90% of Old Scientific Data May Be Lost
Thanks in part to outdated devices, email addresses
By Arden Dier, Newser Staff
Posted Dec 29, 2013 2:30 PM CST
And these old friends are partly to blame.   (Shutterstock)

(Newser) – Scientists rely on raw data to reproduce studies and power new research—it's a foundation of the scientific method. But as much as 90% of data is lost within 20 years, according to a new study that puts at least some of the blame on old technology. Researchers emailed the authors of 516 biological studies published from 1991 to 2011 to ask for the data, but tracked it down in just 23% of cases for various reasons. Usually active email addresses couldn't be found, emails didn't get a response, or the data was lost or inaccessible on outdated drives, Smithsonian reports. Of the studies that were more than 20 years old, 90% of data was inaccessible.

"Some of the time, for instance, it was saved on three-and-a-half inch floppy disks, so no one could access it," study author Tim Vines explains. "The current system of leaving data with authors means that almost all of it is lost over time," he notes, per the Science Recorder. Preserving data is crucial, since no one knows where research will take us moving forward, Smithsonian points out, and Vines makes the case that scientists should be required to post their data online. "Losing data is a waste of research funds and it limits how we can do science," he adds. "Concerted action is needed to ensure it is saved for future research."

More From Newser
My Take on This Story
To report an error on this story,
notify our editors.
90% of Old Scientific Data May Be Lost is...
Show results without voting
You Might Like
Showing 3 of 56 comments
Jan 5, 2014 9:01 AM CST
it there a printed version in the library of congress?hmmmm but!! i rather have a real printed paper book anytime than one of these e-books you read on your whatever you own. and i don't need any electricity to run it. period!!!
Dec 31, 2013 6:58 AM CST
I wonder if the authors of this study uploaded their raw data to the journal that published it?
Dec 30, 2013 9:57 PM CST
This was a small sample looking at a particular kind of data (anatomical measurements of plants and animals). 38% of the respondents just didn't get back to the researchers, so I'd hardly call that 'lost data'. There should probably be an online database for storing this kind of data, just like there is for genetic sequences. The NIH maintains a database called GenBank where researchers are usually required to deposit their genetic sequences if they're involved in a published paper.