The phrase “data not shown” peppers scientific manuscripts, referring to findings that are relevant enough to be mentioned but not to be depicted in a figure or table. This lack of visual evidence seems to contradict the scholarly focus on careful documentation, including detailed materials and methods sections, numerous figure panels and controls, and even lengthy supplementary information.
So why is this practice relatively common? Why not show data if you have it?
There are four main reasons to include the parenthetical “data not shown” (or phrases such as “unpublished observations,” “personal communication,” or “manuscript in preparation”) rather than adding a corresponding figure or table:
######1. Confirmatory results.
Findings that validate past reports are critical to demonstrating the reproducibility of those previous findings and to demonstrating that your research effectively builds on previous research. However, if similar results have been published elsewhere, the journal and/or your readership may not be interested in viewing the actual data.
######2. Negative results.
Results that show no significant effect are also important to the scientific record; if these findings are not cited, the same experiment may be futilely repeated by other researchers in the future. Similar to the confirmatory results mentioned above, your key finding, but not the granular data, may be of interest to readers in this case.
######3. Peripheral results.
The phrase “data not shown” is often cited when referring to data that are not directly relevant to the main topic of the paper. These data include interesting incidental findings, which are often cited in the discussion section, rather than in the main results section. Exclusion of the actual data in this and the previous two cases may be particularly fitting if your paper already contains many figures and the journal has a figure limit.
######4. Future results.
If you plan to publish certain data in a subsequent paper but the data are also relevant to the current manuscript, you may want to briefly refer to these results without showing a figure or table. You may then include a note such as “manuscript in preparation” or even “manuscript submitted.”
###How scholarly publishing is moving away from “data not shown”
Consistent with all four cases, Science Translational Medicine states that “references to unpublished results or data not shown are not permitted to substantiate significant conclusions of the article.” However, the editors of Nature Chemical Biology, among others, have argued against data omission due to the need for complete documentation to facilitate thorough editorial and peer review and subsequent replication.
Other publications, including BMC Medical Ethics, add that results that would traditionally be cited as not-shown data “can and should” be included as supplementary information.
The increasing shift toward online publication, eliminating space limitations, makes including additional data more feasible than in the past, even with the advent of massive datasets. In fact, certain journals, including GigaScience, concentrate on the publication of such large datasets.
In sum, if you are considering referring to findings as “data not shown,” please check whether this practice is permitted by your target journal and, if so, whether your results fit into one of the four categories above. If in doubt, and space permitting, consider including these hard-earned data in a supplementary information section.
Alternatively, you could contribute your confirmatory, negative, and/or peripheral data to a repository (figshare is one well-known example), not only making these data citable but also ensuring that your research findings remain accessible to the scientific community.
With the help of these increasingly common strategies, whether your results are shown or not shown in a manuscript, they will potentially be able to guide and make an impact on future research.