Tyler Cowen has nerves of steel

With no apparent sign of fear or hesitation, he links to the following WSJ article:

Do Girls Speed More Than Boys?
Survey Says Girls Drive More Aggressively; Insurers Up Rates
By JOSEPH B. WHITE and ANJALI ATHAVALEY

Some big auto insurers are raising the rates they charge to cover teenage girls, reflecting the crumbling of conventional wisdom that young women are more responsible behind the wheel.

In a survey of teenage drivers, Allstate Insurance Co. found that 48% of girls said they are likely to drive 10 miles per hour over the speed limit. By comparison, 36% of the boys admitted to speeding. Of the girls, 16% characterized their own driving as aggressive, up from 9% in 2005. And just over half of the girls said they are likely to drive while talking on a phone or texting, compared to 38% of the boys..

The results were "a surprise to many people," says Meghann Dowd of the Allstate Foundation, an independent charitable organization funded by Allstate which sponsored the survey.

While teens fessed up about their own bad behavior, they also said their friends drive even worse. The study found that 65% of the respondents, male and female, said they are confident in their own driving skills, but 77% said they had felt unsafe when another teen was driving. Only 23% of teens agree that most teens are good drivers. This suggests teens recognize in their friends the dubious and dangerous behavior they won't admit to indulging in themselves.

The data was gleaned from online interviews with 1,063 teens across the country. It was conducted in May 2009 for the Allstate Foundation by the TRU division of TNS Custom Research Inc., a Chicago-based youth research and marketing firm. (For highlights of the study, see www.allstate.com/foundation/teen-driving/Shifting-Teen-Attitudes.aspx).

The survey relies on what teens report about themselves, and Allstate Foundation spokeswoman Meghann Dowd says that means the results could be affected by how forthcoming individuals are when answering the survey questions.

(read the rest here)

I don't know how he does it. I started to sweat when I hit the second paragraph and it just got worse from there. I'm not ashamed to admit that I get nervous around studies that:

1. assume that teenage boys and teenage girls are equally likely to tell the truth about their illegal driving habits;

2. assume that teenage boys and teenage girls are equally likely to be aware of their illegal driving habits;

3. have a spokesperson who talks about individuals not being forthcoming but doesn't mention the possibility that this effect might be more prevalent for one gender;

4. don't cross-check their findings against highly correlated hard data like highway patrol citation records;

5. contain at any point any variation of the phrase 'online interview.'

I have to sit down now. I'm getting dizzy...

You can always get the answer you want if you ask the right question

In today's Wall Street Journal op-ed page, Arthur C. Brooks claims that most Americans oppose progressive taxation. This would seem to contradict most recent polling on the subject but Brooks has his own data:

A 2009 survey conducted by the polling firm Ayers-McHenry asked respondents to choose which of the following statements came closer to their views: "Government policies should promote fairness by narrowing the gap between rich and poor, spreading the wealth, and making sure that economic outcomes are more equal"; or "Government policies should promote opportunity by fostering job growth, encouraging entrepreneurs, and allowing people to keep more of what they earn." Respondents chose the second option over the first, 63% to 31%.

Jonathan Chait responds:

Ayers-McHenry is a Republican polling firm, and the question Brooks cites is an advocacy poll, not a legitimate survey of public opinion. Public opinion experts understand that loading the terms of a question can produce almost any result. The question cited in this poll is filled with loaded language to an almost comical degree.

Straightforward measure of public opinion show strong public support for a progressive tax system and higher taxes on the rich. Gallup finds that Americans overwhelmingly think the rich pay too little:

Raising taxes on the rich enjoys broad support, including among Republicans, though this belief is not reflected at all among Republican elites:

The Quinnipiac University poll found that 60 percent of Americans among both major political parties think raising income taxes on households making more than $250,000 should be a main tenet of the government's efforts to tame the deficit. More than 70 percent, including a majority of Republicans, say those making more than $1 million should pay more.

Polling on progressive versus proportional or regressive taxes is hard to come by. The conservative Tax Foundation asked a loaded question last year: "Would you support or oppose the government redistributing wealth by a much higher income tax on high income earners?" The language seems clearly designed to prompt a negative response, but respondents said yes anyway, by a 52%-31% margin. (This is probably why the Ayers-McHenry poll cited by Brooks threw in even more loaded terms in order to produce the desired result.)