WEBLOG

Previous Month | RSS/XML | Current | Next Month

April 29th, 2009 (Permalink)

Poll Watch

An editorial in The Washington Times claims that the President is less popular than he is portrayed by the media:

According to Gallup's April survey, Americans have a lower approval of Mr. Obama at this point than all but one president since Gallup began tracking this in 1969. The only new president less popular was Bill Clinton…. Mr. Obama's current approval rating of 56 percent is only one tick higher than the 55-percent approval Mr. Clinton had…. [F]ive presidents rated higher than Mr. Obama after 100 days in office. Ronald Reagan topped the charts in April 1981 with 67 percent approval. Following the Gipper, in order of popularity, were: Jimmy Carter with 63 percent in 1977; George W. Bush with 62 percent in 2001; Richard Nixon with 61 percent in 1969; and George H.W. Bush with 58 percent in 1989.

The editorial doesn't link to its source for these numbers, but it may have taken them from an article by Judith Klinghoffer. There are two problems with Klinghoffer's analysis:

  1. Klinghoffer seems to have compared apples to oranges: the historical numbers for past presidents are average approval ratings, whereas the number that she cites for Obama―56%―comes from a different question in another survey. Obama's current average approval rating is 65%. However, that is not the worst problem:
  2. Whether Obama's rating is the next-to-lowest or next-to-highest doesn't really mean much, since the margin of sampling error for these polls is the usual ±3%. The difference between the lowest and highest ratings is 12 percentage points, which is significant, but Obama's rating is only significantly different than Clinton's and the first Bush's. About the most that one can say is that Obama is in the front of the pack, rather than trailing it, as Klinghoffer and the D.C. Times have it.

Sources:

April 26th, 2009 (Permalink)

New Book: Annals of Gullibility

Here's Stephen Greenspan on his new book:

In my new book Annals of Gullibility, I analyze the topic of financial scams, along with a great number of other forms of human gullibility, including war (the Trojan Horse), politics (WMDs in Iraq), relationships (sexual seduction), pathological science (cold fusion), religion (Christian Science), human services (Facilitated Communication), medical fads (homeopathy), etc.

That sounds both interesting and topical. Greenspan actually lost money in Bernard Madoff's pyramid scheme, despite being an expert on gullibility! Of course, that makes you wonder what good being an expert did. However, being skeptical or a good critical thinker is no guarantee against getting taken; at best, it lowers the odds of it happening. How did Greenspan and others get taken in?

The basic mechanism explaining the success of Ponzi schemes is the tendency of humans to model their actions (especially when dealing with matters they don’t fully understand) on the behavior of other humans. … Simply stated, the fact that so many people seem to be making big profits on the investment, and telling others about their good fortune, makes the investment seem safe and too good to pass up.

In other words, people―including Greenspan―jumped on Madoff's bandwagon, not realizing that it was headed for a cliff.

This sounds as though it would make a good Book Club book, if only some nice person would send me a review copy.

Source: Stephen Greenspan, "Fooled by Ponzi (and Madoff): How Bernard Madoff Made Off with My Money", eSkeptic, 12/23/2008

April 25th, 2009 (Permalink)

Another Puzzle of Another Day

Clues:

  1. Cranston Smoot, Jr. works at Biggie Burger.
  2. Some Biggie Burger employees enjoy a good game of whist.
  3. Only ovo-lacto-vegetarians read Proust in the original French.
  4. No ovo-lacto-vegetarians both work at Biggie Burger and read Proust in the original French.

Does Cranston Smoot, Jr. read Proust in the original French?

Solution

April 24th, 2009 (Permalink)

Blurb Watch: The Informers

An ad for the new movie The Informers is an education in itself on how to find blurbs for poorly-reviewed movies. With a Metacritic rating of 18, or "Extreme dislike or disgust", and a "Rotten" 16% on the Tomatometer, the ad writer had some hard work to do. Take the first blurb in the ad:

"Part morality tale, part voyeuristic time trip.
Strong stuff!"
-Michael Cieply, The New York Times

This blurb uses a trick that we've seen before: when you can't find a good review to quote, look elsewhere. A.O. Scott actually reviewed this movie for The Times saying, among other things: "It has nudity, sex, drug use and violence. And it’s still boring." The quotes from Cieply, in contrast, are taken from a report on the Sundance movie festival, rather than a review, and are just a description. Here's the second blurb:

'Pitch-perfect. Might make you choke laughing.'-Joshua Rothkopf, Time Out New York

This is another old trick. Though this may be the most favorable review of the movie, it's four stars out of a possible six, which the ad does not tell you. The next blurb is:

"Impresses with style."
-Troy Patterson, Spin

This is a classic contextomy, and here's the missing context: "Catching the book's tone, it impresses with style even as it alienates."

