Calibrating Validation Samples when Accounting for Measurement Error in Intervention Studies