May 2010 Archives

Michele McNeil has a very important post up on the U.S. Department of Education's plans to expand a data-collection requirement in the economic-stimulus bill.

The requirement essentially asks districts to provide information on disparities in expenditures on schools in the same district. Much of those disparities are caused by differentials in educators' salaries that often aren't taken into account.

As she writes, the expanded collection will be part of the Office of Civil Rights' biennial data collection, and might even become an annual requirement.

The odds are that, armed with this information, the administration can make a good case for changing the "comparability" requirement in Title I of the Elementary and Secondary Education Act. It has already signaled that it supports such a change in its ESEA blueprint.

(See here, here and here for background on the comparability requirements if you're coming to this for the first time. The basic idea is that local funding must be comparable among schools before they can receive federal Title I funds.)

This may seem like hopelessly wonky stuff, but think of it this way: If Congress indeed revises the requirement while rewriting ESEA, it would be one of the most far-reaching changes to Title I ever made. Districts, used to budgeting in average rather than actual salary terms, would have to overhaul their accounting systems. They'd be forced to think through how to equalize overall school funding between more- and less-affluent schools. Arguably, this change might have more repercussions than any tweaks made to the annual accountability system.

In a paper released this week by the Center for American Progress, authors John Affeldt and Guillermo Mayer note that a law in California tackling these intra-district funding disparities had bipartisan support, and that we might expect similar broad on Capitol Hill for a comparability change. I'm not so sure. The teachers' unions haven't supported revisions of the comparability language, fearing changes would cause districts to forcibly transfer teachers, and they still carry a lot of clout on the Hill.

Posted by Stephen Sawchuk at 11:29 AM | Permalink | Leave a comment | Recommend

In comments on a federal performance-pay initiative that's poised to expand, the National Education Association urged the Department of Education to require participating districts to prove that they've established a "competitive compensation schedule" before instituting a performance-bonus system.

I've always wondered what the union's vision of an ideal compensation schedule consists of, and this document essentially lays that out.

Apparently, the union feels that teachers' baseline salaries should be equivalent to those seen in other professions, or at least $40,000. No surprises there, since the $40,000 figure has been a core part of NEA's national salary initiative for some time.

But then I came across this line: A competitive salary advancement schedule means earning "significant" wage increases such that one's salary doubles in 10 years.

NEA literature calls for a "short and strong" salary schedule, and one of its bargaining experts told me once that the national office advocates for a compressed schedule, not one that goes on for dozens of years. But I didn't realize that the union had raises of that magnitude in mind.

So, do any schedules out there operate this way? Is it feasible from an economic or budgeting standpoint? What effect could this schedule have on flow of teachers into and out of the workforce? And would districts—and unions—that have bargained much longer schedules be willing to trade them in?

Those are just a few questions that came to my mind. Does anyone out there care to weigh in with some answers?

Posted by Stephen Sawchuk at 1:46 PM | Permalink | Leave a comment | Recommend

Earlier this year, when Florida Gov. Charlie Crist supported SB 6—which would have put all teachers on annual contracts and tied half of their pay and evaluations to student test scores—he was Public Enemy No. 1 for the Florida Education Association.

But, at the last moment, the then-Republican governor changed his mind and vetoed the bill. Public adulation by the FEA followed. And now Crist is reaping the FEA's support in the political arena: FEA took the unusual step of endorsing both Crist, now an independent, and a Democrat for a Senate race this fall at an AFL-CIO event.

It's hard to know exactly how this is going to affect the outcome of that race and whether appreciative teachers will really pony up for Crist inside the voting booth. The Associated Press story linked to above notes that the Democratic candidate, Kendrick Meek, is popular for his support of the state's class-size-reduction mandate .

Still, this comes as a good reminder of why politicians court teachers' unions: The unions have thousands of people in their constituencies who can help get out the vote. Would an independent Crist be a viable candidate without FEA support? Good question.

Posted by Stephen Sawchuk at 11:40 AM | Permalink | Leave a comment | Recommend

Not long ago, the American Civil Liberties Union sued the Los Angeles schools system and California over its policy for laying off teachers. The last-in, first-out system decimated teachers in a number of high-poverty, high-minority schools and, the group charged, deprived students of their state constitutional right to an education. ACLU later won an injunction in the case.

