JCMT Citations 1990-2001

Citations are a commonly used, and often controversial, metric of scientific work. Lies, Damned Lies and Citations! Citation information is simply that — information. This information needs to be carefully considered before drawing any meaningful conclusions. While the number of papers published is a measure of productivity, citations are often referred to as a measure of impact. I like to think of citations as measuring the relevance of the research to what others are doing.

I have been maintaining a database of observatory publications that includes the JCMT telescope among others. This database also includes citation counts that are retrieved from NASA's Astrophysical Database System (ADS). The standard source of citation information is the Institute for Scientific Information (ISI) Science Citation Index. The big advantage of the ADS is that the citation information is free and available via a Web interface. The ADS citation numbers are different from the ISI ones as ADS does not include citations from a number of physics journals and they do include citations from conference proceedings. Crabtree and Bryson (2001) found that the ADS and ISI numbers agree to about the 5% level. Besides the raw citation numbers I also use "final" citation numbers. These are the predicted number of citations a paper will have after 20 years. These numbers are estimated from an average growth curve for CFHT publications (Crabtree and Bryson 2001). The "final" citation numbers are useful when comparing an ensemble of papers published in different years as the number of citations for papers grows naturally with time. While far from perfect, I've found this a very useful approach when comparing average citation counts across many years.

This first figure shows the average citations/paper for JCMT publications between 1990 and 2001. One can see the expected trend of increasing citation counts for older papers. The second figure shows the "final" citations numbers which removes this average trend. Clearly evident starting in 1998 is what I call the Scuba Bump. The average citation rate for JCMT papers jumped significantly because highly cited SCUBA papers were being published.

As mentioned, I also have a number of other telescope publications in my database. The third figure shows how JCMT compares to these other telescope between 1992 and 2001. Again, the tremendous impact of SCUBA papers is readily visible. Note how the average citation rate for JCMT is higher than that of HST. HST pumps out a lot of papers, many of them highly cited. However, on average, a recent JCMT paper receives as many, or more, citations.

I've retrieved the 500 most cited papers from the ADS for several years and can count the number of papers from each telescope in the "Top 500" for each year.

Click here for printable version.