As predicted, this ethics train wreck keep getting bigger.
There was a lot to wince about in Diane Sawyer’s “exclusive” interview two weeks ago with former Congresswoman Gabriella Giffords and her husband, Mark Kelly. The Arizona couple announced their intention to launch a non-profit organization dedicated to more effective anti-gun violence measures, concentrating, predictably, on the prominent features of the maniac’s rampage in Tucson that left Giffords with brain injuries that will impede her for a lifetime. Nothing to wince about regarding the effort, but Giffords’ diminished state—she can speak in only short burst of words, cannot see well out of one eye, and has difficulty walking—is tragic. It reminded me how unconscionable it was that she held her post in the Congress for more than a year when it should have been clear that her disabilities precluded her functioning as a Representative. The disturbing feeling also arose that Giffords, in her current pathetic condition, is now like the children President Obama used as window dressing for his gun-related Executive Orders announcement at the White House, an exploited figure of sentiment and public manipulation being used in the anti-gun wars. Her name was listed as the author of a first person op-ed in USA Today that contained sentences and perhaps thoughts that she cannot possibly compose. Diane Sawyer told us that she will be dragged into Congressional offices with her husband to seek support from her former colleagues, who will be forced, as Sawyer said, to say no “to her face.”
The most substantive wince, however, came from a statement of “fact” by Mark Kelly, who told Sawyer this:
“You know, how do we get to the point where 85 percent of the children in the world that are killed with guns are killed in the United States. That is a sobering statistic.”
Sobering, and obviously nonsense.
Sawyer, as a journalist, had an obligation to correct that misinformation, if not in the interview itself, then after, as she continued an ABC studio discussion about the gun control debate. It was a breach of journalist integrity and responsibility to aid and abet the launch of yet another piece of bad information into an important national policy debate. This is an example of how bias in the news media not only misinforms the public and warps policy, not to mention its leading to untrustworthy journalism from supposed professionals. Why didn’t such a jaw-dropping statistic raise a red flag for Sawyer? She was blinded by pure confirmation bias—an anti-gun advocate herself, driven by less knowledge than emotion, Sawyer somehow found the absurd statistic credible.
I didn’t, and it took me all of 42 seconds to find out what was wrong with it. A study by Erin Richardson and David Hemenway was published in 2010 in what was then called the Journal of Trauma, Injury, Infection, and Critical Care. Using 2003 data from the 23 richest nations (Australia, Austria, Canada, the Czech Republic, Finland, France, Germany, Hungary, Iceland, Italy, Japan, Luxembourg, Netherlands, New Zealand, Norway, Portugal, Slovakia, Spain, Sweden, England/Wales, North Ireland, Scotland, and the US) furnished by the World Health Organization, Richardson and Hemenway determined that 87% of all children aged 0 to 14 killed by firearms in these 23 countries were US children. 80% of all firearm deaths in this group occurred in the US, which had a rate of death by gunfire that was 19.5 times higher than the other 22 countries. Even this conclusion, based on data almost a decade old, cannot be stated fairly as current fact, but at least it is a real number.
But that number is not “85 percent of the children in the world that are killed with guns and killed in the United States,” not even close. A liberal blogger and gun control advocate who is also a scientist was asannoyed by the fake statistic as I was, and did some digging and calculating, expanding beyond the 22 countries in the Richardson and Hemenway study:
“…The study cited could not possibly have been carried out for all 196 countries in the world. However, the UN (specifically the UNODC) collects data on all homicides in a country, and the data was particularly well-filled-out in 2008 (only 18 countries had no data). I calculated overall homicide rates for the world vs the US using data from 2008. Whereas the US had a rate 6.9 times higher than the 22 other rich countries, it had a below average rate for the world. The US share of homicides in 2008 was 3.57%, whereas its share of population (surveyed) in 2008 was 4.66%. 98 countries on the list had higher homicide rates than we do. The US’s rate was 5.4 homicides per 100000 people. Contrast that with Burundi (21.7), Ethiopia (25.5), or Kenya (20.1). Ivory Coast (cote d’Ivoire) had 10,801 homicides in a population of only 19 million people, a rate of 56.9 per 100000. Jamaica: 59.5, El Salvador 51.9, Honduras, for Christ’s sake, 61.3 (4473 homicides in a population of only 7.3 million!).”