Sources:

April 18th, 2009 (Permalink)

Stat Wars

Interest groups with agendas to advance often exaggerate or play down the statistics of a problem, depending upon whether they want to alarm or soothe the populace. Of course, politicians play the same game, and even some journalists. When they are on opposite sides of an issue, a stat war results.

What percentage of guns used in crimes in Mexico come from the U.S.? Is it 90%, as the President has claimed? Or, only 17%, as Fox News has reported? A good guess would be somewhere in between. Find out the answer in Fact Check's latest report.

Source: D'Angelo Gore, "Counting Mexico's Guns", Annenberg Political Fact Check, 4/17/2009

'Has moments so good' - Roger Ebert, Chicago Sun Times
April 17th, 2009 (Permalink)

A Gigantic Contextomy

On the basis of the blurb for the new movie Gigantic, the ad writer can make a fine and original contextomy. This isn't quite it, but "has moments so good"―all you wish for is a second draft. The blurb sounds a bit less than enthusiastic, and the ad writer has failed to put it all in caps and add an exclamation point, which is ad writer 101. "HAS MOMENTS SO GOOD!" That's better, but still a little lukewarm. Perhaps it's not surprising given that Roger Ebert gave the movie a thumbs down and only two-and-a-half stars out of four.

Here's the quote in context:

On the basis of "Gigantic," Matt Aselton can make a fine and original film. This isn't quite it, but it has moments so good, all you wish for is a second draft.

If you're going to contextomize a critic who doesn't even recommend the movie, why do such a poor job of it? With only the above quote to work with, we can come up with a better blurb:

A FINE AND ORIGINAL FILM!
―Roger Ebert, Chicago Sun Times

I have a feeling that the ad writer is just a little too honest for the job.

Source: Roger Ebert, "Gigantic", Chicago Sun-Times, 4/9/2009

April 16th, 2009 (Permalink)

Check it Out, Too

The current issue of Scientific American Mind has a good article by Gerd Gigerenzer, et al., on probabilistic fallacies―including the base rate fallacy―and how they affect our understanding of health statistics.

Source: Gerd Gigerenzer, Wolfgang Gaissmaier, Elke Kurz-Milcke, Lisa M. Schwartz & Steven Woloshin, "Knowing Your Chances: What Health Stats Really Mean", Scientific American Mind, April 2009

April 13th, 2009 (Permalink)

Argumentum ad Litem

I mentioned previously (see the Resource, below) that Dr. Ben Goldacre had won a lawsuit brought against him by a snake oil huckster. What I didn't mention, because I didn't know it, is that Goldacre had been forced by the lawsuit to omit a chapter on the plaintiff from his Bad Science book. Now, Goldacre has made that chapter available free online. Be sure to check it out, as it's a remarkable cautionary tale, and a good answer to anyone who wonders where the harm is in bad logic and bad science.

As I also mentioned before, such lawsuits are a form of argumentum ad baculum―one might call it "argumentum ad litem", i.e., "argument by lawsuit". In this case, the attempt to legally suppress criticism was thankfully only temporarily successful, but Goldacre and the Guardian newspaper were forced to spend a lot of time and money defending themselves. Even though they won the battle, the plaintiff may win the war by successfully deterring others from publicly criticizing him. You might consider the possibility of helping Goldacre pay his legal bills by purchasing his book.

Another familiar fallacy plays a role in the chapter, namely, the appeal to nature, as advocates of vitamins contend that their pills are natural, but drugs are unnatural. While reading it, keep in mind that viruses are also natural.

