I wish I’d had this a few weeks ago, when I was telling students how not to present their data. This is a chart illustrating the effects of stand-your-ground-laws on murder in Florida.
I glanced at that and thought, “Whoa, surprise: the stand-your-ground-laws had a pretty dramatic effect in reducing murder. I did not expect that at all.”
And then I was a bit disappointed: “But they really should have set the Y axis at zero. It’s a bit misleading and magnifies the apparent effect, otherwise.”
And then I did a double-take: “They inverted the freaking Y axis!”
That’s right. It doesn’t show a decline, it shows a dramatic spike in murder after the law was passed. The text in the article actually says that clearly, but the chart was actively selling the opposite message. They’ve since added a corrected chart that actually makes the point clearly, instead of obscuring it.
I took away two points. It’s really easy to lie with graphics, and shouldn’t any evidence-based legal system recognize the consequences of passing a bad law and correct itself?
More from a data visualization expert.
‘cos they’re vicious and dishonest shit-molded-into-a-vaguely-human-forms. I’m only astonished they bothered to correct the graph.
Try this one on for size:
http://cdn2.vox-cdn.com/assets/4296819/BlL4wWICAAA85NH.jpg
Actually, I think that’s a really good idea if you do it right. Show graph to gun nut. Gun nut says – see, clearly an immediate and significant effect. Turn graph up right way. Gun nut now has to rationalise why it isn’t immediate and significant.
They’ll still rationalise it, but you get to laugh at them while they do it.
@ violetknight
Wow I didn’t realise that all the white people in the US live in the south… or wasn’t that what I was supposed to get from the ‘chart’.
I’ve seen people tinker with the Y scale and conveniently chopping off the bottom to exaggerate the change, but I’ve never seen anyone make the graph upside down, that’s real dirty pool.
But the real point this and any other chart on gun deaths says to me is that this issue is way overblown in our body politic given the damage it does. So SYG increased gun deaths from about 550-750 annually in its first year. Florida has a population approaching 20,000,000. So SYG was responsible for the deaths of 0.001% of the population, or one person in 100,000. And while each and every one of those is rightfully tragic for the families of the victims, as public policy considerations go that hardly seems worthy of consuming the amount of political capital we spend on it when there are many many other pressing issues that effect near majorities of the populace.
Oh, incidentally I feel the same way when the gun nuts go off about how society is going to collapse and hoards of brown people are going to raid their homes and kill their children if we take away their guns. They’ve been watching too many Mad Max reruns.
Please note: the business insider article clearly said that there was an increase in gun murders after the law was passed — it was NOT distorting the facts. It was only this bizarre graph that was confusing.
It came from Reuters via the Florida Department of Law Enforcement. I tried to find where the dishonest inversion came from, but haven’t managed to track it down…but it wasn’t Business Insider.
That would require an agreed-upon way to measure the effect of a law, and thus agreement on what laws are supposed to do. I’m sure if we were to argue that the law causes a higher murder rate, the response would be that the murder rate is tangential: laws should increase a philosophical form of freedom which isn’t dependent on how many people can do what they want in practice.
I’ve heard at least one person argue in ernest that the government is limiting its citizens’ freedom if you aren’t allowed to bind yourself in slavery in a contract.
The sheer chatmanship in that first graph is so amazingly brazen, I think it is almost slow-clap worthy.
I would argue, based on comparative evidence of charts, that this one originated from Faux News.
Wow. That was one hell of a doozy of a graph.
Anyone know what caused the original drop off over the course of the 90s? 873 in ’91 to somewhere over 400 in ’99 seems pretty significant.
The main heading on the chart says “Gun Deaths in Florida”, but the subheading says “Number of murders committed using firearms”. The whole point of SYG is you can kill someone with your firearm without it being considered murder.
What is supposed to be the point of the graph again?
Gun nuts don’t care about death increases. The more that die the better the law works because obviously they did something to deserve it. There is no rational argument on the gun side that will change their minds. http://www.esquire.com/blogs/politics/bateman-guns-death-threats-041414
Stalking, death threats and intimidation are routine against those that call for a rational gun culture.
