Pages

Sunday, January 23, 2011

Blog Polls Are Crap!

I have recently returned to posting "Blog Polls" every week on this site.  I have had some reluctance in doing it because they are not really scientific polls.  It is really misleading to call them polls because it implies that they are random and science based.  They are nothing of the sort. I have had some Tweets from folks I respect who have called these "blog polls" crap and I agree...especially if they are represented or understood to be science based random and reliable polls.


WHY BLOG AND MEDIA "POLLS" ARE CRAP
I know the main stream media including major newspapers, radio stations and television websites often to the same misleading techniques as "blog polls."  That hardly justifies the activity and does nothing to absolve the misleading label of "poll."  These mainstream media "polls" are not scientific or reliable either.  Like those on blogs they are at best a relatively unreliable representation of what some readers might think.  We have no idea of the nature and composition of the readership or the respondents in either case.

These "polls" can even be hijacked by groups who what to create an impression of reality that is also misleading by pushing one group or another to flood the "poll" with answers that serves their preferences or purposes. The recent question of who should hold the balance of power if we have a minority Alberta government is perhaps an example of a push by a group to influence results.  I don't know but have suspicions.  Many of my readers are Alberta Party members and many more are Alberta Party curious so it is no surprise that this new party is the dominant choice.  I note a key strategist of the Alberta NDP recently criticized my blog polls Twitter and all of a sudden there is a surge in NDP support for holding the balance of power.  I think this is an example of starting the conversation so long as people see that the survey results are only useful for that purpose and not conclusive of anything.

This is just a signal to readers to use their critical thinking skills and read these surveys like they would horoscopes.  Treat them with the same degree of authenticity.  These so-called "polls" are to random sampling opinion surveys as Dr. Phil is to psychology...infotainment at worst and conversation starters at best.

SCIENCE BASED POLLING UNDER PRESSURE TOO
Opinion polling as a science has come under serious credibility pressures recently too.  This is because the tried and true techniques used in the past to generate a random representative sampling has reliability problems with the rise of cell phones as virtually the exclusive connectivity link of a younger generation and the difficulty to connect to them skews the data.  The move towards self-selecting volunteers to register with pollsters who allegedly represent a demographic or a region is suspect too because the more indifferent or disengaged opinions are not likely to be canvasses.  Then we have caller ID that allows people to filter out unwanted calls or unknown callers.  That again undermines true randomness.  Then we have the fact that as many as 20 calls have to be made before someone will take the call and invest the time to respond to a phone survey.  How randomly representative is the data collected with that kind of randomness?  There are "solutions" but they are not perfect either.  But that is another issue for another post some other time.

DOES HARPER HATE EVIDENCE MORE THAN LIBERALS?
With the Harper government destroying the Canada Census we will not longer have a randomly selected scientifically reliable source of crucial information on Canadians after 2006.  The ignorance of that policy decision will ensure that public policy design in the future will be a crap shoot that will inevitably result in crap public policy.  I believe that is the ultimate political goal here.  The fundamentalist anti-intellectual underpinnings of the Reform roots of the Harper government want to design failure into government so it can be replaced by Darwinian market forces in all cases.  I mention this to show that even quality scientific polling is being undermined by a political ideology that says it is OK for faith to trump facts.  Evidence is tough to rebut so the Reformatory Harper government passes policy to ensure we don't have facts in the first place.  That is even more dangerous to democracy than silly unscientific "blog polls."

IT'S ABOUT STARTING THE CONVERSATIONS
I think the questions posed in a "blog poll" will only be conversation starters in and amongst the readers of that blog in comments, social media and off-line IRL (in real life).  There is no reliable value to be attributed to the responses and folks have to know that.  I will continue to put questions to my readers for response but I will not call them "Blog Polls" any more.  I will call them "Burning Questions" from now.  I hope these Burning Questions continue to serve the purpose to engage citizens in the political culture of our times.  I hope they trigger real conversations in communities, between friends and amongst co-workers and even within families to help focus attention on the political issues and public policy concerns that are shaping our times.

If that is the case, I believe they are worth keeping.  If you have a Burning Question you want me to pose, email it to me.