Correct

I suppose one could argue both ways on this one.
The use of multiple observers helps to improve
agreement or reliability among the observers, thus
improving the strength of the measures used. Construct
validity is improved when the theoretical predictions of a
theory are validated by empirical data in experimentation.
To the degree that multiple observers contribute to such
validation then, yes, they do help to develop construct validity.
But the best way is to have empirical support of theoretical
predictions. So this question is more false than true.