The 2015 EdNext Poll on School Reform
Public thinking on testing, opt out, common core, unions, and more
The American public is displaying its independent streak. Critics of testing will take no comfort from the findings of the 2015 Education Next poll—but neither will supporters of the Common Core State Standards, school choice, merit pay, or tenure reform. The unions will not like the public’s view on their demands that nonmembers contribute financially to their activities. Teachers will be unhappy to hear that public enthusiasm for increasing teacher pay falls through the floor when people are told current salary levels and asked if they are willing to pay additional taxes for that purpose. The Obama administration will be equally unhappy to hear what both teachers and the public think about its proposals to require similar student suspension and expulsion rates across racial and ethnic groups.
These are among the many findings to emerge from the ninth annual Education Next survey, administered in May and June 2015 to a nationally representative sample of some 4,000 respondents, including oversamples of roughly 700 teachers, 700 African Americans, and 700 Hispanics (see methodology sidebar). The large number of survey respondents enabled us to ask alternative questions on the same topic in order to determine the sensitivity of opinion to new information and particular wording. We also posed many new questions in 2015, allowing us to explore opinion on curricular and other issues that have never before been examined in a nationally representative survey of the American public. Results from the full survey are available here, and click here for a graphic display of most findings.
Testing and Accountability
In early 2015, as Congress began rewriting the No Child Left Behind Act (NCLB), no issue loomed larger than the use of student testing to measure the performance of schools and teachers. Media reports featured teachers decrying a scourge of overtesting. By spring, hundreds of thousands of parents had chosen to have their children “opt out” of state tests, garnering the rousing approval of the teachers unions. Out on the hustings, Republican presidential candidates escalated their critique of the Common Core. The movement to put “the standardized testing machine in reverse,” in the words of New York mayor Bill de Blasio, seemed to have legs.
It is perhaps surprising, then, that in July a bipartisan Senate supermajority of 81–17 passed a revision of NCLB that keeps the federal requirement that all students be tested in math and reading in grades 3 to 8 and again in high school. Has the upper chamber ignored the people’s will? Or, is the public’s appetite for the information provided by regular student testing broader and more robust than the media coverage would indicate?
Our polling suggests the latter (see Figure 1). A solid 67% of members of the public say they support continuing the federal requirement for annual testing, while just 21% oppose the idea, with the remainder taking a neutral position. Parental support for testing (66%) is about as high as that of the public as a whole. Teachers are divided down the middle, with 47% saying yes and 46% saying no to continuing the policy.
In 2012, the last time we asked this question, 63% of the public said they supported annual testing, and only 12% opposed. In other words, the shares of supporters and opponents are both slightly higher in 2015 than they were three years ago, with the share taking a neutral position declining from 25% to 13%. This shift could suggest that public opinion has crystallized in the intervening years (but it may also reflect the fact that our survey presented the neutral response option more prominently in 2012). Either way, the backlash against standardized testing appears less potent than opponents claim.
Opting out. The House of Representatives also passed a reauthorization bill requiring that states maintain annual testing regimes, but its version differs from the Senate’s in one key respect: it allows parents to “opt out” of state tests, despite the fact that the federal government does not require that the tests be used to evaluate the performance of individual students. The difference between the two bills looms large, because one cannot assess school performance accurately unless nearly all students participate in the testing process.
What do people think of the opt-out movement? To find out, we asked whether they thought parents should be able to decide whether or not their children take annual state tests. Our results reveal little public sympathy for giving parents this option (see Figure 1). Only 25% of members of the public like the idea of letting parents decide whether their children are tested, while 59% oppose it. Among parents themselves, just 32% favor the opt-out approach, while 52% oppose it. Fifty-seven percent of teachers also dislike the idea, with only 32% giving it their support. In short, as Senate and House negotiators turn to ironing out differences between their bills, the Senate team can argue that its approach to “opt out” (which does not require states to offer that option) is backed by strong majorities of the public and of teachers.
Accountability: who should hold the reins? Another fault line in the debate over the proposed federal education law lies between Congress and the executive branch. As of late July, both the Senate and the House bills defer to the states on the question of how to design their school accountability programs. The Obama administration, backed by civil rights and business groups, wants the feds to have more voice in defining what constitutes a “failing school” and in proposing remedies. But the Senate has nixed the so-called Murphy Amendment, which would require states to identify and intervene in their lowest-performing schools; high schools with fewer than 67% on-time graduates; and any school where disadvantaged or disabled students fall short of standardized test goals for two consecutive years.
Where do people come down on this debate? To find out, we asked our respondents which level of government (federal, state, or local) should play the largest role in three key aspects of the design of school accountability programs:
· Setting education standards for what students should know;
· Deciding whether or not a school is failing; and
· Deciding how to fix failing schools.
When it comes to standard setting, members of the public are evenly divided over whether the federal government or the states should be in the driver’s seat: 43% say the states, and 41% say the federal government, while just 15% suggest that the local government should play this role (see Figure 2). But people clearly want the feds in the back seat when it comes to identifying and improving failing schools. Only 18% of respondents say that the federal government should play the largest role in identifying failing schools, and 20% say it should do so when it comes to fixing them. The percentages of those who say the states should have the lead role in these areas are 50 and 51, respectively.
Given the backing of civil rights groups for a larger federal role in this area, it is worth noting that neither African Americans nor Hispanics differ notably in their thinking from that of the broader public with respect to the role of the federal government in school accountability. Among African Americans, the share favoring federal leadership across the three topics is 46% for setting standards, 23% for identifying failing schools, and 23% for fixing failing schools, respectively. Among Hispanics, the parallel numbers are 44%, 18%, and 29%.
In short, if those in our nation’s capital want to modify federal education policy along lines preferred by the public at large, they will enact a law that resembles the bipartisan bill passed by the Senate. Education secretary Arne Duncan has indicated that the administration will not support a bill that doesn’t strengthen federal oversight of school accountability measures. If it should come down to a presidential veto, defending that action to the public on these grounds would be an uphill battle.
The Common Core
While support for standardized testing remains strong, the debate over the Common Core State Standards continues to divide both teachers and the general public (see Figure 3). Support for using the Common Core, which fell from 65% in 2013 to 53% in 2014, has now slipped slightly further, to 49%. Still, only 35% of members of the public express opposition to using the standards, with the remaining 16% undecided. Democrats (57%) remain much more supportive of the Obama-backed policy than Republicans are (37%).
The latest decline in support for these standards does not arise simply from a politically tainted Common Core “brand.” Among a second group of respondents who answered the same question but without the phrase “Common Core,” support for the use of shared standards across the states slid from 68% in 2014 to 54% in 2015.
It is interesting to note that this year’s difference between those favoring the Common Core standards (49%) and those favoring generic standards (54%) is just 5 percentage points. In 2014, that differential was 15 points. Why? It may be that the debate over national standards has been so energetic over the past year that the public now is more aware of the issue, whether or not the phrase “Common Core” is mentioned.
A third group of respondents were not told the standards would be “used to hold public schools accountable for their performance.” Without the accountability phrase in the question, support for the Common Core falls to just 39%, with 37% opposed. The proportion of people with no opinion increases from 16% to 23%.
Teacher support is also sliding. In 2013, 76% of teachers supported the Common Core—giving it a far greater approval rating than did the general public. But teacher approval collapsed to 46% in 2014 and has now fallen to just 40%. Meanwhile, the share of teachers expressing opposition has risen to 50%, leaving just 10% undecided. Unlike the public at large, teachers are more likely to express support for the Common Core when the survey question does not include the accountability phrase. They divide evenly when the question omits that phrase, with 44% in support and 43% opposed.
The news for proponents of the Common Core is not all bad. Those who favor the Common Core continue to outnumber opponents, by 14 percentage points. Also, the rate of decline in support slowed markedly between 2014 and 2015, perhaps suggesting that opinion on the issue has begun to stabilize. Moreover, the broader public’s opposition to the Common Core appears to rest on a shallow factual foundation. Asked whether or not the Common Core is being used in their local school district, fully 58% of the members of the public admit that they do not know. Only 44% of residents in states that have adopted the Common Core realize that the standards are being used in their school districts; and perhaps more startling, 24% of residents in states that do not have the Common Core believe their districts are using the standards.
Yet among the 34% of the public who report that the standards are being used in their district, respondents who believe the standards have had a negative effect on schools (51%) exceed those who think they have had a positive effect (28%). Twenty-one percent give a neutral response. Teachers and parents, who claim greater knowledge of whether the standards are in use, are just as negative in their assessment of the impact. Seventy-three percent of teachers report that the standards are being used in their district, with 49% of that group reporting negative effects and 32% reporting positive effects. Among parents, 49% say that the standards are being used in their district, with 53% reporting negative effects and just 28% reporting positive effects.
In other words, teachers and parents who say their district is implementing the standards are the ones most likely to offer a critical assessment of their impact. That finding should be of concern to all those hoping to see the Common Core succeed.
Changes in Support for School Reform
In retrospect it looks as if 2014, an election year that swept Republicans into power in Congress and many state capitals, propelled school reform to a high-water mark that has proven difficult to sustain. For three years in a row now, we have asked either identical or quite similar questions on several issues. On a surprising number of them, support for policy changes has slipped in 2015 from peaks attained in 2014, though sometimes the fall is to a level that remains above the one reached in 2013. None of the changes are large, and some of the shifts fall short of statistical significance, leaving it unclear as to whether a true change has taken place. But consider the overall pattern of responses across major parts of the school reform agenda (see Figure 4):
· Charter schools. Support for charter schools has dipped from a high of 54% in 2014 to 51% in 2015, the same level as in 2013. However, the percentage supporting charters remains twice that of the 27% expressing opposition.
· Tax credits for scholarships for low-income students. Support for a tax credit for businesses and individuals who contribute to private-school scholarships for low-income families has also fallen, to 55% from 60% in 2014. (This question was not asked in 2013.)
· Vouchers for low-income students. Backing for the use of “government funds to pay the tuition of low-income students who choose to attend private schools” has fallen steadily—from 41% to 37% between 2013 and 2014, with a further (though not statistically significant) drop to just 34% in 2015.
· Universal vouchers. Public enthusiasm for universal vouchers without regard to income has slipped from 50% in 2014 to 46% in 2015, just a bit higher than the 44% level reported in 2013. (However, these changes are not statistically significant and the comparison is not exact, as the question in 2015 for the first time included the word “all,” clearly presenting vouchers as a universal benefit for every family.)
· Merit pay for teachers. People are not fully embracing policy reforms affecting teachers. Between 2014 and 2015, public support for merit pay has slid from 57% to 51%, about the same as in 2013, when merit pay garnered support from 49% of the population. Even so, just 34% of the population opposes merit pay, with the remainder taking a neutral position.
·Tenure. Between 2014 and 2015, public opposition to teacher tenure has also slipped, from 57% to 51%, just above the 47% level attained in 2013. Nonetheless, current public support for teacher tenure is just 29%, a little more than half the size of the opposition.
One hesitates to read too much into shifts in opinion that are only modestly larger than what a statistical aberration might account for—and in some cases, not even that big. Perhaps the higher levels of support we observed in 2014 reflected temporary shocks to public opinion stemming from events such as Wisconsin governor Scott Walker’s recall election and the landmark Vergara v. California decision that struck down California’s teacher evaluation and tenure laws, both of which took place while our survey was in the field. But school reformers might take the 2015 findings as a red light on the dashboard, a warning that efforts to alter the public’s thinking on education policy may be faltering.
Expenditures and Salaries
In its 2016 budget, the Obama administration has proposed a new billion-dollar federal program, Teaching for Tomorrow, which requests an additional $1 billion in federal funding for services to children from low-income families. It also calls for more money for English language acquisition programs, civil rights enforcement, and special education services. Reporters nonetheless have pronounced the budget “dead on arrival,” as Congress is reluctant to increase spending at a time when the country is running a large fiscal deficit. Consistent with these reports, the House of Representatives has passed a budget resolution that calls for a more than 8% cut in federal spending.
Missing from virtually all the media coverage of these developments are answers to a few basic questions: How much do we currently spend per pupil? How much does the federal government contribute to the total expenditure? And does the public think spending should be increased? To gauge people’s knowledge and views on these matters, we asked our respondents a series of questions concerning school spending.
Americans greatly underestimate the amount of money spent on schools. According to the federal government’s National Center for Education Statistics (NCES), the school districts in which our survey respondents resided spent an average of $12,440 per pupil in 2012 (the most recent data available). But when we ask respondents to estimate per-pupil expenditures in their local school district, they guess, on average, just $6,307, a little more than half actual spending levels.
Our survey found that people are often willing to alter their thinking when given additional information. Before asking our respondents if they thought spending in their districts should be increased, we told half of them what the current spending levels were. The other half were left uninformed. Among those not informed, 58% favor increases in spending. That support drops to 42% when people are told the actual level of expenditures (after having provided their own estimate).
Respondents who most seriously underestimate spending levels are the ones most likely to change their minds when told the facts. When those who underestimate school expenditures by $5,000 or less are told real spending levels, their support for increased spending drops by 12 percentage points. Among those who underestimate expenditures by more than $5,000, the downward opinion shift, upon being informed of real levels, is 20 percentage points. On the other hand, those who overestimate expenditures barely budge in their opinions when told their districts spend less than they thought.
Sources of funding: who pays what? Americans are also poorly informed about the sources of funding for the nation’s schools. We asked half of our respondents, randomly selected, to estimate “what percentage of funding for schools currently comes from each level of government”—federal, state, and local. The question required respondents to make their percentages add up to 100. NCES data from 2011–12 (the most recent available) indicate that the actual levels are 10% for the federal government, 45% for state governments, and 45% for local governments. But people greatly overstate the federal share, estimating it as 32% (see Figure 5). In turn, they believe that state and local governments contribute less than they actually do.
The other half of respondents were asked how much funding should come from each of these sources. The average responses are 37% for the federal share, 35% for the state share, and 28% for the local share. In other words, people think the federal government should assume considerably more of the cost of schooling than its current 10% share, and local government should carry a considerably smaller burden than the 45% share it now bears.
Teacher salaries. To explore national opinion on teacher pay, we randomly divided our respondents into four groups. One group was simply asked whether teacher salaries should be raised. Another was asked whether taxes should be raised to fund salary increases. A third group was first told the average teacher salary in their states before being asked whether salaries should be raised. The fourth group was told the average teacher salary and then asked whether taxes should be raised to fund increases.
In the first group, 63% of respondents favor a pay increase for teachers (see Figure 6). Support falls to 45%, however, when the question (posed to the second group) asks about raising taxes to pay for teacher salaries.
In the third group, informed of current salaries, 45% of respondents support pay increases. And only 32% of people in the fourth group, told teacher salaries and asked if taxes should be raised, support a hike in teacher pay.
In sum, it is hard to say whether the public really wants a salary increase for teachers or not. It all depends on how much members of the public know and whether they are keeping in mind that the increment has to be covered by themselves as taxpayers.
Racial Disparities in Suspension Rates
In 2014 the U.S. Department of Education and the Department of Justice sent a joint letter to every school district in the country, urging local officials to avoid racial bias when suspending or expelling students. Officials were advised that they risked legal action if school disciplinary policies had “a disparate impact, i.e., a disproportionate and unjustified effect on students of a particular race.” In the Fall 2014 issue of Education Next, Richard Epstein, a professor at the New York University School of Law, criticized the action of the two departments, averring that it “forces school districts to comply with a substantive rule of dubious legal validity and practical soundness.” But in June 2015, the Supreme Court, in a Texas housing case, bolstered the departments’ position by holding that statistical evidence of “disparate impact” across racial groups could indeed be used as evidence that a government policy was discriminatory.
What do members of the public—and what do teachers—think of federally mandated “no-disparate-impact” disciplinary policies? And what do they think of such policies if set by local school districts? To find out, we split our sample into two randomly selected groups (see Figure 7). The first was asked whether it supported or opposed “federal policies that prevent schools from expelling or suspending black and Hispanic students at higher rates than other students.” Fifty-one percent of the public opposes such policies, while just 21% backs them. That division of opinion is essentially the same among the second group, who was asked about school district policies of the same sort. By a large margin, the public opposes “no-disparate impact” policies, regardless of whether the federal government or the local school district formulates them.
The division of opinion within the teaching profession approximates that of the public as a whole. A hefty 59% of teachers oppose federal “no-disparate impact” policies, while only 23% favor them.
Differences of opinion emerge along racial and ethnic lines. Among whites, only 14% favor the federal policies, while 57% oppose them. Higher levels of support are observed among African Americans—41% are in favor, 23% against. However, only 31% of Hispanic respondents approve of such policies, with 44% opposed.