If you go back to the Nye-Ham debate you’ll see Ken Ham do the same thing with a graph from a scientific paper. It showed the exact opposite of the point he wanted to make – so he turned it upside down!
The graph appears at the 41:10 mark
I don’t think dishonesty was intended, the creator of the graph has since admitted the graph is misleading. She also says she was going for a blood-dripping-down-the-wall effect.
It’s not clear: Did Reuters produce the graph based on Florida DLF data, or did they republish a DLF graph? I’ve been digging around the Reuters website and can’t find the graph. It’s difficult to imagine why Reuters might do the original graph, other than ineptness. I can’t quite see them having a stake in selling Stand Your Ground. It’s easy to believe the DLF would do it to obfuscate the impact of the law.
The corrected graph was done by a Business Insider reader per the article.
Rawnaeris @#11 — As the article notes, violent crimes declined during the period up to 2005.
trurl @#12 — The labeling of the chart is certainly ambiguous. “Gun deaths” can include suicides and accidents, as well as the fact that an SYG shooting would not be considered legally murder.
4idan: Ah, there it is. Ineptness.
C. Chan appears to be a staff graphics person at Reuters who does many different types of graphics for Reuters news stories.
The article also switches between “gun deaths” and murders? It would be interesting what the chart actually depicts. If the number of “official” murders were to rise, or also the “lawfull” killings would make quite a difference. Also, both variants would be devastating for stand-your-ground, either the law encourages people to shoot in situations which would not be lethal otherwise or the law does not even discourage actual murders.
I’ll just point out two things. First of all, I do find it very difficult to believe that graph was an accident, despite the protestations of the designer. Second, this shows the importance of peer review. This kind of thing would never have made it into a peer-reviewed paper.
@11:
This is a general trend for violent crime in the US over this time period. [Yes, I know that doesn’t answer your question.]
This one will be a perfect example for the “missleading statistics”-lecture next semester…
Ah, ok. General trend actually makes sense. At the risk of making others feel old, I was in elementary during that period of decline, so the entire time I’ve been aware of this shit, it’s been on the upward trend.
It’s not so terrible in my view. As a Dexter fan, I see this graph as art, and it looks like blood flowing down.. More blood, more it goes down. Just it.
@Rawnaeris, Lulu Cthulhu
People have advanced all different kinds of answers to that. From the legalization of abortion about 20+ years before (eliminating from society some of the least-privileged), overall continued economic prosperity, changes in police technology and tactics, and the elimination of lead paint from our environment (presumably making us less violent).
Which would have been trivially easy to achieve by switching the colors around rather than the axis. Oy.
Presenting a chart like this upside-down is as much against Western communication norms as would be printing the text upside down or backwards. I expect that the reason they could even consider doing this, and the reason that it was published at all, is that they know that general readers are ignorant of normal scientific communication tools (such as the typical positioning of x and y axes) and are unlikely to notice the inverted axis. If the chart had not been challenged, I wonder if it would have been corrected.
And here I thought it was going to tell us that the passing of SYG had resulted in a dramatic decrease in the murder rate in Florida because a significant portion of gun deaths in that state were suddenly no longer considered murders.
@richardelguru
Even more disturbing, Asians have disappeared. They’ll come back by 2060, though.
In similar news (;-))
when I lived in Rochester, NY a couple of decades ago, one of the news programmes had a graph as background for the stocks report. When stocks went up the graph zig-zagged nicely from bottom left to top right. When stocks were down they merely turned the graph 90° clockwise.
After seeing this for several weeks (and chortling quietly, though probably nastily) I wrote to them and pointed out that the down graph meant that they were suggesting that at certain times the market had three values, and they actually changed it!
Ok, so the designer tried to be clever and creative, and fucked the graph up by overdesigning it. Seems plausible enough to me. Hanlon’s razor and all that.
To be fair we’d want to look at the types of murders which have been committed since the stand your ground law has been effected in order to draw firmer conclusions. Sometimes murder rates can fluctuate like a random drift. These graphs tentatively suggest a connection.
However the statistics came out, I would oppose stand your ground laws anyway, we already have something called the police, however imperfect, and don’t need Batman wannabes wandering the streets to commit Zimmerman style murders.
“gun deaths” vs “murders” — IANAL, but doesn’t SYG eliminate the “murder” (criminal) aspect of the death? So to combine “murders” and “SYG gun deaths” confuses the issue. What I’d be interested in is seeing the post-SYG numbers broken down into cases that were deemed “murder” and those deemed “not murder” under SYG. And then to ask the killers in the SYG cases, “If there were not a SYG law, would you have killed the person”? I’d be further interested to know, about the SYG cases, in how many cases there was an actual threat worth shooting someone over.
In any case, it’s pretty sad. I’m disgusted that FL gives people license to kill.
@mordred #19: Stand Your Ground laws probably prompt MORE murders. If more people shoot in situations that would not otherwise be lethal, presumably not all of them will be found Not Guilty on the basis of SYG (for example, rates of acquittal for Black people killing White people are lower than for White people killing Black people – in fact, acquittal rates for White people killing Black people are the highest); SYG probably results in an increase in both ‘justified’ and ‘unjustified’ homicides. This article from Frontline summarizes the findings of a couple of studies (and links the originals, of course) looking at the issue.
http://www.visualisingdata.com/blog/wp-content/uploads/2013/04/IRAQ.jpg
That is supposed to be the inspiration or basis for the blood wall idea. Except the X axis is at the top not the bottom, so the Iraq one is still very clear at presenting the data. I know blaming incompetence instead of evil is the way to go generally. I just have very little room left for any benefit of the doubt when it comes to the media and right wing politics.
violetknight wrote:
I don’t know, I think the mass deportation of the hispanics and blacks coming in 2020-2030 should be quite disturbing, too.
Being a data guy, I have a follow up question to ask: How is “murder” defined? Isn’t it the case that George Zimmerman is technically not a murderer, because the court determined he was only “standing his ground”? If that’s the way they look at it, then this graph may even be understating an overall increase in shooting deaths.
Another problem with that graph from comment #2 is that the instinctive take-away is to compare areas, not vertical cross section at a point in time. Looking at 2060, the graphic is conveying 8% of the population are Asian, but because it is the tip of Maine, its area is not at all like 25% the area of Hispanic, which in turn doesn’t at all look like 72% the area of White.
@37: A nitpick:
Zimmerman didn’t actually invoke SYG in his defense. He claimed self-defense, instead. He is still, of course, a murderer.
I agree with 4idan, #15
My first thought when I saw it was, clever image of flowing blood, but perhaps too clever. I think it may have been inadvertently misleading. Calling it dishonest may be unfair.
@35.
For the gun death graph, I did not get an impression of blood dripping down at all. In the Iraq graph the x-axis labeled at 0 is the primary clue that it’s legitimately upside down, but there are other cues as well. Use bar numbers rather than connect the medians somehow makes it clearer that the red and not the white is the foreground color (as well as gives the dribble effect) and printing text over the red area pointing down to the value rather than in the white area pointing up gives the impression of which is the data and which is background.
A failed attempt at blood drip seems a likely explanation.
It doesn’t matter if the defendant invokes Stand Your Ground. The judge’s instructions to the jury made it clear they were supposed to consider it. I wish I could find the story from the Tampa Bay Times where I read it…
Here is another very funny graphical treatment of data:
http://denialdepot.blogspot.ca/2010/11/how-to-cook-graph-skepticalsciencecom.html
It’s also possible that there are more murders because the other person shoots first.
There’s an arms race in the USA unknown in Western Europe.
A bad guy* who risks being shot if he doesn’t will make sure he won’t get killed. A bad guy who risks being shouted at probably won’t.
*just to repeat the trope, not to indicate that it’s only men
Gun homicides are probably a Poisson quantity, so the increase right after the law is about a 4 standard deviation effect–astoundingly significant. Then again, I expect this was the effect they were aiming for with the law.
