I had been mulling a rebuttal to my colleague and friend Jon Rauch’s interesting—but wrong—new Brookings paper praising the role of “hacks, machines, big money, and back room deals” in democracy. I thought the indictments of Chris Christie’s associates last week provided a perfect example of the dangers of all of that, and so of why Jon was incorrect. But in yesterday’s L.A. Times, he beat me to it, himself defending the political morality (if not the efficacy) of their actions, and in the process delivering a knockout blow to his own position.

Bridgegate is a perfect example of why we need fewer "hacks, machines, big money, and back room deals" in our politics, not more. There is no justification whatsoever for government officials abusing their powers, stopping emergency vehicles and risking lives, making kids late for school and parents late for their jobs to retaliate against a mayor who withholds an election endorsement. We vote in our democracy to make government work, not break. We expect that officials will serve the public, not their personal interests. This conduct weakens our democracy, not strengthens it.

It is also incorrect that, as Jon suggests, reformers and transparency advocates are, in part, to blame for the gridlock that sometimes afflicts our American government at every level. As my co-authors and I demonstrated at some length in our recent Brookings paper, “Why Critics of Transparency Are Wrong,” and in our follow-up Op-Ed in the Washington Post, reform and transparency efforts are no more responsible for the current dysfunction in our democracy than they were for the gridlock in Fort Lee. Indeed, in both cases, “hacks, machines, big money, and back room deals” are a major cause of the dysfunction. The vicious cycle of special interests, campaign contributions and secrecy too often freeze our system into stasis, both on a grand scale, when special interests block needed legislation, and on a petty scale, as in Fort Lee. The power of megadonors has, for example, made dysfunction within the House Republican Caucus worse, not better.

Others will undoubtedly address Jon’s new paper at length. But one other point is worth noting now. As in foreign policy discussions, I don’t think Jon’s position merits the mantle of political “realism,” as if those who want democracy to be more democratic and less corrupt are fluffy-headed dreamers. It is the reformers who are the true realists. My co-authors and I in our paper stressed the importance of striking realistic, hard-headed balances, e.g. in discussing our non-absolutist approach to transparency; alas, Jon gives that the back of his hand, acknowledging our approach but discarding the substance to criticize our rhetoric as “radiat[ing] uncompromising moralism.” As Bridgegate shows, the reform movement’s “moralism" correctly recognizes the corrupting nature of power, and accordingly advocates reasonable checks and balances. That is what I call realism. So I will race Jon to the trademark office for who really deserves the title of realist!

One of the great parts of being at Brookings has been the many champions of government reform in the US and around the world who have reached out to visit us here, meet me and my colleagues, and talk about how best to transform government and make it work better for people. The latest was MK Erel Margalit, who before joining the Israeli Knesset started a leading venture capital firm in Israel (and was the first Israeli to make the Forbes Midas list of top tech investors globally). My Brookings colleagues, including Elaine Kamarck, Bill Galston, Natan Sachs and John Hudak talked with MK Margalit about the lessons he learned in the private sector, and about his efforts to bring those lessons to his work in government. 

Coming not long after our meeting with Czech Deputy Prime Minister and Finance Minister Andre Babis, who enjoyed similar success in business and has ambitious reform goals of his own informed by his business career, it was fascinating to talk about what does and does not translate to the government sector. MK Margalit’s focus includes supporting peace and economic development by developing enterprise zones in and around Israel that encourage economic partnerships between Jewish and Arab Israelis and their businesses, and that include Palestinians as well. It was an impressive melding of business and government methodologies. The meeting built on similar ones we have had with other innovators including CFPB Director Rich Cordray, former Mayor and Governor Martin O’Malley, and of course DPM Babis, all of whom have in common innovating to make government function more effectively.

Editor's Note: The introduction to the 2015 Brown Center Report on American Education appears below. Use the Table of Contents to navigate through the report online, or download a PDF of the full report.

TABLE OF CONTENTS

Part I: Girls, Boys, and Reading

Part II: Measuring Effects of the Common Core

Part III: Student Engagement

INTRODUCTION

The 2015 Brown Center Report (BCR) represents the 14^th edition of the series since the first issue was published in 2000.  It includes three studies.  Like all previous BCRs, the studies explore independent topics but share two characteristics: they are empirical and based on the best evidence available.  The studies in this edition are on the gender gap in reading, the impact of the Common Core State Standards -- English Language Arts on reading achievement, and student engagement.

Part one examines the gender gap in reading.  Girls outscore boys on practically every reading test given to a large population.  And they have for a long time.  A 1942 Iowa study found girls performing better than boys on tests of reading comprehension, vocabulary, and basic language skills.  Girls have outscored boys on every reading test ever given by the National Assessment of Educational Progress (NAEP)—the first long term trend test was administered in 1971—at ages nine, 13, and 17.  The gap is not confined to the U.S.  Reading tests administered as part of the Progress in International Reading Literacy Study (PIRLS) and the Program for International Student Assessment (PISA) reveal that the gender gap is a worldwide phenomenon.  In more than sixty countries participating in the two assessments, girls are better readers than boys. 

Perhaps the most surprising finding is that Finland, celebrated for its extraordinary performance on PISA for over a decade, can take pride in its high standing on the PISA reading test solely because of the performance of that nation’s young women.  With its 62 point gap, Finland has the largest gender gap of any PISA participant, with girls scoring 556 and boys scoring 494 points (the OECD average is 496, with a standard deviation of 94).   If Finland were only a nation of young men, its PISA ranking would be mediocre.

Part two is about reading achievement, too. More specifically, it’s about reading and the English Language Arts standards of the Common Core (CCSS-ELA).  It’s also about an important decision that policy analysts must make when evaluating public policies—the determination of when a policy begins. How can CCSS be properly evaluated? 

Two different indexes of CCSS-ELA implementation are presented, one based on 2011 data and the other on data collected in 2013.  In both years, state education officials were surveyed about their Common Core implementation efforts.  Because forty-six states originally signed on to the CCSS-ELA—and with at least forty still on track for full implementation by 2016—little variability exists among the states in terms of standards policy.  Of course, the four states that never adopted CCSS-ELA can serve as a small control group.  But variation is also found in how the states are implementing CCSS.  Some states are pursuing an array of activities and aiming for full implementation earlier rather than later.  Others have a narrow, targeted implementation strategy and are proceeding more slowly. 

The analysis investigates whether CCSS-ELA implementation is related to 2009-2013 gains on the fourth grade NAEP reading test.  The analysis cannot verify causal relationships between the two variables, only correlations.  States that have aggressively implemented CCSS-ELA (referred to as “strong” implementers in the study) evidence a one to one and one-half point larger gain on the NAEP scale compared to non-adopters of the standards.  This association is similar in magnitude to an advantage found in a study of eighth grade math achievement in last year’s BCR.  Although positive, these effects are quite small.  When the 2015 NAEP results are released this winter, it will be important for the fate of the Common Core project to see if strong implementers of the CCSS-ELA can maintain their momentum.

Part three is on student engagement.  PISA tests fifteen-year-olds on three subjects—reading, math, and science—every three years.  It also collects a wealth of background information from students, including their attitudes toward school and learning.  When the 2012 PISA results were released, PISA analysts published an accompanying volume, Ready to Learn: Students’ Engagement, Drive, and Self-Beliefs, exploring topics related to student engagement.

Part three provides secondary analysis of several dimensions of engagement found in the PISA report.  Intrinsic motivation, the internal rewards that encourage students to learn, is an important component of student engagement.  National scores on PISA’s index of intrinsic motivation to learn mathematics are compared to national PISA math scores.  Surprisingly, the relationship is negative.  Countries with highly motivated kids tend to score lower on the math test; conversely, higher-scoring nations tend to have less-motivated kids. 

