This validity is concerned with the relation between theory and observation. If the relationship between cause and effect is causal, we must ensure two things: 1) that the treatment reflected the construct of the cause well and 2) that the outcome reflected the construct of the effect well. Threats to construct validity refer to the extent to which the experiment setting actually reflects the construct under study. For example, the number of courses taken at the university in computer science may be a poor measure of the subject’s experience in a programming language, i.e. has poor construct validity. The number of years of practical use may be a better measure, i.e. has better construct validity. Examples: Inadequate preoperational explicitation of constructs, Mono-operation bias, Mono-method bias, Confounding constructs and levels of constructs, Interaction of different treatments, Interaction of testing and treatment, Restricted generalizability across constructs, Hypothesis guessing, Evaluation apprehension, Experimenter expectancies.

Wohlin, C., Runeson, P., Höst, M., Ohlsson, M., Regnell, B., & Wesslén, A. (2000). Introduction to Experimentation in Software Engineering. [-]