His conclusion?
“The statement Mark Kelly made is probably off by a factor of 20.”
A factor of 20! Yet this bogus statistic was broadcast as fact over ABC to a national audience, and eagerly put in the holster of thousands of anti-gun lobbyists, columnists, bloggers, “opinion journalists” (which is most of them these days) and activists for future use. They don’t care whether the statistic is accurate; they only care that it is so outrageous that it can influence flaccid minds. To them, the ends (rigid gun control) are justified by the means (making their case with false statistics). In a rational, competent, fair and responsible political system, defined as one occupied by rational, competent, fair and responsible elected officials, such a strategy would be futile. We do not have such a system, however, as illustrated by the fact that the thoroughly debunked, misleading and ancient fake statistic showing that “women are valued as worth only 77% of men in the same jobs” was still cited by prominent Democrats and journalists during the last campaign.
As “Reuben,” the blogging physicist says about the 2003 data that the US overall had a gun death rate 19.5 times higher than the other 22 countries, “Ok, good, stop there. Just say ‘compared to other developed countries, our laws are ridiculous and it’s literally killing our citizens’. Great, factual argument.”
But the anti-gun zealots don’t want to risk debating public policy on factual arguments. They are determined to do it with sad puppets like Gabby Giffords, threats, personal attacks, fear-mongering and grossly-hyped fake statistics. After all, these tactics have worked before.
What an embarrassing way to run a country.
_________________________________
Pointer: Althouse
Spark: ABC News,
Source: Idle Analytics, Huffington Post, USA Today
Ethics Alarms 
It’s weird. Perhaps I missed it in the passages that you cited. In one (the first one), it was children killed by guns. The second one, which was supposed to be a correction of the first, just talks about children homicides overall. Which seems to be a different statistic altogether. Does he quote something specifically about children’s homicides by gun somewhere else in the article?
No, you’re right—he’s extrapolating from just homicide data, because children by gun data doesn’t exist. His assumptions are valid, I think—if the nations outside those 22 rich nations have greatly larger instances of homicide, it is a fair assumption that the 85% figure is vastly reduced. Is it by a factor of 20? I have no idea. Whatever it is, the 85% figure is fantasy.
It’s not a fair assumption. Homicide rate does not necessarily correlate with gun death rate. Just look at the U.S. and Japan.
I suspect the 85% number is significantly off (what with all the African child soldiers), but if you’re going to hammer on invalid statistics, using invalid statistics isn’t the best way to do it.
How do invalid statistics get a foothold?
Problem 1: People use invalid statistics and they aren’t called on them.
Problem 2: When people are called on their invalid statistics, they don’t even update their posts where they use the invalid statistics.
Jack,
I agree with you on your point, and I think your assumption is likely correct but I am with tgt in regards to invalid statistics.
It was sloppy of the physicist not to flag his own switch and assumptions, that’s for sure. Regarding the 85%: if Gabby’s hubby states a 2003 stat covering 22 countries as a 2012 stat covering 196 countries, it is by definition a fake stat. There’s no alternative conclusion. The blogger was pulling out statistics to show how unlikely it was that the 85% could hold for the rest of the world. His 20X was an estimate; he needn’t have bothered. My issue: why is an obviously made up stat being broadcast on a news show and not challenged?
“My issue: why is an obviously made up stat being broadcast on a news show and not challenged?”
Because we are inundated with a dozen “facts” every day, people either don’t have the time or take the time to see if they are true. Pundits and advocates know this and use them to inflame passions because it is effective. Why no one calls them on it is because both sides do it, it is now who can tell the best story, not what’s the right thing to do. It seems like we have no power, we used to be able to vote with our pocket book, in this case who we will give our viewing time to but it doesn’t seem to matter anymore. There is a channel or show for everyone, each with their own lies so if I don’t like what one channel has to say I can switch to another. All of them seem to be making money from their niche so there is no reason for them to change. Advertisers either blanket all the channels or do a study to find out what channel their target market is watching. Until there is a financial risk to the show or channel they will not change, it is just that simple.
