Home | You are at

What proportion of papers are submitted to the archive only after peer review?

Written by Tim Brody, last updated August 31 2000 10:09:20.

Using data from the publication rates.

Because quite a few Journal-ref entries contain what looks like dates, but are in fact other values the possible dates have been restricted to between 1950 and 2050. This means that papers that have been published prior to this date (of which there are some in the archive) will not figure in these results.

We can see by looking at the graph of when exclusive submissions have been made to papers, that the percentage of papers with submissions only after their publication date is:
3.76% of published papers.

It is possible that this figure could be much higher, as 80000 of the 130000 papers have no Journal-ref, with a high proportion of these being submitted only once. It is possible that this "set" of papers have only been submitted after being peer-reviewed, and the author has neglected to give the Journal reference.

Using incremental data between Nov 1999 and Jun 2000

grep 'abs' < d_papers | uniquecol 2 | restrictcol '3/^(199911|199912|2000)/' | papers2withjr | wc
920 papers first submitted and only submitted once in the period have been submitted with a journal-ref (i.e. have not been submitted before peer review) - against overall submission of 12742 unique papers with only one submission in the period.

100*(920/12742) = 7.22% of papers submitted in the period with only one submission are peer-reviewed.

Using data from the archive

If a paper has only been submitted once and contains a journal-ref entry then it must have been submitted after being reviewed.

103238 papers have had only one submission. Of these 35097 papers have got journal-ref entires. Therefore the proportion of papers that have been submitted following peer review is 100*(35097/103238) = 34.00%.

Papers submitted in December 1999

grep 'abs' < d_papers | restrictcol '3/^199912/' | uniquecol 2 | wc
2520 papers were submitted in November 1999.

grep 'abs' < d_papers | restrictcol '3/^199912/' | uniquecol 2 | papers2withjr | wc
Of these 112 papers were submitted with journal references (i.e. submitted after peer review).

100*(112/2520) = 4.44% of papers submitted in December 1999 were peer reviewed.

For january:

100*(98/2357) = 4.16%

For hep-ph (dec and jan):

100*(11/688) = 1.60%

For math-ph (dec and jan):

100*(9/68) = 13.24%

For nucl-th (dec and jan):

100*(5/141) = 3.55%

For nucl-th (dec thru jun):

100*(17/445) = 3.82%