Hacker News new | past | comments | ask | show | jobs | submit login

I think you can pretty much reproduce their study from the percentages with the following table (they claim 26 of the 106 did not meet study eligibility criteria, but the total seems to be 79? but then the total isn't consistent between the two categories; I didn't read carefully enough to see if the discrepancy is explained)

   categories    abnormal semen (n=34)   normal semen (n=45)
    talk >1h          14                   9
        <=1h          20                   36
     
    while 
    chging? 
         yes          12                   6
         no           21                   38

          
It's interesting to note that they actually collected more data than < 1h and >1h (they more finely categorized it as 0-30, 30-60, 60-120 and 120+). The freedom to choose the boundary there should count against the p-value.

Of course, it would be nice if they treated hours talking on phone as a continuous variable instead of a discrete <1h and >1h. With the freedom to set that boundary (if they had that), it would reduce the associated p-value.

Either way, it's very unlikely that the cause is RF radiation and not some confounding variable. Radiation damage is pretty much excluded, and I don't think there can be any effect on the impulses of the nervous system, which I think operates in the kHz, very far away from the 800 MHz+ from cell phones.

Talking while charging seems very unlikely to be related to any radiation, but might be associated with talking while driving or a sedentary lifestyle or something like that.




I also noticed that discrepancy in the "talking while charging column"; 21 + 12 = 33 != 34. I briefly looked for an explanation in the text, but I could not find one.




The deadline for YC's W25 batch is 8pm PT tonight. Go for it!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: