It turns out that Houston has a bit of a controversy going on in terms of how they evaluate teachers. The school board there was set to vote on a plan to evaluate teachers based on the Texas high stakes testing results. Teachers who aren't doing well may even be fired. Naturally, this has got some people overjoyed and others less so. It does raise - for me anyway - one issue that I think is worth really delving into. (For the sake of argument, let's assume that standardized testing is a valid way to measure student learning. Obviously there are pros and cons to the approach, but let's put those aside for this post.)
The president of the teachers' union there is quoted as saying, "There are so many factors that influence scores — school climate and leadership, not to mention how students woke up feeling on test day." Let's throw into that hopper things like socioeconomic status, nutrition, health and dental care, and more. I think most everyone would agree that kids don't walk into classrooms as blank slates ready to take in whatever the teacher throws at them. Rather, there are all sorts of things both in and out of school that have an impact on student learning. We know that. So let's take it to the next step.
Who has ultimate responsibility for ensuring that kids are learning?
If the school board approves this measure, they are saying it's the teachers. The teachers' union is saying that the principal and outside factors (like parents) also have some responsibility. All of the arguments have some validity. However, if we accept all of them, then who do we blame when things go wrong or praise when they go right? Can we take children away from their parents if their test scores aren't improving?
If we go down that road, we get into accountability nihlism and that's not ultimately going to help us very much. So then where does the responsibility lie?
In the end, I think we have to answer the question by saying that it's the schools. We can and should do everything we can to improve the living conditions for children in underserved communities to help on that front. We can and should provide as much support for parents as possible to help them raise their children so that they'll be able to succeed in school and in life. We can and should do a whole host of things to help the kids who need it most. But in the end, it's the school's job to educate kids and the buck needs to stop there - fairly or unfairly.
Should we fire teachers because of poor student performance? Not as a first line of defense, but it should probably be an option. As long as our goal is to improve schools first, I think we'll see much less need to punish them.
Showing posts with label accountability. Show all posts
Showing posts with label accountability. Show all posts
Friday, January 15, 2010
Wednesday, August 26, 2009
Where's the Middle Ground and Outrage?
Do you ever get the feeling that we're missing the middle ground when we have educational debates? I do sometimes. It seems like two camps stake out diametrically opposed positions and miss the fact that there's probably a significant amount of common ground in between.
Take promotional requirements for example. In Memphis, the city schools have banned flunking (and then retaining) students in pre-kindergarten through third grade. (If you really read the article it's not quite as touchy feely as it sounds.) Then, in New York City, the Mayor (with his recently renewed control of the schools) is looking to end - unspeak alert! - social promotion in fourth and sixth grade.
It seems like there should be a middle ground here between never retaining a child and always retaining children if they don't meet standardized test score cut offs. What about looking at each child on a case by case basis and determining what's best for that child? Sometimes it may be retention. Other times it may be promotion with heavy interventions and extra help. This all or nothing, black and white stuff just wears a little thin.
Of course, if a child can't pass the standardized tests in New York, maybe they do deserve to be retained. As reported by Gotham Schools last week, the scoring of the ELA test is done in such a way that a student who knew nothing and guessed randomly would likely (57%) pass the test with a level two.
Have I mentioned before that "accountability" definitely falls into that unspeak buzzword bin?
Let's just stop a moment and consider this information, though, because I think it's really stunning. We're basing a whole system - promotional criteria, curriculum, data systems, school report cards, mayoral campaigns, etc - around test scores that are supposed to measure what students know and how much they're learning. But the tests may be beaten by a student who randomly guesses all their answers! I know I've been out of town for the last two weeks, so maybe I missed it, but shouldn't there be outrage about this?
Oy. It's good to be back.
Take promotional requirements for example. In Memphis, the city schools have banned flunking (and then retaining) students in pre-kindergarten through third grade. (If you really read the article it's not quite as touchy feely as it sounds.) Then, in New York City, the Mayor (with his recently renewed control of the schools) is looking to end - unspeak alert! - social promotion in fourth and sixth grade.
It seems like there should be a middle ground here between never retaining a child and always retaining children if they don't meet standardized test score cut offs. What about looking at each child on a case by case basis and determining what's best for that child? Sometimes it may be retention. Other times it may be promotion with heavy interventions and extra help. This all or nothing, black and white stuff just wears a little thin.
Of course, if a child can't pass the standardized tests in New York, maybe they do deserve to be retained. As reported by Gotham Schools last week, the scoring of the ELA test is done in such a way that a student who knew nothing and guessed randomly would likely (57%) pass the test with a level two.
Have I mentioned before that "accountability" definitely falls into that unspeak buzzword bin?
Let's just stop a moment and consider this information, though, because I think it's really stunning. We're basing a whole system - promotional criteria, curriculum, data systems, school report cards, mayoral campaigns, etc - around test scores that are supposed to measure what students know and how much they're learning. But the tests may be beaten by a student who randomly guesses all their answers! I know I've been out of town for the last two weeks, so maybe I missed it, but shouldn't there be outrage about this?
Oy. It's good to be back.
Friday, July 17, 2009
Unspeak Bonanza
Well, it's Friday and the Daily News has come through with an educational unspeak bonanza. Unspeak, you may recall, is the art of making an argument without actually making an argument. It's about picking language that seeks to preclude any disagreement (think "support the middle class," whatever that means). Education is chock full of unspeak and the Daily News tapped into the motherlode this week.
In the article, "test experts" say that the tests being given to city kids are getting easier, which means more kids are passing and moving on to the next grade. In other words, standards are being dumbed down to undermine accountability and that's leading to social promotion. Holy unspeak!
Frankly, I don't habe the background or the data in front of me to agree or disagree with the claims. DOE says that the tests have gotten harder so they're requiring fewer correct answers to show proficiency. Critics say that it's a game to make the schools look better and reflect more favorably on the Mayor (juking the stats, as they'd say in The Wire).
That said, it sure does look bad. According to the story, on this year's test kids needed to answer 28% of questions correctly to be judged proficient. When you remember that these are multiple choice tests with four possible answers (and thus someone who guesses randomly on every question should figure on answering 25% of the questions correctly) that seems pretty low. A slilghtly lucky kid could simply guess his way to proficiency.
I guess the real question here is whether standardized testing is really the best way to assess standards. Unspeak though they may be, I agree that there needs to be some baseline of what kids should learn in school. I just doubt the bubble filling and all the drama and juking that comes with it is really the best way.
In the article, "test experts" say that the tests being given to city kids are getting easier, which means more kids are passing and moving on to the next grade. In other words, standards are being dumbed down to undermine accountability and that's leading to social promotion. Holy unspeak!
Frankly, I don't habe the background or the data in front of me to agree or disagree with the claims. DOE says that the tests have gotten harder so they're requiring fewer correct answers to show proficiency. Critics say that it's a game to make the schools look better and reflect more favorably on the Mayor (juking the stats, as they'd say in The Wire).
That said, it sure does look bad. According to the story, on this year's test kids needed to answer 28% of questions correctly to be judged proficient. When you remember that these are multiple choice tests with four possible answers (and thus someone who guesses randomly on every question should figure on answering 25% of the questions correctly) that seems pretty low. A slilghtly lucky kid could simply guess his way to proficiency.
I guess the real question here is whether standardized testing is really the best way to assess standards. Unspeak though they may be, I agree that there needs to be some baseline of what kids should learn in school. I just doubt the bubble filling and all the drama and juking that comes with it is really the best way.
Friday, April 17, 2009
Pre-Emptive Accountability
Every so often, it's good to hear some good news about education in this country. That's why it was great to hear about the success that Oregon has had in reducing its dropout rate and increasing the number of high school graduates. The state is now up to an 84% graduation rate while it only has a 3.7% dropout rate. (And yes, that leaves about 12% missing. I don't know what the deal with them is.)
As always, though, it's not enough to celebrate success or condemn failure - we need to look at why. That's what's going to help us reproduce the results we're seeing in Oregon.
The article isn't terribly specific on what's changed, but there's a few hints. The accountability director for the state education department credits better tracking of students and increased interventions for the ones who are in danger of dropping out.
What an obvious idea. Find the kids who are at risk and then help them achieve before they reach a crisis stage. It's so simple that it's brilliant. And effective too, as it turns out. It's a little pre-emptive accountability.
It kind of makes you wonder why we don't do this kind of thing more often. It seems like whenever you hear about accountability, it's in a punitive context (closing schools, holding kids back in a grade, etc.) and not in looking at how to help ahead of time. When a teacher or a school system issues a failing grade, it's a failure for the teacher/system too.
If we're serious about real accountability, let's start being a little proactive rather than just swooping in like an avenging angel after it's too late.
BONUS: According to the article, graduation rates tend to rise as the economy declines. Maybe Bush was focused on being the education president after all.
As always, though, it's not enough to celebrate success or condemn failure - we need to look at why. That's what's going to help us reproduce the results we're seeing in Oregon.
The article isn't terribly specific on what's changed, but there's a few hints. The accountability director for the state education department credits better tracking of students and increased interventions for the ones who are in danger of dropping out.
What an obvious idea. Find the kids who are at risk and then help them achieve before they reach a crisis stage. It's so simple that it's brilliant. And effective too, as it turns out. It's a little pre-emptive accountability.
