Are Long Commutes Bad For Democracy?

Policy wonks across the ideological spectrum tend to agree that cities should increase their housing density by altering or eliminating certain zoning regulations. Because city dwellers have higher productivity, more housing allows more people to earn the higher wages that accompany such productivity. Furthermore, as Edward Glaeser’s work has shown, living in cities tends to enhance cognitive capacity, creativity, and cooperation. In short, people in urban centers are closer to good stuff.

Denser cities also lead to shorter commutes, and a new study (pdf) led by UConn’s Benjamin Newman suggests this creates another benefit: Citizens who are more politically engaged.

Newman found that time spent commuting — but not time spent working — leads to declining political participation, and that the negative impact of long commutes is largely focused on the poor.

Our results indicate that, even after controlling for a variety of relevant individual and contextual factors, time spent working exerts no impact on one’s level of participation. An increase in time spent commuting, however, is found to lead to a significant decrease in participation.


We find that political interest serves as a significant mediator between commuting and political participation, but that this mediated effect is itself moderated by individual income. Among lower income Americans, a longer commute leads to a significant erosion of interest in politics, and this in turn leads to significant decreases in participation. Among higher income Americans however, the relationship is reversed; increased time spent commuting among those with the highest income is found to significantly enhance interest, which in turn, increases levels of participation.

The reasoning is that commuting creates psychological strain above and beyond other time-consuming situations, and that this strain depletes cognitive or emotional resources that could be put into political activism. The high salaries the rich earn in return for their commuting time helps mitigate this strain, and thus the effect of commuting is focused on the poor.

If long commutes do disproportionately decrease political involvement, they have the potential to create a negative feedback loop. A lack of housing in urban centers will push the poor further out into the suburbs, which will create a longer commute, which will decrease the political activism necessary to bring about progressive policies aimed at increasing the housing stock in urban centers, which will ensure the poor continue to get pushed out into the suburbs. Building enough housing thus becomes even more urgent.

More broadly, the study highlights the absurdity of the notion — albeit one that’s largely confined to Fox news — that low-income people are illegally usurping political power through things like voter fraud and the nefarious acts of ACORN. In reality, our political and social systems stack the deck against low-income voters, and as a group they require significant assistance to merely reach the same level of influence as wealthier citizens. That the daily struggle to get to work inflicts more harm on the political involvement of the poor is just one more example of it.
Newman, B.J., Johnson, J., & Lown, P.L. (2013). The “Daily Grind”: Work, Commuting, and Their Impact on Political Participation American Politics Research DOI: 10.1177/1532673X13498265


How Can We Make School Less Threatening?

I have a new article in Pacific Standard about the power of threatening information — if you haven’t read it, do it now. (Seriously, go read it.) While the article focuses on how self-image concerns create a disconnect between what a policy’s supporters perceive to be effective arguments and the arguments that are likely to actually persuade their opponents, the desire to mitigate threats and maintain self-worth is important in a number of different domains, including education.

Two lines of research that can be traced back to theories about self-worth have received a good deal of attention from educators and education researchers. The first is Carol Dweck’s work on “mindsets” and the benefits of believing intelligence in malleable. The basic idea is that if you believe your intelligence is fixed at birth, trying and failing is extremely threatening because it’s evidence you are eternally stupid. On the other hand, if you believe it’s possible to increase your intelligence, failure is much less threatening (pdf) because it’s only a snapshot of where you stand in that specific moment. The research of Dweck and her collaborators is couched in a variety of different terms and phrases, but its theoretical basis is in making the act of learning less threatening.

The second and more ostensible threat-related work deals with “stereotype threat,” a problem that can arise when the anxiety of confirming a negative stereotype (e.g. girls are bad at math) causes somebody (e.g. a girl taking a math test) to perform poorly. Stereotype threat presents a slightly different situation because the threat is not strictly posed to the individual, but it’s another popular example of how the school day presents students with an environment from which they might want to disconnect. Unfortunately, while researchers have created a number of interventions that successfully mitigate the threats posed by bad mindsets and stereotypes, not much is being done on a large scale to bring such interventions into the traditional school day.

Perhaps more troubling, there are a number of additional overlooked ways that school can create a threatening environment for students. For example, one of the simplest messages schools send to students is that there’s a world of stuff they don’t know. We take this for granted when it comes to kids, but nobody of any age wants to be told about how much they don’t know. It’s a scary notion. One reaction kids might have is convincing themselves that there’s not actually all that much they don’t know, and a step in that direction is a step away from the idea that classroom learning is a valuable opportunity.

