Maintaining Trust AND Data Integrity - a forum for discussion. INFO HERE.
#IntegrityAndTrust 1. Publish statements on how data integrity was assured
By Pedro R. Peres-Neto
It is great to see how much effort is being put to scrutinize these data and even understand how the reported issues have happened. Perhaps this was a one-time incident, though data issues of less severe proportions are likely to occur from time to time. Perhaps we should take this opportunity to start thinking towards a more robust system to reduce future potential issues. Obviously, the number of papers is increasing dramatically, so data issues (from minor to huge) are likely to increase as a result of data management, lack of scientific rigor or even data fraud. Collating data from different sources beyond single studies, for example, has become common practice in many fields. Simple or systematic ways to improve data integrity and scrutiny may be required by different fields and data sizes.
There are strong signs that the number of co-authors per paper is increasing through time in many fields, including ecology. Perhaps there can be some support to adhere to a publication policy in which co-authors explain which steps they took to assure data integrity. These could be in the form of published statements. One way could be to have more than one co-author (perhaps not all) spend a meaningful amount of time in scrutinizing the data and analyses to assure that (we) co-authors did the best to reduce potential data issues. By no means I am saying that co-authors of retracted papers are to blame. We know all too well that issues like that can happen. What I’m saying is that by having more than one co-author involved from the beginning is one way to increase data integrity control. There can be certainly other solutions and that’s why I’m suggesting that authors publish statements on how they did their best to assure data integrity.
It is time for a discussion on potential solutions to hold and preserve data integrity to the highest standards possibly. Solutions (in my mind) are needed to have the continuing public support in trusting and using research outcomes, and the support of taxpayers that much support our research.