Union Fees for Nonunion Teachers
In June 2015, the U.S. Supreme Court agreed to review an appeals court ruling in Friedrichs v. California Teachers Association, a case challenging a California law that allows public-sector unions to levy an agency fee on all teachers who refuse to join the union. Such fees are allowed in 21 states plus the District of Columbia.
Randi Weingarten, president of the American Federation of Teachers (AFT), defends the law on the grounds that “unions have a right to collect a fair share from the people [they] represent,” regardless of whether the people want to pay, so that the AFT can “ensure that we’re able to speak for all workers.” But teacher Rebecca Friedrichs, the lead plaintiff, contends that collective bargaining is political speech. Thus, she maintains, the required agency fee denies her constitutional right of free speech because the union uses her money to speak for purposes with which she disagrees.
The California law allows individual teachers to request a refund of the portion of their dues that is used for political purposes—helping to elect candidates, lobbying for union-sponsored legislation, or financially assisting like-minded groups. Such costs run into hundreds of millions of dollars, nearly one-third of the dues unions ask school districts to collect. But every teacher, union member or not, still must pay the remaining two-thirds of the fee to help fund collective bargaining. Friedrichs argues that the act of bargaining with public officials is every bit as political as donating to political campaigns.
Our data indicate that a plurality of people—indeed a decided majority of those with an opinion on the matter—agree with Friedrichs (see Figure 8). Only 34% support agency fees, while 43% oppose them, with the balance taking a neutral position. If we exclude the neutral group, then a clear majority, 56% of those with an opinion, say they want to end mandatory agency fees. This finding comports with the public’s overall opinion of teachers unions, as only 30% of respondents say unions have had a positive effect on schools and 40% say they have had a negative effect.
The more surprising results came from the teachers. Only 38% of teachers favor the agency fee, while 50% oppose it, with the remaining 13% expressing no opinion. In other words, 57% of teachers with an opinion on agency fees disagree with the AFT and the National Education Association. Union members constitute 46% of our teacher sample, roughly equal to national estimates of teachers union membership. Only 52% of these union teachers like the agency fee, and the approval rating plummets to 25% among nonunion teachers. These findings should not be extrapolated to say that teachers are turning against their unions more generally. Fifty-seven percent think the unions have had a positive effect on schools, and only a quarter think they have had a negative impact. But most teachers do seem to agree with Friedrichs that they should be able to decide whether to contribute money to cover collective-bargaining costs.
Academic Emphasis in K–12 Education
Have federal testing requirements forced schools to place excessive emphasis on math and reading? Have budget squeezes driven the arts out of the curriculum? Or are science, technology, engineering, and math (known as the STEM subjects) being ignored in favor of “softer” subjects? And, quite apart from striking the right balance among academic subjects, do schools place enough emphasis on cultivating
students’ character and creativity, educating them about global warming, and taking steps to prevent bullying? Finally, has the country’s passion for professional sports led schools to place too much emphasis on athletics?
All these questions can provoke passionate discussion. David Drew, an education professor at Claremont Graduate University, insists that “we have de-emphasized STEM … to the point that people who could have become scientists or engineers … didn’t get the educational experience they needed.” To which Rocco Landesmann, former chairman of the National Endowment for the Arts, replies: “We’re going to try to move forward all the kids who were left behind by No Child Left Behind…. It’s very often the arts that catches them.” Meanwhile, journalist Amanda Ripley says that “it’s worth reevaluating the American sporting tradition. If sports were not central to the mission of American high schools, then what would be?”
What do the people think? To find out, we conducted the first-ever experimental inquiry into such matters. We asked a random half of our respondents to estimate (on a scale from 1 to 7) how much emphasis they think their local schools place on each of several subjects and topics. The second half was asked to use the same scale to indicate how much emphasis should be placed on these subjects.
For every subject except sports, respondents in the second group think the subject should be given more emphasis than their counterparts in the first group perceive it is getting. In other words, the public thinks schools should place more emphasis on just about everything. Perhaps it is just human nature to say that other people should be doing more.
But if everyone wants more of almost everything, how much more varies with the subject and the population being interviewed. As illustrated in Figure 9, the public thinks much more emphasis should be placed on reading and math than do teachers and (to a lesser extent) parents. The public says that math and reading should be given a better than 1-point increment over the 5.2-point emphasis (on the 7-point scale) it perceives these subjects are now given. But teachers think the emphasis needs to be increased by only about half a point in reading and even less in math, while parents would increase the emphasis in the two subjects by no more than two-thirds of a point.
