Questions about groups of people can be vague to fallacious -- the fallacy being that a group can be treated as an individual...

Consider: Are religious people happier?

What does this mean: Who is happier than who?

Happier on average? What is the "average" happiness of a group? Does it make sense to average happiness? If a few people are elated, that balances out one who is tortured? Does it even make sense to average income of individuals to say things about a group? Pareto had some ideas opposing such notions.

Happier than similar unreligious people? This is problematic, when are two people otherwise similar and apt for comparison?

Could it mean that if you pull out a person from the control group and the test group at random, that they would agree, or even a 3rd party could determine, who is happier?

Happier than they were before they found religion? What about religious people who were indoctrinated from birth? In the first case, restricting the sample, and knowing that those who made a 'rational' choice weighed benefits (social, etc...) against costs (tithing...), would this be a surprise?

Treating groups as anthropomorphic is either a 'legal fiction', or done for convenience. The convenience can lure one into fallacy. We can say that Google Inc. dislikes spam, or that the United States has declared war on drugs and terrorism. These usages are ordinary, and when done correctly mean that a named body adopted a policy or made an announcement of some kind through its official channels. To say that Americans are "tired of corruption" sounds safe (who wouldn't oppose corruption?), but isn't (some Americans might argue as "lesser evil" to something else). It is a generalization, and a few steps from a prejudice. And great for endless discussion, but not so great for finding correct answers.

Should we close questions about groups of people that involve an over-generalization?

2 Answers 2


Your are making two points in your question:

  1. The question is vague, which is bad.
  2. The question is treating a group as an individual, which is bad.

The first one is a valid point. In fact, when question are too vague, we close them as "not a real question":

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form.

If you feel the question can be improved, then speak out. Leave a comment asking the user to improve the question or try to persuade other users that it is too vague to be answered.

I can certainly agree that the question should be formulated as "Are religious happier than Group X?"

As to the second point, I don't think that it is true. It is perfectly acceptable to say "Canadians are wealthier than Kenyans." It is implied that we are talking on average. There might be Kenyans richer than some Canadians, but on average Canadians are wealthier. To insist for the addition of the word "average" is pedantry. It simply isn't necessary to understand what is being asked.


There are methods for measuring happiness in social science. I don't offhand know how good or reliable they are (and I don't know of a good way to test that), but it enables social scientists to correlate happiness with income.

So, for questions about relative happiness, we'd probably have to go with that.

For a social scientist, the question "Are religious people happier?" would be legitimate. It would be understood as happier than non-religious people, and it would refer to happiness-measuring techniques applied to different groups, controlling for as many other variables as possible.

I'd be way out of my depth trying to judge if such a study were properly conducted, but if properly done it would be meaningful.

  • 1
    Agreed, with the small caveat that subjective states of mind like "happy" or "in pain", are notoriously hard to measure because of bias, placebo and nocebo, etc. The studies are generally much less accurate than one would think.
    – Sklivvz
    Apr 12, 2011 at 6:31

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .