Shepard (1997)

"It is possible to appraise the construct validity of a test interpretation without considering test use so long as no use is intended. As soon as a use is specified, then the validity investigation, including analysis of effects, must be tailored to the particular application."
Shepard, L. A. (1997). The Centrality of Test Use and Consequences for Test Validity. Educational Measurement: Issues and Practice, 16(2), 5-13.