Meanwhile, teachers would give much greater (+1.7 points) emphasis to the arts than the 3.6 level teachers estimate it is now getting. Parents would give the arts only two-thirds of a point more emphasis, and the general public would boost its emphasis by only 0.8 more points. A similar, if smaller discrepancy is observed among the three groups when they are asked about history.
On other topics, the three groups—teachers, parents, and the general public—are more like-minded. All three think that character development and creativity deserve much more emphasis. But while parents and the general public also want far more attention given to bullying prevention, teachers think the matter only needs modestly more attention. On all these matters, opinion differences among the groups are marginal.
The extent to which public schools should emphasize global warming has become a political issue. In the recent debate over NCLB reauthorization, for example, Democratic senators sought to create a new program allowing districts to apply for funding to help teach about climate change. The Republican majority killed the proposal, emphasizing the degree to which the issue had become a partisan football. As Senator Lamar Alexander put it, “Just imagine what the curriculum on climate change would be if we shifted from President Obama to President Cruz and then back to President Sanders and then to President Trump.”
The partisan divisions in Congress extend to the public at large. Overall, our results would suggest that people want more emphasis placed on global warming—on average, about two-thirds of a point more. This gap is substantially smaller than the difference between what is perceived and what is desired on most other topics. The modest size of the gap masks substantial partisan divergence. Although Democrats and Republicans respond similarly when asked how much their local schools currently emphasize global warming (3.4 and 3.6 points, respectively), Democrats want the topic to be given 1.5 points more emphasis, while Republicans would give 0.3 points less emphasis. In short, Democrats and Republicans have similar views about the extent to which schools currently emphasize this issue, but they have very different preferences about how much schools should emphasize it.
To sum up, everyone wants more emphasis on just about everything, except athletics. The general public—as well as teachers—thinks sports should be given about a third of a point less emphasis than they believe it currently receives. Parents are less dissatisfied with the sports status quo.
The general public is especially eager for more emphasis on reading and math, while teachers see greater needs in history and the arts. Meanwhile, the attention given to global warming has the potential to generate as much polarization among ordinary citizens as it does among the elites in Washington.
Drawing Conclusions from the 2015 Poll
Many more findings from the 2015 Education Next poll are available in the full set of results available at educationnext.org/files/2015ednextpoll. pdf. Among them are:
· People think their local schools do a better job of attending to the needs of girls than of boys, with African Americans perceiving the largest gender differences in the way students are treated;
· A clear majority think 30% of high school instructional time should take place “independently through or on a computer”; and
· Support for school vouchers depends heavily on how a question about them is phrased.
Readers will decide for themselves which results are of greatest interest. In our view, the poll yields four especially important findings:
1) Support for standardized testing remains strong. Both teachers and the public at large oppose the idea of letting parents decide whether or not their children should participate in standards-based testing. About two-thirds of the public supports the federal mandate for testing of math and reading in grades 3 to 8 and in high school, although teachers are divided on this requirement.
2) Support for the Common Core State Standards declined a bit further in 2015, after falling sharply between 2013 and 2014. Among the public at large, support for the Common Core has fallen from a high of 65% to 53% in 2014 and to 49% in 2015. Among members of the general public (though not among teachers), those who favor the Common Core continue to outnumber opponents.
3) Union agency fees are not popular. A plurality of the American public—indeed a decided majority of those with an opinion on the matter—objects to the union practice of charging fees to nonmembers. An equally large share of teachers opposes the agency fees imposed on them by California and 20 other states.
4) A majority of people oppose the federal government’s new policy on school discipline. More than 50% disagree with the Obama administration’s mandate that schools must not expel or suspend black and Hispanic students at higher rates than other students. Just 21% back the idea.
Michael B. Henderson is research director for the Public Policy Research Lab at Louisiana State University. Paul E. Peterson, editor-in chief of Education Next, is professor and director of the Program on Education Policy and Governance at the Harvard Kennedy School. Martin R. West is associate professor at the Harvard Graduate School of Education and deputy director of the Program on Education Policy and Governance at the Harvard Kennedy School.