3 ms·
what you described is prevalent in social sciences. Ton of biases and causation/correlation error and putting blind trust in some arcane statistical analysis wi
by curiousjorge 11y ago
what you described is prevalent in social sciences. Ton of biases and causation/correlation error and putting blind trust in some arcane statistical analysis without knowing what they really mean. Conclusion: statistically significant is the magic word peppered throughout academic literature.
- achompas 11y agoAgreed. Fortunately, excellent social scientists really care about this -- see Andy Gelman's blog for many rants on this topic.
- Balgair 11y agoI'll echo for biology and sports sciences (think moneyball). http://www.wired.com/2009/09/fmrisalmon/ http://www.wired.com/2009/09/fmrisalmon/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182327/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182327/