Turns Out, Science Has Its Own 1% Problem

Nov 25, 2014 at 10:48 AM ET

The world of scientific research has a bigger problem with inequality than most developing countries.

A new study from Cornell found that once researchers get published in one top-notch scientific journal, it becomes much easier for them to get published in more. This trend has created a situation in which the majority of publications and citations are attributable to a small group of boffins.

This snowball effect of sorts is not uncommon, and is referred to as cumulative advantage, or the Matthew effect. The idea is that once a person gains an advantage in a competitive environment, that advantage will continue to grow and compound over time. This concept is encapsulated in the adage “the rich get richer and the poor get poorer.”

The traditional way to measure inequality is with the Gini coefficient, which represents how evenly wealth is spread across a country. A Gini coefficient of zero represents total equality, and disparity increases as you get closer to 1. The Cornell researchers used this same measure to figure out the inequality among researchers in the world of scientific journals. After they looked at nearly 412,500 publications spanning 23 different publications, they found that there was an average Gini coefficient of 0.48 for publications and a whopping 0.73 for total citations. To put these in perspective, Brazil, a country notorious for its income disparity, has a Gini coefficient of roughly 0.53.

So yeah, a relatively small group of scientists is hogging all the accolades. Some argue that there’s an upside to the inequality because it rewards scientists for quality research and creates a situation where the best and the brightest get published frequently. But on the flip side, the researchers found that there was a decreasing trend in the citation impact with each subsequent publication over the course of the careers of the scientists. This would imply that their work was less and less impactful over time, which makes sense.