At the time, I wondered if layoffs would be increasingly viewed through this type of civil-rights lens. Now along comes an analysis from the Center on Reinventing Public Education that analyzes teacher-experience data from the 15 largest districts in California.

In 2008-2009, the paper states, teachers at risk of layoffs were indeed concentrated in the poorest schools. Schools in the highest poverty quartile, it found, were likely to see 25 percent more teacher layoffs than the quartile with the lowest concentration of poor students. Even more troubling, schools with the most minority students could lose 60 percent more of their teaching staff than those with the fewest such students.

Ammunition for civil rights groups? We won't really know, unless the ACLU mounts a larger-scale lawsuit or other groups like the Citizens' Commission on Civil Rights take this up in the courts.

Even then, this is potentially complicated territory. One of the important civil-rights voices, the Leadership Conference on Civil Rights, counts the American Federation of Teachers and the National Education Association as members. Neither of those teachers' unions is enthusiastic about ending seniority-based layoffs as part of the $23 billion jobs bill now winding its way on Capitol Hill.

And in fairness to the unions, you'd need good teacher evaluations to institute an alternative layoff system and most states and districts don't have those.

Still, having this kind of data out there must be uncomfortable for the Obama administration. Education Secretary Arne Duncan has pushed for the jobs bill but hasn't endorsed conditioning the aid on layoff reform.

Posted by Stephen Sawchuk at 10:34 AM | Permalink | Leave a comment | Recommend

Over the past year or so, we've talked a lot about principal, administrator, and peer reviews of teachers here at Teacher Beat. But what about students themselves? Can they offer valuable insights into which teachers are the most effective?

Quite a few school districts, New York City among them, perform comprehensive reviews of the school environment that include student-survey information. Typically, though, these surveys take place at the building level, rather than at the individual classroom or teacher level.

But a few instances of classroom-based surveys, including questions about specific teachers, do exist: The Knowledge Is Power Program schools use it as part of their"healthy schools" project, for instance.

And at least one national study is trying to gauge whether student reviews are correlated with other measures of teacher effectiveness.

The Bill & Melinda Gates Foundation's Measures of Effective Teaching study (click here for my writeup in EdWeek), will include a classroom-based student-survey measure in which students report on conditions in their own school, such as whether their teacher uses multiple explanations of topics and makes learning interesting. The measure is based on the survey component of the Tripod Project, founded by Harvard University's Ronald F. Ferguson.

Of course, using this type of feedback for research or general informational purposes is still quite different from including it in a formal teacher evaluation. So now I pose the question to you: Do you think it might be appropriate to solicit student feedback when reviewing a teacher's performance?

Posted by Stephen Sawchuk at 11:41 AM | Permalink | Leave a comment | Recommend

According to colleagues and sources who attended the Education Writers' Association recent conference, National Education Association President Dennis Van Roekel took the opportunity during a panel discussion to reiterate that it's "absurd" for districts and states to think that it's possible to use "one test on one day" to evaluate a teacher.

I am genuinely perplexed: I have not heard of a single district that's proposing to use just one test score to evaluate a teacher. The whole point of value-added measures of student growth is that they require at least two scores at different points in time to get a gauge of teacher effectiveness. Some states and districts require at least three or more discrete scores, and a few of the growth models out there, like the one used in Tennessee and elsewhere, take into account every test data point available on a student.

Even then, because these models are imperfect, experts say they must be combined with multiple other measures, such as reviews of practice conducted by administrators or peers.

The big disagreements over teacher-effectiveness legislation have concerned how heavily to weigh the VA-model component with these other measures.

Let me hasten to add that I'm in no way trying to minimize all of the thorny issues associated with incorporating student achievement into reviews of teachers.

But I recall that not all that long ago NEA's single test-score line managed to really tick off House Education and Labor chairman Rep. George Miller. Isn't it a sign that it's time to update a talking point when even lawmakers start to roll their eyes in response?

Posted by Stephen Sawchuk at 4:00 PM | Permalink | Leave a comment | Recommend

I've heard a lot of people question the wisdom of teacher transfers based on the reasoning that a teacher who's effective in one school setting might not be as effective in another, due to differences in the student population, the culture of the school, or the pedagogy/curricula.

