Deltoid

Daniel Davies has some criticism of a Steve Milloy Fox News column that purported to debunk a study that found that sugary drinks were linked with weight gain and diabetes. Milloy has a column on Fox News where he regularly disinforms his readers. Today I’m going to look at a Steve Milloy effort titled Gun Control Science Misfires, where he attacks two studies that he clearly has not even read. Milloy writes:

Dr. Kellerman claimed in a 1986 New England Journal of Medicine study that having a firearm in the home is counter-productive. He reported “a gun owner is 43 times more likely to kill a family member than an intruder.”

Kellermann did not report that, nor did the study find that. Kellermann actually reported that:

For every case of self-protection homicide involving a firearm kept in the home, there were 1.3 accidental deaths, 4.6 criminal homicides, and 37 suicides involving firearms.

Note that most of the self-protection homicides were not of intruders.

How did Milloy happen to misquote Kellermann? Well, he didn’t bother to read Kellermann’s paper but relied on this article by Miguel Faria. And Faria didn’t read Kellermann either, but relied on this article by Edgar Suter. The same misquote seems to have been spread far wide, appearing in a law review article, an amicus brief, and a report from the Statistical Assessment Service (apparently written by Iain Murray).

Pro-gun writers often complain about gun-control advocates using this study to make the misleading claim that gun misuse is 43 times as common as defensive use, but if you search you find that just about every such reference is made by pro-gunners objecting to the statistic.

Milloy then attacks another Kellermann study:

In a 1993 New England Journal of Medicine study, Dr. Kellerman again reported guns in the home are a greater risk to the victims than the assailants. In addition to repeating the errors of his prior research, Dr. Kellerman used studies of populations with disproportionately high rates of serious psychosocial dysfunction such as a history of arrest, drug abuse and domestic violence. Moreover, 71 percent of the victims were killed by assailants who didn’t live in the victims’ household, using guns presumably not kept in the home.

This is wrong from beginning to end. Once more it is clear that Milloy did not bother to look at Kellermann’s study, instead relying on Faria who is relying on Suter. The 1993 study used a case-control design, completely different from the 1986 study, so even if there were errors in the that study, the 1993 one does not repeat them. Suter just doesn’t know what a case-control study is. Ironically, Milloy does know what a case-control study is, and if he’d read it, might have been able to come with a coherent critique. For example, Milloy complains that the population studied was disproportionately dysfunctional, but that is not an error. The cases were people who were murdered, and yes they were disproportionately dysfunctional but that is the nature of murder victims. And 71% of the victims were not killed with guns from outside. A glance at the tables in the survey shows that claim to be false.

After dismissing Kellermann’s work as junk science, Milloy gets to John Lott. Instead of citing third hand criticism, he cites Lott’s findings with not a word of criticism:

laws that permit the carrying of concealed weapons are associated with a 69 percent decrease in death rate from public, multiple shootings such as those that occurred in Jonesboro, Arkansas and Columbine High School.

Oddly enough, when writing about a study by Cummings that linked safe-storage laws with reductions in accidental shootings Milloy wrote:

This was an ecologic epidemiology study, meaning the conclusion is based on very “macro” comparisons of groups of people. The study involved no data about individuals, just groups. Traditionally, these studies are only useful for forming hypotheses for further testing, not irrefutable facts.

Now Lott’s study was an ecologic study at the group level and Kellermann’s case-control study was at the individual level, but Milloy uncritically accepted Lott and trashed Kellermann. I wonder why? Furthermore, when I asked him why he had criticized Cummings while posting an uncritical summary of Lott’s work he didn’t defend Lott but tried to pretend he wasn’t endorsing Lott’s findings:

That wasn’t my summary… but quotes from the article.

But here he does endorse Lott’s findings.

Here’s the most telling thing about Milloy—you can tell what his conclusions about a scientific study will be without having to look at the methodology of the study. If he doesn’t like the conclusions he will find some grounds, no matter how specious, for dismissing the study as “junk science”.

Update: SayUncle comments on a draft of this post that I accidently posted:

The problem with Kellerman’s study is he compares self-protection gun deaths to other gun deaths, which discounts the self-protection that does not result in the death of someone.

Kellermann clearly notes this problem in his 1986 study, which also discounts gun misuse that does not result in death. This was one of the reasons why he did the 1993 study which does measure self-protection that does not result in death.

In my previous post I mentioned Daniel Davies’ demolition of yet another dodgy Steve Milloy article. Milloy attacked a recent JAMA study that found:

Higher consumption of sugar-sweetened beverages is associated with a greater magnitude of weight gain and an increased risk for development of type 2 diabetes in women, possibly by providing excessive calories and large amounts of rapidly absorbable sugars.

Todd Zywicki, who endorsed Milloy’s piece as a “devastating critique” has mounted a defence of Milloy. Unfortunately it is clear that Zywicki has not read the article (subscription required) or even the abstract describing the study.

Zywicki dismisses concerns about Milloy’s character raised by Davies and John Quiggin as not relevant to the question of whether the JAMA study is a good one or not. It isn’t, but Zywicki hasn’t read the JAMA study. Instead he is relying on Milloy to accurately describe it. Milloy’s character suggests that his description of the study might be misleading; and in fact it is misleading and as we will see below, Zywicki has been mislead.

Zywicki writes:

Milloy says that the once the researchers “statistically adjusted their results for bodyweight (a risk factor for diabetes) and for caloric intake (a proxy measure of consumption of sweetened foods other than soda), the 83 percent increase [in type 2 diabetes prevalence] dropped to an even more statistically dubious (and soft-pedaled) 32 percent increase.” Now it seems to me that Milloy is obviously correct here—bodyweight and non-soda caloric intake seem to me to obviously relevant to trying to isolate the marginal effect of the increased soda consumption. So the 83 percent figure is really an irrelevant number

Milloy has mislead Zywicki into thinking that the 83% increase in diabetes was partly caused by confounds—that the women who drank more sugary drinks also happened to be heavier and eat more and that these factors are what caused most of the increase. But right in the abstract they clearly state (my emphasis):

After adjustment for potential confounders, women consuming 1 or more sugar-sweetened soft drinks per day had a relative risk [RR] of type 2 diabetes of 1.83 (95% confidence interval [CI], 1.42-2.36; P<.001 for trend) compared with those who consumed less than 1 of these beverages per month.

So what is the 32% figure that Milloy tried to pass off as the “real” increase? Well, they found that increased soda consumption was associated with weight gain and weight gain is known to be a risk factor for diabetes. The 32% increase is the extra risk factor for soda consumption on top of the increase from the weight gain from drinking more soda. It would only be the real risk if sugar-sweetened drinks did not cause weight gain, but they do.

This is not the only matter that Milloy has misled Zywicki about. Zywicki writes:

Milloy similarly notes that the study does not control for genetics or lifestyle issues

But the study did control for genetics and lifestyle issues (my emphasis):

We evaluated whether the association between sugar-sweetened soft drink consumption and risk of diabetes was modified by BMI, physical activity, and a family history of diabetes using analyses stratified by these variables and by modelling interaction terms.

Milloy even accuses the authors of “scientific misconduct” for not mentioning another study that Milloy alleges contradicts their results. But that other study was not about soft drink consumption but about overall sugar consumption. The new study suggests that consuming sugar in a drink where it is more rapidly absorbed may increase the risk of diabetes. This is hardly contradicted by results that suggest that sugar intake including that in solid food is not a risk factor. Zywicki endorses the serious charges that Milloy makes without checking whether they are accurate.

Update: The Washington Times has also published Milloy’s misleading article and Reason’s Nick Gillespie was also taken in. Matthew Yglesias reckons that libertarians should just argue that they have a right to unhealthy food instead of trying to debunk the science that shows them to be unhealthy.

Update 2: Nick Gillespie links here (thanks!) and to a Tech Central Station article by Jon Robison that criticizes the JAMA study. Robison, like Milloy, tries to pass of the 32% increase in diabetes, which is the extra risk after accounting for the effect of weight gain, as the total increase in risk. Robison also asserts that “Epidemiologists generally agree that relative risks less than 2 should be ignored or at least viewed with extreme skepticism”. In fact, epidemiologists do not “generally agree” with this. I explain why in this post.

The Sydney Morning Herald reports (emphasis mine):

In late 2002 a casual lecturer employed by the University of Newcastle to teach at its affiliated school in Malaysia, Institut WIRA, gave zero marks to 15 students in a class of 50, citing “deliberate, serious plagiarism”.

The lecturer, Ian Firns, found that several students had copied large sections of their essays from a paper available on the internet without acknowledging the source. In the comments he wrote on the papers before handing them to the business school, Mr Firns spelt out the web address.

However, the ICAC Commissioner, Peter Hall, SC, heard yesterday that Mr Firns’s discoveries were ignored. All the essays passed—with grades up to 84 per cent—when marked again, after Mr Firns’s comments and the web address had been hidden with white correction fluid.

Paul Ryder, then the dean of business and the head of the Newcastle Graduate School of Business, told the inquiry yesterday that in ordering the remarking he acted on the advice of his deputy, Robert Rugimbana.

Dr Rugimbana had told his administrative staff to white-out Mr Firns’s comments. He told the commission he decided “there was no prima facie case evidence” of plagiarism, though he did not look at the website Mr Firns cited.

Asked by Ms Ronalds why he had did not check the web address, Dr Rugimbana said no academic would consider a web address as evidence of plagiarism. He likened Mr Firns’s explicit directions to someone saying “There’s the library”.

I don’t think that they were actually corrupt but it is disheartening that people this grossly incompetent got into positions of authority at my alma mater. I should just send my B Math back to Newcastle.

Discussion on this affair last year (including a acomment from Firn) at Troppo is here.

In the latest installment in the U of Newcastle plagiarism scandal, Ronald MacDonald tries to outdo Robert Rugimbana (emphasis mine):

The deputy vice-chancellor of research and internationalisation, Ronald MacDonald, said yesterday that he had originally believed the plagiarism claims by Ian Firns. Mr Firns had written to him in late February 2003 protesting at the handling of the matter.

In an email, Mr Firns wrote that “it is particularly galling to know that the top mark for this assignment was awarded to an identified cheat”.

However, Professor MacDonald said that he had changed his opinion shortly afterwards, and accepted Dr Ryder’s view that Mr Firns was making exaggerated accusations out of “spite”. He held that view despite the lecturer having provided specific web addresses for the sites from which students had lifted slabs of material.

Professor MacDonald said it would have been beyond his capacity to judge the plagiarism allegations for himself, because the essays were to do with business. “I’m a physicist.”

Of course, that’s not even faintly plausible. He should have said: “I’m an administrator”.

McKitrick has added a correction his page describing his paper that purports to find economic signals that I posted on here. McKitrick admits to mixing up degrees and radians but claims:

There was a small error in the calculation of regression coefficients in our paper. Our conclusions were not affected by this problem

As I noted in my post, correcting the error halves the size of the economic signal in the warming trend, reducing it from 0.16 (out of 0.27) to 0.09. McKitrick’s correction states:

Outside the dry/cold regions the measured temperature change is significantly (previous: primarily ) influenced by economic and social variables.

That’s quite a difference, so how can he say that their conclusions were not affected? Well, all the conclusion says is that there were socioeconomic effects, without mentioning their size. The size of the effects, which change substantially, are only mentioned in the body. And the “bombshell” nature of the paper touted by Michaels et al in their TCS article depends on socioeconomic effects being the primary cause of the warming trend, something that McKitrick has now retracted.

McKitrick has also failed to correct or even acknowledge another serious problem in his paper—he has not corrected his standard errors for clustering. This is required because his socioeconomic variables are all the same for the stations in the same country. This means he will find some variables to be statistically significant when they are not really so.

Nor has McKitrick explained why he decided to take the cosine of the absolute latitude in the first place. Calculating it correctly makes no difference to the model, while calculating in incorrectly makes the model fit worse. There does not seem to be any theoretical or empirical justification for this change to his model. As John Quiggin observes:

a trawl back through the files makes it pretty clear that this error was not exactly an innocent mistake. It seems pretty clear that McKitrick tried some regressions with (absolute) latitude as the explanatory variable, didn’t like the results he got and switched to the cosine (note that, if you were starting here, you wouldn’t need to take the absolute value, since cosine is a symmetric function). Because of the degrees-radians mistake, this variable came out insignificant, as desired, and McKitrick didn’t do the checks that would have revealed the error. Asymmetric error-checking is a standard problem with cherry picking, as illustrated by the work of John Lott.

Lott has teamed up with Kevin Hassett to study whether economic reporting is biased. The paper, Is Newspaper Coverage of Economic Events Politically Biased?, concludes, surprise, surprise that the newspapers are biased against Republicans.

The trouble with their study is that the economy was stronger under Clinton than under either Bush, so of course the reporting of the economy under Clinton was more positive. Lott and Hassett claim to have controlled for this with a multivariate analysis but you should only find this persuasive if you have complete confidence in the competence and integrity of the authors. When building such models there are so many choices you can make that it easy to get the results that you want to see. In this case it is particularly easy, since all you have to do is leave out a relevant variable so that the state of the economy is not fully controlled for and you will get results like Lott and Hassett report.

We probably should not have complete confidence in Lott and Hassett. Lott’s previous attempt to show that the media was biased seems to have involved cherry picking of models, and careful selection of models and data to get a desired result seems to be a constant characteristic of his work. As for Hassett, I’ll let Lott tell you about him. This is Lott’s anonymous review of Hassett’s Bubbleology:

Despite Mr. Hassett’s track record with his previous book “Dow 36,000,” I saw him appear on CNBC during the early morning show and thought that he did well enough that I should buy the book. He promised that you could use his book to figure out what stocks were overvalued and which ones weren’t. A pretty important topic given the current market environment. However, after reading this short book I have no idea of how to actually rank stocks on the 1 to 6 scale that he uses. He doesn’t actually provide concrete examples, only that he says that he put together this ranking and it worked really well. My other problem is that if this approach works so well how come he didn’t use it when his “Dow 36,000″ book came out when the stock market was at its peak. Some explanation would have been useful for why Hassett, who is marketing this book as a full proof approach to spotting bubbles, wasn’t able to use this approach himself over just the last couple of years to warn people and predict which stocks were going to crash, a period when he was supposedly writing this book. Claiming that you use a not clearly stated formula to identify overvalued stocks after they have already crashed seems like a scam to me.

“seems like a scam”, indeed.

The New York Times has published what Lott calls a “hit piece” on their study. The Times gives their study more credence than it deserves (Atrios is disgusted that they even did a story on him), but they do eventually mention Mary Rosh and the 36,000 Dow prediction and finish with an apposite quote from Brad DeLong:

To even base a story on Lott’s work at this point in time is to demonstrate a pronounced bias toward right-wing hacks

Update: The AEI is holding an event to promote the study and the official George W Bush blog is also pushing the study. The gang at Lawyers, Guns and Money get stuck into Lott and Hassett here and here.

Update 2: Brad DeLong points out that the NY Times reporter dropped the ball by letting Lott’s deceitful statement that “the things he had said in the guise of Ms. Rosh were, indeed, truthful” stand. You can see them here and count how many are plainly untrue.

By popular request I’m going to comment on Lott’s LA Times oped on the assault weapons ban. Basically, I agree with Lott here. As I noted earlier the ban doesn’t make sense.

However, Chris Mooney has a point when he writes:

Providing balance is one thing, and it’s something op-ed pages should strive for. But if op-ed editors can’t get the other side from a credible expert—which Lott emphatically isn’t—they shouldn’t just publish anybody for the sake of having different perspectives represented.

It’s not as if the LA Times couldn’t have found someone credible to argue against the ban. For example, see Matthew Yglesias.

The Syney Morning Herald has the latest evidence from the plagiarism scandal at the University of Newcastle (my previous posts are here and here) (my emphasis):

The University of Newcastle sat on its “most serious” plagiarism allegations for more than four months, only acting when a television show peppered the institution with questions, the Independent Commission Against Corruption heard yesterday.

But the vice-chancellor, Roger Holmes, told ICAC it was coincidental that one day after the media inquiries, the university finally began investigating a lecturer’s detailed claim that 15 post-graduate business students had copied slabs of essays from the internet. …

The university set up an inquiry after the Herald and Newcastle Herald broke the story in March 2003, but Professor Holmes said yesterday that despite the plagiarism allegations being potentially the “most serious” in the institution’s history, the terms of reference left it “optional” to find out whether students had cheated. …

When Professor Lamond handed down his report in March last year, it was apparent that he had exercised his option not to investigate the allegations of plagiarism, ICAC heard yesterday.

The ICAC commissioner, Peter Hall, SC, asked the vice-chancellor what he then did to find out whether plagiarism had occurred.

“We didn’t do anything at that particular time,” Professor Holmes said.

This lack of action led Mr Firns to write to the vice-chancellor in April, calling for a “credible” review of the papers, according to documents tendered at ICAC.

In replying, however, the vice-chancellor told Mr Firns: “Do not bother me again“, the commission heard. “I regret saying that … it was off the cuff.”

Immediately after the Sunday television program covered the story in August 2003, Professor Holmes issued a statement saying the university vigorously investigated claims of soft marking.

Yet, when asked by Mr Hall to name one fact supporting this assertion in relation to this “most serious” of cases, Professor Holmes could not.

Mr Hall said that the vice-chancellor appeared to have been guilty of a “dereliction of duty”.

Asked to explain his lack of action, Professor Holmes said that “there’s got to be an explanation [but] I don’t have an explanation.”

Words fail me.

It has now been two years since I asked him for evidence that he had conducted a survey. The original email is here. In all that time, the only evidence for a survey that he has been able to come up with is David Gross’s story.

The AEI now has a video of the Lott/Hassett show alleging media bias against Republicans. (My previous post is here.) What is wrong with their study can be explained very briefly. First look at the graph on the left of unemployment rates. Now, here’s what Lott and Hassett say:

“In the case of unemployment, 44 percent of the headlines under the Clinton administration were positive while that same number was only 23 percent under Bush II. By comparison, the average unemployment rates were fairly similar, 5.2 percent under Clinton s eight years and 5.5 percent under Bush during the sample. There is also a great deal of overlap (3.9 to 7.1 percent under Clinton to 4.2 to 6.4 percent under Bush II).”

What they fail to mention and what is obvious from the graph is that under Clinton the unemployment rate decreased from 7.1% to 3.9%, while under Bush it increased from 4.2% to 6.4%. Maybe, just maybe, that’s why the headlines were more positive under Clinton. In fact, there seems to be evidence of bias against Clinton—why were only 44% of the headlines about unemployment positive when it just kept going down and down to the lowest levels in decades? Oh, and don’t expect to see a graph of the unemployment rate anywhere in their paper or presentation.

Now, they claim to have controlled for level and trends in unemployment in their analysis, but of course they have not. The only control they have for trend is the change since the previous quarter and it is obvious that changes over longer terms will affect the reporting. Do Lott and Hassett believe that no-one ever compares the unemployment rate with what it was a year or two before?

Another look at the graph of the unemployment rate will reveal the futility of the whole exercise. Is it not obvious that unemployment did something fundamentally different under Clinton than under either Bush? To get a meaningful comparison you would need to compare Clinton with a Republican administration where unemployment declined for eight years.

Meanwhile, Donald “Stalker” Luskin has leaped to Lott’s defence. His defence is more than a little odd. First, he calls DeLong “Jabba the Economist”. I don’t see how alleging that DeLong looks like a character from Star Wars proves that he is wrong about Lott. And if DeLong is cast as Jabba the Hutt in that scene, it conjures up unfortunate images of Mary Rosh as Princess Leia in a bikini. Second, he asserts that DeLong only attacked Lott because Hassett allegedly showed him in paper published over ten years ago. Somehow he didn’t notice that Lott and Hassett were different people. Third, he claims that DeLong is a hack because he didn’t tell the NY Times reporter that he was only criticizing Lott to revenge himself against Hassett.

King at SCSUScholars displays a wilful refusal to consider the possibility that there is anything wrong with Lott’s research, dismissing it all as proof of the NY Times liberal bias.

The story of the University of Newcastle plagiarism scandal continues:

Professor Marimuthu was 18 days from signing a new five-year agreement that increased the annual fee payable by the Institut WIRA in Malaysia, where the campus was based, from $50,000 to $300,000.

He asked the offshore program co-ordinator to review the problem essays “more generously” and warned recruitment could suffer “if word gets out” about large numbers of students failing, the commission heard yesterday….

Despite all students having received anti-plagiarism instruction in the course outlines, he had not considered this to be a case of cheating.

“I was of the view that it was weak referencing.”

He had formed this view without examining the papers for plagiarism, and had never investigated whether plagiarism had taken place, as had been alleged by Ian Firns, a lecturer.

When the commissioner, Peter Hall, SC, asked if he could “point to any fact” that showed why he never acted to protect the academic standards of Institut WIRA and the university by investigating this “grave” case, Professor Marimuthu said “it did not occur to me”, and the 15 students had done well in other subjects.

He also told the commission that never in his career had he detected plagiarism, and that he lacked the computer skills necessary to look for stolen prose on the internet.