The same is true for responses to the statements, “I do mathematics because I enjoy it,” and “I look forward to my mathematics lessons.”  Countries with students who say that they enjoy math or look forward to their math lessons tend to score lower on the PISA math test compared to countries where students respond negatively to the statements.  These counterintuitive finding may be influenced by how terms such as “enjoy” and “looking forward” are interpreted in different cultures.  Within-country analyses address that problem.  The correlation coefficients for within-country, student-level associations of achievement and other components of engagement run in the anticipated direction—they are positive.  But they are also modest in size, with correlation coefficients of 0.20 or less. 

Policymakers are interested in questions requiring analysis of aggregated data—at the national level, that means between-country data.  When countries increase their students’ intrinsic motivation to learn math, is there a concomitant increase in PISA math scores?  Data from 2003 to 2012 are examined.  Seventeen countries managed to increase student motivation, but their PISA math scores fell an average of 3.7 scale score points.  Fourteen countries showed no change on the index of intrinsic motivation—and their PISA scores also evidenced little change.  Eight countries witnessed a decline in intrinsic motivation.  Inexplicably, their PISA math scores increased by an average of 10.3 scale score points.  Motivation down, achievement up.

Correlation is not causation.  Moreover, the absence of a positive correlation—or in this case, the presence of a negative correlation—is not refutation of a possible positive relationship.  The lesson here is not that policymakers should adopt the most effective way of stamping out student motivation.  The lesson is that the level of analysis matters when analyzing achievement data.  Policy reports must be read warily—especially those freely offering policy recommendations.  Beware of analyses that exclusively rely on within- or between-country test data without making any attempt to reconcile discrepancies at other levels of analysis.  Those analysts could be cherry-picking the data.  Also, consumers of education research should grant more credence to approaches modeling change over time (as in difference in difference models) than to cross-sectional analyses that only explore statistical relationships at a single point in time. 

Downloads

• Download the report

Tom Loveless, a nonresident senior fellow in Governance Studies, explains his latest research on measuring achievement of American students.

“The bottom line here: the implementation of the common core has appeared to have very little impact on student achievement,” Loveless says. In this episode, he discusses whether the common core is failing our students, whether AP achievement is indicative of student success, and the role of principals as instructional leaders.

Also in this episode: Get to know Constanze Stelzenmüller, the Robert Bosch Senior Fellow in the Center on the United States and Europe, during our "Coffee Break” segment. Also stay tuned to hear the final episode in our centenary series with current and past Brookings scholars.

Show Notes:

The Brown Center Report on American Education

Brookings Centenary Timeline

Subscribe to the Brookings Cafeteria on iTunes, listen in all the usual places, and send feedback email to BCP@Brookings.edu.

The Supreme Court today decided that Congress meant what it said when it enacted the Affordable Care Act (ACA). The ACA requires people in all 50 states to carry health insurance and provided tax credits to help them afford it. To have offered such credits only in the dozen states that set up their own exchanges would have been cruel and unsustainable because premiums for many people would have been unaffordable.

But the law said that such credits could be paid in exchanges ‘established by a state,’ which led some to claim that the credits could not be paid to people enrolled by the federally operated exchange. In his opinion, Chief Justice Roberts euphemistically calls that wording ‘inartful.’ Six Supreme Court justices decided that, read in its entirety, the law provides tax credits in every state, whether the state manages the exchange itself or lets the federal government do it for them.

That decision is unsurprising. More surprising is that the Court agreed to hear the case. When it did so, cases on the same issue were making their ways through four federal circuits. In only one of the four circuits was there a standing decision, and it found that tax credits were available everywhere. It is customary for the Supreme Court to wait to take a case until action in lower courts is complete or two circuits have disagreed. In this situation, the justices, eyeing the electoral calendar, may have preferred to hear the case sooner rather than later to avoid confronting it in the middle of a presidential election.

