Techdirt. Stories filed under "surveys"

If You Want Two-Thirds Of Americans To Agree That Violent Video Games Are More Dangerous Than Guns, All You Have To Do Is Ask The Right Americans

Tim Cushing — Tue, 12 Feb 2013 08:42:54 PST

If you need some handy numbers to argue that violent video games are more dangerous than guns, Public Policy Polling has just delivered a gift-wrapped poll result especially for you. In the middle of a long poll attempting to suss out potential front runners for the 2016 elections, PPP decided to toss in a question comparing violent video games and guns.

There you have it. Violent video games are a "bigger safety threat" than guns, according to two out of three respondents. Seems pretty open and shut. Everyone cross out the word "gun" on your pet piece of legislation and replace it with "video game!" The nation is saved!

Many of you may be reaching for your guns/lower jaw/commenting implement. Before we start firing off mouths/angry wall o' text screeds/bullets, let's have a look at the methodology.

PPP surveyed 800 voters nationally from January 31st to February 3rd. The margin of error for the survey is +/-3.5%. We oversampled 416 Democratic and 508 Republican primary voters with margins of error of +/-4.8% and +/-4.4% respectively. In Iowa between February 1st and 3rd we interviewed 313 Democratic and 326 Republican primary voters with margins of error of +/-5.5% and +/-5.4% respectively. This poll was not paid for or authorized by any campaign or political organization. PPP surveys are conducted through automated telephone interviews.

This certainly seems above board. So, why do the results seem so surprising? Well, maybe it's the prevailing demographics of those surveyed. As pointed out in the comments at Kotaku, there are two factors that skew the results.

1. 72% of the respondents are older than 45.

2. The "violent video games" question was only posed to Republican primary voters.

Now, this data pretty much agrees with the stereotypical view that older people and Republicans trust guns more than they trust violent video games. Sure, there are plenty of outliers along the way, but the Republican Party has generally fought gun control laws, and older people are generally more distrustful of recent technology. In fact, given a narrow enough demographic, you could probably get poll results that indicates that "most Americans" believe cellphones are a bigger safety threat than depleted uranium.

So, what PPP has actually done is gift-wrapped a set of numbers useful for preaching to the converted. All it does is add to noise that surrounds this heated topic. Considering there's nothing else resembling that question in the other several dozen pages, one wonders why the question appears at all. Truly bizarre.

PPP Release National 207 (PDF)PPP Release National 207 (Text)

Permalink | Comments | Email This Story

Evidence That UK Needs Mandatory Porn Filters? Informal Survey Done At One School

Glyn Moody — Wed, 5 Sep 2012 10:33:48 PDT

In the UK there is currently a campaign and associated petition from the organization "Safety Net: Protecting Innocence Online", which calls for mandatory Net filtering of pornography -- people would need to opt out of the system if they wanted to view this material. The justification -- of course -- is the usual "won't someone think of the children?" Here's the pitch:

Every day children and young people are accessing mainstream pornography on the internet, including the most hardcore, violent and abusive images. Evidence clearly shows pornography has a detrimental impact on children and young people including premature sexualisation, negative body image and unhealthy notions about relationships. This cannot be allowed to continue.

Nick Pickles from Big Brother Watch looked into what that "evidence" might be, and found something rather interesting: One of the key statistics relied upon by the campaign is that "1 in 3 10 year olds have seen pornography online". They do recognise it was published in Psychologies Magazine in 2010, but the appearance is given that this is a serious statistic. It’s also used in their 'Key Facts' briefing.

When you dig a little deeper however, that definitely isn't the case. The full section in the magazine reads:
"We've had plenty of letters from concerned readers on this very topic, and when we decided to canvass the views of 14- to 16-year-olds at a north London secondary school, the results took us by surprise.

Almost one-third first looked at sexual images online when they were aged 10 or younger."
So, the statistic -- […] at the heart of the petition's press release -- is based on one magazine's anecdotal research at a single school.
Actually, it's even more ridiculous than that. That "statistic" states "[a]lmost one-third first looked at sexual images online when they were aged 10 or younger." But as is well known, UK newspaper titles like Rupert Murdoch's "The Sun" carry "sexual images" -- pictures of topless women -- every day. Given the large circulation of the those titles, it's far more likely that children will have seen "sexual images" there, rather than online, and that their attitudes to women will have been harmed more by this kind of relentless objectification than by isolated images they come across on the Internet. And yet strangely no one is calling for Rupert Murdoch's newspapers to be censored.

It's a classic demonization of the Internet that ignores the broader context, and is based on the flimsiest of pretexts. Worryingly, the UK government is sending out clear signals that it supports this campaign regardless. It's currently conducting a consultation on "Parental Internet controls", which closes on September 6. It's extremely poorly worded and clearly biased in favour of the idea of making blanket censorship the default.

If such Net blocks are brought in, legitimate sites will inevitably be blocked by mistake, but it's not so clear that the objectives of protecting children will be achieved. With blocks in place, parents may be lulled into a false sense of security, and so fail to supervise their children's online activities adequately, which will leave the latter exposed to greater not lesser risks. Meanwhile, young people will find ways to circumvent the blocks -- or just buy a copy of "The Sun".