Why is an obviously invalid response to the stat being broadcast on your blog and not fixed? When you repeat the statistic, you are responsible for fixing the error. You can’t just push blame off on the creator of the statistic.
“The physicist” did state his switch. He didn’t emphasize the switch, but he did clearly state it.
To your main question of why the stat is not being challenged: why is Fox News still on the air at all? Why is the NRA allowed to speak into a microphone? Your points and others in the comments are on track: a widespread lack of journalistic integrity (including many on both sides) and a lack of people checking the ‘facts’ we are bombarded with every day. It is a complicated world, and most important issues cannot be distilled down to an 8th grade level.
From Rueben’s blog post: ‘the US had a rate 6.9 times higher than the 22 other rich countries [for total homicides]”. If I got my math correct, this works out to about 87% of all homicides in the rich countries happening in the US, a remarkably good match to the gun homicide stat of 80%. Therefore, it is not a huge jump to assume homicide stats and gun death stats will be correlated for all countries. (Still might not be valid, but please prove why it is not valid if you disagree.)
The fake stat given was that 85% of all children killed by guns were in the US. There is no such statistic.
Wow, neither of my comments posted in the intended locations. 😦
I was meaning to respond to tgt and point out that using total homicides as a proxy for gun deaths seems valid, since the US share of gun death rates and the homicide rates appear reasonably correlated. (I was defending both this blog post and your heavily-cited source blog.)
I agree that the 85% statistic was misrepresented by Mark Kelly as applying to the entire world rather than the set of countries selected. I would not go so far as to say it was ‘fake’ as ‘fake’ implies intent to me and there is no proof given of intent to deceive. Given that he also got the number slightly wrong (85 vs 87%) I would guess he did not clearly remember the number or its source… ‘brain farts’ happen. I do agree that the journalist should have responded to the stat with something along the lines of, “85%? I find that unbelievable. Can you go into further detail on that number?” which might have caused Mr. Kelly to qualify his statement. Given how difficult it can be to remember details on the spot in an interview, I’d put the burden of this mistake mostly on the journalist.
If you say “the world” when you know it’s 22 countries, then that’s intentional.
Entertainingly, I assumed your post was a response to me in an errant location, but since jack responded, I decided to let it go.
Now, to respond directly:
First, there is no gun homicide stat anywhere. There’s a “gun death rate” stat and a “homicide rate” state. The gun death rate stat doesn’t even correlate to gun homicide rate for each country.. From wikipedia, for countries that broke things down fully, I see everything from 5% to 85% of gun related deaths in a country being homicide related. (Source: http://en.wikipedia.org/wiki/List_of_countries_by_firearm-related_death_rate).
We also know the homicide rate stat doesn’t correlate to gun homicide rate. If you compare the previous chart with this chart (http://en.wikipedia.org/wiki/List_of_countries_by_intentional_homicide_rate), you can see that roughly 1 in 20 Japanese homicides are with a gun. In the U.S., it’s 3 out of 4.
The evidence goes against your claim, but this isn’t a surprise.That the two percentages originally in question are 7% apart is not strange correlation. If you took any random percentage, there a 1 in 7 chance it would be just as close to 80% as 87% is. And this is before we get into error ranges.
Despite going through all that trouble, you should also know that burden of proof was on you to prove the two stats were related. Correlation of summations is not evidence that the sub stats are in any way similar.
Why should the comparison be limited to developed countries?
Development has as much to do with crime as the first letter of a country’s name.
This is not true, at least for intentional homicide. As can be seen in the first chart on this page: http://diegobasch.com/more-charts-murders-gdp-inequality, some developing countries have low murder rates and some developing countries have high murder rates, but just about all developed countries have low murder rates (in comparison to developing countries with high murder rates).
