We’re Losing 80% Of Our Scientific Data Because of Hotmail Email Addresses And Floppy Disks (And Other Outdates Storage Devices)
Everyone assumes that some amount of old scientific data gets lost along the way. However, no one knew how bad it really is. A new study in Current Biology evaluated the rapid pace at which we are losing the data, exposing a potentially devastating trend.
Like Us on Facebook
Here are some statistical figures to set you thinking on the enormous loss to science due to careless handling of data. Someone needing data from previous studies while doing their research will be able access them if those studies were published within the last two years, according to the researchers. However, stretch the post-publication span to 20 years, and you will have nearly 80 percent data lost. The primary reasons for the missing info, according to the researchers, are mundane issues such as obsolete storage devices and old email addresses.
Timothy Vines of the University of British Columbia and the lead author of the study found it surprising that nearly all data are lost within two decades of their publication, even as he is realistic enough to claim "nobody expects that you'd be able to get data from a fifty year old paper."
Vines and his colleagues examined research papers that used simple and specific data to arrive at their conclusion. They zeroed-in-on data on "length measurements of plants and animals" because the method involved in these measurements has remained unchanged for decades, so it was much easier for the study team to make fairly straight comparisons over time.
The study team's analysis found that there were 17 percent lesser chances, each year, of finding an original data set for any one of those papers. In other words, as you move year over year, the chances of finding the original data diminished by 17 percent each year.
Each year, an enormous amount of unique scientific data is generated, especially from publicly funded scientific researches. They are essentially irreplaceable, because recreating the studies and regenerating the data would be prohibitively costly. Much of the blame for the loss of data rests with the current system of leaving data with the author, according to Vines. Researchers, and journals in particular, should ensure that "it is saved for future researches" Vines said.
There's a simple solution, the study says: authors should share their data on a public archive even before they get published.
Photo above courtesy of Shutterstock.
© 2012 iScience Times All rights reserved. Do not reproduce without permission.