5 ms·
"the image is dominated by red and blue, a sign that scientists are more likely to cite the research that conforms to their outlook." Or a sign that the green
by hardlianotion 11y ago
"the image is dominated by red and blue, a sign that scientists are more likely to cite the research that conforms to their outlook."
Or a sign that the green lines were drawn first, and test red and blue overlaid.
- Ensorceled 11y agoAgreed. This is a terrible graphic. I have no idea what's going on for any data point except the few on the edges.
- IshKebab 11y agoWas about to post the same. It's hard to continue reading the article when the basic premise is so obviously flawed.
- ajholyoake 11y agoThis could have been improved by clustering using a d3-like force layout - The current node layout obfuscates the graph terribly!
- yk 11y agoCame here to mention that. Additionally these findings are easily explained if we assume that salt has some positive and some negative effects occurring in different parts of the body. Say salt has some positive effects for the liver and negative effects for the heart, then a paper on liver disease will mostly cite positive studies about salt, that is it cites studies on the liver, and a paper on heart diseases will cite mostly negative.