Construct validity is the extent to which the measurements used, often questionnaires, actually test the hypothesis or theory they are measuring. Construct validity should demonstrate that scores on a particular test do predict the theoretical trait it says it does.
There are two subsets of construct validity: convergent construct validity and discriminant construct validity. Convergent construct validity tests the relationship between the construct and a similar measure; this shows that constructs which are meant to be related are related. Discriminant construct validity tests the relationships between the construct and an unrelated measure; this shows that the constructs are not related to something unexpected. In order to have good construct validity one must have a strong relationship with convergent construct validity and no relationship for discriminant construct validity.