Problems with Publish or Perish

Research is a combination of collaboration and competition. We collaborate in order to discover new things, to understand the processes of nature at a higher level. We compete not only because we wish to be pioneers and be the first with our foot in the door, but also because limited funding leaves 90% of all labs in the dust without a government grant. Laboratories are judged primarily on the ability of its members to publish good science. While this makes sense, it also gets us into trouble.

In cancer science, many “discoveries” don’t hold up. Researchers at Amgen took 53 groundbreaking publications and attempted to reproduce the results. They even performed some of the same experiments in the laboratories where they were originally done in an attempt to decrease any possible variability. Only 6 of the 53 papers actually had reproducible results.

The publish or perish mentality is obviously failing. Most claims that are published never get repeated, and research progresses based on the assumption that previously done work is true. While trust is a good thing, the thought alone that nobody will repeat your experiments can lead scientists to skew results and publish false inferences that they may have believed to witness.

I believe the problem stems from our funding system. My PI is an excellent story teller, and one thing he tells me is that grant funding applications had a far higher acceptance rate 20 years ago than now. This makes sense given the fiscal situation of our government. To correlate, a sharp rise in retractions has lead some to push for an ethics reform. The rise in scientific misconduct can be traced back to increasing competition for funds.

Most people would lie (or skew data), given the opportunity, especially if it brings them one step closer to that million dollar grant. There is urgency in science; if you don’t publish a result as soon as you find it, then a competing lab next door will snatch it away. Rarely double checked results and limited funds brings out the worst in science. Competition may breed the best research, but when pushed to the limit, it can also bring out the worst. An increase in science funding may aid in reducing the number retracted publications. In the end this would save us all time, and hopefully a truly landmark discovery can be made to help save lives. Come on lobbyists, push for science funding!