Wednesday, April 25, 2007

Education on the Agenda

Check out the new Strong American Schools website. This group is pushing for a substantive education debate in the '08 presidential race as opposed to the usual vague generalities. Amen to that! An unusual idea, it's like a presidential campaign, but the candidate is education. And with the help of Roy Romer's prestige and big money from places like Broad and Gates, it looks like it's going to make some waves in '08.

Tuesday, April 24, 2007

Must Read Drop-Out Study

Senators Ted Kennedy (D-MA), Richard Burr (R-NC), and Jeff Bingaman (D-NM) yesterday introduced the Graduation Promise Act (GPA), which is being promoted by a coalition including the Center for American Progress, Jobs For the Future, Alliance for Excellent Education and The National Council of La Raza.

But before anybody moves on graduation rate policy everyone ought to all read this AMAZING! literature review on reducing drop-outs (with at a glance summary here), just published online by the Center for Public Education, and authored by education consultant Craig Jerald, who, if you don't know his work, is a man to watch in ed policy.

Though it was released completely without fanfare, the review shows that there is a lot more knowledge (including rigorous experimental studies) on preventing drop-outs than most people think. Jerald must have a vault of research studies in his basement because I've never seen a bibliography so diverse or comprehensive. From it he draws valuable and myth-busting insights like:

1) What schools do matters, even for the most at-risk students- in a randomized study using Check and Connect in Philadelphia, schools cut 4 year drop-out rates for high-risk students (low-income African-American males with disabilities and single-parent families) by 1/3rd and 5 year rates by 1/2.

2) Decreasing drop-out rates does not automatically increase graduation rates; in one Check & Connect study 4 year drop-out rates fell from 58 to 39 percent, while 4 year grad rates were up only 1 point (30% vs. 29%). To be effective programs have to address both monitoring/counseling and instruction.

3) Easy = boring - several studies found that making classes more challenging classes helped kids graduate and similarly unchallenging classes led to higher drop-out rates.

4) For programs to work they have to be intensive - low intensity, counseling, tutoring, or self-esteem workshops all had no effect on drop-out rates; successful programs essentially involved a caring, committed adult monitoring student attendance and grades almost daily, coaching students in academic and life skills, and intervening immediately if they start to fall off; as you can expect this kind of intensive approach can be expensive.

This is just the tip of the useful-info-iceberg and I highly encourage you to invest the time to read the full lit review and not just the at-a-glance summary. Coming soon, what I think this all means for grad rate improvement policy...

Thursday, April 12, 2007

On the subject of support among teachers...

This news from MoJo should be a wake up call to those who think NCLB will implement itself. NCLB will do little to improve education if teachers don't believe in or support it.

The Angel in the Details - Performance Pay that Works

Though everyone's sure to be blogging about this one, I couldn't help but jump on the bandwagon to mention something so worthy of enthusiasm. The Center for Teacher Quality has just come out with a report on teacher pay, which among other things advocates a salary range from $30K - $130K for teachers, incentives to work in high-poverty schools, performance pay for individual teachers and small groups of teachers who work collaboratively, and creating a career ladder with novice, professional, and expert designations for teachers----with corresponding levels of pay, responsibility, and authority. The study was led by an diverse group of 18 effective teachers, giving its conclusions that much more legitimacy.

What's great about this study is not just that it advocates things like performance pay and challenge pay--which are easy to advocate, but harder to implement--but that it puts enough nuance into its formulation of these ideas to offer some usable solutions. For example, one of the most frequent criticisms of performance pay is that it sets teachers in competition with each other and undermines the collaboration and teamwork necessary to good teaching. Though this argument can be overblown, some common, but ill-conceived formulations of performance pay, such as Florida's recent attempts, do produce irrational results. Florida offered performance bonuses to the top 25% of teachers in every school. Certainly this does de-incentivize the sharing of instructional know-how that would help fellow teachers advance to top performance, because if you help others rise they might surpass you and jeopardize your bonus. Likewise the Florida plan simultaneously rewards tons of bad teachers--even if every teacher is terrible in a school someone is definitionally going to be in the top 25%--and neglects very good teachers who teach in schools full of very good and some truly outstanding teachers.

