The only valid conclusion to be drawn out of this 'study' is that push polling works. If you get people to affirm a view and then you ask then a logically relat

The only valid conclusion to be drawn out of this 'study' is that push polling works. If you get people to affirm a view and then you ask then a logically related question they are likely to bend the answer to be logically consistent with the first statement. For them to go on and try correlate these results with a 'past study' stretches things beyond dubious.