Supplementary material can present a problem for a few reasons:
- Supplementary material is not always indexed, so citations that only appear in supplementary material (and not the manuscript) may not be counted by citation services (e.g., Seeber, 2008; Weiss et al., 2010; Rafferty et al., 2015)
- Supplementary material may have an issue with long term storage (i.e., links break, the repository may not be maintained perfectly, etc.; Evangelou et al., 2015)
- Reviewers may not spend as much time with supplementary material as the main manuscript (e.g., Pop and Salzburg, 2015).
- Production costs (time spent) for the editors and copy editors increase per article (if they even look at this material; Shutler and Murray, 2016)
- It is inconvenient/difficult for readers to bounce between the manuscript and the supplementary material (e.g., Shutler and Murray, 2016)
This is enough of a problem that some journals have taken steps to curtail the usage of supplementary material (see for example Borowski, 2011)
Kenyon and Sprague (2014) looked at supplementary material use in 60 different environmental science journals from 2000-2011. Included in the analysis by Kenyon and Sprague (2014) is Geology, which in 2011 had 80% of articles contributing supplementary material. One nice thing about GSA journals is all of the supplementary material is archived in one place — making it very convenient for looking at trends in supplementary material.
Here I am going to extend the analysis of Kenyon and Sprague (2014) for the entire history of Geology, from 1975-2015. Coupled with my previous work on the number of papers published per year, below is a time series of papers published per year, and number of Geology supplements published in the GSA data repository:
Percentages are shown below. In 2015, ~90% of Geology papers had supplemental material in the GSA data repository:
This percentage data can be fit with a logistic function. If the data above adheres to a logistic function (asymptotically approaching 100%), 100% of Geology articles will have supplementary material by ~2025-2030. I would put my money on a faster collapse to 100%.
As I have mentioned previously, there is a trend in rising authorship for length restricted articles — Engelder (2007) presented data from Geology (GSA) and in a previous post of mine I looked at GRL (AGU). One plausible reason for this is that more work must be done (by more people) to make it into these prestigious letter-length journals. The rise in supplementary material may be similar — so much research needs to be done to write short articles that the work cannot fit into the mandated length of the paper.
My sense is that other Earth and Geoscience journals have this issue with supplementary material — I’m just using Geology because GSA does the best job of recording all the supplementary material in one repository.