Whatever the Court’s motives for taking the case, their willingness to hear the case caused supporters of the Affordable Care Act enormous unease. Were the more conservative members of the Court poised to accept an interpretation of the law that ACA supporters found ridiculous but that inartful legislative drafting gave the gloss of plausibility? Judicial demeanor at oral argument was not comforting. A 5-4 decision disallowing payment of tax credits seemed ominously plausible.

The Court’s 6-3 decision ended those fears. The existential threat to health reform from litigation is over. But efforts to undo the Affordable Care Act are not at an end. They will continue in the political sphere. And that is where they should be. ACA opponents know that there is little chance for them to roll back the Affordable Care Act in any fundamental way as long as a Democrat is in the White House. To dismantle the law, they must win the presidency in 2016.

But winning the presidency will not be enough. It would be mid 2017 before ACA opponents could draft and enact legislation to curb the Affordable Care Act and months more before it could take effect. To borrow a metaphor from the military, even if those opposed to the ACA win the presidency, they will have to deal with ‘facts on the ground.’

As government funding for disability insurance is expected to run out next year, Congress should re-evaluate the costs of the program.

Nine million people in America today are receiving Social Security Disability Insurance, double the number in 1995 and six times the number in 1970. With statistics like that, it’s hardly surprising to see some in Congress worry that more will enroll in the program and costs would continue to rise, especially since government funding for disability insurance is expected to run out by the end of next year. If Congress does nothing, benefits would fall by 19% immediately following next year’s presidential election. So, Congress will likely do something. But what exactly should it do?

Funding for disability insurance has nearly run out of money before. Each time, Congress has simply increased the share of the Social Security payroll tax that goes for disability insurance. This time, however, many members of Congress oppose such a shift unless it is linked to changes that curb eligibility and promote return to work. They fear that rolls will keep growing and costs would keep rising, but findings from a report by a government panel conclude that disability insurance rolls have stopped rising and will likely shrink. The report, authored by a panel of the Social Security Advisory Board, is important in that many of the factors that caused disability insurance to rise, particularly during the Great Recession, have ended.

• Baby-boomers, who added to the rolls as they reached the disability-prone middle age years, are aging out of disability benefits and into retirement benefits. 


• The decades-long flood of women increased the pool of people with the work histories needed to be eligible for disability insurance. But women’s labor force participation has fallen a bit from pre-Great Recession peaks, and is not expected again to rise materially. 


• The Great Recession, which led many who lost jobs and couldn’t find work to apply for disability insurance, is over and applications are down. A recession as large as that of 2008 is improbable any time soon. 


• Approval rates by administrative law judges, who for many years were suspected of being too ready to approve applications, have been falling. Whatever the cause, this stringency augurs a fall in the disability insurance rolls.

Throughout the past year, we have heard paid leave debated in state houses and on the campaign trail. I am all in favor of paid leave. As I have argued elsewhere, it would enable more people, especially those in lower-paid jobs, to take time off to deal with a serious illness or the care of another family member, including a newborn child. But we shouldn’t stop with paid leave. We should also consider shortening the standard work week. Such a step would be gender neutral and would not discriminate between the very different kinds of time pressures faced by adults. It might even help to create more jobs.

The standard work week is 40 hours -- 8 hours a day for five days a week. It’s been that way for a long time. Back in 1900, the typical factory worker spent 53 hours on the job, more than a third more hours than we spend today. The Fair Labor Standards Act was passed in 1938, and set maximum hours at 40 per week. Amazingly, more than three quarters of a century after passage of the FLSA, there has been no further decline in the standard work week. Not only has the legal standard remained unchanged, but 40 hours has become the social and cultural norm.

