Surveys and Sequence

July 8, 2012
Posted by Jay Livingston

Push polls are an extreme example of the problems inherent in surveys, even surveys that are not apparently tendentious.  You ask a seemingly straightforward questions, but respondents may not be answering the question you think you asked.  That’s why I tend to distrust one-shot surveys with questions that have never been used before.  (Earlier posts on this are here and here).

Good surveys also vary the sequence of questions since Question #1 may set the framework a person then uses to think about Question #2. 

“Yes, Prime Minister” offers a useful example – exaggerated, but useful in the research methods course nevertheless.

HT: Keith Humphrey

Au Canada

July 5, 2012
Posted by Jay Livingston

After the Supreme Court’s Obamacare decision, some conservatives announced Twitterers joked that they were so fed up they were moving to Canada (Buzzfeed story here.)

Maybe they were doctors thinking about administrative costs:
A 2010 Health Affairs study found that doctors in Ontario, a Canadian province, spent $22,205 each year dealing with the single-payer agency, compared to the $82,975 American doctors spend dealing with private insurance companies, Medicare and Medicaid.
Or maybe they were patients in need of costly tests:
An MRI that costs, on average, $1,200 in the United States comes in at $824 north of the border.
(from a WaPo blog by Sarah Kliff.)

The Flag

July 4, 2012
Posted by Jay Livingston

Five years ago, Pew Research did a survey on patriotism.  The study was released with this title:
Who Flies the Flag? Not Always Who You Might Think
A Closer Look at Patriotism
Pew asked people if they displayed the flag at home, at work, or on their cars. Most of the demographic breakdowns were what you and Pew would expect.  Flag flying was more common among Whites, males, and Republicans. Apparently what surprised Pew (or its headline writer) was the regional breakdown.

As Pew says, “Notably, significantly more Northeasterners and Midwesterners fly the flag than do residents of the South or the West.”

Not to get to Clintonesque here, but maybe it depends on what  “the flag” means.  People in the South are in fact more likely to display a national flag. But the flag these patriots display is not the one in the picture above, the flag of the USA.  It is the flag of a country that fought a war against the USA – a war that killed a greater proportion of the population of the USA than has any other war in our history. (Even the absolute number of USA dead and wounded is second only to World War II.)

UPDATE:  A 2011 Pew survey found that 8% of the total sample said they displayed the Confederate flag.  If we assume that all these were in the South and add them to the 58% of Southerners in the 2007 survey who said they displayed “the flag,” the South still trails the Northeast in flagwaving, though the 3-point difference is within the margin of error. (HT: @ConradHackett)

Name, Race, and Class

July 3, 2012
Posted by Jay Livingston

The protagonist of Max Shulman’s 1957 novel Rally Round the Flag, Boys is Lt. Guido DiMaggio. He never had andy particular talent for baseball, but practically since he could walk, people were certain that any boy named DiMaggio must have baseball in his blood, so he was encouraged to play and play often.  As a result, he turned into a pretty good outfielder.*

Do names make for destiny because of the way people respond to them?  Freakonomics (2005) says it ain’t so, Joe.  Levitt and Dubner, writing about Black names, conclude that once you control for social class, names make no difference.  In the Freakonomist world, teachers, landlords, and employers are like Steven Colbert – they don’t see race. 
On average, a person with a distinctively black name . . .does have a worse life outcome . . . . But it isn't the fault of his or her name. . . .  The kind of parents who name their son Jake don't tend to live in the same neighborhoods or share economic circumstances with the kind of parents who name their son DeShawn. And that's why, on average, a boy named Jake will tend to earn more money and get more education than a boy named DeShawn. DeShawn's name is an indicator–but not a cause–of his life path.
I was skeptical about this when I read it years ago.  What about all those field tests for civil rights groups?  What about those black college grads who finally wise up and send out their resumes as D. William Green after DeShawn W. Green gets nothing but rejections? 

The problem is that we don’t know whether people are responding to “DeShawn” as a marker of race or marker of class or both. 

Now, S. Michael Gaddis has taken a step towards untangling the race and class in names.  He finds that some Black names are associated with more education, some with less. The same goes for some distinctively white names.  Nearly four out of five Jalens, for example, are Black, but 61% of Jalens have gone to college.  Ronny is mostly white and mostly dropout.

Gaddis went job hunting over the Internet using these names.  He looked at who was offered an interview and at what salary range.  On both outcome variables, race and class both made a difference. 
Moreover, the race- and class- based penalties compound for low-SES black males.  In other words, Jalen and DaQuan are both disadvantaged on the job market compared to Caleb and Ronny, but DaQuan is by far the most severely disadvantaged.  Worse yet:  the situation between white and black candidates does not change whether they are graduates from less selective schools like UMass and UC Riverside or highly selective schools like Harvard and Stanford.
Gaddis has a brief write-up of his research here.

HT: A tweet from Philip Cohen.

* Readers of the first chapter of Malcolm Gladwell’s Outliers may hear an echo in this story.  Gladwell is writing non-fiction about hockey and age; Shulman is writing fiction about baseball and names.  But the “culling” effect  is similar.