And regardless of whether these problems turn out to be big or small, the perception among teachers that performance pay is against, not for, them is as much of problem as are the purported consequences of such programs on teacher collaboration and the inaccurracy of the program in recognizing success. The point of performance bonuses is to motivate teachers to perform better, recognize teachers who do, and retain quality teachers in the school system. In other words we want happy, hard-working teachers. If even great teachers distrust performance pay because of the way its designed, the program will be fail to motivate or retain. And if the average teacher, who is neither amazing nor incompetent, is pissed off about it you are not helping motivation or retention among the vast majority of teachers.

A typical approach some have proposed to these problems is school-wide bonuses for student achievement. But this idea just sets up a typical prisoner's dilemma system of incentives. If I work hard and no one else does I get no bonus. If everyone else works hard and I don't I still get a reward. So what incentive (other than the intrinsic ones, which existed prior to the peformanc pay program) would motivate me to work any harder? Collective accountability rarely produces incentives for individuals. Also, schoolwide bonuses often include all staff, administrators, secretaries, janitors etc. While I have any objection to paying the lunch lady well, but no amount you put in her pocket is going to raise student achievement.

But the CTQ recommendation heads objections off at the pass. By allowing all teachers to be eligible to receive bonuses based on absolute not relative performance it undercuts the harmful competition argument. And by offering bonuses for collaborative teams of teachers based on the performance of the students they teach together, it actually incentivizes the sharing of best practices and mutual support that opponents say peformance pay undermines. And pay for small teams minimizes if not eliminates the collective action problems associated with school-wide programs. As astute students of collective action know small groups create a regularized flow of communication about participation in group activities and relationship-based accountability for group success. Lastly, small-groups provide opportunities for leadership among a much broader range of school staff, and most importantly among staff who are still anchored in the classroom unlike most school administrators.

A combination of universal eligibility-bonuses for individuals and small, collaborative groups is simple elegant solution to one element of the problem of building teacher buy-in for performance pay. Read the full report for a treasure trove of such ideas. Kudos.

Don't Get Me Wrong...

Update: It occurred to me after reading the Gadfly's thoughts on the WaPo article referenced in my last post that some might interpret my remarks below either as advocating "drill and kill" methods of teaching or as admitting that NCLB is producing such methods on a mass scale. Both assumptions would be incorrect. As E.D. Hirsch has spent a lot of time demonstrating teaching to the test is not the best way to help students pass the test; teaching to high and meaningful standards is. As the principal of a famous, "success story" public high school says, "If you teach to advanced, proficiency takes care of itself."

The reason I harped on this point about why drill and kill test prep would still be an improvement for failing schools was because I wanted to show that the worst-case scenario version of NCLB would still be a good thing, or at least I suppose, the lesser evil. Like democracy, NCLB is the worst system of school accountability, except for all the others. So, for the sake of argument, I momentarily assumed that NCLB was as bad as its enemies say it is. I leave it to a soon-forthcoming post to talk about what I think the real consequences of NCLB are for pedagogy, and also what I think a vision of truly inspiring teaching looks like in the era of standards and accountability. Until then I challenge the David Keyes's and Dan Brown's of the world to prove that even their straw-man NCLB has (or would) produce worse results than the system that existed prior.

Tuesday, April 10, 2007

A Case for the Basics and Against "pre-NCLB" Mythology

This op-ed in the WaPo literally made me physically ill. In it a second grade teacher argues that No Child Left Behind has created a higher-level thinking skills gap between poor and minority students and their more advantaged peers.

Anyone who has spent any time in an inner city public school (like the one I grew up in) knows that schools have given poor and minority kids less challenging assignments, lower quality teachers, and less of just about everything that would conduce to developing higher order thinking skills since long before anyone ever dreamed up No Child Left Behind. David Keyes, can you produce one scrap of evidence that the public education system was doing anything to teach critical thinking to disadvantaged students before Congress decided to shine an ugly spotlight on the disgraceful results? Wake up.

