If anyone really wants to find out if there is such a thing as electrosensitivity, that can be established with a double-blind test. Suppose someone says that they can sense the presence of a wi-fi router. Conceal such a router in a remote location free of other electronic devices that might be blamed, with a remotely operated power switch. Researcher A introduces a subject to the room and explains that the router may be switched on shortly, or not, depending on a coin toss that a colleague will shortly perform. An hour or whatever later, researcher A asks the subject whether s/he thought the router was turned on. repeat, as often as time or the subject permits. It's important that the subject and the colleague B never see each other, so there are no subconscious voice tones or body language cues passed on. Similarly, that A does not know or even suspect the on/off status at any time before the subject leaves the test. It's best if A is not in the presence of B at all during the tests.
If electrosensitivity exists, the subject(s) will be right significantly more than 50% of the time. If it doesn't, they won't.
Personally I'd put the router in a dark but accessible place with a photocell or a piece of unexposed photographic paper, to catch cheats. (Not a webcam, given what is being tested!) Methinks statistical analysis will reveal far more people claiming electrosensitivity in countries with generous social security systems ....