Source: Ben Goldacre, "Matthias Rath―Steal this Chapter", Bad Science, 4/9/2009

Resource: Too Good to Check it Out, 9/6/2008

April 10th, 2009 (Permalink)

"I'm not a firefighter, but I play one on TV."

Slate's "Explainer" column answers the burning question: Do firefighters believe 9/11 conspiracy theories?

Here's a question about that question: Does it mean "all" firefighters or only "some"? If the former, then of course firefighters don't believe in such theories; if the latter, then of course firefighters believe in such theories. No doubt at least one firefighter is a conspiracy theorist, and at least one is not. Why do we need an article to answer this question?

Here are some other questions the "Explainer" might answer while he's at it: Do dentists believe 9/11 conspiracy theories? Is there a "Dentists for 9/11 Truth" group? Ask a stupid question, and what are you going to get?

Apparently, the question was inspired by an actor who is a 9/11 conspiracist and plays a firefighter on a television show. But even if you could find a firefighter somewhere who shares his beliefs, so what? There's no reason to think that firefighters know more about what happened on 9/11 than anyone else. The fact that many firefighters were killed that day is tragic, but it doesn't somehow invest other firefighters with expertise. Of course, a firefighter who was actually there that day and survived is a witness, but the article gives no indication that any such survivor is a conspiracist.

This is an example of a common type of appeal to misleading authority: we see it in the way that the media treats the parents of a sick child as if they are experts on the causes and treatments of their child's illness. I suspect that this attitude is partly the result of sympathy for the parents' plight so that if, say, they begin babbling about how their child became autistic due to vaccines, we hesitate to disagree with them. Similarly, our admiration for the courage of firefighters and respect for the great sacrifice they made on 9/11 may lead us to humor them if they start spouting conspiracy theories.

The real experts on 9/11 are physicists, engineers, historians, and others trained to evaluate evidence, not firefighters. And certainly not actors who play them on TV.

Source: Christopher Beam, "Heated Controversy", Slate, 4/8/2009

April 8th, 2009 (Permalink)

Straw Poll

John Dickerson has an article in Slate on the President's tendency to attack straw men. As I've mentioned previously, straw man arguments are par for the course in politics. However, the article does tend to show up Dana Milbank's claim that the previous holder of the office attacked straw men more than other politicians.

Dickerson himself claims that Obama sees grey where Bush saw everything in black and white. However, just as some of the earlier examples of Bush's straw man arguments were black-or-white fallacies, Obama also often paints in stark contrasts. For instance, when Obama said "Some of what's been said in Congress is that there seems to be a set of folks who just believe that we should do nothing" about the economy, he was attempting to portray the choice in black-or-white terms: either we pass the specific legislation that Obama supports, or we do nothing. Of course, there are many other courses of action possible, but Obama claims it's his way or the highway. Even given the unrealistically limited set of choices, hasty and ill-considered action may be worse than doing nothing.

At the end of the article, Dickerson engages in some exaggeration of his own:

According to a recent New York Times/CBS News poll, Obama is as popular as ever. And his Republican opponents in Congress received their lowest approval rating in the entire span of history in which that question has been asked.

By "popular", I assume Dickerson is referring to Obama's "approval" rating (Question 1). Obama's 66% approval rating in the poll is within the three percentage point margin of error (MoE) of his previous ratings as President, so it's correct to describe it as unchanged. However, his disapproval rating has climbed from a low of 15% two months ago to 24%, which is well outside the MoE. "Popularity" would seem to have as much to do with disapproval as approval.

By "approval rating" of Republicans in Congress, I guess that Dickerson is referring to a "favorability rating" of the Republican Party (Question 31). The "favorable" rating is indeed at it's lowest percentage point since the poll began measuring it, namely, 31%. However, it's not significantly changed since July of 2007, and the "unfavorable" portion of the rating was actually higher in the previous poll, though again not significantly.

