Friday, April 17, 2015

Voice of San Diego discussion on teacher evaluations

I agree with much of what Heather Poland says in the comments section of yesterday's Voice of San Diego article, but I have to agree with Sherry S. regarding teacher evaluations.  Too many students are failing; something needs to be done. I am appalled by teacher union complacency with the status quo.

Heather says, "We already DO evaluate teachers. It is not a perfect measure, I don't think anything could be..."

Not perfect?  It's a complete joke!   Of course nothing can be perfect.  That's no excuse for maintaining the current failed system. 

Teacher evaluations could easily be much more effective. 

We could start with the obvious: observations by unbiased observers instead of, or in addition to, principals.

Most principals don’t even bother to observe teachers, and why should they?  Few principals have the skills or inclination to work closely with teachers who aren’t part of the clique that controls their school.  

Heather Poland describes the situation accurately: “We have seen how some principals just want teachers who are quiet and do what they say, regardless of their teaching ability. We have also seen great teachers who are also outspoken, be targets.”
And I think student scores should be part of teacher evaluations.  I think teachers should be given more credit for improving the scores of poor kids than for improving the scores of rich kids.

I agree with Heather that teachers should be paid more. At least, the best teachers should be paid more—lots more, maybe twice as much as a regular teacher.  And I don't even think that the less competent teachers should be fired.  The best teachers should be teaching most of the lessons in ALL of the classrooms—but they don’t have to stay all day in one classroom. Less competent teachers could be doing reinforcement of lessons and other activities that they are perfectly well-equipped to handle.

P.S. Regarding school uniforms: 

At my elementary school, children were required to wear black leather-soled shoes. It was a disaster during physical education class and recess. Those poor kids slipped and slid on the asphalt. It was much better when they just wore sneakers to school. But only the ruling clique of teachers had any input on how our school was run, so the problem never got fixed. 

Voice of San Diego
April 16, 2015


Heather Poland

I agree that an unbiased third party should be involved in evaluating teachers. The system now is not a joke, it can work well, but it is up to the principal to make it work. I would like specifics on just how it is a joke? I'm not sure the general public actually understands what goes into the evaluation?

I disagree that most principals do not observe teachers. They do. Yes, some more than others, but they do. Also not every site has a clique that "controls the school". Some schools do have dysfunctional climates, and that definitely needs to be addressed.

How do you know students are failing? There is a big misconception about this...

If you are going to compare country rankings, well, most countries don't educate every single person, like we do. Most western countries also don't have the poverty we do. When you control for poverty, we are actually at the top.

Maura Larkins

Heather, you say that the current teacher evaluation system "can work well", but that it is up to the principal to make it work.

Well, then, Heather, if it only works when there's a good principal to make it work, it doesn't really matter what the system is, does it?  It only matters who the principal is.

You say, "I'm not sure the general public actually understands what goes into the evaluation." 

You yourself already pointed out that principals can be very arbitrary in their evaluations.  So, again, it all depends on the principal, doesn't it?

Also, you say that principals DO observe teachers.
I assure you, Heather, that you are just plain wrong about this.  Many principals don't do observations.  I have gone years without having a principal do an observation, and so have many other teachers.
Of course not all schools are run by teacher cliques.  But when they are, obviously, the principals work with those cliques.  I stand by my statement that few principals have the skills or inclination to work with ALL the teachers in their school.  Of course, even the principals who aren't interested in what is going on in each classroom are plenty interested in the teachers who are part of any clique that might be running their school.

You write, "How do you know students are failing? There is a big misconception about this." You're joking, right, Heather? You think it's acceptable that our colleges have to teach remedial math and reading and writing to the graduates we produce? You think our drop-out rate is acceptable? You make clear that you think we shouldn't be expected to educate poor kids because some other countries don't.  But we're the richest country in the world!  It's not okay that we're creating an  underclass in an economy where employers require more and more education.

Here's a related story from the San Diego Reader:

...Kristin Phatak and Heather Poland, educators and parents who believe that students lose too much class time on test preparation and test taking, hosted the party to encourage parents to opt their kids out of standardized tests.

Heather Poland has two children who attend a local elementary school. She teaches in San Diego Unified School District, is an administrator for a national teachers’ association called BATS, and a California contact for the national United Opt Out group.

Poland said she administered a field test for the new standardized test last year. 

“My students struggled with this test,” said Poland. “One reason was because it’s online and they didn’t have pencil and paper to write things down. Many of them had trouble with the keyboard. The instructions were fairly complicated and even more so for English language learners…. The test will be graded on a curve, which condemns many students to fail.” ...

