My fellow contributors all make strong points around the question of policymakers placing a specific focus on minorities in education – in this case issues related to the performance of Hispanics. By now, the statistics around poor academic performance and high drop-out rates are familiar, too.

With Hispanic learners the fastest-growing population in U.S. schools, we must make sure their learning environments are relevant and engaging to them. So, the short answer is that schools – all schools – must provide the best education possible to the students they serve, regardless of who they are.

Too many students languish in failing schools, especially those from economically disadvantaged backgrounds – which are often linked to minority communities, as our editor notes. The main order of business is to transform our failing schools.

In order to ensure continuous improvement of all schools, federal policy should allow states the flexibility to dedicate federal resources for school improvement activities in the lowest performing 15 percent of schools in the state. This will help states target resources strategically and prevent the next tier from falling further behind.

Federal resources should also support ongoing school improvement activities in low-performing schools demonstrating significant progress. Education policy should not restrict resources to the limitations of a grant cycle, but help schools transition to long-term sustainable outcomes.

I also agree with Tom Vander Ark on the issue of digital learning. We can’t continue to ask our kids to power down in the classroom when that’s not their reality away from school. With the proliferation of technology, we can create learning experiences that are relevant and customized for students, enhancing the overall quality of learning.

After giving students the tools to succeed and working to develop the schools they deserve, we must have high expectations that historically underserved students can do the work.

As part of KnowledgeWorks’ historic high school reform work in Ohio over a six-year period, we saw a 29 percent increase in the graduation rate of African-American students. During the same period, the state’s overall graduation rate rose 2.3 percent. Our work – which included the implementation of smaller schools and Early College High Schools -- was concentrated in the state’s most at-risk high schools with rising minority populations, shrinking family incomes and decreases in family education.

With new studies out that show billions of dollars added to the economy by cutting the high school drop-out rate in half for just one class, we literally cannot afford to allow any of our kids to slip through the cracks – but especially the “least of these.”

Well - sadly - it continues.

Monty predictably doesn't want to attribute the significant rise in NAEP scores in the 2000s to accountability. Of course not, that would go against the mission of his organization.

Instead, without any proof, he simply asserts the rise must have been due to larger social and funding issues. What might those be, Monty?

Total education funding in the nation went up pretty evenly during the 80s-90s and the 90s-2000s, with the exception of occasional small changes due to booms and dips in both decades.

Social changes? I get desegregation being influential in the late 70s and early 80s, but Monty can't even mount a hypothesis for what they might be in these decades, much less offer proof of impact of any of them.

I appreciate Sandy’s offer to find common ground, though I am quite uncertain what that common ground is.

First, to NCLB and the rate of improvement on NAEP. Since the bill was signed in Jan 2002, it could not much have affected the 2001-2 school year. Hence NCLB could not have impacted the period from 2000-02, the two reading test years. It may be that NCLB influenced the third year of the 2000-2003 period between the two math tests, but at best that would be very modest. In any case, since NCLB was signed NAEP gains have markedly slowed in the two subjects that are the focus of the law. This hardly speaks well for NCLB. As for common ground, I still do not understand how Sandy thinks NAEP deserves credit for gains made before the law was implemented.

If the federal law has not had a beneficial effect, perhaps state accountability actions have, Sandy says, and asks if that could form common ground.

Sorry, but it does not.

Back around the turn of the millennium, researchers were trying to figure out whether high-stakes, test-based accountability led to improving schools (as measured by NAEP scores in reading in math, which is inadequate but at least something). They sought to contrast states with such accountability with states that did not have it. This was not a simple task then, and cannot be done at all now since NCLB forced all states to have this form of accountability.

There were dueling research reports, with steadily increasing sophistication of statistical techniques. One interesting phenomenon was that several researchers who initially concluded that high-stakes accountability did lead to faster rates of score increases on NAEP changed their minds as more data came in. Both researchers at RAND as well as Martin Carnoy and Rebecca Loeb revised their views from concluding that high stakes produce improved scores on independent tests to concluding the evidence was at best mixed. Others, such as David Berliner, Audrey Amrein and colleagues concluded that high-stakes testing did not improve outcomes on other tests.

The primary competing hypothesis attributes the cause of NAEP gains to wider social factors. Thus, in analyzing an earlier period of significant NAEP gains by Blacks, the Civil Rights Project attributes it to the impact of desegregation. In many states, gains have followed increased education funding.

In my state of Massachusetts, markedly increased spending on education focused on low-income districts was accompanied by increased testing (but in only a few grades prior to NCLB). NAEP scores, already high, rose ahead of other states, including in the NCLB period. More recently, however, as aid to these communities has shrunk, MA’s NAEP scores have flattened. BTW, scores for Hispanic, Black, ELL and low-income students have been flattening or have been stagnant for some time in MA, and MA deserves its top-of-the-nation score ranking largely because it has relatively few students of color and less poverty compared with most states.

In sum, it is far more likely that NAEP gains are due to larger social and school funding issues than to accountability. Indeed, it is not at all clear that accountability itself contributed to gains on NAEP. (State test score inflation, however, is most likely due almost entirely to high-stakes, test-based accountability). On top of that are the well-known negative consequences flowing from test-based accountability (which may have been “unintended” but were certainly forewarned).

Sandy and I will have to find any common ground somewhere other than on the question of state accountability systems.

I close by noting that neither I nor FairTest oppose accountability. Indeed it is necessary, the public deserves a regular accounting, and done well accountability could help improve schools. But such accountability must look very different than test-based accountability, as the Forum on Educational Accountability (which I chair), among others, has been working on for years now.

I'm sure readers are as tired as I am of the back and forth with Monty Neill.

But though I remain utterly amazed at his continuing sleight of hand with NAEP numbers, this time, dear readers, I believe we can finally can find agreement!

But, first, let’s at least clear up a matter or two:

1. Monty likes to say that the math results were better from 2000-2003 than they were from 2003-2009. Let's recall once again that NCLB passed in December of 2001 and became law in January of 2002. For the life of me, I can't figure out how Monty keeps including all of 2002 and a good part of 2003 in the PRE-NCLB period?? The calendar suggests strongly that that period was POST-NCLB.

2. But, much more important, I want to reach out to Monty to see if we can find common ground here. I don't give all the credit for 2000-2009 improvement to NCLB. I do give most of it to consequential accountability, the movement that brought standards, testing, and consequences to a majority of the states by 2000 and then, through NCLB, to all of the states by 2002 and subsequent years.

States dramatically expanded the use of testing and accountability in the late 1990s. That was the policy rage of the time. So, Monty, if we can’t agree on NCLB, let’s at least agree that the huge growth of testing and accountability by 2000 contributed significantly to the dramatic NAEP gains for disadvantaged students in the 2000s.

I thought I'd try to answer some of the more general questions, adding to my post (below) which focuses on the real evidence NAEP provides about Hispanics.

First, NCLB has occurred when gaps in wealth have widened, rates of children in poverty have grown, and the U.S. has ignored this and its consequences for children and schools, in order to focus on test-based accountability. Schools serving low-income students mostly get far less than schools serving middle and upper-income youth, when the low-income youth should get more.

Second, the havoc wreaked on education by test-based accountability has hurt learning opportunities and results (see my post just below this one). Steve is correct that there have been and no doubt are teachers who don’t teach. But the “choice” should never have been between test-based schooling and allowing failures to continue unimpeded: there were and are other choices.

Third, NCLB managed to define “highly adequate teachers” as those who merely possessed certification, even if it was their first year in the classroom. But even this PR move was weakened when the Education Department said “teachers” didn’t even need to be licensed. When federal courts said that clearly contradicted NCLB itself, Congress hastily passed the weakened version into law. Congress needs to change this if it expects to prevent unprepared people from parading through the classrooms of low-income youth in a revolving door process.

Fourth, NCLB said to schools: “Almost inevitably, between now and 2014, you will fail to make AYP and start facing sanctions. But the sanctions do not rest on anything known to actually help schools get better, and schools in real need aren’t going to get the extra resources they need.” When Duncan got Race to the Top and School Improvement Grant money, he tied those to forced options (“models”) that lacked evidence they would help.

In short, NCLB foisted bad ideas and practices on all schools, with particularly harmful consequences for schools serving low-income youth, Hispanics and Blacks, and English language learners, as seen in declining rates of improvement on NAEP tests. While funding increased, it remained way short of what was and is needed, failing to provide additional resources that could be used in helpful ways.

