Hey wiki comics is advertising some pharmacy links you can find below. Sorry for any inconvenience. Hope you can understand... Links are below: Tadalafil Citrate | generic cialis 10 mg | tadalafil citrate 10mg | tadalafil citrate 5mg | generic cialis 40 mg |

Business - Written by on Wednesday, June 30, 2010 15:36 - 5 Comments

Tim Bevins
The usefulness and validity of surveys and data

I recently had a “discussion” with a friend of my wife’s who argued, in essence, that data from surveys are invalid unless the data are perfect. By perfect, she appeared to mean that no “statistically valid” survey exists because it cannot represent every potential view or experience of every person in a target population.

Specifically, I was talking about surveys about Gen Ys and the conclusions being drawn, by us and by others, about them – generalizations really, not conclusions, e.g., that Gen Ys are closer to their parents than other generations; more altruistic/less altruistic; more resistant to authority; more interested in finding their own way; and so on. My wife’s friend countered that the survey could not accurately represent some people who probably are not surveyable, e.g., the Gen Y children of illegal migrant workers in Missouri, so no conclusions about Gen Y or even generalizations about them are valid because they are not likely to reflect the views of or apply to this subset of Gen Ys who are likely not going to be reached by a survey.

My understanding of her position is that unless 100% of the people you are targeting with a survey are included, the survey data have no meaning. Okay, I admit to losing it here in that discussion and just withdrawing. There was no argument I could muster against “perfect or nothing.”

I won’t argue that surveys reveal “The Truth” (if that exists), but data from statistically valid and well constructed surveys do reveal a reality that may or may not apply to everyone but that does exist.

I also admit that it was hard for me to argue my position because I am already suspicious of how accurately Gen Y surveys actually reflect the Gen Y population. I tend to rely on anecdote – which has little validity – as a touchstone to what the surveys reveal. I have two Gen Y children so I tend to ask myself whether survey conclusions about Gen Ys work for them. Sometimes yes, sometimes no. My children are more open by far than I was about their lives, but they do not reveal everything, nor do I want them to.

All this is really a lead-in to saying that survey data – about Gen Ys or anything else – at the very least provide something to think about and consider, even if the surveys and the data are always imperfect. To simply declare that we can know nothing unless we talk to everyone is too easy and lets everyone of us off the hook when it comes to digging into the why of everything. That said, I tend to validate survey data by the organization that does the work; like everything else, some are just much better than others at this. We at nGenera will continue to survey because we are curious.



5 Comments

You can follow any responses to this entry through the RSS 2.0 feed. Responses are currently closed, but you can trackback from your own site.

A stats guy
Jun 30, 2010 16:51

Your friend doesn’t understand statistics at all. Can you tell her to stop slandering my goddamn profession due to her utter ignorance. Stats never used to be about more than 1000 samples from a population. It was rarely about surveying an entire population. Most of the stats you do is the based on samples (fair or not) the fact is that is all the statisticians did back in the days.

Grr, that’s very annoying. You don’t invoke the name of stats to ignore what it is about, randomness, sampling and variability!

If stats was about completely population measures we’d just call it measurement!

Tim
Jul 1, 2010 10:03

Thanks, stats guy. I will keep your post in Word version to remember how to respond. I appreciate your passion about stats and data. It’s a learning experience and that’s how I see it.

Interesting History Behind Folding Rocking Chairs | Chinese Table and Chair
Jul 4, 2010 12:24

[...] Wikinomics – The usefulness and validity of surveys and data [...]

Bruce Harpham
Jul 6, 2010 23:07

Even noisy data of uneven quality can be very useful. In the June 2010 issue of WIRED, Thomas Goetz writes about Surgey Brin’s efforts to promote a new model of medical research that starts with very large volumes of data. Rather than the traditional, highly controlled clinical model of collecting data from patient, Brin’s approach lets patients self-report data and then go from there. The article, Sergey’s Search: Can Google’s Cofounder Find a Cure for Parkinson’s Before He Gets the Disease, is a great read.

The more general point that one ought only to seek out perfect information is a very limiting notion. If that concept was widely adopted, the only kind of advances we could make would be in pure mathematics (and possibly some aspects of physics). Properly conducted surveys can be very informative even if they have flaws.

Chris Keer
Oct 17, 2010 10:26

Anyone know any good paid survey sites that actually pay?

Now available in paperback!
Don Tapscott and Anthony D. William's latest collaboration, Macrowikinomics: New Solutions for a Connected Planet. Learn more.

Business - Oct 5, 2010 12:00 - 0 Comments

DRM and us

More In Business


Entertainment - Aug 3, 2010 13:14 - 2 Comments

Want to see the future? Look to the games

More In Entertainment


Society - Aug 6, 2010 8:19 - 4 Comments

The Empire strikes a light

More In Society