It kind of makes you wonder why we don't do this kind of thing more often. It seems like whenever you hear about accountability, it's in a punitive context (closing schools, holding kids back in a grade, etc.) and not in looking at how to help ahead of time. When a teacher or a school system issues a failing grade, it's a failure for the teacher/system too.
If we're serious about real accountability, let's start being a little proactive rather than just swooping in like an avenging angel after it's too late.
BONUS: According to the article, graduation rates tend to rise as the economy declines. Maybe Bush was focused on being the education president after all.
Tuesday, February 24, 2009
Accountability and Shane Battier
When you look at it from the right angle, everything relates to education. Take the article about basketball player Shane Battier from the February 13th issue of New York Times Magazine. On the surface, it seems like it's just about a basketball player. But I read it as an insightful commentary on the weaknesses of our standardized system of accountability and measuring success.
The Battier article is interesting if you're a fan of basketball. If not, I probably wouldn't recommend it. The gist of the article is this: while Shane Battier's statistics (points, rebounds, steals, assists, etc) aren't all that great, he does so many unmeasured things that make him a great team player. While the data on his individual achievement doesn't necessarily show it, he makes the teams he plays on much better and the teams he plays against much worse.
So now you know that you should appreciate Shane Battier much more the next time you watch the Houston Rockets play. But let's pivot and see how this affects education.
If we were to view points, rebounds, and all that as a standardized test score, Battier wouldn't be doing all that well. He probably wouldn't be meeting standards (certainly not for a starter) and it's unlikely he'd be showing much yearly progress. You might even go so far as to label him a failing player.
But we know from this article that Battier is a highly valued player and member of his team. If we based all of our judgements on his scores we would completely miss that.
This isn't to say that if you're looking for a player to draft out of college or trade for that you shouldn't be looking at things like points and rebounds. I am saying that you shouldn't base everything on a statistical line. Similarly, I don't know if we can look at schools, teachers, and students based solely on that line. There's too great a chance of missing something fundamental.
Speaking of things that over-rely on statistical data, Arne Duncan has said that he's open to renaming the No Child Left Behind Act. Almost immediately, Eduwonk opened a contest to re-name the law. It's definitely worth checking out. My favorite has to be: Double Back Around To Pick Up The Children We Left Behind Act.
UPDATE: Turns out this post had pretty much already been written by Matthew Ladner. Great minds ...
The Battier article is interesting if you're a fan of basketball. If not, I probably wouldn't recommend it. The gist of the article is this: while Shane Battier's statistics (points, rebounds, steals, assists, etc) aren't all that great, he does so many unmeasured things that make him a great team player. While the data on his individual achievement doesn't necessarily show it, he makes the teams he plays on much better and the teams he plays against much worse.
So now you know that you should appreciate Shane Battier much more the next time you watch the Houston Rockets play. But let's pivot and see how this affects education.
If we were to view points, rebounds, and all that as a standardized test score, Battier wouldn't be doing all that well. He probably wouldn't be meeting standards (certainly not for a starter) and it's unlikely he'd be showing much yearly progress. You might even go so far as to label him a failing player.
But we know from this article that Battier is a highly valued player and member of his team. If we based all of our judgements on his scores we would completely miss that.
This isn't to say that if you're looking for a player to draft out of college or trade for that you shouldn't be looking at things like points and rebounds. I am saying that you shouldn't base everything on a statistical line. Similarly, I don't know if we can look at schools, teachers, and students based solely on that line. There's too great a chance of missing something fundamental.
Speaking of things that over-rely on statistical data, Arne Duncan has said that he's open to renaming the No Child Left Behind Act. Almost immediately, Eduwonk opened a contest to re-name the law. It's definitely worth checking out. My favorite has to be: Double Back Around To Pick Up The Children We Left Behind Act.
UPDATE: Turns out this post had pretty much already been written by Matthew Ladner. Great minds ...
Tuesday, January 13, 2009
A Stitch in Time
The St. Petersburg Times (Florida) ran a story last week about the effects of the Florida policy that has mandated that thousands of third graders repeat third grade after failing to meet the standards the first time. Apparently, minorities are more likely to be held back and that those students who are held back tend to do much better in school as a result of their retention. If you just read the first two (or even 12) paragraphs, it seems like a solid blow against that unspeak bugaboo, the dreaded social promotion.
For the purposes of this post, let's set aside the issues of minorities being more likely to be retained. This is apparently true even when you compare students of similar socioeconomic backgrounds and academic performances. That's interesting, but not what I'm looking at today.
Instead, I want to draw your attention to the 13th paragraph of the story which slips in that Florida law mandates that retained children get extra help including 90 minutes of extra reading instruction each day. No wonder the kids do better as a result. They get an extra hour and a half of reading instruction every day.
