Peg Tyre, Author of The Good School, Answers Your Questions

This week, we solicited your questions for Peg Tyre, education journalist and author of The Good School: How Smart Parents Get Their Kids The Education They Deserve. You responded quickly, and so has she, with answers to a handful of your education-related questions, ranging from textbooks, to No Child Left Behind.

This turned into a smart conversation on a topic that affects all of us. Education policy and reform is certainly something we’ll keep coming back to on the blog. Thanks to everyone for participating.


Q. How important is preschool, and how much variation is there between preschools? Having just enrolled my 3-year-old daughter in preschool, I think the opportunity to get out of the house and socialize with other children is the main benefit to her development–not so much the topics that they learn about.

As for your comment that “teaching to the test” tends to emphasize a lowest common denominator: separating students into standard, honors, and AP versions of courses tends to alleviate this a bit, doesn’t it? While the teachers may still be teaching to the test, the higher-achieving students will be taught to a harder test, and usually on a more specific discipline (i.e. “Calculus” rather than “Math”), so they do get that more in-depth knowledge. -Kip

A. Okay — there are really two questions here. First one: preschool is not, at its best, simply social — although there are many important social lessons to be learned. Preschool teachers also need to be embedding important early math and pre-literacy skills in age appropriate play. Lots of talking, rhyming, singing, clapping out beats of words in a sentence as well.

Question Two: Really important distinction here. I’m writing about standardized tests — the ones all our kids take to determine how well they are doing and how well the school is doing (and in some places — if their teacher should get a bonus). Not end of the year exams for classes — which are a different beast all together. Until people understand the difference, our efforts at education reform will go… nowhere.

Q. Speaking of standardized testing, how do you feel about the (presumed) decline of American students test results (particularly mathematics) compared to other countries, even developing countries? Does this mean the other countries are just teaching to the test better, or is the overall teaching (or learning) better? –David

A. Good question. The international benchmark tests are built around standards which are not available to the people who give out the tests. (teachers and schools.) The state standards, which control the No Child Left Behind/Race to the Top tests are not only available, companies sell workshops and products on how to teach to them.

There is no teaching to the test on the international benchmark exams.

Kids do better in Singapore because they have better instruction. I know… ouch. But accurate.

Q. What do you make of adolescents who perform poorly in high school but excel in college? Or students who performed well in high school and struggle in college? And what does this mean for our school systems? –Marty

A. Another great question. I would refer you to an oped I wrote in the New York Times about grade fog. What happens, especially in low and middle-income communities, is a horrible betrayal of students and their families. Students are given A’s for C work– and what seems like a harmless boost actually deceives parents into thinking their children are being well-prepared for college. They take out loans, send Junior off to college where s/he sinks like a stone. Tragic.

Q. I’ve read your rationale for why “teaching to the test” is bad, but let’s agree that testing is going to stick around and teachers and school systems will continue to be judged by test results. More of the masses are happy with this configuration than others available. They want a tool to be able to develop critical thought about their local school system and the teachers in it. How do we improve the system? Do we develop the test differently? Do we grade the test differently, collectively. Do we grade the test differently between communities (based on socio-economic factors, for instance)? –Todd Sullivan

A. Such a thoughtful comment and question. I agree with you wholeheartedly that we, the masses, have a deep desire to assign a clean clear number (test scores) to this complicated undertaking we call education. The thing is, we think we have it and we don’t. And that is very dangerous.

Plenty of smart people in Washington know this needs to be fixed. And the federal government has set aside $350 million and asked two companies to come up with a new generation of tests to measure success in teaching/learning called the Common Core (a set of national curriculum standards, for those of you who are not following this rather arcane matter.) They have said they are going to use what is called in test jargon as “authentic assessment” which means actually doing chemistry experiments instead of answering three multiple choice questions about a chemical reaction etc. They are going to supply real-time information (as opposed to take the test in March and get the results in Oct, once the kids have moved to the next grade.) I haven’t seen a prototype of the new generations of tests — I hear they will depend a great deal on a sophisticated technological infrastructure — and right off the bat, I”m not sure how our cash-strapped states will afford that. We’ll see.

Q. Looking back, what would you say have been the greatest accomplishments and failures of the No Child Left Behind Act? –Ernest

A Well, I’m not a NCLB hater. I know, I know. Everyone hates this law but me. But here’s the thing: it was a bipartisan effort to focus the nation’s attention on the achievement gap between low-income kids and middle class kids — which until that time was a much ignored problem — and once you really look at the numbers, a national disgrace .

That said, the limitations of the law grow more obvious to me every day. The standardized tests, which have become the litmus test for school achievement, are not the instruments of measurement that we believe them to be. We want to believe they measure academic achievement but they don’t. And these days, the tail is wagging the dog. Standardized tests are deeply flawed, though control our schools’ curriculum.

Mike B

Back in the Q phase of this Q&A I should have mentioned that study that shows that if you band American schools by poverty level and then compare them to countries with similar poverty levels that our schools to better on those International Benchmark Tests than every other country. So for a suburban school with 3% child poverty those students will score higher than students in Finland and an urban school with a 20% child poverty rate those students will score higher than their peers in Mexico. Long story short our schools aren't the problem, its the poverty. Fix that and you'll fix the learning.


Within a competitive system be it due to the standarized tests or even when considering college/universitt applications it is not the "poverty" that flaws the system its the income gaps.
Stop poverty but still keep the rich MUCH better than the "not-quite-poor" and you will be left with the same problems.