What’s going on here? Economists predicted that as we became more prosperous we would choose to work fewer hours. That hasn’t happened. Instead we have kept on working at about the same pace as we did earlier in our history, but have poured all of the gains from productivity growth into ever-higher levels of consumption – bigger houses, more electronic gadgets, fancier cars. With increased prosperity, people are buying more and more stuff, but they don’t have any more time to enjoy it. A reduction in the standard work week would improve the quality of life, especially for those in hourly jobs who have benefitted hardly at all from economic growth in recent decades.

Two-earner couples would also benefit. Among couples between the ages of 25 and 54, the number of hours worked increased by 20 percent between 1969 and 2000, from 56 hours to 67 hours (for both husband and wife combined). As Heather Boushey notes in her new book, Finding Time, we no longer live in a world where there is a “the silent partner” in every business enterprise, the iconic “American Wife,” who takes care of the children and the millions of details of daily living. With a shorter work week, both men and women would have more time for everything from cutting the grass to cooking dinner with no presumption about who does what. Although much of the debate this year has been about work-family balance, empty nesters or singles without young children might also welcome a shorter work week. For them it would provide the chance to follow their dream of becoming an artist, a boat builder, or the creator of their own small business.

Shorter hours could have another benefit and that is more jobs for workers who would otherwise be left behind by technological change. Many economists believe that as existing jobs are replaced by machines and artificial intelligence, new jobs will be created in technical, management, and service fields. But will this happen fast enough or at sufficient scale to reemploy all those who now find themselves without decent-paying work? I doubt it. A shorter work week might help to spread the available jobs around. Germany and other European countries, along with a few U.S. states used this strategy during the Great Recession. It kept more people on the job but at shorter hours and reduced unemployment. Using a similar strategy to deal with automation and long-term joblessness, although controversial, should not be dismissed out of hand.

Of course, shorter hours can mean lower total pay. But in one typical survey published in the Monthly Labor Review, 28 percent of the respondents said they would give up a day’s pay for one fewer day of work per week. Any new movement to reduce the work week would need to be phased in slowly, with flexibility for both employers and employees to negotiate adjustments around the standard. Yet if done correctly, the transition could be accomplished with little or no reduction in wages, just smaller raises as a bigger slice of any productivity improvement was invested in more free time. When Henry Ford reduced the work week from 6 to 5 days in 1926, he did not cut wages; he assumed that both productivity and consumption would rise, and his example encouraged other employers to follow suit.

I am not talking about reducing hours for those of us who want to spend long hours at work because we enjoy it. We would still be free to work 24/7, tied to our electronic devices, and no longer knowing exactly when work begins and ends. A new hours standard would primarily affect hourly (nonexempt) employees. These are the people in the less glamourous jobs at the bottom of the ladder, many of them single parents. Right now they finish work exhausted only to come home to a “second shift” that may be equally exhausting. A reduction in the standard workweek would almost certainly improve the quality of life for these hard-pressed and overworked Americans.

By all means, let’s enact a paid leave policy, but let’s also debate some even bigger ideas – ones that could lead to greater work-life balance now, and more job opportunities in the longer run.

Editor's note: This piece originally appeared on The Washington Post's In Theory Blog.

When Franklin Roosevelt delivered his second inaugural address on January 20, 1936 he lamented the “one-third of a nation ill-housed, ill-clad, ill-nourished.” He challenged Americans to measure their collective progress not by “whether we add more to the abundance of those who have much; [but rather] whether we provide enough for those who have too little.” In our new paper, One third of a nation: Strategies for helping working families, we ask a simple question: How are we doing?

In brief, we find that:

• The gulf in labor market income between the haves and have-nots remains wide. The median income of households in the bottom third in 2014 was $24,000, just a little more than a quarter of the median of $90,000 for the top two-thirds.
• The bottom-third households are disproportionately made up of minority adults, adults with limited educational attainment, and single parents.  
• The most important reason for the low incomes of the bottom third is a “work gap”: the fact that many are not employed at all, or work limited hours. 

The work gap

The decline in labor force participation rates has been widely documented, but the growing gulf in the work gap between the bottom third and the rest of the population is truly striking:

While the share of men who are employed in the top two-thirds has been quite stable since 1980, lower-income men’s work rates have declined by 11 percentage points. What about women?

