In the name of science

In my opinion, as long as the subjects give consent and if he knows the possible consequences the research is still in line. Likewise, if the subject gives consent but does not know the possible consequences the research is out of line.

Then... the results wouldn't be accurate. There have to be certain areas that are blinded sometimes in experiments. I learned in my Psych class that sometimes even the researchers can't know what the experiment is or the experiment would fail. (Because of certain selective biases.)

Look at my own personal experiment on SP: Random Poll
Basically, I gave people four choices to test what people would choose the most often. I conducted that experiment because according to statistics, the probability would be 25% in all categories. However, my hypothesis was that it would be polarized towards one of the choices, so I wanted to confirm that. (It's also why I'm pissed at the mods on SW for moving that to LoLMS where nobody would see it.)

In a way, my experiment worked:

77 people took the vote

17 people voted A 22.08%
14 people voted B 18.18%
32 people voted C 41.56%
14 people voted D 18.18%

However, it also failed, because I failed to determine the reason people chose C most often. (due to lack of replies).

You can definitely see from this though, that if I told people beforehand what the experiment is, then the results would have been skewed.