I can suggest one way to improve the situation with math and other science subjects: get rid of multiple choice exams with 10+ questions. 3-4 medium to hard questions will provide better way to judge the level than many simple questions. The actual answer to such questions is not even that important; what is more important is a way how a person is trying to get to answer.

Teaching to this type of questions will require changes in educations; teachers will be forces to teach math in a methodical way slowly building a basis for solving problems instead of jumping from one chapter of test preparation to another.


I agree, esp for students with little focus.
It is the quality not quanity.
It is seeing that they get it in 3-4 questions for if there is 20 and they get 10 right, but their focus is gone then get the next 10 wrong(due to focus-frustration) or do not do them they still fail.

William Nuesslein

"[NCLB] was a bipartisan effort to focus the nation’s attention on the achievement gap between low-income kids and middle class kids — which until that time was a much ignored problem..."

What were the 1954 "Brown vs. Board of Education" case and the decade of school busing court orders in the 1970's about?

Anybody who could narrow the achievement gap would have gotten tons of favorable attention. In fact, there was a fellow in Los Angeles who had success in teaching mathematics to Hispanic children. A movie was made about him. But his great success seemed to be just a chance event.

It shocks me that people would put put any credence in the education policy George W. Bush and Ted Kennedy. They are two of the dumbest people to ever walk on Earth.


"What were the 1954 “Brown vs. Board of Education” case and the decade of school busing court orders in the 1970?s about?"

Please don't confuse Race with Income Level. There are low-income white kids and high-income black/hispanic kids and the school achievement gap holds for low vs high-income minority children.

NCLB was a good idea and while I don't agree with it's execution or the president that helped pass it I can support the attention it has drawn to education. The successor to NCLB should be an improvement on it.

Enter your name...

"I’m writing about standardized tests — the ones all our kids take to determine how well they are doing and how well the school is doing (and in some places — if their teacher should get a bonus)."

Um, the AP tests, which the OP mentioned as the example, *are* standardized tests. You know, "every student takes the same test under the same conditions"? That's what a standardized test is. How you use the test (say, to decide how well the school is doing) is not what makes something a standardized test.

Enter your name...

Another plug for the oped she links in the third: I loved it when it came out, and it's worth reading again. It discusses the serious disconnect between most school's grading systems -- your grade goes up if you bring a box of Kleenex to school, and down if you look bored during class -- and learning.

I wish that my local schools would adopt the system described there, in which students get separate grades for "life skills" (like being pleasant and turning in assignments on time) and "knowledge" (learning the academic material).


Okay, I'm a layman, but I've always thougt there was a stunningly simple answer to the "teaching to the test" conundrum. Let's say that coming out of 8th grade, there are 120 mathematics or English or history concepts or facts you want that child to have mastered.

Simply develop a large database and have tests generated randomly covering a majority of the concepts/facts. The teachers will have to teach to the universe, not to a target. Last year's or last semester's test won't look like this one.

This is done with the SATs, why not with all these tests? We wind up saying that district is terrible because their kids only got 98.3 percent on the math test, but our district is great because we got 99.1. Should the scores be more like 82 percent...more SAT-like?

It isn't the test -- it's teachers teaching to the test. So make it a true test and don't let them know.

PS: and certainly don't let them sit around with the test booklets afterwards and an eraser like they did in Chicago and Atlanta.


Enter your name...

That doesn't address the critics' concerns. There are two big concerns about teaching to the test:

1. That if there are 120 things in the curriculum/on the test, then you'll only teach those 120 things, and not the 780 other things that are more important to the critics. In other words, the system works as designed in making teachers teach the official curriculum, and the critics are mad because official curriculum doesn't prioritize their pet subjects.

2. That the test questions will be too simple (or the teachers will expect them to be too simple), so the teachers will only teach simple aspects of the 120 things in the curriculum. For example, the test will be all multiple choice, so the English teacher will teach grammar (which can be checked in multiple choice) but not how to write a coherent paragraph, which can't. The solution here is to spend more money on test grading, just like the SAT did when they added a writing section.



I don't know about you, but my standardized exit exam in New Jersey had an essay portion.


I'm a retired elementary principal who supports two aspects of NCLB: it holds schools accountable for ALL students and it set targets (defined by each state) for increasing achievement levels. I don't support the punitive aspect of NCLB; yet our school managed to increase student achievement/learning 6/7 years, meeting California's targets and chipping away at NCLB's in the face of sanctions and in the absence of rewards, as many other low-income schools are doing. And it's due to good teaching, teachers who are willing to change their ways, and staffs who believe kids can achieve. For example, our students learned test-taking skills and strategies, which are lifelong skills, as well as learning the content of the tests; we 'taught to the test' b/c the test reflects the Standards, which is what the kids should know and be able to do at their grade. We involved the students in goal-setting and in being responsible for their learning, too. Among other things, these strategies help. The main flaws in the standardized testing movement are it's focus on narrow assessments and curriculum, and the unintended consequence of 'leaving behind' faster learners and creative thinking, the core of our entrepreneurial economy. A well-rounded curriculum provides learners opportunities to problem-solve, to create, to apply, and evaluate content through experience - a constructivist approach - rather than the overemphasis on behaviorism in standardized testing. Kids in high-performing schools get both; those in struggling schools get only the latter.