[Maura Larkins comment: Grading on a curve only condemns students to fail if they perform significantly worse than average. By definition, grading on a curve PREVENTS the majority of students from failing.

You imply that most of your students had trouble with this test so it would seem that almost ALL of them might fail if the test weren't graded on a curve.  

Are you saying that you think that the very lowest students should pass?  Why? Why would you say that?  You want a test that kids can pass whether or not they have mastered the skills being tested?]

...[Heather Poland] said, "I've always incorporated critical thinking into my classroom."

Here's another thread from the Voice of San Diego comments:

George Sheridan

San Diego Unified has much more money this year than anytime in the last several years. And it will have more money again next year and the year after. It's true that state funding of schools in California is significantly lower than in most other states. It's also true that we have significant challenges with high rates of poverty and high numbers of students learning English as a second language - factors that have a major impact on student achievement. But more money than last year means the district does not have to cut existing programs to fund raises which are long overdue.

Merit pay for teachers is not an original idea. It has been tried repeatedly over the last fifty years, and it has never worked. Of course it is subject to cronyism. Of course test scores are a terrible measure of teacher quality. But the biggest reason is that good teachers are not motivated primarily by money. Teachers want to be paid fairly. But there are very few teachers who would - or could - work harder or give students more one-on-one attention in exchange for higher pay. The average teacher now is working more than fifty hours per week.

There is no secret sauce for improving education. But everyone knows what good schools look like. We need to make sure all our students have access to the resources typical in the best schools, including fully staffed libraries, art and music and drama and career education and classes small enough for one-on-one attention. Really good schools work, not only because they have the resources to provide opportunity for all, but also because the educators work as a team. Merit pay is destructive of that necessary collaboration.

Maura Larkins

There are two obvious ways to combat cronyism  in teacher evaluations: use unbiased outsiders to do observations, and use student standardized test scores.
Test scores aren't a perfect measurement, but they are surely not affected by cronyism.  Over a period of years, student scores are a reliable measure of teacher effectiveness.  And the top 15% of teachers can be revealed even more quickly by student test scores: their students are consistent year after year.

Studies have shown that graduates of the best colleges have been shying away from the education field more and more over the past few decades.  As other fields open up, more women are choosing to become doctors and lawyers and executives, etc.  We need to bring some of these women back to teaching, and high pay--equivalent to the pay of those careers I just mentioned--would likely lure some excellent candidates to the teaching profession.

There are huge differences between teachers.  Some are much, much better teachers than others.  They should NOT be paid the same.  We should institute effective teacher evaluations for many reasons, such as identifying which teachers should be given extra help.  The best teachers should be utilized differently from mediocre teachers: they should be teaching more of the lessons and doing fewer of the activities that can be handled by their  less-skilled colleagues.

George Sheridan

 Almost everyone would like a better teacher evaluation system. The California Teachers Association has proposed what Professor Linda Darling-Hammond says would be the best system in the country. It would focus on formative assessment (the kind that says "Here's how you can improve") and would include peer observation and feedback. Obviously this won't work if people's pay is tied to those evaluations.

Your statement about test scores over a period of years is not supported by data. The best predictor of student test scores is the zip code where the students live.

Even in a given school, teachers do not usually have comparable classes. The best and most dedicated teachers often are assigned or volunteer to teach the most difficult students. As a result, their test scores may be lower than those of other teachers. In the same way, the very best hospitals often have the highest mortality rates, because they treat the sickest patients.

The psychometricians who design these tests caution that they are not valid and reliable for purposes of assessing teacher quality. Nonetheless, evaluation by test scores has been tried in a number of states in recent years, with entirely predictable results. The Teacher of the Year, found "highly effective" one year, is rated "unsatisfactory" the next on the basis of test scores.

Teacher pay needs to be raised across the board. Teachers are underpaid compared to other professionals with comparable levels of education. They are underpaid compared to other workers with comparable levels of responsibility. Teachers in SDUSD are underpaid compared with colleagues in neighboring districts. Many are taking home less money now than seven years ago. Raising teacher salaries won't fix every issue in the schools. But continued failure to raise those salaries will make things worse and worse.

Maura Larkins

 I am appalled that you would make this statement: "Your statement about test scores over a period of years is not supported by data."

You must know that the student test data used to evaluate teachers is the "value added" by the teacher to the students' test scores.

You are absolutely wrong when you deny that the top 15% of teachers (as well as the bottom 15% of teachers, by the way) can be spotted by the consistency of the value they add each year to their students' test scores, while the middle 70% of teachers tend to have dramatically fluctuating value-added scores.  It takes a number of years to figure out how well most teachers are actually performing.

