Wednesday, February 27, 2013

Surveys Can Be Flawed - Don't Freak Out

I'm working on my hand/eye coordination. Honest!
A January 2013 Harris Poll has been making its internet rounds this week, stating that 58% of American’s believe that violent video games do affect children.

Things to keep in mind: This was an online poll conducted through Harris Interactive, that has a limited reached versus, say People Magazine.  The poll was taken by 2,278 people believed to be over the age of 18. The age cannot be verified because this is an online poll. The press release issued by Harris Interactive mainly focuses on what I’ve stated in the first sentence. It kind of overlooks the other, positive things, which were found out: 

69% believe that video games can be good for children in helping develop hand/eye coordination and other skills.

73% strongly agree and 90% somewhat to strongly agree that Parents should be the chief regulators of what games their kids play.

56% believe there should be no government regulation on games and content ala the First Amendment.

But we really don’t see that on Fox News or other stations when the title from Harris Interactive says ‘Majority of Americans See Connection Between Video Games and Violent Behavior in Teens.’ Well gee. Thanks for being objective there guys. The rest of your findings are moot when you have a headline like that.

This isn’t just an opinion from someone who sides with video games, but as a scholar who recognizes the flaws in doing such surveys and studies. The pool of those who answered the survey was limited not just in number, but in ethnicity, religion, gender, and various cultural differences. And because it was done online, there is no sure-fire way to ensure that the people completing said survey are who they say they are. That 18 year old young white man who hates video games might actually be a 48 year old black female trying to skew the results.

There is also an inherit bias in surveys. No matter how hard you try to make it as open as possible, there is going to be an underlying current, even if you’re completely unaware of it. You want to know what people’s opinions are about a new product you’ve made, but you don’t want it to be a failure right out of the gate. So you tweak your questions to make your product sound awesome. Unintentionally, perhaps. But it’s a bias that dates all the way back to the Bobo Doll experiment regarding television violence.  It’s something that I’ve brought up several times. An inherit bias was placed the moment the research began and it tainted the results.

So take the survey findings with a grain of salt. The scholar in me knows that this study is not fact, but many others may not understand why. Until we can produce a survey that every US American can answer that has unbiased questions, we’ll never have true results.


Post a Comment

Thank you for taking the time to leave a comment.

We ask that you please do not include any offensive, sexist, or derogatory language - otherwise your comment will be removed.