Another threat posed by the school day involves other students. Even if a student has an adaptive “growth” mindset with regard to intelligence, they’re still getting potentially threatening information about where their intelligence ranks within their peer group. And when something poses a threat, people respond by derogating it. For example, in one study (pdf) meat-eaters rated vegetarians less favorably after imagining how the vegetarians would view them. The meat-eaters’ moral standing faced a threat from the vegetarians’ judgment, and so the meat-eaters responded by characterizing the vegetarians in a more disparaging way. When it comes to learning, it seems plausible that if a student sees school as a threat to their social or intellectual standing, they’ll respond by downplaying its importance or relevance.

The knee-jerk reaction to the threat posed by peer comparison is to blame standardized testing — after all, being directly told that you’re “not proficient” is the epitome of a threatening situation. But I think a focus on testing misses the forest for the trees. Kids are clever. They know who’s raising their hands, who’s getting questions right, and who’s doing a good job on class assignments. Even if we get rid of 80% of the accountability-driven standardized testing kids are going to know where they stand. Regardless of how much testing there is, each day of school presents an opportunity for kids to be reminded that other students are achieving goals they still haven’t reached.

I suppose my broader point is that it would be beneficial to shift some of the focus on buzzwords like “motivation” and “engagement” to the underlying psychological factors that are likely to produce them. I’m hopeful that in the next 20 years we’ll see legitimate efforts to re-design the traditional school model we’ve had for the last half-century, and it would be great if such efforts started from a point of creating an experience that poses less of a threat to a student’s self-image or self-worth. This might involve giving students more choices or independence to make peer comparisons less obvious. It might involve changing the way we talk about education to emphasize how young students are consistently successful in putting their education to good use. Frankly, I don’t have a slam-dunk research-supported idea for a less-threatening core model. That’s why the previous to sentences are painfully broad. Nevertheless, the right design or innovative changes could do a lot to encourage kids to approach school less defensively.
Minson, J.A., & Monin, B. (2011). Do-Gooder Derogation: Disparaging Morally Motivated Minorities to Defuse Anticipated Reproach Social Psychological & Personality Science DOI: 10.1177/1948550611415695

Blackwell, L.S., Trzesniewski, K.H., & Dweck, C.S. (2007). Implicit Theories of Intelligence Predict Achievement Across an Adolescent Transition: A Longitudinal Study and an Intervention Child Development DOI: 10.1111/j.1467-8624.2007.00995.x

The Search For Rampant Voter Fraud Continues

And still nothing:

Although voter ID laws have become a hot topic of political debate, existing scholarship has failed to produce conclusive evidence concerning the existence or frequency of electoral fraud, especially the type of fraud that would be prevented by photo identification laws and signature verification protocols for voting by mail. We propose a new method of measuring election fraud, especially identity fraud, that is superior to previous measurement efforts because it measures actual instances of fraud rather than waiting for conclusive proof of fraud produced in a criminal prosecution. We test our method in multiple jurisdictions, including two known cases of electoral fraud, and we find no additional cases of fraud. We speculate that public access to voting and registration records play an important role in preventing this type of election fraud, suggesting that these practices are perhaps more important than voter ID laws in preventing election fraud.

How Stop-and-Frisk Affects White People

A confluence of unrelated events — New York City’s mayoral election, NYC Police Commish Ray Kelly being floated as a potential Dept. of Homeland Security Chief, a big judicial ruling — have put stop-and-frisk policies under the microscope. (Also, let’s not forget Joan from the Upper West Side.)

Much has been written about the negative impact such policies have on Black and Hispanic communities. And much has been written about the constitutionality of randomly searching people on the street. But how does stop-and-frisk affect those who aren’t being stopped or frisked. Or as they like to say in the halls of the Capitol, “how does this policy affect White people?”

A new study set to be published in Law and Human Behavior suggests that racial profiling makes white people more likely to engage in illicit behavior. The study, which was led by Georgia Southern’s Amy Hackney, used an experimental setting in which groups composed of White and Black students were tested on their ability to complete difficult anagrams. Students had access to an answer key and graded their own exams, and so they had ample opportunity to cheat. The manipulation occurred right before the test began, when the experimenter stared directly at two students and explained that cheating would not be tolerated. He then asked both students to move their desks closer to the front of the room. In one condition the two students were Black. In another condition the two students were White. A third condition with no profiling functioned as a control.

Examining two types of students (White and Black) in three different conditions produced produced six measurements of cheating frequency. They were all fairly similar, but there was one exception. White students cheated more when they saw Black students being profiled. They cheated at a higher rate than black students who saw Whites profiled, Whites who saw Whites profiled, or either group when there was no profiling.