Why would you deny this? You yourself admit, "The Teacher of the Year, found "highly effective" one year, is rated "unsatisfactory" the next on the basis of test scores."

You also say, "The best predictor of student test scores is the zip code where the students live."  Of course it is, but no one is suggesting that teachers should get credit for how high the scores are to start with.  Test scores are used to figure out how much the teacher IMPROVED the students' test scores.

I would have agreed with you if you had said that it is harder to raise the scores of kids who are below grade level.  But that problem can easily be fixed by tweaking the formulas for calculating a teacher's "value-added" score.

First you say, "But there are very few teachers who would - or could - work harder or give students more one-on-one attention in exchange for higher pay."

Then you say, "Obviously this [evaluation plan] won't work if people's pay is tied to those evaluations."

Aren't you contradicting yourself?

At first you seem to say that teachers always give everything they've got, no matter what they're paid.

Then you seem to be saying that there is some connection between pay and performance.
Or perhaps you meant something else.  Are you suggesting that merit pay would cause evaluators to falsify their evaluations of their peers?

In fact, even if pay isn't tied to peer evaluations, there are other motives for falsifying peer evaluations.  School politics exerts amazing influence over the ethics of teachers.  Teacher culture is not unlike high school clique culture.  And the situation is further complicated by genuine friendships between teachers.

So I'm not at all convinced that CTA has the best evaluation system.
Why not have observations by teachers from outside the school district, or at least outside the school?
I do think that there's no reason to fire teachers based on evaluations.  I think teachers who can't manage to achieve a high level of effectiveness should have different responsibilities and different pay than the most highly-effective teachers.

George Sheridan at the 2015 HRC Foundation’s Time To THRIVE Conference February 19, 2015
NEA Executive Committemember George Sheridan addresses the Time To THRIVE Conference in Portland, OR...

Published on Apr 3, 2014
George Sheridan appreciates your support in NEA Denver 2014]

 George Sheridan

A formative assessment system can only work if all parties are free to be completely honest about any weaknesses. It would take a very brave (some would say naive) teacher to tell an evaluator about his or her own weaknesses, knowing that the evaluation would determine the teacher's pay. But without such honesty, an evaluation system will do little to help the teacher improve.

The possibility of rating teachers based on the value they have added to student test scores was an interesting hypothesis ten or fifteen years ago. It has been thoroughly discredited by all major groups involved in education research and evaluation. See for example this article in Educational Leadership, and the citations at the end of the article.

 Maura Larkins note:

 Here's the board of directors of ASC.  It seems to be an organization of school administrators--including a "chief strategy officer"--with a couple of associate professors and business people  thrown in.  When you get school administrators together, I've noticed, you generally have a bunch of people who want to influence policy rather than conduct open inquiry.  

2014–15 Board of Directors and Executive Director

Nancy Gibson, President

Associate Professor
Concordia University
19W 233 Old Tavern Road
Oak Brook, IL 60523

Becky J. Berg, Immediate Past President

Marysville School District
4220 80th Street, NE
Marysville, WA 98270

Marie Adair 

Executive Director
New Jersey ASCD
3001 East Chestnut Avenue, #F49
Vineland, NJ 08361

Ronal Butler

Networking & Engineering Technologies, Inc.
2750 Killarney Drive, Suite 205
Woodbridge, VA 22192

Susie Carr

Retired Assistant Superintendent, Whitehall City Schools
Galloway, OH 43119

Jon Chapman

Chief Strategy Officer
EverFi, Inc.
3299 K Street, NW, Suite 400
Washington, DC 20007

Joshua Garcia

Deputy Superintendent
Tacoma School District
601 S. 8th Street, 8th Floor
Tacoma, WA 98405

David Mathis

Saluda County Schools
404 N. Wise Road
Saluda, SC 29138

Matt McClure

Chief Learning and Financial Officer, Cross County School District
LEAD 21 Program Coordinator/Facilitator, Arkansas Tech University
346 Highway 321
Beebe, AR 72012

William Potts-Datema

Acting Senior Advisor
Division of Adolescent and School Health
Centers for Disease Control and Prevention
1412 Sever Creek Drive
Lawrenceville, GA 30043

Lorraine Ringrose

Retired Principal, Patricia Heights Elementary School
114 Dormie Park Crescent
Vernon, British Columbia V1H 1Y7

Pam Vogel

East Union Community School District
1916 High School Drive
Afton, IA 50830

Judith Zimmerman

Associate Professor Emeritus
Bowling Green State University
17669 Ravine Drive
Elmore, OH 43416

Judy Seltz

Executive Director
Alexandria, Virginia

Maura Larkins' response to George Sheridan
Well, my goodness, you didn't mention that your evaluation system relied on teachers reporting their own weaknesses!