Kevin Carey points out the backwards logic of Keyes' argument.

But I'm willing to go it a step further. Even if NCLB produces the boogey-man version of public education its critics claim it will, with a narrow curriculum, and constant test-prep, it will still be an improvement. I don't think for a second that this state of affairs would be either desirable or acceptable. But it would be better than the status quo ante. That's because contrary to the imaginary world in which this author and people like Dan Brown live, poor and minority students were getting neither critical thinking, nor basic-skills training prior to NCLB. If, because of NCLB, they at least get basic skills training, they'll at least be getting half of what they need instead of none.

As an example, a colleague of mine recently went to visit an elementary school in Baltimore which serves large numbers of disadvantaged students and which had posted high test scores. She was dissappointed to find that school was indeed focused very narrowly on reading and math to the neglect of subjects like social studies and in general did a lot of drilling of basic skills. She did some spot checking of the students and discovered happily however that despite not knowing things like where China could be found on the map, the students could read passages they had not seen before quite fluently and were rather proud of their ability to do so.

When I asked her what she would recommend to a parent considering this school versus other public schools in the area, she said unequivocally "I would tell them to send their kids there. It isn't ideal, but at least the kids can read." Having visited many other elementary schools in Baltimore she was well aware that while students in most of those schools do not know the geography of China, they also struggle to even decipher basic written passages.

The mistake many well-intentioned people make is thinking that NCLB is creating a ceiling on education; the reality is that it has merely established a floor; a school that doesn't teach children to at least read and do math, is no longer acceptable. And whatever criticisms people have of standardized tests they are actually extremely effective at measuring whether students can do such basic things.

That however is the mistake. The lie critics are telling themselves (and the educator community is currently awash in this form of self-deception) is that you can teach critical thinking to children who lack basic skills. If you haven't already read enough of this from people like Dan Brown listen to David Keyes from the WaPo article:

"Schools often use test-prep programs to try to raise test scores. The problem with these programs is that they teach the skills covered on tests, and only these skills. Poor and minority students spend hours repeating "B buh ball" and two plus two equals four. Every hour spent drilling basic skills is an hour not spent developing the higher-level thinking skills that are emphasized in wealthier school districts."

Does Keyes think a student who doesn't know that "two plus two equals four" will be able to do higher level math like Algebraic problem solving, or that a student who is struggling with "B buh ball" will be able to read a book about the American Civil Rights Movement? He is right that "every hour spent drilling basic skills is an hour not spent developing higher-level thinking skills." But what he seems oblivious to is that these hours learning basic skills must precede the hours spent developing higher-order thinking. Luckily for them, many affluent and white students walk in the school-house door with the basic skills already in place or half-way there. They've essentially had tutors in their homes for the past several years. But for the kids who weren't so lucky, the schools have to do some major catch-up work, and they have to do it fast or the defecits in the basics will prevent them from building on this foundation and they will end up even further behind. And sometimes playing quick catch-up requires some serious drill and kill. But please don't talk to me about how impoverished a form of education this is if your proposal is to teach a "rich curriculum" to students who can't even read their textbooks.

I am no cheerleader for No Child Left Behind. But I am grateful to it for helping us at least start to get to square one. I will be so happy to see us get from 1 to 2 and 3 and beyond. But the fact of the matter is that there has to be a minimum below which we will not let anyone fall. Students go through school not being able to read and understand the newspaper nor figure out how much a 15% downpayment on $150,000 house would be; (I know; I am currently working with an 11th grader who cannot do either.) Someone with that kind of defecit in basic skills will not make up for it because they can "make music" or "tell vivid stories." David Keyes, are you going to tell your graduating seniors to put those skills on a resume? Of course that would assume they were able to write one in the first place.

Wednesday, April 4, 2007

Can Schools Save Drop-Outs?