We theorized that heightened surveillance of members of a minority group would increase illicit activity in the majority group—that it would have a reverse deterrent effect. We found that White participants in the Black-profiled condition cheated more than participants in any other condition. Although cheating on a test of this sort is not a crime, it is a dishonest behavior that is a particularly serious transgression in academic settings. These results indicate that racial profiling could increase crime among nonprofiled groups, having a counterproductive effect.

Hackney reasoned that seeing others get profiled increases your feeling of impunity, but there doesn’t have to be an overwhelming feeling of impunity for profiling to affect non-profiled groups. Even if people aren’t thinking, “I’m white, therefore I can get away with this,” they still may come to believe their actions are more justifiable. Eventually a bar fight becomes mischief rather than assault, or stealing that piece of art your friend will love is an act of kindness rather than breaking and entering. Perhaps you manage to rationalize stealing a marble rye from an old lady.

When talking about stop-and-frisk it’s easy to get caught up in the intricacies of crime statistics and constitutional law. But it’s important to remember that these policies influence social norms and the way people see themselves. When people of different races are treated differently based on conjecture — even when that conjecture is based on cursory data — it has an impact on the way people see the world. Ending stop-and-frisk might not fix our culture of white crime, but it would create an image of society that’s marginally more conducive to social progress.
Hackney, A.A., & Glaser, J. (2013). Reverse Deterrence in Racial Profiling: Increased Transgressions by Nonprofiled Whites. Law and Human Behavior DOI: 10.1037/lhb0000038

Rethinking the Marshmallow Experiment

As policymakers begin to pay more attention to “non-cognitive skills” and “grit” there’s been renewed interest in Walter Micshel’s famous marshmallow experiment. In the original experiment young children were left alone in a room with a marshmallow (or a cookie) and told they would receive two marshmallows if they didn’t eat it until the experimenter returned. The ability to hold off on eating the marshmallow was found to be predictive of a variety of positive life outcomes, such as educational attainment and health.

The standard take on the experiment is that it reveals a child’s level self-control. If a child wanted to eat the marshmallow right away, they would do it. The fact that some attempted to wait but couldn’t is a sign that they lack a certain ability possessed by those who were able to wait.

Or perhaps not. In a new paper Penn’s Joseph McGuire and Joseph Kable propose that attempting to wait and then giving in may reflect perfectly rational decision making. Their reasoning is fairly straightforward. Generally the assumption is that as the time you wait increases, the remaining wait time decreases. When you’re in line at the DMV and you think it will be 20 minutes, every minute that passes gets you closer to the front of the line. But that’s not always the case. Perhaps the line is moving so slowly that after two minutes have passed you come to believe it will take a total of 30 minutes. If you were initially willing to wait for 25 minutes, deciding to wait but then giving up after two minutes will be a rational decision that has nothing to do with self-control. This is similar to deciding to wait for both marshmallows but then giving up halfway through. What may initially seem like a lack of self-control could actually be a rational decision based on how much waiting you think is left.

It has often been misreported, but in Mischel’s original experiment he didn’t tell the kids he would be back in 15 minutes, he said he would be gone for a “while” or a “long time.” This means it could have been rational for a child to give in. Fox example, imagine a child believed “a while” meant 10 minutes, but also thought it could mean 45 minutes. If the child was willing to wait for 15 minutes, the decision to not initially eat the marshmallow is rational. But once the 10 minute mark passes, the child may begin to believe it will be an additional 35 minutes, and at that point the rational decision is to eat the marshmallow. Thus giving in does not necessarily signal a lack of self-control.

Of course McGuire and Kable are quick to point out that their theory is not directly applicable to the marshmallow experiments, but more generally to adult situations in which not persisting is rational behavior rather than a sign a person lacks self-control. In fact, they conducted an experiment in which people were asked about reaching certain goals related to dieting, training for a mile run, studying for the LSAT, or practicing piano. In each case McGuire and Kable found that as more time elapsed without the goal being reached, participants increased their predictions of the remaining time it would take to reach the goal. So for example, people who imagined attempting to lose weight thought it would take more additional to time to reach their goal on day 45 than on day 1.

The implication is that attempting to draw conclusions from a person’s persistence or lack thereof is not so clear cut. They may lack self-control, but they could also be making a rational decision based on a revised expectation of how much time remains.