OK, welcome to my course on how to look for stolen prose on the Internet. I’m Tim Lambert and I’ll be your teacher. Type a phrase from the essay into Google and press return. That’s it, collect your diploma on the way out.

You know, the newspaper stories have been so damning that I thought that maybe the reporter was taking stuff out of context to make Ryder and company look like fools, but you can read the transcripts yourself and see that these actually make these people look even worse.

The hearing continues on Friday with Ian Firns who promises fireworks:

I daresay that my evidence will give rise to some further media comment. I will be directly contradicting parts of Ryder’s testimony in particular - so one of us may end up being charged with perjury!

Paul Watson calls for the dismissal of English, Ryder, MacDonald and Zeffane (Rugimbana has already left, and Holmes is retiring soon). Seems fair, except that Zehane (who remarked the essays) doesn’t deserve to be sacked since the evidence of plagiarism was concealed from him. And if the University wanted to try to recover its reputation and demonstrate a commitment against plagiarism it should rehire Ian Firns.

Glenn Reynolds writes

Here’s some helpful advice for CBS: “A source lies to you, and you find it out, you burn him. Period.”

Damn straight. So, Professor Reynolds, who was the anonymous source who lied to you, falsely claiming that Steve Levitt was “rabidly antigun”?

Crime statistics for the last 12 months in NSW have been released. Most crime categories have decreased significantly.

Indecent assault, act of indecency and other sexual offences	Down by 11.9%
Robbery with a weapon not a firearm	Down by 19.7%
Break and enter - dwelling	Down by 9.4%
Break and enter - non-dwelling	Down by 17.6%
Motor vehicle theft	Down by 8.7%
Steal from motor vehicle	Down by 13.7%
Steal from retail store	Down by 17.4%
Steal from dwelling	Down by 5.1%
Steal from person	Down by 17.8%
Fraud	Down by 12.3%
Murder	No significant trend
Assault	No significant trend
Sexual assault	No significant trend
Robbery without a weapon	No significant trend
Robbery with a firearm	No significant trend
Malicious damage to property	No significant trend

Now if, hypothetically, you were out to show that crime had increased here because of the 1996 gun laws you would be faced with a problem. Normally with sixteen crime categories you can find one that increased and you can then run with that. But there weren’t any significant increases this time. What to do, what to do?

Fortunately the crime statistics are broken down into region and subregion. If you scroll down in the report you find a table giving the crime categories for each part of Sydney. Fourteen rows and sixteen columns means that there are 224 cells in the table. The table is a sea of negative numbers. Crime is down everywhere and in every category. But wait! There is one, just one, positive number in the whole table: Robbery with a firearm increased by 34% in inner Sydney, from 123 to 165. So what do we find on Lott’s blog?

With a reported 34 percent increase in armed robberies in Sydney during just the last 12 months, some have been driven to try and stop the attacks. (I don’t have the numbers handy at the moment, but armed robberies have been going up dramatically for the last six or so years in Australia.)

Actually it was firearm robberies in inner Sydney that went up 34%. Armed robberies in Sydney declined by 16%. And armed robberies in Australia have also been going down. To be fair, we must give the Sun-Herald an assist on this cherry pick since they pulled out the 34% figure:

Latest Bureau of Crime Statistics figures show all major crime rates as stable or in decline except for armed robbery, which has undergone a 34 per cent spike in inner Sydney in the past 12 months.

But leaving out the part about all major crime rates being stable or declining was all Lott’s work. As was the false claim that armed robberies were increasing nationwide.

Summary: Lott and Hassett have not analyzed their data correctly—it actually shows no evidence that headlines are biased against Republicans. (My previous posts are here and here.)

The essence of Lott and Hassett’s case that newspapers are biased against Republicans is given in their presentation:

“In the case of unemployment, 44 percent of the headlines under the Clinton administration were positive while that same number was only 23 percent under Bush II. By comparison, the average unemployment rates were fairly similar, 5.2 percent under Clinton s eight years and 5.5 percent under Bush during the sample.”

Their argument is that Clinton got more positive headlines and that this is not explained by differing economic conditions. To test this they do a whole pile of regressions, but these just obscure what is going on. To understand what is happening, all you have to do is look at some headlines of stories reporting the unemployment rate. Here are some typical examples:

• Unemployment rate falls to 5.4%
• US Economy Creates 144,000 Jobs; Unemployment Rate Drops Slightly
• Missouri unemployment rate holds at 5.5 percent
• Colorado unemployment rate steady

Almost all of the headlines for stories reporting on the latest unemployment rate told you whether the rate had increased, decreased or stayed the same. Lott and Hassett counted a headline about unemployment as positive if it reported that the unemployment rate had gone down. So if headlines just follow the pattern above, what percentage of positive headlines would you expect to see? You won’t find the answer in their paper—they don’t even include a control for whether unemployment increased.

I downloaded the relevant unemployment data from the Bureau of Labour Statistics and calculated what percentage of the time the unemployment rate decreased.¹ The results are in the third column of the table below. For example, it shows that under Bush I, 21% of the time, the rate decreased. The second column shows the percentage of positive stories that Lott and Hassett report in their paper. For Bush I, this is 20%, almost exactly the number you would expect if all headlines just reported whether the rate had gone up or down. Clinton doesn’t do as well, with only 42% positive headlines despite unemployment decreases 48% of the time, a 6 point gap, while Bush II does even worse with an 11 point gap. Clearly there is no bias for or against Republicans here—the gap for Clinton is exactly half way between that for Bush I and Bush II.