People often can't see their own weaknesses, so you have a problem here.

But I certainly agree with you that a self-evaluation system should ABSOLUTELY NOT BE CONNECTED TO PAY!

The truth is that most people would be loath to reveal their weaknesses to any but their closest friends.  This sounds like a pretty bogus evaluation plan.  It sounds more like a mentor plan. It needs to be accompanied by an effective empirical evaluation.

I suppose we'll all shrivel and die before CTA comes up with a meaningful evaluation plan.
And I wouldn't put much stock in an article by ASCD. Judging from its board of directors, it looks like an association of school administrators.

Maura Larkins

By the way, regarding that "Teacher of the Year" you mentioned who was found effective one year and then unsatisfactory on the basis of test scores: what measure was used to find the teacher effective in the beginning?  It wasn't by any chance a principal who happened to be a political ally, was it?  The finding wasn't based on popularity or charisma, was it?  In fact, in Los Angeles Unified it was discovered that MANY of the charming teachers who had been deemed superior by their cronies turned out to add little or nothing to their students' test scores. It's lovely for a teacher to be popular, but parents have a right to know when popular teachers aren't actually delivering results.

Gina Amodeo 

Maura, are you saying that principals don't need to be accountable for executing the evaluation properly?  If administrators used the resources available to them to evaluate and mentor teachers (and dismiss them if necessary), then we wouldn't be having this discussion. 
Additionally, I don't think Heather is saying that we shouldn't be expected to education "poor" kids.  I think she is saying that comparisons to other countries should be adjusted by the poverty factor, based on the fact that we do educate all children.

Also, regarding your language that teachers "produce" graduates.  This sounds like you are applying the corporate model of production to education, which is inappropriate.  Teachers teach to the needs of each individual child.  And each individual child comes into school with different resources, background, and baggage.  The best predictor of success in school is parent income and education level, not because that's where the best schools are, but because those students come "pre-loaded" with experiences and skills that children born in poverty do not have.  Children born in higher income households usually have early access to books, parents that read to them and talk to them, exposure to engaging educational experiences, and healthy nutrition.  Children raised in lower income communities have lower nutrition and other kinds of life experiences that might include having a parent in jail, neighborhood violence and crime, siblings in gangs, or an unstable home (e.g. homelessness).

Do you expect these students to learn at a faster rate in order to catch up with those students who enter school already knowing how to read?  Instead of blaming teachers, we should be thanking teachers who are teaching a class of 30 kids with 30 different sets of needs and experiences

You've come up with some amazing ideas that you seem to want to attribute to me.  How carefully did you read what I wrote?

I would love to have principals be held accountable for their performance, but how do you propose to hold them accountable for evaluating teachers effectively when you have no way of measuring whether or not their evaluations were accurate?  You need some objective measurement of teacher performance.

You can't evaluate principals properly when you can't evaluate teachers effectively.

An obvious means of evaluating principals is the same "value-added" system being used to evaluate teachers.  In other words, we'd measure how much improvement a principal brought about in teacher performance as measured by student test scores.

Schools produce graduates, Gina.  It has nothing to do with a corporate model of production.  You produce good results all day long, I'm sure, in your own life.  That doesn't mean you're applying a corporate model.  I really think it's better if we focus on ideas rather than trying to find political agendas in perfectly good words like "produce". 

I expect teachers to make improvements in their students' educational levels. You've belabored the obvious about differences in children.  As I said before, it is harder to raise scores of low-income kids than to raise scores of high-income kids, but that's no excuse for teachers to insist that no one should hold them accountable for bringing about measurable academic gains in their students.


Maura, are you one of these plants?:

To Dennis M. Doyle:
Dennis, you should know exactly who I am.  I worked in Chula Vista Elementary School District when you were Assistant Superintendent, before you went to National School District.  You inspired me to try to do something about serious problems in schools.  Do you seriously not remember me?????

I'm actually not a big fan of Michelle Rhee.  I was disgusted that she tried to raise scores in DC schools by focusing on the middle class kids, whose scores are easier to improve, while abandoning the poor kids.  What thoughts do you have about Michelle Rhee?

My ideas are completely my own.  What are your ideas, Dennis? Do you have some criticism of something I've said? If so, why you don't share your ideas with us? 

P.S.  It occurs to me that you are in a position to confirm that when you were at CVESD many principals failed to perform even minimally-adequate observations of teachers.  For a short time in the early 90s CVESD principals were required to do full-lesson observations complete with detailed notes and a follow-up interview, then the requirement was mysteriously dropped.  Please tell us why that requirement was dropped.