I also think there’s a second, unrelated issue with standard marshmallow test. In many situations the utility of self-control is not merely in how it helps you avoid desirable but costly behaviors (eating the marshmallow), it’s in how it helps you persist in undesirable but beneficial behaviors (fitting parts together at your assembly-line job.) In most practical situations avoiding the distraction is only half the battle. You then need to actively engage in the task you’re fighting the urge to be distracted from. It would be great if everything in life was fascinating, but many things — studying, exercising, cleaning your apartment — require actively engaging in a somewhat dull behavior rather than merely avoiding a behavior.

Imagine that instead of sitting quietly and not eating the marshmallow, participants also had to engage in a dull task. For example, a blue circle and a red circle repeatedly appear on a computer screen. Participants have two seconds to click on the blue circle, and they have to do it until the experimenter returns. Would the results be different? Would there be kids who could wait all day for a second marshmallow, but wouldn’t last five minutes if the task required active participation in a boring activity? I’m not sure. But it seems like such an experiment would do a better job of measuring whether a person can operationalize their self-control to accomplish beneficial tasks.

So here’s the new marshmallow experiment I want to see: People are told exactly how long they’ll have to wait and are given assurances that what they are being told is the truth. That way the initial decision to persist can’t turn into a rational decision to give in. Then, while participants wait they have to continuously engage in an extremely dull task. Will the results be different? What do you think?
McGuire, J.T., & Kable, J.W. (2013). Rational temporal predictions can underlie apparent failures to delay gratification. Psychological Review DOI: 10.1037/a0031910

Are New York’s New Tests Better, or Just Harder?

The big drop in scores on New York’s new Common Core exams have sparked discussion about whether the tests are different in the *right* way. Did the tests actually evaluate the “higher-order” thinking skills kids are now supposed to be learning, or are the low scores the result of questions that were harder or just plain bad, but not conceptually different? In other words, did the tests do what they were supposed to do?

At this point there’s no way to get a definitive answer, but the data on how each grade’s performance changed relative to the other grades may be of some use. In theory, younger students have spent less time learning complex math and taking high-stakes exams, and so they should have an easier time adjusting to change. Look at the test question below. It’s not hard, but it initially makes your brain hurt.

If you went through elementary school without encountering “weird” questions like that, you might be a little thrown off. So older kids should have a harder time adjusting to exams that test for new higher-order skills. But what does the data say?

Below are the math proficiency rates of New York City 3rd through 8th graders over the last 4 years. (I’ve focused only on math because it seems like students are more likely to have set ways of thinking about and math problems, and thus the shift to “Common Core math” ought to present bigger adjustment. But I could be wrong.)

As you can see,  it’s hard to make out any kind of trend because of the large across-the-board drop this year.

chart 1 new

However, the chart below presents the same data, but now each grade’s proficiency rate has been replaced with its performance relative to the overall 3rd-8th grade (i.e. “All Grades”) proficiency rate. Note that the numbers below don’t represent raw percentage point changes, but the rate of proficiency relative to the 3rd-8th grade average. So for example, in 2012, the 4th grade proficiency rate of 65.7% represented a 9.6% increase over the 3rd-8th grade proficiency rate of 60%. But in 2013, the 4th grade proficiency rate of 35.2% represented an 18.9% increase over the 3rd-8th grade rate of 29.6%.

Overall, what you see are large jumps for the two youngest grades and large declines for the two oldest grades (although the drop for 8th grade puts them at a level similar to 2010.)

chart 2 new

Below is the same data, but now 3rd & 4th grade and 7th & 8th grade have been averaged into single measures to emphasize the change. Once again, the x-axis represents the proficiency rate of all 3rd-8th grade students.

In 2013, 3rd and 4th graders went from being 2.3% better than the 3rd-8th grade average to being 15.4% better.

chart 3 new

Finally, below is the same chart as above, but with data from all of New York State rather than just New York City. The pattern is the same.

chart 4 new

It would be interesting to know whether the same thing occurred last year in Kentucky, the first state to pilot new exams based on the Common Core. I poked around a bit but couldn’t find any handy state level aggregates. (Feel free to let me know or leave a comment if you have them.)

Of course these numbers don’t necessarily mean that the tests successfully evaluated new skills, or that the younger kids did better because they weren’t acclimated to the old tests. It’s possible that for whatever reason the 3rd and 4th grade tests were objectively easier. It also possible that this year’s 3rd and 4th grades were particularly strong cohorts, although the fact that this year’s top performing 4th graders were pedestrian 3rd graders last year suggests that might not be the case. Most importantly, it’s only one year of data, and so it might just be statistical noise.

