Good idea. Thanks.
That's what I was mumbling about. What it seems to say is that there was absolutely no exposure data from Group B (non-complainers) with GSM at either 900 or 1800. There was some data at 2100. Therefore, since nobody in Group B felt anything, they'll just generate some numbers based on the relative level of complaints between the two groups. It's a little better than fabricating data, but not by much. I never could figure out how they correlated their "wellness" scores. Lots of detail on the procedure, but without the raw data to verify that the proceedures were followed, they could have just cooked the numbers and nobody could tell.
Ok, that makes sense. Still, it's kinda dumb stuffing the probe directly in front of the antennas, which is guaranteed to produce weird measurements, affect the antenna patterns, and affect the exposure levels.
If they had recorded those two, I could get a minimal indication if the subject was lying on the wellness tests. They dropped it probably to avoid correlating wellness scores with nervousness. If I were doing it, I would probably dump the LCD touch screen, and run the test orally, so I could use a voice stress analyzer.
I suffer from all the above even without RF exposure.
There's another problem. The list is too long. People taking such tests always are in a rush (to get out of there). If the list is presented on paper, one typically starts at the top, checks off a few items, and then thinks "that's enough". If the list is presented on an LCD screen, one at a time, where one is asked "Yes/No" to each of the symptoms, the opposite happens. Near the end of the list, people tend to feel guilty pushing no all the time, and throw in a few yes answers for balance. Somewhere in my pile of papers is a study done on such "check all that apply" lists, which demonstrates the effect.
Oddly, the most common real RF complaint I can recall is hearing tones or noises. That makes some sense with the 217Hz GSM pulse rate. I've seen ear rings respond to this. Also, an aluminum coated fire fighters jacket. It's subtle, it's real, but it's not on the list.
What I think would have made more sense is to ask the participant "What do they feel different from when they entered the room" and have the researchers tabulate the results in general catagories.
No difference, unless there are some resonance effects. I wonder if any of the participants brought their own cell phones. It wasn't mentioned in the test procedure. Table 6.1 shows the order of frequency testing, which looks symmetrical in both time and block order. Therefore, I don't think there would be a time effect (i.e. we're late and I want to get out of here). Perhaps the Agilent RF generator made different noises on 2100 MHz just outside the room. I can't think of anything better.