Middle- and upper-income women have increased their work rates by 13 percentage points. This has helped maintain or even increase their family’s income. But employment rates among lower-income women have been flat, despite reforms of the welfare system and safety net designed to encourage work.

Many on the left point to problems like low pay and lack of access to affordable childcare, and so favor a higher minimum wage and more subsidies for daycare. For many conservatives, the problem is rooted in family breakdown and a dependency-inducing safety net. They therefore champion proposals like marriage promotion programs and strict work requirements for public benefits. Most agree about the importance of education.

Time to debate some more radical solutions 

It may be that the standard solutions to the problems of the bottom third, while helpful, are no longer sufficient. A debate about whether to make safety net programs such as Food Stamps and housing assistance conditional on work or training is underway. So are other solutions such as subsidized jobs (created by some states during the Great Recession as a natural complement to a work-conditioned safety net), more work sharing (used in Germany during the recession), or even a universal basic income (being considered by Swiss voters in June).

Downloads

• Table of Contents
• Chapter One

• {CD2E3D28-0096-4D03-B2DE-6567EB62AD1E}, 9780815727958, $32.00 Add to Cart

Israel apologized to Turkey today for the May 2010 incident on board the Mavi Marmara naval vessel, part of a flotilla to Gaza, in which nine Turks were killed from Israel Defense Forces fire. The apology came during a 30-minute telephone conversation between Israeli Prime Minister Benjamin Netanyahu and Turkish Prime Minister Recep Tayyip Erdogan, orchestrated by President Barack Obama, who was ending his 3 day visit to Israel and the Palestinian Authority. Erdogan accepted the Israeli apology, and the leaders agreed to begin a normalization process between Israel and Turkey, following the past three years, when relations were practically at a standstill. (Last December, I wrote about the beginnings of a Turkey-Israeli rapprochement, and discussed more of the policy implications here.)

This development allows the two countries to begin a new phase in their relationship, which has known crisis and tension, but also cooperation and a strong strategic partnership.

The U.S. administration played a key role behind the scenes in creating the conditions that paved the way for an Israeli apology and Turkish acceptance. Undoubtedly, a close relationship between Turkey and Israel-- two of America’s greatest allies in the region-- serves United States’ strategic interests globally and regionally. At a time when the Middle East political landscape is changing rapidly, it was imperative to end the long impasse between Ankara and Jerusalem.

Over the past year, Turkey and Israel have also come to realize that repairing their relationship and re-establishing a dialogue is at their best interest, as they face great challenges in their immediate vicinity (first and foremost, the Syrian civil war).

United States officials emphasized that this is the first step in a long process. Nevertheless, the parties will have to make a great effort to overcome years of distrust and suspicion if they want the relationship to work. No one is under the allusion that relations will go back to what they were in the “honeymoon” period of the 1990s but modest improvement can be made. It will not be an easy task, and for that to happen it is essential that the parties not only talk to each other, but also listen to one another and begin to respect each other’s sensitivities. In order for this rapprochement to be successful, United States will have to continue to oversee discussions between Turkey and Israel, and remain heavily engaged in this process.

The narratives on U.S. development aid to Pakistan—as well as Pakistan’s own development policy discussion—frequently invoke the conventional wisdom that more education and better economic opportunities result in lower extremism. In the debate surrounding the Kerry-Lugar-Berman bill in 2009, for instance, the late Ambassador Richard Holbrooke urged Congress to “target the economic and social roots of extremism in western Pakistan with more economic aid.”

But evidence across various contexts, including in Pakistan, has not supported this notion (see Alan Kreuger’s What Makes a Terrorist for a good overview of this evidence). We know that many terrorists are educated. And lack of education and economic opportunities do not appear to drive support for terrorism and terrorist groups. I have argued that we need to focus on the quality and content of the educational curricula—in Pakistan’s case, they are rife with biases and intolerance, and designed to foster an exclusionary identity—to understand the relationship between education and attitudes toward extremism.