Follow me @glynmoody on Twitter or identi.ca, and on Google+

Permalink | Comments | Email This Story

How To Create A Survey That Makes A Problem Seem Bigger Than It Is: 'Do You Know Anyone...'

Mike Masnick — Wed, 16 Mar 2011 21:13:17 PDT

There are all sorts of ways to lie with stats and surveys. One of the more obnoxious ones is used in a new "study" about "mobile etiquette" which claims the problem is reaching epidemic levels mainly based on an online survey. The accuracy of online surveys are already suspect enough, but this form of survey is done in a sneaky way that makes the problem look potentially much worse than it is:

Nine out of ten people who responded to the online survey said they have seen people misuse their mobile devices, according to the survey.

Right. It didn't ask people about their own activities, but just if they've seen other people. That's almost meaningless. Taken to an extreme version, say there was one (highly visible) mobile jackass, and 99 other people surveyed saw that one guy. Well, based on this survey, you could claim that mobile etiquette is in a terrible state.

I'm not arguing that there isn't necessarily a problem with how some people handle mobile etiquette. There could very well be just such a problem, but a study that sets out to investigate the issue with a survey like this one really tells you nothing.

Permalink | Comments | Email This Story

If You Ask The Question In A Certain Way, 61% Of Americans Say They Support An Internet Kill Switch

Mike Masnick — Thu, 28 Oct 2010 13:44:00 PDT

Slashdot points us to a story claiming that 61% of Americans "support an internet kill switch." Of course, this is a topic that's been hotly debated lately, with some attempt at passing laws that aren't really a "kill switch," but merely a coordinated way to reroute internet traffic in the event of some sort of "attack" (broadly defined) from a particular country.

That 61% number certainly sounded pretty high, and I was doubly skeptical when I read that the study came from Unisys, a security company who clearly stands to profit from greater "worries" about the still apparently bogus concept of "cyberwar." And, of course, people always point out that you can get a survey to say pretty much anything you want, depending on how you ask the question. So I went digging to see if I could find exactly what question Unisys (and its partner Lieberman Research Group) used to get this result. It took a bit of searching, but here's the question:

If there were clear evidence of a malicious cyber-security attack by a foreign government against our military, civilian government, electrical grid, financial systems, or other critical infrastructure, should the President have the authority to take control of or effectively shut down portions of the Internet to mitigate a crisis?

First of all, that's a big, big "if" right at the beginning there. Second, all of this assumes that an attack on the military, the government, the electrical grid, the financial system or other "critical infrastructure" could actually come via the internet. This isn't a reason to support an internet kill switch. It's a reason to get people to ask more reasonable questions, rather than broadbased scary questions, without highlighting the corresponding concerns, civil liberties issues and other worries. If you make any question "scary" enough, you can get people to agree with you, but that hardly means that people would actually want such a kill switch if they understood (a) the likelihood of such an attack, (b) what such a "kill switch" would actually mean, and (c) what alternatives there are.

In other words, this is pure propaganda from Unisys, rather than any bit of meaningful data.

Permalink | Comments | Email This Story

Fun With Stats: Do 80% Of Australians Really Support Gov't Censorship Of The Internet?

Mike Masnick — Wed, 24 Feb 2010 08:55:00 PST

Ah, the fun you can have with survey data. It's well known that if you get to control how survey questions are worded, you can get large groups of people to support almost anything. Take, for example, what's happening down in Australia with the fight over the government's latest plan (after many previous attempts) to force filters on the entire internet, requiring ISPs to block a secret list of sites the gov't doesn't like. A lot of people are up in arms about this blatant censorship. But supporters of the plan can eagerly point to a recent survey that got 80% of people to say they were in favor of "having a mandatory Government Internet filter that would automatically block all access, in Australia, to overseas websites containing material that is Refused Classification." 80%? That seems really surprising.

But, it shouldn't be if you saw how the question was asked.

SG points us to the news about how that particular question was prefaced by asking people to read the following "definition" of what kinds of content would be blocked:

child sexual abuse
bestiality
sexual violence
gratuitous, exploitative or offensive sexual fetishes; and
detailed instructions on or promotion of crime, violence or use of illegal drugs

You read that list -- especially given what it starts off with -- and who's going to say "no, I want to make sure that's available"? Not that many people. Obviously, they're going to say they support blocking it. But that doesn't really mean they support government censorship. It seems quite likely that what people want is for the government not to focus on useless and expensive censorship campaigns, but to focus that same effort on catching those who are actually breaking the law and stopping them from engaging in the illegal activities. In fact, the very same study found that 91% don't like the fact that the government will be censoring the internet via a secret list. Basically, people don't want government censorship, but that doesn't mean they support the things in the list. It's just that they would prefer the government do something real to stop it, rather than brushing it under a digital rug.