Once again we need to take another look at NAEP scores, since Sandy Kress has used them to make the case for NCLB continuing to use high-stakes tests to hold schools “accountable.” The reality is that under NCLB, there has been almost no reading gain, the math gains are modest and like reading, have slowed down since NCLB was implemented. For Hispanic youth:

Grade 4 reading gain from 2002 until 2009: 4 points – less than one half of one point per year. Sandy used data from 2000, before NCLB even existed: from 2000 to 2002, the gain was 11 points.

Grade 8 reading gain from 2002 to 2009: 2 points -- less than one quarter of a point per year. The previous grade 8 reading test was 1998, and the gain from 1998 to 2002 was 4 points (test with accommodations, as is case in years following), or 1 point per year.

Grade 4 math gain from 2003 to 2009: 5 points. The gain from 2000 to 2003, however, was 14 points, with again the pre-NCLB period covering most of the gains Sandy wants to attribute to NCLB.

Grade 8 math gain from 2003 to 2009: 9 points. The gain from 2000 to 2003 was 6 points. Even in what amounts to the best case scenario for Sandy, the rate of gain in the NCLB period has slowed down compared with the previous period.

I have made this point several times in this column: since NCLB, the rate of gain in both subjects, both grades, for almost all categories of students, has slowed. It is clearly true for Hispanics.

So, for minimal to modest but slowing gains, it is worth turning the nation’s schools into test-prep programs. I find it amazing that some people still think so.

And Steve, it is true that this country chose to use test-based accountability instead of the sorts of programs that actually improve education, as many other countries have done (and which test from not at all to only 3 grades). But is not the real choice to demand of Congress that it scrap test-based ‘accountability’ and go the proven route of these other nations?

That would entail reducing required testing to 3 grades; dropping AYP and related sanctions (as Duncan does want to do); rejecting the evaluation of teachers by test scores (which Duncan wants to do despite the overwhelming evidence it is a bad idea); and switching from the unsound, unproven choices imposed on the lowest-scoring schools to the sorts of programs and approaches that do have some evidence.

Just last week, U.S. Census data revealed that about one in four school-aged children is Hispanic. In fact, low-income students and students of color together comprise a majority of students in America’s K-12 schools. So when black and Latino twelfth-graders perform at about the same level as white eighth-graders, it is not just a “minority” problem; it’s a national problem.

The sad truth is that we have an ugly history of ignoring the needs of low-income students and students of color. For generations, our nation’s schools have systematically taken the students who come to school with less, and then given them less of everything needed to spur high achievement. Low-income and minority students are still far less likely than their white and more affluent peers to be taught by teachers with an academic background in the subject matter they are teaching. They are still far less likely to be exposed to a rich and challenging college-preparatory curriculum. And the schools that educate them still consistently receive less state and local funding than do schools with fewer poor students and students of color—even when those schools are in the same district.

But here’s the good news: Despite the too-often crippling inequities our schools visit on these students—and despite dead-wrong myths that nonetheless continue to be spread about minority parents, falsely asserting that they don’t care about their children’s educations—the vast majority of Hispanic and African-American parents aspire to send their children to college. And while the gaps between poor, black, or Latino students and their white or more affluent classmates are still too wide, they are narrower now than they’ve ever been. Thanks to the hard work at schools like Bethune Elementary in New Orleans and Elmont (N.Y.) Memorial Junior/Senior High, there is clear evidence that when low-income students and students of color are taught at high levels, they achieve at high levels.

The achievement gap is real and it threatens our collective future. But it can be addressed by eliminating the systemic inequities of our education system. The federal government can and must provide the policies, guidance and incentives to ensure that this happens—in every state, district and school in the country. Doing so will leave these students—and our nation—in a better position to achieve and protect the American dream.

President Obama spoke Monday at a town hall hosted by a Spanish-language television network at a Washington, D.C., high school with predominantly Latino students. He was asked by a student, “Could you reduce the amount of tests?” and he responded, according to the official transcript on the White House website, that we should test less: “let’s figure out whether we have to do it every year or whether we can do it maybe every several years” and in a “less pressure-packed atmosphere.” (Search on "My name is Lisa" and read the whole response -- it's worth it.)

Secretary Duncan was there as well but didn’t chime in with, “Er, Mr. President, our policy is actually to do the exact opposite. We’re ramping up the tests and the pressure, particularly on schools serving disproportionately Latino students.” It’s unfortunate that the administration’s policy is so contrary to what the president himself seems to think wise.