Some Discouraging Stats from Philly:

Likelihood of dropping out given certain life problems:

-Abused or neglected: 71.3%
-In Foster Care: 75.2%
-Juvenile Justice placement: 90.1%
-Gave birth during high school: 68.3%

-Number of high school age children with these problems in Philadelphia: 13,393

Source: Unfulfilled Promise: The Dimensions and Characteristics of Philadelphia's Dropout Crisis, 2000-2005

Some Encouraging Stats From Philly:

Effects of the "Check and Connect" identification and intervention program on a high-risk population in Philadelphia:
-No intervention (control group) Drop-out Rate (5 year): 94%
-Treatment Group Drop-out Rate (5 year): 42%
-Improvement due to program: Decrease 5 year drop-out rate by 1/2

-Effects of "Career Academies" alternative schools program on a high-risk population in Philadelphia:
-No intervention (control group) Drop-out Rate (4 year): 32%
-Treatment Group Drop-out Rate (4 year): 21%
-Improvement due to program: Decrease 4 year drop-out rate by 1/3

Source: "Keeping kids in school: Lessons from research about preventing dropouts" by Craig Jerald. Research summary available through the Center for Public Education. (White paper forthcoming).

Some food for thought From Philly:

-Cost per student of the "Check and Connect" program: $1600 per year
-Number of students in need of such programs in Philadelphia School District: ~85,000 to 90,000
-Cost for all Philadelphia's drop-outs of the "Check and Connect" program: ~$135 - $145 million
-Percent of students nationwide who fail to graduate: ~25% - 33%
-Number of students nationwide who fail to graduate: ~12 - 16 million
-Cost of check and connect nationwide: ~$19 - $25 billion
-Current federal Title I expenditures: 12.7 billion
-Increase necessary to save these students: Doubling federal expenditures

Take home message:

It can be done, but it's going to cost us.

More on this later.

Tuesday, April 3, 2007

The Little Credential That Couldn't

Though I focused on class-size, there was another great story of waste in the new Clotfelter, Ladd & Vigdor study: Master's degrees don't matter. Kevin Carey over at the Quick and the Ed writes this one up well and draws the fire he was looking for from teacher's union defenders of MA differentials. Check it out.

Monday, April 2, 2007

New York Squanders Opportunity to Innovate

After the blogathon that was my second post, I was a bit tuckered out, but this piece of news out of New York was just too much to bear.

The theme of the article is the state budget overall and it talks mostly about who is getting how much. The good news is that Spitzer has successfully pushed New York in the direction of Fair Student Funding (FSF). But buried within that big budget is the following gem:

The Bloomberg administration also praised the lifting of the state’s limit on charter schools [to 200], but there were compromises on that front, too, including a provision that automatically unionizes [emphasis mine] the employees of any charter school serving more than 250 students in its first two years.


Without wading too far into a discussion of Charter schools generally, it is fair to say that the whole idea of charter schools is based on allowing some schools to run themselves differently in order to innovate new ways to help kids. Some ideas will succeed and some will fail, but the point is trying something new to SEE if it will work. Because they are not unionized or governed by typical regulations, charter schools can experiment with policies like longer school days and school years, merit pay, or differential pay for teachers in shortage areas like math and science (or for that matter higher pay for teachers combined with larger class-sizes).

And anyone who has heard of the amazing success of some charter schools like KIPP knows that some of those ideas (particularly more time in school) have worked really, really well. Friends and enemies of charters alike recognize that one of the biggest differences between charter schools and district schools is the lack of unionization in charter schools. And one does not have to be an opponent of unions to admit that the collective bargaining process slows down and restrains the ability of a principal to try radical, new approaches. With automatic unionization, charters end up being not much more able to innovate than normal district schools.

It is another debate entirely to discuss whether the benefits of experimentation are worth the costs. But to allow experimentation only under the conditions that a school not actually be allowed to experiment is equivalent to banning the practice in the first place.

To say "You can have more charter schools, as long as you run them like district schools" is essentially to say "There will be no such thing as charter schools in New York."

P.S. On another fun note, the budget deal also requires New York to cut class-sizes. While I'll stay out of this fight for the moment, on account of New York really pushing the limits of potential non-linearity on class-size (frequent 35-40 person classes or so I hear), I doubt rather severely that the reductions will be targeted to those classes only.