"PSR published the results of the exit poll after the closing of the election centers. Our results gave the advantage to Fateh (42% compared to 35% for Hamas). The actual results showed Hamas gaining the advantage with 44% versus Fateh's 41%. In a previous statement, PSR expressed the believe that the error in the prediction was the result of the large number of rejections on the day of elections. A total of 3560, or 17%, of the voters in the sample, refused to be interviewed on the day of election. The rejection, PSR believes was the result of deliberate incitement against pollsters . In order to overcome the problem, we have re-weighted the data to reflect the actual outcome of the elections for the lists in the districts and at the national level. The following results and analysis are based on the re-weighted data which reflect the actual outcome of the elections with the advantage going to Hamas rather than Fateh."

http://www.pcpsr.org/survey/polls/2006/exitplcfulljan06e.html

It is possible that the high numbers of "rejections" were from people who didn't want to admit they switched votes for Hamas. The survey I am referring to I don't believe had that problem. I don't have the time to hunt it up, however.