So far there's been precious little research literature on this topic, but a new working paper from the National Bureau of Economic Research indicates that there are good reasons to investigate it in depth.

Analyzing a set of matched student-teacher data in North Carolina between 1995 and 2006, researcher C. Kirabo Jackson found that what he calls "match quality"—the factors that make a teacher more productive in one school setting than another—can account for up to a quarter of the observable "teacher effect," i.e., how effective a teacher appears to be at raising his or her students' academic achievement.

He also found that the teachers studied tended to be more effective in mathematics after they moved to a new school. That finding suggests that they are actively seeking out schools with a better match for their talent. Teachers were also less likely to leave their current school when the "match quality" was high.

"Part of what we typically interpret as a teacher quality effect is in fact a match quality effect that may not be portable across schooling environments," Jackson writes in the paper.

This is potentially an important thing to consider, especially in the context of the "equitable distribution" of teachers. Typically, that phrase refers to policies and incentives to encourage teachers who are the most effective to transfer to—and stay in—schools with high percentages of disadvantaged or minority students, or to keep highly effective teachers from transferring away from such schools.

After all, if a large part of teacher effectiveness is related to match quality, and if teachers themselves are responding to match quality in their choice of workplace, then equitable-distribution policies probably need to take into account which teacher-school pairings are the most successful.

Posted by Stephen Sawchuk at 3:17 PM | Permalink | Leave a comment | Recommend

Colorado's tenure and evaluation-reform bill passed with most of the core details intact, making what are probably the most aggressive Race-to-the-Top-inspired teacher-quality policy changes to date. The law includes a requirement that teachers be deemed "effective" three years running to earn tenure and a provision that would cause teachers to revert to probationary status if they have two successive "ineffective" ratings. (An appeals process will be granted to such teachers.) Effective teaching will be defined by making student-achievement growth at least half of the evaluation.

New York officials, in the meantime, have struck an agreement that would base 40 percent of the evaluation on student achievement. It also specifies that state standardized-test scores won't be more than than 20 percent of the evaluation with local measures forming the other 20 percent. The state legislature still must approve it.

The New York example is significant for another reason, though. Maybe I'm making a mountain out of a molehill, but it's the only one I know of so far that's mandated a student-achievement weight of less than 50 percent. That figure is so ubiquitous out there—Tennessee, D.C., Florida, and Louisiana all have it or have proposed it, while Rhode Island and Indiana proposed 51 percent—that a lot of people think it's part of the Race to the Top guidelines. (It isn't.)

Department of Education officials claimed there was no "magic number" to succeed in the teacher section of the Race to the Top application, but somehow all these lawmakers have fixed on one regardless.

There is not a whole lot of precedent for 50 percent. Some of the most famous extant evaluation/pay models, such as the Teacher Advancement Program, typically base only a quarter to a third of evaluations or pay on student achievement. UPDATED: The TAP folks wrote in to remind me that it's 50 percent when you consider both school-based and individual value-added measures; in the previous sentence I was referring only to individual VA. Sorry if this threw anyone off.

I can think of only one state so far that's left the student-achievement figure up to locals: Illinois. Even that law, though, has a trigger that would let the state board decide the figure for districts that dawdle in setting one with their local unions. Want to bet what it will end up being?

Posted by Stephen Sawchuk at 2:52 PM | Permalink | Leave a comment | Recommend

As goes Colorado so goes the nation? Right on the heels of the news that the Centennial state's National Education Association and American Federation of Teachers affiliates are on different pages about tenure legislation in that state—which is seen as a key element of its Race to the Top bid—an analogous situation is developing in Maryland.

The state is working on a plan to make student-achievement growth part of a new system of teacher evaluation. In its RTTT application, the state plans to make student growth 50 percent of the teacher rating, although test scores wouldn't be more than 35 percent of the overall growth figure.

The NEA affiliate says that the plan violates the recently passed law that laid the groundwork for changes to evaluation. But the Baltimore Teachers Union, an AFT affiliate, supports the state's application.

The Washington Post's Michael Birnbaum has the full scoop.

Posted by Stephen Sawchuk at 11:29 AM | Permalink | Leave a comment | Recommend

The term "job embedded" professional development is a big one in our field these days. It reflects the common-sense and increasingly influential idea that post-baccalaureate teacher training should be responsive to teachers' day to day practices in school and their own students' needs. The term has also shown up in a lot of recent legislation, including the economic-stimulus bill.

