In recent years, social scientists have tried to find out whether important decisions are shaped by subtle biases. They've studied recruiters as they decide whom to hire. They've studied teachers, deciding which students to help at school. And they've studied doctors, figuring out what treatments to give patients. Now, researchers have trained their attention on a new group of influential people — state legislators.
Christian Grose, a political scientist at the University of Southern California, and graduate student Matthew Mendez wanted to see if state legislators were equally responsive to their constituents. For part of their experiment, the researchers sent emails to 1,871 legislators in 14 states with large Latino populations, asking the politicians what kind of documentation they needed to vote. They randomly assigned legislators to get the emails, but some emails came from a man named Jacob Smith, and others came from a man named Santiago Rodriguez.
"No one had really looked at sort of what underlies legislator behavior," Grose says. "Is there the possibility that legislators' own biases regarding race and ethnicity might rear their heads and that legislators might ignore Latino constituents more than white constituents?"
NPR's social science correspondent Shankar Vedantam dug into the study. Here are some highlights with some additional context.
On the study's findings
There was a difference, and the difference had a partisan tinge to it. Democrats responded about the same to both names, but Republicans were more likely to respond to the man with Anglo name rather than the Latino name. ...
Grose told me that he and Mendez decided to look a little bit deeper at the data and they found something very interesting. In many ways, the difference was less between Republicans and Democrats and more between some Republicans and other Republicans.
The researchers analyzed whether the Republicans in these states sponsored or co-sponsored voter ID laws. Now these are laws that are designed to reduce voting fraud. Critics of these laws have said that they disenfranchise minorities and others who are trying to vote for Democrats. Grose said that there was very strong correlation between Republicans who had failed to respond to the Latino constituent, and the ones who sponsored such laws.
"Republicans who support voter identification are different than those Republicans who did not support voter identification," Grose says. "Among those Republicans who did support voter ID laws, the Latino constituent was very unlikely to receive a response from their elected official. The difference was almost 40 percentage points, which is just one of the largest gaps I have ever seen."
On how to interpret the study's implications
An implication of the study is that the same bias that caused legislators not to respond to a Latino constituent also drove them to sponsor voter ID laws ... but let me put it into context in a couple ways. The first thing is, lots of legislators — both Republicans and Democrats — did not get back to either Jacob or Santiago. So if a legislator is unresponsive, it does not automatically mean that he or she is biased.
Second, this research does not establish cause and effect when it comes to voter ID laws. It's fair to say the Republicans who sponsored such bills seem to be biased when it comes to responding to the Latino name versus the Anglo name.
But we don't know if that bias is what prompted them to sponsor the voter ID laws. That might be an inference, that might be a correlation, but it's not a proven fact.
STEVE INSKEEP, HOST:
Let's get a new data point in the debate over voter ID laws. Those laws are a huge issue as the fall election nears, with down-to-the-wire court battles in several states. Advocates of those laws - mostly Republicans - say, they prevent voter fraud. Critics - mostly Democrats - say, the laws are designed to keep Democratic-leaning young people, poor people and minorities from voting at all. That's the debate. Now we have some research into the behavior of legislators who sponsor voter ID laws. NPR's social science correspondent, Shankar Vedantam, is covering the story. Hi, Shankar.
SHANKAR VEDANTAM, BYLINE: Good morning, Steve.
INSKEEP: What's the new research?
VEDANTAM: Well, it's part of a broader area of studies in the social sciences that looks at the question of bias, Steve. And when most of us think of bias, we think about conscious bias. A lot of this new work has focused on a different kind of bias - unconscious bias. There's been research showing unconscious bias is affecting how police officers relate to crime suspects or how doctors treat patients. And, you know, you and I, Steve, we did a story looking at how faculty members at universities respond differently to students of different groups. Many of these earlier studies have used tests that measure unconscious biases, and they find the stronger the unconscious bias, the stronger the bias in behavior. There's a new data point now, and it looks specifically at the behavior of state legislators who support voter ID laws. This study doesn't specifically test for unconscious biases, but it's in line with these other studies that suggest unconscious bias might be at work.
INSKEEP: Unconscious bias - so that's the idea that someone is not overtly prejudiced, doesn't believe that they're racist, but because of something going on in their mind, they treat people of different races in different ways.
VEDANTAM: That's exactly right, Steve. And there's been a lot of research to support that idea over the last 20 years. This new study was done by Christian Grose. He's a political scientist at the University of Southern California. He did the study with Matthew Mendez because he wanted to find out if the same phenomenon is at work with state legislators. Here he is.
(SOUNDBITE OF ARCHIVED RECORDING)
CHRISTIAN GROSE: No one had really looked at sort of what underlies legislator behavior. Is there a possibility that legislators own biases regarding race and ethnicity might rear their heads and that legislators might ignore Latino constituents more than white constituents?
INSKEEP: OK. So they're asking if Latinos are pushed aside by legislators, which is, by definition, going to be a controversial topic. What did the research find?
VEDANTAM: Well, Grose and Mendez sent out emails 1,871 legislators in 14 states with large Latino populations. These would be states like California, Texas, Florida, New York, Illinois. And including in this group were legislators who had either sponsored or cosponsored voter ID laws. So about one-third of all Republicans have sponsored or cosponsored voter ID laws. The other two-thirds have not. And Democrats overwhelmingly do not support these laws.
All the e-mails asked the legislators whether a driver's license was needed to vote, but there was a catch. Some of the e-mails came from so a man named Jacob Smith, and some of them came from a man named Santiago Rodriguez. Now, no state actually requires a driver's license, so the correct answer to all these e-mails was a simple no. But Grose's question was would legislators respond differently to the Anglo name versus the Latino name? And if they do, then a racial bias might be at work because the only thing different between the e-mails was that one of them had an Anglo name, one of them had a Latino name.
INSKEEP: So legislators are being asked for help in these emails, and the question is whether they will respond differently to people with different names. What did the research find?
VEDANTAM: Well, you might have expected Democrats to respond more often to the Latino name because Latinos tend to vote Democratic. But Grose found the real difference was between Republicans who sponsored or cosponsored voter ID laws and Republicans who didn't. Here he is.
(SOUNDBITE OF ARCHIVED RECORDING)
GROSE: Among those Republicans who did support voter ID laws, the Latino constituent was very unlikely to see a response from their elected official. The difference was almost 40 percentage points, which is just one of the largest gaps I've ever seen.
INSKEEP: OK. Could unconscious bias then explain why some Republicans are sponsoring voter ID laws in the first place?
VEDANTAM: You know, that could be an implication of this research, Steve, but I don't think the data actually makes that case. I think the value of the study, Steve, is that that this is a data point. And in all the other fields I told you about - medicine and universities and police departments - there are major efforts underway to try and prevent unconscious biases from affecting decision-making. The hope of this research is that by showing people about the biases they may not know they had, people who want to do something about it will be able to do so.
INSKEEP: Shankar, thanks, as always.
VEDANTAM: Thank you, Steve.
INSKEEP: That's NPR's Shankar Vedantam. You can find him on Twitter at @HiddenBrain. Find this program at @MorningEdition and at @NPRInskeep. Transcript provided by NPR, Copyright NPR.