President	Percent of stories that are Positive	Percent of months with monthly declines	Gap between Positive stories and monthly declines	Percent of months with quarterly declines	Percent of months with yearly declines
Bush I	20	21	-1	17	0
Clinton	42	48	-6	65	92
Bush II	22	33	-11	28	13

Why is there a gap for Clinton and for Bush II? The correct way to answer this is to go back to the news stories and look at the ones that did not have positive headlines when the unemployment rate went down. I suspect that the ones that weren’t positive included some negative news as well. For example, if unemployment fell, but jobs also fell, Lott and Hassett would count this as “mixed” rather than positive. The incorrect way to answer this is to run more regressions². The actual words in the headline tell you why it was positive or not. Lott/Hassett’s methodology throws away this information and then tries to work out why the headline was positive by doing lots of regressions. This approach seems to be driven solely by the need to express the data in a form suitable for the application of linear regression.

The last two columns of the table show how unemployment had changed over the previous three months and over the previous year. This probably gives a better idea of trends in unemployment since it smooths out small fluctuations. These show that the focus on the short term in the headlines made the headlines better for Bush I and II and worse for Clinton. Clinton only got 42% positive headlines even though there was a one-year decline 92% of the time, while Bush I managed 20% positive headlines despite never having a one-year decline in unemployment.

1. You can download the data and calculations in this spreadsheet.

2. The title of this post comes from the saying “If the only tool you have is a hammer, then everything looks like a nail.”. If the only tool you have is linear regression … Update: D squared reminds me that they actually used tobit regression, which isn’t even the right kind of regression for their data. So I’m going to make regression equal hammer in my analogy, and say that they didn’t even use the right kind of hammer.

Update 2: I searched Factiva to see what the headlines were when a decrease in unemployment did not generate a positive headline and in those cases it seems that while unemployment had fallen, the number of jobs had also fallen. The headline then either reported both things, which they count as a “mixed” headline or the fall in jobs, which is a negative headline. I added data for the change in the number of jobs to my spreadsheet and worked out how many positive headlines you would get if both unemployment had to fall and jobs have to increase to get a positive headline. The results are in the table below.

My new model predicts the results for Clinton and Bush II almost perfectly. For Bush I, the previous model fits better. It appears that after Clinton came to power headlines switched from just reporting the unemployment rate to reporting the unemployment rate and the number of new jobs. And once again we see no evidence of partisan bias in the headlines.

President	Percent of stories that are positive	Percent of months where you expect a positive story
Bush I	20	13
Clinton	42	45
Bush II	22	23

For a webzine that has “Where free markets meet technology” in its masthead, Tech Central Station sure seems to have little faith in the ability of free markets to provide consumers with what they want. Consider this column by Glenn Reynolds. Reynolds reckons that bookstore employees are driving customers away:

Even my hard-core lefty colleagues have noticed the wall of Bush-bashing books that are prominently displayed at the entrance to every bookstore in town, and to my surprise, one of them told me the other day that it was turning him off. He hates Bush, and will certainly vote for Kerry (unless Kerry looks like such a lost cause that his temptation to vote Nader wins out) but he’s still taken to buying his books from Amazon. All those Bush-bashing books, he said, are just too depressing to encounter every time he visits the bookstore. He buys books to escape politics.

Now here’s the thing: bookstores need to sell books or else they will go out of business. So if they are building walls of anti-Bush books, it’s because people are buying those books and that’s the way to sell them. They will also conduct market research to find out how they can get customers into the store, so are likely to have more basis for their decisions than Reynold’s one anecdote. And if they weren’t doing these things, there would be a business opportunity for someone to start a bookstore with no anti-Bush wall and clean up. Reynolds seems to believe that the free market has failed to provide consumers with the sort of bookstore that they want.

Reynolds continues:

On the right, of course, people are a lot unhappier. And they’re unhappier still since word got out that Borders employees were actually bragging about hiding copies of the anti-Kerry book, Unfit for Command. (You can see a cached Google page here. Example: “Just ‘carelessly’ hide the boxes, ‘accidentally’ drop them off pallets, ‘forget’ to stock the ones you have, and then suggest a nice Al Franken or Michael Moore book as a substitute. Borders wants those recommends, remember? . . .I don’t care if these Neanderthals in fancy suits get mad at me, they aren’t regular customers anyway. Other than ‘Left Behind’ books, they don’t read.”)

Unfortunately, Reynolds seems to have fallen for a posting by a sock puppet. The forum administrator reports

According to IP Address records, the person who posted the “let’s hide the book” message was also posting under another name, responding to the messages he posted under his first name. Under his second ID, which identified him as a free market conservative, or “Classical Liberal”, he said he was shocked that Borders employees would not tolerate diversity of opinion, and that the messages (he) posted under his first pseudonym proved that there was a problem. This suggests to us that this was a troll who was trying to manufacture controversy. His account(s) have been deleted. He was a new user.

This information was posted at the forum a week before Reynolds wrote his column, but he does not seem to have checked. Oh well, at least he didn’t base his column on forged memos.

Update: I emailed this post to both Glenn Reynolds and Atrios. Atrios linked here. Reynolds posted a correction but did not link or acknowledge my email. It is possible that somebody else emailed him first, but the timing of his correction (a couple of hours after my email and a couple of days after his column) suggests otherwise. More interestingly, he has posted some comments from his readers speculating about why bookstores are prominently displaying anti-Bush books. Amazingly, in neither his comments or the other comments he posted is the explanation I gave above (and emailed to Reynolds) considered: the anti-Bush books sell and make money for the bookstore.

