<body><script type="text/javascript"> function setAttributeOnload(object, attribute, val) { if(window.addEventListener) { window.addEventListener('load', function(){ object[attribute] = val; }, false); } else { window.attachEvent('onload', function(){ object[attribute] = val; }); } } </script> <div id="navbar-iframe-container"></div> <script type="text/javascript" src="https://apis.google.com/js/plusone.js"></script> <script type="text/javascript"> gapi.load("gapi.iframes:gapi.iframes.style.bubble", function() { if (gapi.iframes && gapi.iframes.getContext) { gapi.iframes.getContext().openChild({ url: 'https://www.blogger.com/navbar.g?targetBlogID\x3d11857647\x26blogName\x3dM.K.+BRAATEN\x26publishMode\x3dPUBLISH_MODE_BLOGSPOT\x26navbarType\x3dBLUE\x26layoutType\x3dCLASSIC\x26searchRoot\x3dhttp://mkbraaten.blogspot.com/search\x26blogLocale\x3den_US\x26v\x3d2\x26homepageUrl\x3dhttp://mkbraaten.blogspot.com/\x26vt\x3d-662957104341791521', where: document.getElementById("navbar-iframe-container"), id: "navbar-iframe" }); } }); </script>

SES POLL is misleading

Many of todays papers are saying that the Liberals are bouncing back from a their low points in public perception based on this SES research poll that was released yesterday. In my opinion, this is a seriously misleading question. And using elementary econometrics theorem I will attempt to explain why its misleading.

Firstly, the question is an indirect, stated preference type of question and it could produce a wide array of opinions. Stated preference data can be flawed because it produces a moral hazard to the person responding to the question.

For example, if I asked you:

"How much would you be willing to pay to save kids in war torn countries?"
a) 100
b)50
c)0

Most likely you, as a moral person, would choose the highest amount you could possibly think of and choose 100$.

However if the question asked:

"How much would you pay right now to save the kids in war torn countries?"
a) 100
b)50
c)0

What would you choose? Well, this decision is more complicated, in that you actually have to pay right now. Therefore, in this decision there will several factors that will influence your decision such as if you just got paid, is your rent due, etc; these moral hazards directly affect your decision, wheras in the previous question, they dont, because its more hypothetical then practical. As a result, your true preference is revealed, because the factors affecting your decision are calculated into your response. However your stated preference is in your answer to the first question.

Stated preference responses are based mainly on what you would like to do, or what currently is going on in a particular instance (ie what current political party is in power).

So, getting back to the SES question. This is what they asked the respondents:

For those parties you would consider voting for federally could you please rank your top two current local preferences?

Like in the aforementioned example, the natural response to the question of helping war torn children was $100 because it is the most obvious choice; its a stated preference. However, in this SES question, on average, what do you figure the 'natural' and obvious choice would be? Most likely to the average person it would be the governing party. People's true natural preference, wouldn’t be the Conservatives because they aren’t in power. They are the opposition, the un-natural variable in this question. Therefore, in a stated preference question, they are the losers, because in current reality, they are not the governing party.

The question doesn’t directly tell the person reading/answering the question to rank them one and two, just for them to rank their top two parties. This is where their natural preference causes them to likely choose the Liberal party as #1 and the Tories as #2, this is also their realistic ranking in the current power structure in Ottawa so again, it’s a natural choice to choose the Liberals and Tories as 1 and 2, respectively.

A better question would be: As of today, which political party would you vote for?

However, this SES poll is an another example of misleading statistics. What is even more stupid, is that all the newspapers picked this poll up and are now using it to prop of the Liberal party of Canada.

M.K. Braaten
« Home | Next »
| Next »
| Next »
| Next »
| Next »
| Next »
| Next »
| Next »
| Next »
| Next »

5/06/2005 12:47:00 PM

Looks like the media has an aversion to polls showing the Conservatives ahead (the Pollara poll). How are we bloggers supposed to win the war against the main stream media?    



5/06/2005 05:41:00 PM

I have managed to get a lot of people interested in blogs for real news. I did this by personally emailing friends en-mass about interesting items in a paraphrased form and citing the links/authors.

I try to never miss an opportunity to SHOW how biased the media is through actual stories and so on.

Finally, I get on my soap box and make sure to tell everyone at any opportunity that if they're relying on MSM, they're getting spoonfed an agenda.

In this country we will get past MSM. It's already happening. I know people who now actively avoid CBC (one of the worst offenders).    



5/06/2005 09:44:00 PM

Let the MSM say what they want.

The only poll that counts is the one on election day. Just ask George Bush    



» Post a Comment


Listed on BlogShares