A hidden variable theory is a QM interpretation that postulates that there are particle properties that cannot be measured, or are unknown, but that, none-the-less affect the behavior of the particle. The notion that something cannot be measured, but has noticable effect, is of course, self-contradictory which is, really, what the problem with these interpretations generally comes down to in one way or another.
Generally, hidden variable interpretations are divided into local hidden variable interpretations where these unknown quantities are tied to a particular particle, and globabl hidden variable interpretations where the state is universal. Local hidden variable interpretations have been largely falsified by the combination of Bell's Theorem, and a series of experiements, the most famous of which is Aspect et al. The most well-known global hidden variable interpretatation is apparently Bohmian mechanics.
There are a number of good resources on the net for this topic.
N.B. Lurch's post is confusing:
It is the idea that the behavior of quantum particles is actually determined by even smaller particles (or something) that we've not yet detected. Probably the most popular example is quantum entanglement.
This is poorly phrased, it would probably be better to say something like:
Hidden variables are one of several interpretations used to explain quantum phenoma such as the behavior of entangled particles.
Yet, altering the state of one of the pair can have an effect on the other.
This is a common misconception or, at best very misleading. It is only true for an unusual notion of effect. (If you pull a shoe out of a shoe box, does noticing it is the right show cause the other shoe in the box to be the left one?)