My hope is that the administration and Congress, as well as state and local policymakers, maintain their stated commitment to addressing the needs of schools serving Latino students – but that they do so in a way that collaboratively engages with those communities. A school is part of a community, and the school’s strength depends on the community’s strength. There’s a role for top-down demands, but there’s also a role for bottom-up solutions.

It’s absolutely essential that policymakers, corporate leaders, educators and community members pay particular attention to the needs of all groups as we try to raise the college-going and graduation rates in the U.S. In fact, the Hispanic Scholarship Fund was founded on the belief that the country prospers when all Americans have access to the opportunities a college education can afford.

We are glad to have the President’s voice and support on these issues. We recognize that there are many barriers that often keep Latinos from earning a college degree, and we are working to overcome them through a broad range of outreach and education programs that help students and their families navigate collegiate life, from gaining admission and securing financial aid to finding employment after graduation.

We have launched Generation 1st Degree, a campaign aimed at encouraging and inspiring at least one person in every Latino household to graduate from college. These first-in-family graduates can be the spark that creates a college-bound culture within the Hispanic community.

Generation 1st Degree builds on sense of responsibility to the family that rests at the center of Hispanic-American life. HSF’s vision is for the U.S. Latino degree attainment rate to increase from 19 to 60 percent by 2025.

HSF estimates that percentage change will result in a significant increase in Latino lifetime earnings—from the current $24 trillion to $47 trillion (in current dollars) by 2025 if the goal is met.

This will be an effective way to address economic insecurity and poverty because it will help create new opportunities for financial growth that will not only benefit the Hispanic community, but all Americans.

Hispanics have been an integral and important part of the American experience from the beginning of this country. Now, we’re ready and able to do even more to help our families, our communities and our nation succeed.

Mr. Kress and Mr. Vander Ark make points worth noting. Mr. Kress even has the numbers to prove the point that accountability provides a structure that correlates with positive achievement.

While Mr. Kress’s data is convincing, and Mr. Vander Ark’s ideas are sensible, I see another positive post-accountability effect in schools: educators thinking differently and working better.

Sixteen years ago, I could walk into almost any elementary school and find at least one primary teacher who did little more than give kids coloring assignments. Today, the Crayola Curriculum is no more.

Sixteen years ago, I could walk into almost any high school and find at least one teacher for whom showing popular movies was a key instructional strategy. Today, the Movie of the Week approach does not exist.

My associates and I have worked in over 500 schools with more than 20,000 teachers since accountability began in the late 1990s. Are all of these schools better now? Probably not. But most are, and almost all are trying. Sixteen years ago, I saw few coordinated efforts toward school-wide improvement. Today, this is the rule rather than the exception.

Accountability doesn’t cause learning, but it encourages educators to move toward positive results-based changes in the structure and culture of their schools. Comparing 1995 with 2011, I see a different world in education today. There are parts of this world I do not like. But I prefer it to the way things were.

Through external accountability, many educators—and kids!—have developed habits that have lead to internal accountability. This is accountability that lasts. But if external accountability is withdrawn from most of our schools, and especially for the kids we need to be most accountable to, internal accountability will wane, and the structure and culture of many schools will cling to the status quo or begin to decline.

Accountability isn’t the answer. But it is an important part of the answer—for now.

Removing accountability for our highest performing schools makes sense. As long as they continue to demonstrate their effectiveness, we should give them the freedome they need to execute responsible innovations that sustain excellence.

For our lowest performing schools, the changes they need are so significant, and so long overdue, that measueres beyond conventional accountability are required. I am willing to support this even when I am uncomfortable about it.

Removing accountability for schools in the middle is irresponsible. Most have made gains with accountability in place and still have far to go. I see no proof that these schools would be better off were they to be freed of accountability.

I don’t like standardized testing. I don’t like how districts, schools, and teachers react to it. I really don’t like relying so heavily on a single mechanism that reduces complex human interactions and experiences to numbers between zero and one hundred.

At the same time, I don’t like thinking about what I experienced sixteen years ago either. More to the point, during these past sixteen years, I don’t think anyone has implemented at scale a successful substitute for test-based accountability. When they do, I’ll probably be quick to jump on their bandwagon.

Until then, however, I choose to moderate my distaste for standardized testing. I’d rather use my time and energy to help teachers and kids improve achievement through better instructional practice (which produces better test scores at the same time). I don’t see how complaining about any accountability system, when no viable alternative is present, makes things better.