@slatham #43
Is that supposed to be a satirical blog? If they’re trying to “debunk” global warming like that, it’s just a joke. It hurt to read more than a couple paragraphs of it, but I made it to the end, and it still doesn’t make sense.
It’s a failed chart, but if you’re color blind to red it reads just fine. (That was my initial guess, but it’s not likely that someone would hire a color blind person to do their graphics.)
Denialdepot is indeed satire. It is in in the joke, unlike Tony “Micro”Watts, who is the joke.
Not so much lead paint, as lead additives in petrol (“gas”). The evidence is pretty strong: not only does legislation banning lead additives correlate with drops in violence some 20 years later in numerous countries, there is a highly plausible mechanism: lead poisoning specifically causes damage to the prefrontal cortex, which is known to lead to poor impulse control.
Let’s be careful here… I don’t see this as particularly strong evidence that SYG laws increase gun violence. I certainly think such laws are unjust and immoral, and so oppose them on those grounds… and it is at least prima facie plausible that they increase gun violence, and this is a data point in favor of that. But I think you’d be hard-pressed to say that this alone proves it, especially since other states that have passed similar laws have not experienced this trend (or at least not as markedly). My intuition tells me the spike is at least part-fluke. But one really can’t say.
As to that graph… yikes. I saw that earlier in my FB feed. Still trying to figure out how that came about. It’s a Reuters graphic, in a Business Insider article that doesn’t seem at all to want to obscure what it shows. Very confusing.
I can’t find it now, but yonks ago, some AGW denier produced my all-time favourite misleading graph. I don’t recall now what it was charting, but when displayed in the conventional manner, with the X-axis parallel to the bottom of the page, and the Y-axis parallel to the sides, it showed the whatever indisputably increasing — which the denier was denying was happening.
The solution? Tilt the graph! The upwards increasing trend was printed parallel to the bottom of the page (so it looked horizontal, i.e., “stable”), and hence the axis were at a non-right angle to the edges of the paper / display.
That most recent number (721) is nearly three times the number of all homicide victims for the whole of Germany (281) in the same year.
OSAN murders are still murder. Trayvon Martin was murdered just as much as Emmett Till was…in fact, it was basically the same thing all over again.
There seem to be lots of people who really believe this.
It gets worse: Germany has 80.2 million inhabitants, Florida 19.6.
@53: “OSAN murders are still murder. Trayvon Martin was murdered just as much as Emmett Till was…in fact, it was basically the same thing all over again.”
No, murders under Stand Your Ground are ethically still murders, but technically they are not murders, inasmuch as Zimmerman was judged “not guilty.” That’s what I’m saying — while you think and I think that Zimmerman should be considered a murderer under any sane and just set of laws, he may not be counted under whatever crazy Florida standards were applied to this graph.
Oh for fuck’s fucking sake, you know better than that if you can switch on a computer.
Both killings have every relevant trait of a “murder.” The fact that a bunch of pigfucking racists let the murderer off does not change that, and the only chance we have of changing the culture that enabled it is to not torpedo the attempt to do so by shitwhining about “technically under law.”
Gun nut and trainer of gun nuts here.
Back when Kill the WitnessesStand Your Ground was proposed we knew this would happen and said it was just about the worst thing for self defense and gun owners since the Sullivan Act. It turned the whole notion of self defense on its head and started handing out hunting licenses.
Self defense as it’s been understood for centuries? Great.
Permitting it as a defense in certain well-defined cases involving deadly force? Wonderful. It’s a basic civil right going back as far as any human culture has laws.
Castle doctrine? Likewise
Guns? Absolutely, in their limited place.
“I’d rather shoot someone than run away” or “I gotta right to be here. I think maybe I was afraid, so I’ll blast everything in sight” or “I’m scared of darkies. Oh crap! It’s a Negro! Shoot! Shoot! Shoot!”
No. Not just “No” but “Fuck no, you should be in jail”
Oh, and PZ, before the words “gun fondler” ooze from your lips again, consider that most pistol barrels are at most four inches long and less than half an inch across. If that’s a Monster Phallus O’ Doom to you it’s your poor wife I feel sorry for.