I don’t trust crime statistics from countries other than the US. We may be the only country in the world that views accurate crime statistics to be valuable. To everyone else, they are PR exercises, as trustworthy as the genealogies of Dark Ages kings. As an example, Ontario released their crime statistics shortly before the Canadian government did one year. That was a mistake. It turned out that more people were murdered in Ontario than all of Canada combined! Canada had been late in telling Ontario how many murders to report. This means that the murder statistics in Canada are probably 1/2 to 1/3 of the ‘true’ value. This is also true of European countries as well.
Do you have a source for your example about Ontario and Canada? Was it recent? Are you sure that the Ontario number was not referring to all homicides, while the total number in Canada was only referring to murder (which does not include manslaughter or non-culpable homicide)?
It was in the 80’s (when I was in Michigan and got Canadian news regularly) and no, it wasn’t a difference in the definition. Most countries don’t want it to look like their crime rates are worse than others. This leads to a ‘fudging’ of the data. Once one country does it, everyone has to do it.
http://archive.newsmax.com/archives/articles/2001/3/21/205139.shtml
You would think from our news media and those in Europe that the US is a horribly violent society, but by this report, the US has a violent crime rate half that of Canada and behind Austria, South Africa, Sweden, Belgium, Finland, Netherlands, Luxembourg, and France. The statistics are from the UN and the European Union.
http://www.dailymail.co.uk/news/article-1196941/The-violent-country-Europe-Britain-worse-South-Africa-U-S.html
I live in Ontario and have never heard of a case where the Ontario government reported more people being murdered than the Canadian government. I wasn’t really paying attention in the 1980s, so I cannot say that it did not happen. Even if it did, though, I don’t know how the fact that data was misreported once in the 1980s could possibly imply that current reported murder rates in Canada are 1/2 to 1/3 of the “true value”.
The Daily Mail article you quote seems to imply that Britain (or at least the Conservative Party) actually wants to report a violent crime rate higher than other countries, given that they consider certain crimes to be violent crimes that other countries do not. I question the data or the analysis used in the report, however, because the conclusions of any study that reports Austria as being a more violent society than South Africa must be taken with a rather large grain of salt.
The really bad part of this incident is that Mr. Kelly has bachelor’s and master’s degrees in engineering. He is an astronaut. He know better. He KNOWS these numbers are not true, but he stated them all the same. Jane Sawyer is a journalist and my not be bright enough to know that this can’t possibly be true, but Mr. Kelly knows it is a lie, but the ends justify the means and he is going to punish the American public for what happened to his wife
Michael R. Before you go condemning Mr. Kelly, how much interview experience do you have? I know from my experience that it is easy to mess up a stat or fact in an interview and not even notice. Nowhere in this article or your comments has anyone proved any intent on Mr. Kelly’s part.
I have a lot of interview experience (and so does Kelly). If you go on national TV to talk about your new gun control group, and you clearly state on national TV a key stat involving gun use and abuse, you are accountable for any error, and responsible for getting it right. He came in with that stat, obviously. This wasn’t as if Sawyer out of the blue asked what George Sisler’s lifetime average was with the Browns, and he was off by a few points. Intent? He intended to give a statistic, on national TV. He has an obligation to get it right. If he doesn’t, he has an obligation to withdraw or correct it. If he gives the wrong stat and doesn’t correct it, I presume he intended to deceive.
Oh, so now people understand my point about crime statistics! The values used in the article are reported to come from the EU and the UN (only the values for the UK might come from the British government through the EU).
“Lies, damn lies, and statistics”.
Any time you see “gun deaths” or “gun homicides” beware. Or is it suddenly OK to murder people with knives, axes, and chainsaws?. When you start looking at homicide rates, the US is higher than other industrialized countries (4.8 vs 2.8 for Liechtenstein, 2.2 for Finland, and 1.2 for the UK in homicides/100,000) but the difference isn’t as pronounced. When you take into effect that the US counts bodies and the UK counts convictions, the difference narrows even more. When you see that the top countries have rates 10-20x the US rate (Honduras is over 90!), it becomes obvious how ridiculous this Kelly’s statement is.. Kelly has multiple engineering degrees. He knows intentionally a BS statistic.