Maura, I am sorry to say but you have me mixed up with someone else. I am not "smart" enough to be an administrator.

I am just someone who feels strongly about public education and its importance to our society.


So, what is your last name, Dennis, if it's not actually Doyle?  We're not supposed to be anonymous here.  All the links to commenters' profiles seem to have gotten broken when VOSD set up its new website.  But since you yourself brought up the issue of "plants", I am surprised that you have chosen to continue to give only your first name.  In fact, I'm surprised that someone concerned about the integrity of the comments section would have chosen a first-name-only username.  If you look at the other usernames, you'll see that many, if not most, of us have given at least give the initials of our last names, if not our full last names.

It's interesting that you seem to think that school administrators are notably "smart".  I doubt that a teacher would have written that statement.  In fact, I can't think of anyone who would have made that statement--except a school administrator.

For years many people have marveled at the spectacularly-low GRE scores of school administrators. I have long suspected that the average teacher is smarter than the average school administrator.

Here's something I've never heard anyone mention: if we recruited more teachers from the highest-performing tier of college graduates, we'd get an added bonus: smarter school administrators--since almost all school administrators started out as teachers.

If you feel strongly about public education and its importance to our society, why don't you express some of your thoughts about teacher evaluations or some other issue I've addressed here?  Why do you limit yourself to merely suggesting that I might be a "plant" and that school administrators are smart?

Each of your assertions about Value-added measurement of teacher effectiveness is contradicted by the research reported in this article.

(to George Sheridan)

You have linked to an opinion piece, not a research article.  At least the author of the article is honest about having cherry-picked a couple of "worst-case scenarios" and relied on "five articles" published by AERA.  And what is AERA?  "Within the AERA community of education researchers, members belong to one or more of...over 155 special interest groups (SIGs)."

And you yourself have made an amazing blanket statement, Mr. Sheridan.  You claim that each of my assertions is contradicted by the article!

I take it that you are claiming that Value Added scores do not fluctuate wildly for most teachers, as I stated?

Sorry, Mr. Sheridan, the article clearly backs me up on this.  Valerie Strauss writes, "[T]he five articles detail problems with the instability of VAM results “year to year...'"

You are claiming that the top 15% and the bottom 15% of teachers do NOT have more consistent Value Added scores?

Again, the article does not back you up.

The author correctly states that there are problems with VAM, but what else is new?

The author has not attempted to be even-handed in her piece.  She makes misleading claims about what VAM supporters say. Obviously, no formula can adjust an individual child's score for the fact that he has a stomach ache.  The author is dishonest to imply that VAM claims to be able to adjust a child's score for stomach aches.

All statistical analysis requires a large-enough sample so that a stomach ache won't affect the outcome.  (Of course, if an entire classroom had food poisoning, the analysis would be thrown off.)
Both Valerie Strauss and you, Mr. Sheridan, must surely know that when you take the scores of an entire classroom you can get a composite score that is reasonably reliable despite the fact that some kids have stomach aches.

Shame on both of you for ignoring this basic principle underlying all statistical conclusions.

And I don't see anything in the article that contradicts my claim that teachers need to be evaluated by unbiased observers in order for evaluations to be accurate.

You are opposed to any type of effective evaluation of teachers, aren't you, Mr. Sheridan? It's not just about test scores, is it?


Maura, no subterfuge with my name although I am a little apprehensive posting because of news like this:
My knowledge is with SDUSD. Don't know hardly a thing about South Bay schools.
I apologize if the "plant" comment upset you. It was meant to be tongue in cheek. My humor doesn't get through the keyboard sometimes. Just wanted to show you the depth in which the oligarchs so called "reform movement" is reaching to upend public education.

Maura Larkins

VOSD instituted a policy a few years ago of not allowing any commenter to be anonymous, but obviously you are able to conceal your last name.  In fact, I noticed in February that your profile had been disconnected, Dennis.  I reported the problem to VOSD, and next thing I knew, ALL the profiles of all commenters were disconnected.  I was told that the relaunch of the VOSD website caused the problems.

I'm waiting for VOSD to either reconnect the links to commenter's profiles, or to announce a new policy allowing anonymity. 


Anonymous said...

Wow, you have a lot of time on your hands! Don't forget to find the video of me speaking to the board earlier this week!

Maura Larkins said...

Dear Ms. Poland:

Are you saying that only people with too much time on their hands are interested in the subject of standardized testing?

Why would you post comments on the Internet or speak to reporters if you didn't want anyone to pay attention to your efforts to get parents to opt out of standardized testing?

Is it not okay for me to disagree with you about the current state of education?