My latest analysis with data from the March 2013 Pew Global Attitudes poll conducted in Pakistan sheds new light on the relationship between years of education and Pakistanis’ views of the Taliban, and lends supports to the conventional wisdom. The survey sampled 1,201 respondents throughout Pakistan, except the most insecure areas of Khyber Pakhtunkhwa and Baluchistan. This was a time of mounting terror attacks by the Pakistani Taliban (a few months after their attack on Malala), and came at the tail end of the Pakistan People's Party’s term in power, before the May 2013 general elections.

On attitudes toward the Pakistani Taliban, or Tehrik-e-Taliban Pakistan (TTP), 3 percent of respondents to the Pew poll said they had a very favorable view, 13 percent reported somewhat favorable views, while nearly 17 percent and 39 percent answered that they had somewhat unfavorable and very unfavorable views, respectively. A large percentage of respondents (28 percent) chose not to answer the question or said they did not know their views. This is typical with a sensitive survey question such as this one, in a context as insecure as Pakistan.

So overall levels of support for the TTP are low, and the majority of respondents report having unfavorable views. The non-responses could reflect those who have unfavorable views but choose not to respond because of fear, or those who may simply not have an opinion on the Pakistani Taliban.

The first part of my analysis cross-tabulates attitudes toward the TTP with education and income respectively. I look at the distribution of attitudes for each education and income category (with very and somewhat favorable views lumped together as favorable; similarly for unfavorable attitudes).

Figure 1. Pakistani views on the Pakistani Taliban, by education level, 2013

Figure 1 shows that an increasing percentage of respondents report unfavorable views of the Taliban as education levels rise; and there is a decreasing percentage of non-responses at higher education levels (suggesting that more educated people have more confidence in their views, stronger views, or less fear). However, the percentage of respondents with favorable views of the Taliban, hovering between 10-20 percent, is not that different across education levels, and does not vary monotonically with education. 

Figure 2. Pakistani views on the Pakistani Taliban, by income level, 2013

Figure 2 shows views on the Pakistani Taliban by income level. While the percentage of non-responses is highest for the lowest income category, the percentages responding favorably and unfavorably do not change monotonically with income. We see broadly similar distributions of attitudes across the four income levels.

But these cross-tabulations do not account for other factors that may affect attitudes: age, gender, and geographical location. Regressions (not shown here) accounting for these factors in addition to income and education show interesting results: relative to no education, higher education levels are associated with less favorable opinions of the Pakistani Taliban; these results are strongest for those with some university education, which is heartening. This confirms findings from focus groups I conducted with university students in Pakistan in May 2015. Students at public universities engaged in wide ranging political and social debates with each other on Pakistan and its identity, quoted Rousseau and Chomsky, and had more nuanced views on terrorism and the rest of the world relative to high school students I interviewed. This must at least partly be a result of the superior curriculum and variety of materials to which they are exposed at the college level.

My regressions also show that older people have more unfavorable opinions toward the Taliban, relative to younger people; this is concerning and is consistent with the trend toward rising extremist views in Pakistan’s younger population. The problems in Pakistan’s curriculum that began in the 1980s are likely to be at least partly responsible for this trend. Urban respondents seem to have more favorable opinions toward the Taliban than rural respondents; respondents from Punjab and Baluchistan have more favorable opinions toward the Taliban relative to those from Khyber Pakhtunkhwa, which as a province has had a closer and more direct experience with terror. The regression shows no relationship of income with attitudes, as was suggested by Figure 2.

Overall, the Pew 2013 data show evidence of a positive relationship between more education and lack of support for the Taliban, suggesting that the persisting but increasingly discredited conventional wisdom on these issues may hold some truth after all. These results should be complemented with additional years of data. That is what I will work on next.