The Third-variable problem refers to the possibility that two (2) variables appear to be related when, in fact, they are both influenced by a third variable that causes them to vary together. Moreover, it is the possibility that a correlation between variables A and B is due to the influence of an unknown third variable rather than to a causal relationship between A and B.
With other words the Third variable problem refers to the possibility that variables not measured in a study are the real cause of the relationship between the variables measured in the study; not knowing from a correlational observation whether a change in one variable caused a change in a second variable or whether a third variable caused a change in both.