Sorry, I maintain that processing data that is full of (known) systematic problems or data that is known to be insufficiently sensitive to detect the goal is a drain on limited resources.