How to Interpret and Manipulate a Political Poll

By Patricia L Johnson 

Looking at the following Press Release from Quinnipiac University you would think Barack Obama’s approval rating in Ohio (a major swing state) has dropped from 62% to 49% during the period from May 2009 to July 2009.  You would think that because that is basically what Quinnipiac is telling us.





President Barack Obama gets a lackluster 49 – 44 percent approval rating in Ohio, considered by many to be the most important swing state in a presidential election, according to a Quinnipiac University poll released today. This is President Obama’s lowest approval rating in any national or statewide Quinnipiac University poll since he was inaugurated and is down from 62 – 31 percent in a May 6 survey.

Unfortunately, a poll is not any different than an article, either online or in a newspaper, magazine.  While an article is swayed by the bias of the individual (s) writing it, poll numbers are swayed by the exact questions asked, the number of individuals polled, whether or not they are registered voters, and what political party, if any, they prefer and their location.

The July 7, 2009 press release from Quinnipiac is a prime example.  The first paragraph deals entirely with President Obama’s drop in the polls in Ohio.  How does the average reader know whether or not Obama’s rating did actually drop in that state?  Our first step in answering that question is to look at the actual poll.

The press release indicates a total of 1,259 Ohio voters [483 Democrats and 445 Republicans] were surveyed from June 26 – July 1, with a +/- 4.5 % margin of error on Democrats and a +/- 4.7% margin of error on Republicans [not sure why there would be a higher percentage of errors on Republican voters, so we’ll leave that question for another day].  The other 331 individuals making up the total of 1,259 were probably either Independents or chose not to state their political preference. 

The first question we should ask, on this poll, is why there were not an equal number of Democrats and Republicans polled.  If they only had 445 Republicans to poll; then they should have only polled 445 Democrats; not 483.  Another interesting fact to note on this poll is they are not indicating the actual number of voters in each category, only percentages, so we have no way of knowing whether or not their percentage calculations are correct.

The approval rating for President Obama stems from question number 24 on the Quinnipiac poll; the prior 23 questions all dealt with local politicians.  The fact the poll numbers for this particular question are split by Rep [Republican], Dem [Democrat], Ind [Independent], Men, Wom [Women] and WtBrnAgn Evnglcl [White Born Again Evangelical]; as well as into six geographic areas Cntrl [Central], NrthE [Northeast], NrthW [Northwest], SthE [Southeast], SthW [Southwest] and WstCnt [West Central] shows us how polls and poll numbers can be manipulated.

24. Do you approve or disapprove of the way Barack Obama is handling his job as president?

                     Tot    Rep    Dem    Ind    Men    Wom    Evnglcl

Approve 49% 19% 85% 38% 39% 57% 33% Disapprove 44 75 11 48 53 36 59 DK/NA 8 6 3 14 8 7 8

Cntrl NrthE NrthW SthE SthW WstCnt

Approve 50% 52% 55% 46% 36% 47% Disapprove 43 39 40 44 60 47 DK/NA 6 10 6 10 4 6

TREND: Do you approve or disapprove of the way Barack Obama is handling his job as president?

                     Jul 7   May 6   Mar 18  Feb 6
                     2009    2009    2009    2009

Approve 49 62 57 67 Disapprove 44 31 33 16 DK/NA 8 7 11 17

Just a cursory review of the poll will tell you that an extremely large percentage of Republican men, who are probably white, born again, Evangelicals, living in the [Southwest]  disapprove of the way Barack Obama is handling his job.

Do you want the trend for Obama’s rating to drop further with the next poll?  Not a problem, simply interview more white, BAE Republican men in the southwest. 

It’s time for the voters in this country to wake up when it comes to poll numbers.  The only way you can tell whether or not any given poll is an accurate representation of the political views of individual voters is to research the poll.

Since the Quinnipiac poll is also broken down by geographic region, we have to ask whether or not this particular poll question was strictly limited to Ohio voters. If it is breaking down the various regions in Ohio, that is certainly another way of biasing a poll.  You find the geographic area with the results you want and simply interview more voters in that particular geographic area.

The following statement made by Peter A. Brown of the polling institute about this poll is interesting.  

"The economy in Ohio is as bad as anywhere in America. These numbers indicate that for the first time voters have decided that President Barack Obama bears some responsibility for their problems," said Peter A. Brown, assistant director of the Quinnipiac University Polling Institute

After President Obama has been in office for a total of five months;  Brown is suggesting the current President is responsible for some of the economic problems in Ohio.  Unfortunately, it doesn’t work that way.  When the prior administration left office no one knew the total economic damage that was caused by their economic policies.  It will be many, many more months before all the data is brought to light and the damage assessed.

We, as American voters, have continually been brainwashed into believing this or that based on someone else’s opinion – don’t rely on someone else’s opinion, form your own based on the facts.  The only way you are going to be able to access the facts, and form an opinion, is to look at the actual data being discussed in the poll.

Please Note:  This article is not meant to insinuate the Quinnipiac poll is biased or manipulated, it is simply an example of what possibilities are available to pollsters.

Patricia L Johnson is a former special assignment writer/photographer and co-owner of the Articles and Answers news and information site.

This entry was posted in News and politics and tagged , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s