A commenter on my earlier post on John “I hate guns” Howard wondered: “If Latham wins, will the public generally credit this issue?” Some shooters have a plan to try to make it an issue:

You Can Send John Howard And Canberra A Message From NSW Shooters

• Stop victimising sporting shooters
• Crime control, not gun control
• Stop confiscations and buybacks
• Restore our rights

In this election the Coalition of Law Abiding Sporting Shooters is standing candidates under the Outdoor Recreation Party banner to send a message to Canberra about gun laws. All candidates are sporting shooters standing on a pro-shooting platform. A vote for the Outdoor Recreation Party, in this election, is a vote for shooters rights.

In the Senate (Upper House): Vote 1 Outdoor Recreation Party above the line. Preferences will help to keep out a Green.

In the House of Representatives (Lower House): Vote 1 Outdoor Recreation Party in Eden-Monaro, Greenway, Page and Dobell, and then put Labor ahead of the Liberals and Nationals, Democrats and Greens. Preferences directed this way will help to drive a wedge between the parties and break down their anti-gun bias.

Will their plan work? Well, I don’t think they’ll get that much support in the urban seats, but in Eden-Monaro and Page they have a chance of getting a decent number of votes. You can bet that the major parties will be looking very closely at where those preferences go. Both seats are very merginal so even a couple of percent of the vote could make the difference. If they do cost Howard two seats, it will certainly discourage what plans he may have for further gun bans, even if he wins the election.

Sandor is collecting blogger’s results on the political compass quiz. My own collection is here (now over 500 blogs!). Actually I think the Political Survey quiz is better, but hey, you can try both, enter your results into my table and send them to Sandor if you are so inclined.

60 Minutes (the Australian version) has done a story on the use of deadly force in self-defence.

Unfortunately it sufers from the same problem that many of these stories do. Tony Martin (who was convicted of murdering a burglar) gets to present his account of what happened which makes it look like he acted in self-defence.

TARA BROWN: What led to you firing the gun?

TONY MARTIN: Fear.

TARA BROWN: Were you under attack at that point?

TONY MARTIN: I thought I was, yes, or was about to be attacked. Because the people downstairs, as I went down the stairs, put a torch on me. You might think that’s quite mild, but I can assure you at any point if somebody put a torch on you, it’s quite alarming, to say the least.

Unfortunately, for Martin, it was obvious to the jury that his account was untrue. They visited the house and could see from where the shot had hit the wall that Martin could not possibly have fired from the stairs. It is this sort of reporting that makes people think that killing in self-defence is not lawful. I covered this matter in excruciating detail in this post. This additional comment by Martin gives an indication of why the jury decided that Martin had set out to kill a burglar:

I’ll say this—without sounding callous—it wasn’t unfortunate for them. It was unfortunate for me what happened because I ended up with a murder charge.

The other stories suffer from the same problem—we are just hearing one side of the story from people who obviously want to argue that acted in self-defence. Perhaps they did, but we cannot conclude that unless we hear both sides of the story.

The Power Line blog informs us that the Kerry campaign has mounted a “terrorist attack on Australia”:

“We all know that Kerry’s sister is over in Australia telling the Aussies to vote for their [leftist] candidate if they want to be safer from terrorist attacks; that they need to pull their troops out of Iraq, and not help the U.S. (because that’s why they will be, and have been attacked). So, how is this NOT a soft terrorist attack on Australia from the Kerry campaign? She goes over there, and with words instead of bombs, terrorizes the Aussies to vote liberal; to sway the election.”

Well the existence of this “terrorist attack” on us may come as a surprise to most Australians so I thought I would investigate to find out exactly how we were being terrorised without our knowledge.

It all seems to have started with some comments Diana Kerry made to The Australian’s Washington correspondent:

“Australia has kept faith with the US and we are endangering the Australians now by this wanton disregard for international law and multilateral channels,” she said, referring to the invasion of Iraq.

Asked if she believed the terrorist threat to Australians was now greater because of the support for Republican George W. Bush, Ms Kerry said: “The most recent attack was on the Australian embassy in Jakarta—I would have to say that.”

Now her comments are pretty unobjectionable as well as being true. It is hardly a secret in Australia that invading Iraq made is a bigger target for terrorists. Howard concealed the fact that this was the advice of his intelligence experts even before the war, but the truth came out.

Now look at what Charles Krauthammer misrepresents the meaning of Kerry’s comments. Krauthammer claims that the recent bombing of the Australian embassy was intended to affect the upcoming Australian election.

The terrorists’ objective is to intimidate all countries allied with America. Make them bleed and tell them this is the price they pay for being a U.S. ally. The implication is obvious: Abandon America and buy your safety. …

She said this of her country (and of the war that Australia is helping us with in Iraq): “[W]e are endangering the Australians now by this wanton disregard for international law and multilateral channels.” Mark Latham could not have said it better. Nor could Jemaah Islamiah, the al Qaeda affiliate that killed nine people in the Jakarta bombing.

Yes, Krauthammer equated Mark Latham, who may become our next Prime Minister with the terrorists. Krauthammer is despicable. His equation is based on not one, but two, serious misrepresentations.