Nevertheless, it’s worth thinking about what it would mean if the “newness” of the test had a non-uniform impact across all six grade levels.  One takeaway is that we should expect scores to rise more than we might otherwise expect because kids will either get acclimated to the new tests or be replaced by kids who were never exposed to the old ones. In other words, even if the 2019 cohort of 8th graders has exactly the same raw knowledge as this year’s cohort, we should expect the 2019 cohort to do better. That means that Mr. or Mrs. New Mayor will get a little credit they probably don’t deserve.

More broadly, if familiarity does have an impact it would be yet another reason to be patient when evaluating new standards, tests, or instructional paradigms. In addition to all the obvious moving parts, there is psychologically-based inertia that has to be overcome, and it will take time until everybody is mentally up to speed with the new system. Teachers have already made it clear they don’t feel they’re there yet, and perhaps certain students also need more time to adjust before we can properly judge them. In that sense the charts above merely reinforce what the smart people have been saying all along. If you’re looking to draw sweeping conclusions after one or two years of testing, you’re out of luck.

Momentum: Not Just For Silly Pundits

Watch TV for long enough and you’ll come across a talking head expounding on the awesome power of momentum. For example, sports commentators often praise terrible decisions (e.g. taking a sure field goal instead of going for a touchdown) based on the need to “keep momentum going,” as if scoring points has some significant intrinsic value beyond the benefit of increasing your point total. (Grantland’s Bill Barnwell has been doing the yeoman’s work of tamping down on silly momentum narratives.)

The power of momentum also emerges during political campaigns, when pundits routinely feel the need to conjure momentum out of thin air. If a particular candidate has a good polling day, it’s surely a sign he’s on his way to the top. If he gets a string of positive news cycles, there’s never any doubt that things might change. Not surprisingly, Nate Silver has been quick to debunk such narratives.

So sports and political commentators clearly buy into the myth of momentum, but the question remains: Do people who aren’t paid to fill airtime with dubious narratives believe that momentum is a real thing? That is, do people believe that improvement is bound to lead to more improvement?

A new study led by NYU’s Nathan Pettit suggests the answer is yes. Pettit and his team conducted a series of five experiments, each a variation on the same basic design. In the initial experiment participants were told about a person occupying a given position in the hierarchy of a work group (e.g. 6th out of 10). Some participants were told the person had risen to that position (from 8th to 6th), some were told the person had fallen to that position (from 4th to 6th), and some were merely told the person was 6th (the control condition.) When asked to judge the person’s prestige, participants who were told he had moved up to 6th rated him significantly higher than participants in the control condition (momentum!), while those who were told he had fallen to 6th rated him significantly lower than participants in the control condition (nomentum!)

Similar findings occurred in experiments involving college rankings (a school that moved up to 11th was rated as more prestigious and advised to increase tuition more than a school that dropped to 11th), product rankings (luxury watches that had risen in a magazine’s rankings were seen as more prestigious), and players in a trivia league (players with the same ranking were judged differently depending on whether they had recently risen or fallen to their spot.)

On some level, the findings are nothing more than a specific manifestation of recency bias. People tend to place too much weight on the most recent change (e.g. falling from 4th to 6th) and not enough on previous occurrences (the circumstances that led to being ranked 4th). There’s also a failure to account for regression to the mean. When somebody moves from 8th to 6th, the new position is seen as the “true” position rather than a deviation from the true position. Thus it’s presumed the person is more likely to stay at 6th or move higher rather than fall back to 7th or 8th.

But the findings do show that even if pundits don’t have a great grasp of probability, their mental failings are not unique. Everybody seems biased toward thinking that success will lead to more success.

Does any of this matter? When it comes to a football game, the answer is surely no. A team gains no advantage if a string of positive plays (momentum!) makes fans think they’ll win, and so it’s never a good idea to take a sure field goal over a 50% chance of scoring a touchdown.

On the other hand, in elections people like to vote for winners, and so perceived momentum could influence votes. Similarly, if you’re trying to decide whether to go after a likely but small improvement in status or a less-likely but larger improvement, it may be wise to consider the appearance of momentum. For example, imagine you’re deciding between running for Vice President or President of an organization. If you ever plan on running again, the effect of perceived momentum could tilt the decision toward running for VP and taking the sure but smaller status increase.

In general, the study adds to the pile of evidence that shows people are fairly lousy at judging things. The trick is to find ways to take advantage of it. That klezmer band that fell from 2nd to 6th in the latest coolness ratings? Now is the perfect time to book them for your kid’s bar-mitzvah.
Pettit, N.C., Sivanathan, N., Gladstone, E., & Mar, J.C. (2013). Rising Stars and Sinking Ships
Consequences of Status Momentum Psychological Science DOI: 10.1177/0956797612473120