Unfortunately, for journalists like me, it's also a clunky, jargon-y term that doesn't do a very good job of conveying what the practice actually consists of. Fortunately, the National Comprehensive Center for Teacher Quality, which is housed at Learning Point Associates; the Mid-Atlantic Comprehensive Center; and the National Staff Development Council step in to fill the void with a new brief delineating just what this type of support looks like and how it can be supported by schools and districts.

The paper lists examples of the practices, some of which are done individually, and others with teams of educators. They include including action research, evaluating student work using protocol, case discussions, coaching, and lesson study.

Now, if only we could come up with a better name... let's hear your suggestions!

Posted by Stephen Sawchuk at 12:09 PM | Permalink | Leave a comment | Recommend

Oh, snap!

There's a bit of a disagreement between Colorado teachers' unions over the tenure bill making its way through the state legislature. Why? Because the state National Education Association affiliate won't support the bill but the state American Federation of Teachers, which is admittedly much smaller, has endorsed it.

The AFT has decided to throw its weight behind the bill because of new amendments, expected to pass, that would: insert a due-process procedure for teachers that revert to probationary status after poor evaluations; maintain seniority as a tiebreaker when laying off "effective" teachers; include teacher input in placement decisions; and give "recall rights" to teachers who are laid off.

The Colorado Education Association says that since it represents more teachers its opinion is the one that should really count. "They do not represent the teachers of Colorado; the CEA does," a spokeswoman said. That must be heartening for the 3,000 AFT teachers in the state. Teacher voice, anyone?

Posted by Stephen Sawchuk at 11:58 AM | Permalink | Leave a comment | Recommend

The popular Teach For America program is returning to Motor City, where 100 new graduates will apply for positions in the school's beleaguered public schools. This is a big deal for the organization, which hasn't had any placements there since 2003. Robert Bobb, the emergency financial manager for Detroit Public Schools, has been a supporter of the move.

The teacher-training group put out a four-paged press release announcing the news, and everyone but your mother is quoted in it: Bobb, TFA founder Wendy Kopp, University of Michigan professor Deborah Ball, and Governor Jennifer Granholm all praise the move.

Noticeably absent from the release, though, is Keith Johnson, the president of the Detroit Federation of Teachers. He is threatening to secure an injunction to stop the move if it means displacing some of the 2,000 teachers who have received pink slips.

Sounding a bit as though he's picked up some talking points from the National Education Association, a big critic of TFA, Johnson argues in this story that the move would "devalue" the teaching profession because the TFA candidates haven't received as much training as other teachers.

Posted by Stephen Sawchuk at 10:17 AM | Permalink | Leave a comment | Recommend

All eyes are on the tenure and evaluation bill in Colorado, which has passed the state Senate and is in the House right now. The reason for the attention is not just the Race to the Top subtext, but because the bill faces some hefty opposition from the state teachers' union.

Just how aggressive tenure-reform efforts can be before they spontaneously combust remains an open question. The recent Florida situation is a good example. There, a reform bill generated such outrage among teachers that Gov. Charlie Crist vetoed it. Crist, now running for the Senate as an independent, can probably count on some teacher votes.

The Colorado bill is somewhat less aggressive than the Florida bill, but still has plenty that makes the union nervous. It puts many more parameters around the work of the governor's commission, which is to hammer out a new teacher-effectiveness approach for the state.

For instance, the bill would require student-achievement growth to form at least 50 percent of a teachers' evaluation. It does not specifically tackle pay, as the Florida bill did, but it would direct the state education department to create a career-ladder system.

And while the bill would not dismantle tenure, tenured teachers earning multiple consecutive "unsatisfactory" evaluations would revert to probationary status and could be dismissed.

You're missing a lot of the story, though, if you think the bill is just about teacher issues. In fact, it would also put a lot more pressure on principals by basing at least 66 percent of their evaluation on a combination of growth in student scores and increases in teacher effectiveness.

And finally, the bill would require "mutual consent" for the placement of teachers, an issue that is also cropping up in Denver.

See these links for background information on the bill and a national overview of the issues associated with tenure reform here.

Posted by Stephen Sawchuk at 9:57 AM | Permalink | Leave a comment | Recommend