How Subtle Class Cues Can Backfire on a Resume (2016)

Re:How Subtle Class Cues Can Backfire on a Resume (2016) by red_admiral on Sunday March 26, 2017 @04:19AM (#13959764)

"close to statistical significance" is one of those phrases I dislike because it's almost designed to deceive. To the trained statistician it means "not statistically significant" but to the average reader it suggests there's something more there than there should be.

What we can safely say is that this study did NOT support the hypothesis that there is gender privilige for men within the lower classes in landing top law jobs, and that it would be interesting to repeat this study with a larger sample size.

[ Parent | on HN ]

Re:How Subtle Class Cues Can Backfire on a Resume (2016) by hn_throwaway_99 on Sunday March 26, 2017 @10:57AM (#13961084)

While I agree with your conclusion, I have a problem that people think "statistically significant" is some sort of golden, binary rule, instead of "statistically significant at X% confidence level". It's not like 95% confidence level was ordained by God, and it leads to p-value hacking. I'd argue that if you come out at 94% or 96%, both of these warrant further study.

[ Parent | on HN ]