The average person (and to be fair, most psychologists) thinks of intelligence as the innate, fundamental characteristic of a person to think across all cognitive areas. However, this concept is not easily falsifiable and therefore arguably exists outside the realm of science.

For example, say I wanted to come up with a concept called "sportsness" which is the ability to be good at sports. I could test a bunch of people in a battery of sports-related tasks, and I'd probably get a nice bell curve where some people have high sportsness across all tasks and others have low sportsness across all tasks.

But does that prove the existence of sportsness? Or did I just measure a spurious correlation caused by the fact that some people are just more likely to be playing many different sports than others, or that some body types may lead to being better at sports related tasks, or some people are just better at handling the pressure of athletic performance tests, or some combination thereof? Of course most would say the latter, but then maybe some would defend the concept of sportsness by saying sportsness is just an emergent property of those things or something like that. But then is sportsness useful as a concept at all? You get the idea.