"Sociological imagination is a sociological term, coined by the American sociologist C. Wright Mills in 1959, describing the process of linking individual experience with social institutions and one's place in history." (Wikipedia)
Theres's some ambiguity to this definition (which makes it a rather bad definition IMHO).
Is it
"process of linking
individual experience with
social institutions and one's place in history"
or perhaps
"process of linking
individual experience with social institutions and
one's place in history."
To be honest though, neither make much sense to me, but perhaps the latter is somewhat more plausible. The former doesn't specify
what kind of experience is linked with social institutions and one's place in history; the latter specifies that it's experiences with social institutions that are linked to one's place in history.
It's a bad definition no matter how you try and dissect it, but reading the wiki article about it, I understand it like this:
People have their image of the society defined by their experiences regarding living in said society.
Experience. A person lives in some slum in a country. Another person lives comfortably in their own home, with every basic need satisfied. These individuals will have vastly different image of the society, even though in the broad sense they live in the exact same society... which means that our image of society is at least partially based on what we see it to be.
I can't really think of any other meaning for this term. Furthermore I question the sensibility of using such a term altogether, because all it does is confuse me by mentioning imagination, when it isn't imagination, it's just subjectivity of opinions depending on different experiences.

Defining whether or not sociology sucks is questionable since it isn't an exact science. On my personal relevance scale I would grade it about on the same level as psychohistory or something like that. Whether it sucks or not is hard to say. If it's considered as psychology applied to masses, but unlike normally in things involving probability and chance, masses are even harder to predict than individuals. What I mean is that if you throw a dice once you'll get a number from one to six, but if you throw two die, the likelyhood of getting a seven is higher than any other combined outcome (one out of six throws). But with human masses you can't apply psychology in this fashion. It's much more unprecise. And the fine gentlemen with their very fine hats can argue and define human mob behaviour all they want, it'll never make sense.