Sheldon Adelsons anti-online gaming lobby group, the Coalition to Stop Internet Gambling (CSIG), recently touted a poll on Pennsylvanias appetite for online gaming expansion. As is normally the case with polling data, the wording of the questions garnered precisely the results CSIG was hoping for. The poll in question was conducted by Harper Polling between April 22 and April 27, with 513 registered Pennsylvania voters polled via landline and cell phones. Here is why this poll should be filed in the grain of salt file. A little background on Harper Polling Harper Polling is a right-wing polling company founded by Brock McCleary. McCleary currently serves as Harper Pollings President. Prior to Harper Polling, McCleary had most recently worked as the Polling Director and Deputy Executive Director of the National Republican Congressional Committee during the 2012 election cycle. That being said, skepticism of Harpers polling results goes beyond its partisanship. In an article from May of 2014, New York Times prediction guru Nate Cohn called Harper Polling a newcomer to the land of cheap, partisan, automatic polling, and said of its methodologies in previous polls, Harper underrepresented urban voters inconsistently weight for age. Cohn said of Harper (and similar polling outfits on both sides of the political spectrum), They may not be reliable for precise measurements of public opinion. FiveThirtyEight.coms Nate Silver grades Harper Polling as a C+ in his pollster ratings. Results of PA online gambling poll were expected Given that Harper Polling doesnt have the confidence of the experts, the results from the recent online gaming poll shouldnt be overly surprising and fall within an expected range on this issue. A 2013 poll by Quinnipiac had opposition to online gaming expansion at 62% in the same poll, a full 70% of Pennsylvanians were opposed to further land-based expansion as well. Gambling expansion issues have always polled poorly historically, particularly when the poll was commissioned by an anti-gaming group expecting certain results. This illustrates why the crafting of the language is so critical in polling. If youre curious as to how biased the poll was, simply scroll down to the final page of the results and read the list of messages Harpers pollsters asked the respondents. These are almost word-for-word the talking points of CSIG and anti-gambling crusaders. This polling ploy is clearly evident when we consider Harpers first question about legalizing online gaming in Pennsylvania showed 73% of respondents opposed, but the same question posed just two questions later saw opposition rise to 83%. What changed 10% of respondents minds in the span of two questions? The answer is Question 2 of the poll which reads: First, notice scenario two mentions key problems and potential abuses but scenario one doesnt mention key benefits or safeguards. This careful construction would lead the average citizen (who has scant knowledge of online gaming) to the impression that the risks outweigh the rewards. Scenario one paints online and land-based gaming as the same; scenario two depicts them as different and sneaks in a warning about potential issues. Second, touting the results as 68% find online gaming different than land-based gaming when respondents were instructed to choose the scenario that was closest to their opinion is disingenuous. Particularly when just two scenarios were supplied. Many people likely fall into a gray area between the two. Finally, had Harper added a few lines to scenario one, detailing the revenue and consumer protections the regulation of online gambling would introduce, the swing may have been ten points in the opposite direction. Takeaway Harper Polling designed this poll with a single objective in mind: To obtain the desired results for whomever commissioned and paid for the poll.