NASIG Newsletter

22nd CONFERENCE
STRATEGY SESSION 

How Does Digitization Affect Scholarship?
Roger C. Schonfeld, Ithaka
Reported by Buddy Pennington

Roger C. Shonfeld of Ithaka gave an insightful glimpse into the preliminary results of a research study assessing the impact of digitization on scholarly research.  The ongoing study involves impacts on citation rates for three disciplines including economics, history and biological sciences.  Roger reported that the study has conclusively shown an impact on citation rates for economics.  The study has not progressed to the point where conclusions could be drawn for the disciplines of history and biological sciences.

Scholarly publishing can be viewed as a two-sided market where journal publishers serve as an intermediary between authors and readers.  A national faculty survey conducted in 2006 highlighted the differences between authors and readers in terms of what they value most in an academic journal.  When asked what authors look for when choosing a journal to publish their research, wide circulation was the most important criterion.  That is to say that journal impact matters most to authors.  So how has journal impact changed with the recent increase in journal digitization?

The Ithaka study involved examining the citations to 100 journals for each of the three disciplines within Thomson’s ISI citation databases.  The years examined included 1980 to 2005.  Regression analysis was used to determine correlations between digitization and changes in citations to the selected journals for these disciplines. 

Preliminary results on the impact of backfile digitization, using a sample set of journal volumes from 1956-1968, indicated that there is a relationship between digitization and an increase in citations to those backfile journals.  Inbound citations increased from 7% to 17% after digitization, and the study demonstrated that this impact grows steadily over time.  Different channels or platforms had different effects.  For one platform, the increase was in the 3%-15% range while it was 8%-18% for another platform. 

Preliminary results on the impact of digitization of current issues, using a sample set of journal volumes from 1995-2005, showed a significant effect.  However, the results were more complicated than the backfile data and require additional analysis before they can be reported with any degree of confidence.  Roger did note that the publisher is not always the optimal distribution channel in terms of citation impact and that longer embargo periods decrease the citation impact for that particular platform or channel. 

Roger concluded by stating that Ithaka will continue with the research project in order to assess any statistical variation between the disciplines.  He would also like to look at whether the year of publication matters in terms of digitization’s impact on citation rates.  Is the relative impact greater for older materials?