Instead, I continue to develop and implement my own assessment models—approaches to identifying learning that are less corrosive of the human spirit and more empowering to individuals ready to own that power and to use it wisely, approaches that capture the most worthy aspects of human growth and educational achievement, and that encourage educators, parents, children, and our government to work together in attaining the things that matter most.

I encourage others to do the same.

NCLB has taught us, whether you agree or disagree with the law in principle or logistics, that it is vitally important to disaggregate student performance data. If we are to educate all students equally well, we must be fully aware of achievement patterns and trends among different groups of children in order to diagnose their needs and efficiently address their learning problems.

Disaggregating by race, ethnicity, special education status, poverty, etc., also helps us to identify the most vulnerable children and provide the needed supports early in their school careers.

It is a startling fact that based on national achievement measures, Hispanic (Latino) students post the lowest performance across all subjects, have the lowest attendance rates, and are largely absent from AP classes and tests nationwide. Given the aforementioned, it is not surprising that Latinos are almost three times as likely to drop out of school than their African-American peers, and have the lowest high school graduation rates. This is true for Latinos, whether they go to school in urban or rural areas.

The above facts are very concerning given that Latinos represent the single, largest student minority population in public schools, today, and their numbers are growing exponentially. It’s often been said that within the next generation or two, the United States will be a “majority minority country.” This tells us that solving the low performance issues of Latino students not only benefits everyone but it is essential to maintaining our nation’s economic health and ability to compete globally. By closely monitoring and paying special attention to the academic growth of this particular population of students, the closer we will get to creating productive citizens, who are college and career ready, and are able to take their rightful place in our society. It’s a win-win for all.

The educational needs of Latino students in this country are not unlike the needs of any other student. Access to quality early childhood education, effective teachers, and well-resourced classrooms are just some of what we know students need to be academically successful. 40% of Latino children are learning English as a second language and meeting those needs is critical.

In an effort to ensure our nation continues to maintain its competitive advantage in the world, President Obama has set the bar high by making it his goal that the country will lead the world again in the proportion of college graduates by 2020. I am confident that GEAR UP, a program I created 12 years ago, will help the President and our country meet this goal. We will only be successful by dramatically increasing the college completion rates for Latino, Native American, African American and Asian American students.

There have been numerous positive outcomes associated with GEAR UP programs throughout the country, including higher test scores and higher college enrollment rates compared to non-GEAR UP students. While there have been academic gains for low-income students and students of color alike because of GEAR UP, specific programs developed within GEAR UP have served Latino students especially well. GEAR UP Colorado, for example, developed a class for bilingual students, which led to nearly 90% of its students earning college credit while in high school. Furthermore, in the state of Washington, 77% of Latino students in GEAR UP enrolled in college compared to only 44% of non-GEAR UP Latino students in the state. GEAR UP programs around the country also offer important resources to parents regarding college, the application process, and financial aid awareness. Recently, in California, Glendale’s GEAR UP program provided three hands-on, multilingual FAFSA workshops for parents and students, which resulted in 90% percent of attendees completing the FAFSA application online and provided $1,000 scholarships to one student from each workshop.

The success stories of GEAR UP programs around the country are endless and GEAR UP often serves as the only college preparation assistance in many high poverty schools. As we continue to move toward balancing the budget, we must continue to make essential investments in all of our students, especially if we are to meet the President’s 2020 goal, a goal I am confident we can achieve.

The success stories of GEAR UP programs around the country are endless and GEAR UP often serves as the only college preparation assistance in many high poverty schools. As we continue to move toward balancing the budget, we must continue to make essential investments in all of our students, especially if we are to meet the President’s 2020 goal, a goal I am confident we can achieve.

What role do minorities, Hispanics in particular, play in the education debate? Consider a few facts:

• Hispanics are the fastest growing demographic segment in K-12 education, with the USDOE projecting a 36% increase in this population from 2007-2019; this compares to a 4% decrease among both white and black students over that span (source)
• There has been a persistent proficiency gap between whites and other student groups, including Hispanics, since the 1970s; however, this has been shrinking in recent years (source)
• Hispanics have a significantly higher dropout rate – 18.3% in 2008 – than whites (4.8%) or blacks (9.9%) (source)
• Those who do graduate are significantly underrepresented in the STEM majors – in fact, NACME found that just 4% of minority high school graduates had taken enough science and math to be fully qualified for admission to engineering study (source)

In other words, the fastest-growing segment of the K-12 student population has a higher dropout rate, persistently below-average level of achievement, and is mostly unprepared for postsecondary study in math and the sciences. Considering static or declining funding levels for schools and districts going forward, this is a significant education problem; looked at through the lens of millions of Sputnik-generation STEM-capable workers about to retire, it is a severe workforce and competitiveness problem.

