Research has uncovered a semantic cue effect for short inconsistent word problems; the presence of functionally related objects makes them easier for pupils than the presence of categorically related ones. This study builds on this result by investigating other types of problems. Variants of problems differing in a semantic cue were solved by equally-abled groups of pupils. Item Response Theory was used to determine their latent ability. An independent two-sample t-test was used to determine differences in the difficulty of word problems. The above semantic cue effect was confirmed and the findings distinguished two ways a semantic cue manifests itself (as a reminder or as a helper) in other types of word problems. The study determined the variables supporting the manifestation of the semantic cue (the order of data and the presence of a complicating variable). More research is needed to elaborate the model.