Hacker News Reader

Would need to see the full article to find out the size of their sample, but their confidence intervals are quite tight for a study of this nature, and they add in the caveat of further research being needed. This should be a least a warning correlation for cell phone users, and will make me rethink some of my usage habits.

View on HN

Or the 5th sentence of the abstract:

> Questionnaires accessing demographic data and characteristics of cell phone usage were completed by 106 men referred for semen analysis

And this article isn't recommending a change in behavior, only further study:

> Investigation using large-scale studies is thus needed.

"Investigation using large-scale studies is thus needed."

Which would be the conclusion of pretty much any result that they could have had. I could have given you that recommendation even without conducting a study at all.

this study is virtually useless from statistical point of view. Sampling bias, lack of control for confounding effects and other issues mean you should not in any way use this study as a guide for any changes in behavior. Sure, I understand, they did add a caveat that more study is needed. Just saying, do not use this study as a reason to do anything differently

Reply on news.ycombinator.com