Sources:

Resources:

Reader Response (4/22/2009):

Regarding your post about Obama's use of straw men. I think it is fair to say that he uses straw men less than Bush for a couple of reasons.
  1. First of all, I do not fault Bush for rather innocuous exaggerations common to all politicians, including Obama. I recognize that some who make this argument are guilty of cherry-picking their examples, focusing on Bush's straw men and ignoring Obama's. But that is not what my argument is based on.
  2. Bush had an entire policy built around a straw man/false dichotomy. His infamous line, "you're either with us or against us in the fight against terror," implied that anyone who disagreed with his policies was promoting terrorism. That this view was taken to heart was evident through the administration's dismal interaction with the international community.
  3. Obama, when not reading from a script, actually includes nuance, and acknowledges that his opponents may have valid criticisms to make. The example that comes to mind was his first presidential press conference, during the Q&A session. While I certainly have not spent the time to demonstrate that Bush did not do so in similar circumstances, at the very least they were overshadowed by more extreme statements.

Anyway, I don't think anyone who is serious about this is claiming that Obama is a completely different kind of politician. Just that in general he seems more willing than most to include qualifiers and shades of gray. Without detailed, controlled analyses of the political speeches of recent presidents, it is of course impossible to state with certainty anything one way or the other, but I think at the very least, the impression that Obama can acknowledge mistakes and legitimate criticisms, even if it's a false impression, is a helpful step forward in the political discourse. If it is seen as important and laudable, that can lead more people to take that into consideration when forming their political opinions.―Daniel Hawkins

The trouble with the claim that one politician is worse than another in attacking straw men, black-or-white thinking, or other logical fallacies, is that it's a subjective judgment call. As such, it tells you more about the judgment of the person who makes the call than about the politician. That's one reason why I try not to make such judgments―at least, not publicly―because they would exceed the limits of my knowledge.

In general, people tend to notice the flaws in other people's arguments more readily than in their own, especially the arguments of their ideological or political opponents. Frequently, it's the fact that we disagree with the conclusion of an argument that leads us to examine the reasoning closely enough to spot weaknesses; if we agree with the conclusion, we're liable not to pay much attention to the argument itself.

This is an unfortunate situation, but it's not entirely unreasonable. We live in an atmosphere of arguments, most of which we barely notice, and even logicians don't have the time or energy to carefully evaluate all of them we come across. It's not irrational to only stop to examine those that we have independent reasons for thinking must be bad in some way, and which are in danger of leading us astray if we do accept them. Unfortunately, this leads to the tendency to uncritically accept arguments that reinforce our prejudices, which is a specific form of what psychologists call "confirmation bias". As a result of this cognitive bias, we tend to notice every fallacy committed by those we disagree with, but ignore or excuse those from our own side.

How could one measure the fallacies committed by different politicians while eliminating, or at least minimizing, this kind of bias? First, gather a random sample of the politicians' arguments, edit them in order to conceal the politicians' identities, and then have more than one person trained at the task sift through the arguments for instances of fallacies. Ideally, the judges would have varying political views so that their individual biases would cancel out.

However, even though such a study is theoretically possible, I don't see what scientific value it would have. What would be the use of a conclusion such as: "George W. Bush showed more black-or-white thinking than Barack Obama, but Obama attacked more straw men"? It might have political value if it went all one way, but what value would it have for the study of logical fallacies?

Furthermore, the claim that Bush engaged in black-or-white thinking more than other politicians is not supported by the evidence usually cited. Specifically, the "infamous" with-us-or-against-us quote is the product of quoting out of context, which is why I have never cited it as an instance of the fallacy. Here's the quote with some additional context:

…[W]e will pursue nations that provide aid or safe haven to terrorism. Every nation in every region now has a decision to make: Either you are with us or you are with the terrorists. From this day forward, any nation that continues to harbor or support terrorism will be regarded by the United States as a hostile regime.

It's clear in context that Bush was not claiming that there was no third alternative, but announcing a policy of forcing other nations into taking sides. This is certainly questionable as a policy, and in retrospect doesn't appear to have been effective, but it's not a logical fallacy. In other words, what's wrong with it is not a matter of logic, but of policy.

This is not to say that Bush did not commit examples of black-or-white reasoning, nor attack straw men. The two articles linked to in the Resources above give genuine examples of both types of fallacy (I've restored the link, which had gone dead, to the Jennifer Loven article).

Moreover, I'm not arguing that Bush was no worse about attacking straw men or engaging in black-or-white thinking than Obama is. Rather, I'm specifically not taking a position on that issue, since the only evidence that could decide the matter is the type of study I outlined above, and that has not been done so far as I know. Perhaps you're right and Bush was worse, but the argument to that conclusion is very weak, if not actually fallacious.