@Anuran
You just had to do the second post to show how much more you value your hobby than people huh?
The corrected graph is still misleading. It equates all “gun deaths” with “murders”. Yes there is an increase in deaths, but how many of them were accidental, self-defense or “stand your ground” cases? That data is not there.
No Ingdigo Jump, I just had to say it because PZ has weird psychosexual hangups about gun owners and never misses an opportunity to spew them.
As for the rest, well, my beliefs on self defense and the use of force including deadly force are right in line with the legal traditions of the entire world for the past 3000 years. And I believe that while regrettable and to be avoided if at all possible the life of someone who wants to rape or kill innocent people is worth less than that of those innocents. Maybe you believe otherwise, in which case I hope to the gods you don’t have children or a spouse are anyone who relies on you.
Lies, damn lies, and statistics.
@Anuran
You’re a deplorable little shit aren’t you
Typical isn’t it? Better have a gun or else you’re not a real man.
Gee wonder why PZ things gun nuts have psychosexual issues?
Which is why some psychologists think a gun is a penis substitute.
I have ? Says the guy who admits to being a gun nut and compares a handgun to a penis. I’m the guy who sees the gun as a mere tool and wonders why people fetishize guns rather than socket wrenches or screwdrivers.
Why aren’t these crazy collectors all obsessing over keeping a house full of power drills, for instance?
Because you have to actually know what you’re doing to kill someone with a drill, of course. Not that they wouldn’t take just as much pleasure in it…
anuran @ v61:
Ok, then, let’s take your idea at face value, and we can see if you actually support what you think you support.
Determine, for yourself, what the ratio of worth between bad guys and good guys is. This is to say, how many good guy deaths are worth every bad guy death.
Now look up the statistics for the likelihood that a privately owned firearm will kill a good guy versus a bad guy. If there is a disparity, then in your own reckoning, there is something wrong.
And if you insist on falling back on “Well, I’m a responsible gun owner, unlike all those other idiots out there”, just remember every single one of those idiots is saying the same thing about themselves, pointing to you. Why the automatic assumption that your situation is so super-duper perfect blessed snowflake different, if you think that?
Just do the math.
How many innocent lives must be blown away to support your wish for a shot at the bad guy?
There was never intended to be a causal relationship between Stand Your Ground (SYG) and the reduction of the murder rate, so the truly misleading thing is that there is even a graph that purports to show a correlation between the two. THAT’S what makes the graph an exercise in futility.
What SYG was intended to do was to merely remove the requirement for a victim of violent crime to flee.
According to a Tampa newspaper that did a rough study of SYG, by 2012, there had been 119 homicide cases in which SYG was applied as a defense.
Critics of Florida’s SYG like Al Sharpton have labeled it a racist law in that 78% (It’s actually 73%) of total suspects leveling a SYG defense who kill black victims are set free, while just under 60% who kill white victims are set free. That statistic is probably very intentionally misleading – Al tends to enjoy stirring the racial division pot.
What Al and others DON’T reveal is that the vast majority of suspects in homicides involving black victims are themselves black. I went case by case, using the data the Tampa paper used, and I found that in the period covered up to 2012, there were 27 black homicide victims in cases where 31 black suspects invoked the SYG defense. 21 suspects were either acquitted, were granted immunity from prosecution, or didn’t have any charges filed against them at all. That’s 68% that were set free.
When doing the same thing with white on white homicide, where the suspects invoked a SYG defense, there were 61 homicides and 61 white suspects. Of those 61 suspects, 38 (62%) were set free.
During the same period, there were very few black on white (6), and very few white on black (8) homicides where the suspect presented a SYG defense. Statistically speaking the % who were set free in those cases is not very telling, since the number is so small, but of 6 black suspects, 4 were released (67%) and of the 8 white suspects, 7 (88%) were set free.
In short (too late), Mr. Sharpton, Sybrina Fulton, et al. are barking up a tree that doesn’t warrant being barked up on. And the article referenced above would have just as strong an argument if it tried to compare the relationship between instant replay in the NFL and the number of forward passes occurring in an average game.