First, that JI was trying to influence the election in Australia. Well, if that’s what JI was doing then they wanted Howard to win because there is no question that the bombing helps Howard’s chances for re-election. But that’s not the election they were trying to influence. Krauthammer somehow neglects to mention that the presidential election in Indonesia (you know, the place where the bomb went off) was just eleven days after the bombing.

Second, that Kerry is telling Australia to withdraw from Iraq. But all she did was truthfully answer a question about Australia being at greater risk. Krauthammer insults Australians’ courage and intelligence. Apparently he thinks we are too stupid to realize the risk unless Kerry tells us and he thinks we are cowards who will run away if we ever find out the truth.

Krauthammer has another misrepresentation at the beginning of his column:

Which is why Australia is with us today in both Afghanistan and Iraq.

But Australia doesn’t have troops in Afghanistan. We did, but withdrew them at the end of 2002. This does not mean that Australia turned tail and ran. Similarly, Latham’s plan to withdrew some (not all) of Australia’s small remaining force in Iraq is not running away either. They’ve already stayed there far longer than was originally planned and they do have to be withdrawn eventually. It should not be a big deal—folks like Krauthammer who try to paint a withdrawal of a couple of hundred soldiers as a major victory for the terrorists are the ones creating a PR victory for them.

And of course the usual collection of warbloggers chimed in, adding their own exaggerations, fabrications and distortions. They accused Kerry of being a terrorist (see top of this post). They made up a story that “His shrewish sister is running around in Australia“. (No, she talked to the paper’s Washington correspondent). They called her claim “ludicrous” because the Bali bombing was before the Iraq invasion. (But if you look at the quote, it was the embassy bombing she mentioned). And on and on.

In a post that seems to parallel this one Tim Dunlop also takes Krauthammer to task.

Update: Powerline adds that Kerry’s sister “lives in Australia”. No, and she wasn’t even in Australia.

There’s an election coming up in Australia. I haven’t blogged about it because other people are doing a much better job than I ever could. A special plug though, for Tom Vogelgesang, who is running for Senate on a libertarian platform and has the good judgement to run a blog, choose to study computer science at UNSW and live in Maroubra.

However, I will comment on what the two rival candidates for Prime Minister have to say about guns. Prime Minister John Howard (Interview on 2GB 17 Apr 02: listen)

we will find any means we can to further restrict them because I hate guns. I don’t think people should have guns unless they’re police or in the military or in the security industry. There is no earthly reason for people to have … ordinary citizens should not have weapons. We do not want the American disease imported into Australia.”

(Via John Tingle. Whole interview is here).

Mark Latham, Leader of the Opposition (letter to the Shooters Party):

In conjunction with its State colleagues, Labor will work with sporting and recreational shooting organisations to control the criminal use of firearms without adversely affecting legitimate sporting and recreational shooters.

I strongly disagree with Howard and agree with Latham here. I think law-abiding sporting and recreational shooters should be allowed to have guns and banning such ordinary citizens from having weapons is wrong. Howard is also the architect of the 1996 gun ban which I have already said was bad policy. Now I had already decided to support Latham over Howard because of other issues, but it’s still nice to be supporting the party with the better policy on firearms issues. John Tingle, MLC for the Shooters Party writes:

In my opinion, it’s hard to believe we could be any worse off under Latham than we are under Howard, and, because I’m an optimist, I believe from my conversation with him and from the letter, and from frequent contact I had with him in my radio days, that if he becomes the next Prime Minister, a very real and present threat will be lifted from Australia s LAFO’s [Law Abiding Firearm Owners].

Lott has an article in the National Review Online where he claims that the Washingtonian DC handgun ban caused crime increases:

Crime rose significantly after the gun ban went into effect. In the five years before Washington’s ban in 1976, the murder rate fell from 37 to 27 per 100,000. In the five years after it went into effect, the murder rate rose back up to 35. During this same time, robberies fell from 1,514 to 1,003 per 100,000 and then rose by over 63 percent, up to 1,635.

I’ve graphed the homicide and robbery rates for the ten years on either side of the law so you can see how Lott cherry picked his numbers as usual. (Data is from here.) Notice how the crime rates fluctuate from year to year. If you choose one year at random to represent the situation after the law was passed their is a good chance that it will be unrepresentative. Of course, Lott didn’t just choose one year at random. He chose 1981, which just happens to be the year that had the highest homicide and robbery rates of the ten following years.

The law was also in effect for only part of 1976, so that year is not a good choice to represent the situation before the law.

If you look at the graphs you will see that homicides tended to be lower after the law and robberies were about the same. Of course, just looking at the graphs only gives a rough idea of the possible effects of the law. This has been studied by several researchers. Loftin at al (NEJM 325:1615-1620) found significant decreases in firearms homicides and no significant change in non-firearms homicides. Kleck et al (Law & Society Review 30(2):361-380) disputed their findings, arguing that the law had no effect. Whoever is correct, there is no published support for Lott’s claim that the law caused crime increases.

Over on his blog, Lott presents a graph showing crime rates. And it looks good for his case—it’s V shaped, with the bottom of the V right when the law was passed. Except that he graphed the overall crime rate. The majority of crimes are non-violent ones like larceny, where handguns are not a factor, even offensively or defensively. His graph isn’t even remotely relevant but if he graphed something relevant like homicide it wouldn’t support his case.

I don’t think I’ve ever written a “go and read this post” post, but go and read this post. Please.