Source: George W. Bush, "Transcript of President Bush's Address", CNN, 9/21/2001

April 7th, 2009 (Permalink)

Check it Out

John Allen Paulos' latest Who's Counting? column contains a propositional logic problem that you might want to try solving, but he doesn't include the solution so far as I can find. If you want to tackle the problem and check that you got the right answer, click on the solution:

Solution

Source: John Allen Paulos, "Seduction 101 for Logicians and Economists", Who's Counting?, 4/5/2009

April 3rd, 2009 (Permalink)

"Democratic" Doublespeak

"Such [Newspeak] words, for instance, as joycamp (forced-labor camp) or Minipax (Ministry of Peace, i.e., Ministry of War) meant almost the exact opposite of what they appeared to mean."―George Orwell, "Appendix: The Principles of Newspeak", 1984 (p. 135)

Slate gives us a brief history lesson about the use of the terms "democratic republic" and "people's republic" in the names of communist dictatorships. Examples include the current "Democratic People's Republic of Korea", which manages to combine both in its name, and the former "Democratic Kampuchea", which was the name of Cambodia under the Khmer Rouge.

"People's Republic" is the kind of doublespeak that George Orwell satirized in his novel 1984 by such names as "the Ministry of Truth" for the government propaganda agency, and "the Ministry of Love" for the secret police―that is, it's a name that means the opposite of what it says. Here's what Orwell had to say specifically about "democracy":

It is almost universally felt that when we call a country democratic we are praising it: consequently the defenders of every kind of regime claim that it is a democracy…. Words of this kind are often used in a consciously dishonest way.―George Orwell, "Politics and the English Language"

Check it out.

Sources:

  1. Juliet Lapidos, "The Undemocratic People's Republic of Korea", Slate, 4/1/2009
  2. George Orwell, Orwell's Nineteen Eighty-Four: Text, Sources, Criticism; Irving Howe, editor (1963)
  3. George Orwell, "Politics and the English Language" (1946)
April 1st, 2009 (Permalink)

Puzzle of the Day

A man went into the hardware store in the small town where he lived and spoke to a clerk. "How much for seven hundred?" he asked, pointing. "Three dollars," answered the clerk. "So, they're a dollar apiece?", the man asked. "That's right," said the clerk. "I'll take three," said the man. How many blocks did the man live from Main Street?

Solution

Source: Martin Gardner, Aha! Insight (1978). The puzzle is based on one from page 73.

Solution to Another Puzzle of Another Day: No. Venn Diagram

There are different ways of proving this, but one of the easiest is with a Venn diagram:

The first step is to notice that the second clue provides no useful information about Cranston, because the class of those who enjoy a good game of whist does not occur in any other premiss. So, the class of whist-players can be left out of the diagram, leaving three classes: Biggie Burger employees (B), ovo-lacto-vegetarians (V), and Proust-readers (P). The third clue tells us that all P are V―not the other way around, as many people are inclined to think; for instance, "only men are fathers" means "all fathers are men", not "all men are fathers". So, on the diagram the part of the P circle outside the V circle is shaded (blue). The last clue says that the part of the V circle inside both the B and P circles is empty, so it is shaded (red). Finally, the first clue tells us that Cranston (c) is inside the B circle. We don't know whether he is in the V circle, so two Cs are placed in the B circle, one inside the V circle and one outside, connected by a line to show that there is only one Cranston but we don't know where he is.

Now, we have diagrammed all of the useful information in the clues. What answer does the diagram give to the question? It shows Cranston (c) is outside of the Proust-reading circle (P). Therefore, Cranston Smoot, Jr. does not read Proust in the original French.

Solution to the Paulos Problem: Invalid

Solution to the Puzzle of the Day: This is another puzzle based on the use/mention distinction; see the Resources below for previous use/mention puzzles and an explication. The man went into the hardware store to buy numerals to put on his house. He asked how much "700" was, because that was his house number. The clerk told him that "700" cost $3, which meant that the numerals were $1 apiece. Since his house is number 700 that means it is in the seventh block from Main Street.

Resources:

Previous Month | RSS/XML | Current | Next Month