One promising area: enlisting community support to improve student outcomes. National organizations promoting local engagement like US FIRST, National Academy Foundation, and Communities in Schools all have data highlighting their impact on underserved populations. I’ve also seen numerous case studies showing how mentoring initiatives, career/college awareness programs, and other models are making a difference in the lives of underserved students. There are even examples where indirect programs, such as CEO/principal mentoring programs, are helping to create better operations and more positive environments in schools serving high-minority and disadvantaged populations.

While they may not singlehandedly solve the challenges we face here, smart community engagement programs can make a dramatic contribution to improving outcomes for underserved populations, and are conspicuously absent from current discussions on the matter.

Nearing a quarter of the school populations, Hispanic students are general not well served by U.S. K-12 education. Here’s five stat policies that would help:

1. Close persistently failing schools. It looks like any conceivable ESEA deal will eliminate or weaken NCLB’s school accountability framework leaving it up to states to enact the ‘good school promise’ to low income families. That’s not reassuring given the national track record of tolerating chronic failure. It’s not popular, but closing bad schools is the only thing that seems to work.

2. Open good schools. This is something dozens of organizations know a lot about—at least about the old fashion prep school model: common intellectual mission, rigorous core curriculum, strong supports.

3. Expand online opportunity. It’s now possible to offer every student in American access to quality college prep courses and effective teachers. States should ensure that every student has access to multiple online learning providers and choice to the course level.

4. Weighted funding. One ugly remnant of America’s anachronistic approach to governance is funding that reflects community wealth rather than need. This must change if we are to narrow the preparation gap (the percentage of college-ready students). States can simultaneously shift to personal digital learning with portable need-based funding—that way everyone stands to benefit even though states will spend less on affluent kids.

5. Gift of time. Kids that start with academic disadvantages need more time. Mobile learning technology makes it quite affordable to customize learning and double academic time. But it will require that we shift from age-cohorts slogging through print for 180 days to a long year (or year round) competency-based system where there is no failure, just more time and support.

These equity-seeking recommendations are the backbone of the Digital Learning Now report.

Arne Duncan recently showed up at a gala for the American Association of People with Disabilities, proclaiming, "I remain your champion, your advocate, and your servant."

What he didn't tell the group was that his Blueprint for re-authorizing NCLB would remove disabled students from accountability in all but the worst 5% of the nation's schools.

The simple fact that NCLB requires school improvement where disabled students aren't progressing has led to unprecedented education gains.

Though NAEP math scores for 4th grade students with disabilities (SWDs) had decreased in prior years, they went up a remarkable 23 points from 2000-2009. This represents an advance of over 2 grade levels.

For 8th grade SWDs, the advance in math in the 2000s was 19 points, almost a full 2 grade levels.

While 8th grade reading for all students has been static, gains for 4th grade SWDs in the 2000s have been substantial, again, 23 points, over 2 grade levels.

Arne Duncan wants to get rid of accountability for disabled kids, and yet he says he's their "champion."

Who's kidding whom?

So, now, President Obama is going to give a speech this week about the importance of educating young Hispanic children.

I wonder if he will acknowledge in those remarks that his education Blueprint removes Hispanic students from accountability, except for those in the worst 5% of the nation's schools.

Will he tell the audience that Hispanic 9 year olds made unprecedented gains in math on the NAEP during the 2000s. Those gains were 21 points, an advance of over 2 grade levels?

Will he tell the audience that these young Hispanic students (who are the concern of his speech!) also made a gain of 14 points in reading during the 2000s, a gain of a grade level and a half?

Will the President explain why he's abandoning a policy that forces schools that fail Hispanic students to improve?

Who's kidding whom?

This administration is planting its policies in the ground in the form of a parade of irresponsible waivers and an abandonment of accountability. Once these policies take root, the scoreboard starts measuring disabled kids, Hispanic kids, and all kids on their watch.