Excel's formatting feature has always been a mixed bag. On the one hand, it can save valuable time, but on the other, it can make changes you don't want. Usually, that just results in a couple of extra clicks, but in the scientific field, consequences have been more dire.
An analysis by Melbourne-based Institute Baker IDI has found that out of 3,597 published papers, 704 had Excel-caused errors. The issue stems from gene symbols, one example being “SEPT 2,” which Excel changes to “September 2”.
Assam El-Osta, one of the paper's three researchers, told the BBC that the errors were found mostly in supplementary data sheets. The pages contain “important supporting data, rich with information,” and solving these errors is time-consuming.
Excel's gene renaming was first cited by scientists in 2004, and the Baker IDI study suggests that the problem has increased by an annual rate of 15% over the past five years.
However, a Microsoft spokesperson suggests that this is down to incorrect setting configurations:
“Excel is able to display data and text in many different ways. Default settings are intended to work in most day-to-day scenarios,” she told the BBC. “Excel offers a wide range of options, which customers with specific needs can use to change the way their data is represented.”
Director of the European Bioinformatics Insititute Ewan Birney thinks that bad scientific practices are to blame, however:
“What frustrates me is researchers are relying on Excel spreadsheets for clinical trials,” he says, recommending the application only for “lightweight scientific analysis.”
Excel's autocorrect feature works fine for their target audience, and the issue has been known for over a decade. The fault, then, lies not with the program itself, but with the scientific community's lack of configuration. With such high stakes, institutes should be publishing settings guidelines, or using a different application entirely.