Permalink | Comments | Email This Story

BPI Survey Suggests Spotify Hasn't Magically Decreased Desire For Unauthorized Music Access

Mike Masnick — Fri, 18 Dec 2009 16:35:14 PST

There was some buzz earlier this year concerning reports that new streaming apps, like Spotify, somehow decreased unauthorized access to music. And yet, a new study from BPI suggests unauthorized access to music continues to grow, despite the rise of authorized services like Spotify. Now, there are some caveats. BPI isn't exactly known for being entirely accurate with data and these results are from an online survey. While you would think that fewer people would admit to unauthorized access in an online survey (people don't like to fess up), counteracting that is the fact that BPI has incentives to suggest the issue of piracy is a big deal, as it's pushing hard to force ISPs to kick people offline for file sharing. Still, what strikes me as interesting is that BPI still keeps insisting that this is a "problem," without any evidence that this is true. The only real "problem" is the failure of the record labels that BPI represents to adjust their business models. If they did that, there wouldn't be much of a problem at all. But, the labels don't want to do that. They want the government to rescue them and to pretend they can keep doing business they way they always did.

Permalink | Comments | Email This Story

Just Because People Say They'll Pay For Something, It Doesn't Mean They Will

Mike Masnick — Tue, 17 Nov 2009 20:56:17 PST

I've been ignoring this one, but people keep submitting it. BCG came out with a report over the weekend on a survey it did, claiming that about half of all people would pay for online news. It was amusing to see people react to this, as some reported it as "most won't pay for news" and others reported it as "oh my goodness, a lot of people will pay for news." Of course, the reality is that this is just a survey of what people say they'd pay for -- and history has shown that surveys are notoriously poor indicators in terms of getting people to accurately reflect what they will and will not buy. Besides, just a day later, a totally different survey claimed that 80% of people wouldn't pay for news online. The answer is that no one knows how many people would pay for content online, but I'd bet that the number is lower than what both of these surveys predict, and we'll see that soon enough.

Permalink | Comments | Email This Story

Why Was It Bad For The Chicago Tribune To Find Out What Readers Wanted?

Mike Masnick — Mon, 4 May 2009 04:31:15 PDT

There was a recent dustup in Chicago as news came out that the Chicago Tribune had been running reader surveys on certain news stories before they were published. This has upset a bunch of folks at the Tribune who claim that it somehow "breaks the bond between reporters and editors in a fundamental way." The Tribune has apologized and claimed the whole program was a mistake. But, for the life of me, I'm having trouble figuring out what's the problem. Actually finding out what your community wants? Getting feedback? Being more interactive and engaged with the community? Listening to them? Being open and recognizing a story is a living document? These are all things that any newspaper should be doing these days.

Permalink | Comments | Email This Story

Do Half Of Americans Really Want The Gov't To Regulate The Internet?

Mike Masnick — Thu, 26 Jun 2008 12:36:11 PDT

You can prove almost anything you want with surveys if you word them correctly. Remember a couple months ago, when we were discussing a study that suggested people in China were "happy" with the government's regulation of the internet? At the time, I noted:

Before people act surprised about this, take a step back and recognize that if you did the same study in the US, asking if the government should be protecting children from "bad stuff" online, many Americans would naturally say yes.

And, now it looks like a "public opinion" firm is proving that correct, leading to the publicity-generating headline suggesting that half of all Americans support regulating content online the same way content is regulated on TV or radio -- with 73% of all people saying that it should be illegal to be a jerk online. But is that really representative of what people think? Not really. At least the firm also reveals the actual questions -- from which it's easy to see why the answers came out as they did.

The survey starts out talking about the infamous Megan Meier-Lori Drew case, providing very little of the actual context of what happened, but instead suggesting that Drew sent cruel messages to a young girl to make her commit suicide. That's not accurate at all, according to most reports. While Lori Drew did set up the fake MySpace page, she did it to keep tabs on what Meier was saying about her daughter, not to harass her. And, the harassing "message" was sent by another young girl who thought it was a good way to get Megan to not want to speak to the fake profile again -- rather than to push her to commit suicide. But, none of that context is explained. Instead, survey takers are told that Drew set up a fake MySpace page with which to harass a girl, which resulted in the girl's suicide. They're then asked if they think that should be a crime -- to which many people obviously respond yes.

Immediately following this, they ask people if the FCC should regulate content on the internet, like it does on TV and radio. And, of course, since people are now in the mindset of thinking just how awful some content can be online, they're much more prone to say yes. In other words, the results are totally bogus. If the opinion firm, instead, went in the other direction and opened with questions about internet censorship, and spoke about how aggressive regulation of content online was preventing people from getting all sorts of information that could be useful to them, and then asked how they felt about FCC regulation of online content, I'd imagine a lot fewer would think it was okay. And, just for contrast's sake, we can point to another study from a few years ago that asked people about regulating content on TV. In that study over 90% said they didn't think content should be regulated on TV. That study is obviously questionable too, but if you combine both, you'd have 90% of people saying that TV shouldn't be regulated, but 50% saying that internet content should be. Something doesn't add up.

Permalink | Comments | Email This Story