PZ @ 65:
Because society considers people who obsess over power tools to be oddballs, but those who obsess over firearms to have huge balls.
Slight difference.
It makes you into a good husband and father who people can depend on, y’see.
Just ask anuran.
anuran:
Go. Read.
anuran @58: The rifle used to slaughter children in Newtown has a 16 inch barrel. Just saying, because length seems to be important to you.
anuran:
So what? Do you have evidence to support the efficacy of your beliefs about self defense and the use of force?
Ah, that worn out trope.
Hero.
Protector.
Head of household.
The one who will shoot the rapist, thus saving his wife’s life.
The one who will calmly make the perfect shot to take down a gun toting killer.
You’ve got to stop drinking the NRA Kool-Aid.
Macho fantasies aside, how much protection do guns offer?:
anuran @61:
So what’s the running body count on nasties versus innocents in the US? Or anywhere else for that matter.
Anri:
Owners of firearms have made variations of this statement many times. My question to them: How does anyone else know you’re a responsible gun owner? Are we supposed to take your word for it? In the US we have high levels of gun violence. One can easily find story after story after story about irresponsible gun owners. In light of that, expecting people to take you at your word that you’re really, truly a responsible gun owner is not a reasonable expectation. Show me a psychological evaluation and *then* I might believe you.
*is tired at 2 am*
*pops in*
*drops boring joke about “anuran” and “anencephalic”*
*pops out*
Jumper Guy:
I don’t know if that’s what it was intended for, but the Florida statute reads:
(bolding mine)
I wonder how many people invoking SYG laws did so reasonably. How did they assess the risk of personal injury or death?
The courts have made abundantly clear that they have no intention of ever enforcing the “reasonable belief” clause.
Jumper Guy:
Oh yeah, are we supposed to trust that your numbers are accurate? After all, your comment included no citations to the source of your figures.
Is this your source?
Azkyroth:
That’s scary.
Gun fetishist gotta gun fetish!
Anuran sez:
Cue up the tiny violins, ladies and gents. I hear a cry of “they persecute me because I loves mah gun!” in the background.
The sad truth is that the distinction between the above and “traditional” “self-defense” is blurry, at best.
Projection AND dick size shaming. What a swell person.
It’s like whenever guns come up, anuran becomes like a freshly minted troll that is completely ignorant of Pharyngula’s standards.
Not only do you fail at armchair psychology, you fail at just plain reading.
And obviously tradition means right! I suppose you believe in God and patriarchy too then?
Anuran, Infallibly Moral Super Cowboy.
Aaaaaand go fuck yourself. And make sure that when do so, that the safety is on.
We had a gun nut here in Lake County Illinois, who had a chance to chase an intruder out of his house, but elected to corner the perp into going deeper into the house and then shot him. Sounds like gun nuts everywhere. They are judge and jury, on the spot, without any reflection of the consequences. Arrogant assholes….
Jumper Guy:
Why do you think that black on black crime in a institutionally white supremacist society that has an obviously racially biased result in the use of SYG as a defense is NOT an example of a racist law?
I know this was a bit back in the thread, but it nearly made me choke. This got into nanoletters (it’s a very good, if technical, journal). On any given day there are tons of really shitty and misleading figures published in solid journals.
Tony, I think your source sounds pretty durn similar to what Jumper Guy is babbling about.
And Jumper Guy is glibly dismissing the most significant parts of this. Look at the racial interactions here.
First Accused, Second Victim:
White/White: Justified 32, Convicted 23
White/Black: Justified 6, Convicted 1
White/Hispanic: Justified 4, Convicted 1
Black/Black: Justified 19, Convicted 11
Black/White: Justified 4, Convicted 3
Black/Hispanic: Justified 2, Convicted 0
Hispanic/White: Justified 2, Convicted 3
Hispanic/Black: Justified 3, Convicted 0
Hispanic/Hispanic: Justified 2, Convicted 0.
