and continuing to press a current research agenda through
presentations and follow-up papers. Identifying the precise importance
of all these factors requires much more detailed information about
various dimensions of networking than available in the data on
publications and citations. Our results do suggest two directions of
implication. The first is to confirm the conclusion of Robert Merton and
successive sociologists of science that citations are not an especially
pure measure of scientific impact and may be affected by strategic
considerations or information costs. The other is that networking may
have a significant impact on the dissemination of research findings and
diffusion of valuable knowledge. This second bears on the organisation
of research and can have implications for public policies towards the
production of basic research.
ReferencesAizenman, Joshua and Kenneth Kletzer (2008), “The Life Cycle of
Scholars and Papers in Economics – the ‘Citation Death Tax’,” National
Bureau for Economic
Research, Working Paper 13891, March.
Diamond, Arthur M. (1986), “What is a Citation Worth?” Journal of Human Resources, Vol. 21, pp. 200-215.
Ferber, Marianne A. (1988), “Citations and Networking,” Gender and Society, Vol. 2, No. 1, March, pp. 82-89.
Hamermesh, Daniel S., George E. Johnson and Burton A. Weisbrod (1982),
“Scholarship, Citations and Salaries: Economic Rewards in Economics,” Southern Economic Journal, Vol. 49, No. 2, pp. 472-481.
Jaffe, Adam and Manuel Trajtenberg (2002), Patents, Citations, and Innovations: A Window on the Knowledge Economy, Cambridge, MA: MIT Press.
Laband David N. and Michael J. Piette (1994) “Favoritism versus Search
for Good Papers: Empirical Evidence Regarding the Behavior of Journal
Editors” The Journal of Political Economy, Vol. 102, No. 1., pp. 194-203.
Laband David N. and Robert D. Tollison (2000) “Intellectual Collaboration” The Journal of Political Economy, Vol. 108, No. 3., pp. 632-662.
Merton, Robert K. (1973), The Sociology of Science, Chicago: University of Chicago Press.
Posner, Richard A. (2000), “An Economic Analysis of the Use of Citations in the Law,” American Law and Economics Review, Vol. 2, No. 2, pp. 381-406.
Sauer Raymond D. (1988) “Estimates of the Returns to Quality and Coauthorship in Economic Academia,” The Journal of Political Economy, Vol. 96, No. 4., pp. 855-866.
Stigler, George and Claire Friedland (1975), “The Citation Patterns of Doctorates in Economics, Journal of Political Economy, Vol. 83, No. 3, pp. 477-507.