I had a look at the start of this survey, and it seemed to justify my long-held view that "social sciences" are really pseudo-sciences. It wants to know, for example, whether I decide issues with reason or intuition. The obvious answer is that it depends on the issue!
Questionnaires are but one tool that is at the disposal of "social siences" as you call it. And while it is true, that some questions can be understood ambiguously, this can also be tested for. Tendencies how respondents in certain subgroups answer can be identified and used to create tools that can better discriminate between those subgroups.
And that is exactly what usually happens. You get to see 20 Questions, but these 20 Questions used to be 50 or 60 Questions that have been administered to a thousand or more people. And only the items that can differentiate between whether someone is likely a member of that subgroup are kept in the final scale.
Additionally, as I said, those are self-report items, which CAN be influenced by people who fill out the questionnaire, because they know what the study is about. This is by no means the only way to get your data. There also many studies where you are you divide all participants into groups. One is the control group, where you establish the baseline. And the other group or groups get some kind of manipulation, where you try to trigger a response that is different.
For example: There is a brilliant study (Strack, F., Martin, L. L., & Stepper, S. (1988). Inhibiting and facilitating conditions of the human smile: a nonobtrusive test of the facial feedback hypothesis.
Journal of personality and social psychology,
54(5), 768. ) where the experimenter let the participants rate how funny cartoons are. The clue being that they had to do some easy drawing tasks. In the control case with their non-dominant hand and in the two experimental cases they had to hold the pen either with their lips, thus making a more frowning face, and in the other case with their teeth, thus making a smile.
When after the tasks they had to rate how funny cartoons are, people in the lips condition rated them to be less funny as the control group, whereas people in the teeth condition rated them to be funnier than the control group did. Thus raising the question: Do we smile because we are happy, or are we happy because we smile?