It seems that people tend to be Super Justified when killing minorities! And definitely not justified when killing white people. Unless they happen to also be white people. Then it is slightly more acceptable.
mordred @19
As I understand it, the figures for “gun deaths” are used in aggregate to avoid any problems caused by different definitions in various statutes in different states. It’s obvious that planned and premeditated murder is treated legally much the same everywhere. As soon as you get into the niceties of what does and doesn’t count as legitimate defences of provocation or of self-defence and all the other legal argy-bargy you find gun deaths being classified as other than murder from jurisdiction to jurisdiction. Statisticians don’t want comparisons between states/countries to be made more difficult by the legal boundaries around the concept of murder rather than focusing on the numbers of deaths caused by guns.
fredrikjanson @22
If you want a bit of humour, a couple of other graphic presentations from DenialDepot should help out.
http://denialdepot.blogspot.ca/2010/01/this-week-of-global-cooling.html
http://denialdepot.blogspot.ca/2010_09_01_archive.html
http://denialdepot.blogspot.ca/2009/09/arctic-sea-ice-staggering-growth.html
Most importantly, for students who don’t understand how essential it is to present information clearly and accurately, it’s worth getting them to read the comments. Even for the most ludicrous misrepresentations, you can find people turning up in the comments taking the graph/ic seriously. (Of course, the commentariat in that group were inclined to make matters worse at the expense of the poor benighted naive rather than explain what was going on.)
So it’s a good demonstration of just how important it really is for any chart, graph or other graphic to display information accurately. And also to be as uncluttered as possible – which might make it difficult to be accurate – so the ability to write a clear and unequivocal caption is equally vital for many presentations.
The sample set is too small to begin to draw that conclusion. While I suspect the analysis is reasonable because of what I know about general statistics in the matter, it really might not hold up in a larger sample. You can get a sense for this by examining the Bayesian analysis of a simple coin flip experiment. In that super simple case, it takes over 50-100 flips to begin to know that the coin is fair. A court case has so many particular details which make it unique, it’s nothing like a coin flip. Detecting racial bias with any degree of certainty would need a much larger sample set than this.
That sample set exists in the criminal justice system at large and it shows racial bias. That is, normal self defense law is already enforced in a racist manner. Trying to detect a marginal increase in racial bias in sentencing due to SYG will require a larger dataset than has been collected.
I worked with a guy who told me that he sold his gun collection after he realized that he was indeed fondling them.
As for the added safety of owning a gun, no. In several other areas, safety devices just lead people to do more dangerous things—anti-lock brakes, airbags and bicycle helmets come to mind. Guns seem likely to really provoke macho behavior and going into danger, and certainly have a reputation and record of just that.
You have a citation for that? It doesn’t smell right to me.
This Harvard study by Jonathan Lee concluded that mandatory motorcycle helmet laws resulted in a reduction of risky driving behavior in motorcyclists.
No, I don’t have a citation. That was based on a couple of college professors in classes, and an on-line article or two. They all had sources, but I can’t find them now.
Tony! The Fucking Queer Shoop! @ 88:
Presumably, it also showed a reduction in two substantially risky motorcycling driving behaviors –
1) Driving a motorcycle without a helmet, and
2) Being involved in an MVA on a motorcycle without a helmet.
‘Cause those are pretty risky things to do.
The phenomenon is called Risk Compensation.
@Nick Gotts 49
Ah, right. Misremembrance probably attributable to my recent home purchase.
I disagree. The punishment for trespassing, or for outright burglary, shouldn’t be death.
Maybe if you’re wearing a helmet, you feel more like you’re doing something dangerous?
Similarly, carrying a gun constantly reminds you that there might be opportunities to defend yourself or others from violent crime, so you overinterpret everything you see in that direction…?
*sigh*
Well, if you’re pro-gun, this chart might be a way to argue that SYG laws have been good for America. I saw a graph similar to this in a Persuasive Writing Course in college. The book we were using in the class explained how people deceive and mislead when citing numbers by changing the scale or flipping the axes. The name of the book was, “How to Lie with Statistics!” by Darrell Huff. It was written in 1954….
“So SYG was responsible for the deaths of 0.001% of the population, or one person in 100,000. ”
Much like 9/11.