First of all, this is retention done right. Forcing kids to repeat a grade without providing extra help just proves that old definition of insanity (repeating the same action but expecting different results). If you're going to retain kids, you have to do something different to try to help them. An extra 90 minutes of reading instruction would do that trick.
If we were serious about educating kids instead of just showing how tough and "accountable" we're being, we might even say that it makes sense to try to be a little proactive about this stuff. Every school has a pretty good idea of who the kids are who aren't going to pass the test. Teachers know who's struggling in the classroom. So instead of waiting for the kids to fail and then retain them with the extra help, why not just give them the extra help. Intervene earlier to help close the knowledge and achievement gaps before they're really able to open. It just makes sense.
For the purposes of this post, let's set aside the issues of minorities being more likely to be retained. This is apparently true even when you compare students of similar socioeconomic backgrounds and academic performances. That's interesting, but not what I'm looking at today.
Instead, I want to draw your attention to the 13th paragraph of the story which slips in that Florida law mandates that retained children get extra help including 90 minutes of extra reading instruction each day. No wonder the kids do better as a result. They get an extra hour and a half of reading instruction every day.
First of all, this is retention done right. Forcing kids to repeat a grade without providing extra help just proves that old definition of insanity (repeating the same action but expecting different results). If you're going to retain kids, you have to do something different to try to help them. An extra 90 minutes of reading instruction would do that trick.
If we were serious about educating kids instead of just showing how tough and "accountable" we're being, we might even say that it makes sense to try to be a little proactive about this stuff. Every school has a pretty good idea of who the kids are who aren't going to pass the test. Teachers know who's struggling in the classroom. So instead of waiting for the kids to fail and then retain them with the extra help, why not just give them the extra help. Intervene earlier to help close the knowledge and achievement gaps before they're really able to open. It just makes sense.
Monday, January 12, 2009
An NCLB Epiphany
Michael Petrilli, a self-proclaimed "true believer" in the No Child Left Behind Law and former Bush administration official who worked on implementing the law itself, has an interesting piece in the National Review where he concludes that "NCLB as enacted is fundamentally flawed and probably beyond repair."
In short, Petrilli has come to the same conclusion that educators across the country have been screaming about for years now. Namely, that requiring "highly-qualified teachers" sounds great on paper, but isn't so sensical in practice, that mandating state-defined proficiency leads to a race to the bottom, that focusing on accountability through testing leads to a narrowly focused curriculum/test prep factory, and that there aren't enough good schools around to ensure that every child could go to one if we allowed for school choice.
Amen, Michael.
That being said, he still holds to five ideas that he says are central to NCLB and that he does still support, even if the law doesn't.
1. All children (even the poor ones) have the ability to learn
2. Accountability helps schools and individuals improve
3. Good teachers are needed for good education
4. Giving parents choice has positive benefits
5. Improving education is a "national imperative" in which the federal government can have a productive role
I agree wholeheartedly and without reservation to 1, 3, and 5. I suspect that one would have a hard time finding anyone in the educational world (even the non-"reformers") who disagree with 1 and 3. Even 5 would get you a pretty good degree of agreement. It's the even numbers up there that present some potential sticky points.
As I've written on this site over and over again, "accountability" is a great word that it's hard to argue against, but doesn't really mean anything. Does it mean testing? High stakes testing? High stakes for students? Teachers? Schools? What does it actually mean? It's well and good to say that people should be accountable for teaching and learning. In theory, I'm for that. I just haven't seen a good system yet for making that happen in a way that doesn't lead to the problems that even Petrilli outlines.
As for school choice, again, Petrilli put his finger right on the issue: even if we give everyone a choice of where they'll be going to school, there aren't enough good schools for every child to be in one. If there were, then every child would already be in a good school. It's just basic logic on this. So instead of starting with the idea of school choice, let's start with the idea of making the schools better. Then everyone benefits. And I don't think anyone would disagree with that goal.
In short, Petrilli has come to the same conclusion that educators across the country have been screaming about for years now. Namely, that requiring "highly-qualified teachers" sounds great on paper, but isn't so sensical in practice, that mandating state-defined proficiency leads to a race to the bottom, that focusing on accountability through testing leads to a narrowly focused curriculum/test prep factory, and that there aren't enough good schools around to ensure that every child could go to one if we allowed for school choice.
Amen, Michael.
That being said, he still holds to five ideas that he says are central to NCLB and that he does still support, even if the law doesn't.
1. All children (even the poor ones) have the ability to learn
2. Accountability helps schools and individuals improve
3. Good teachers are needed for good education
4. Giving parents choice has positive benefits
5. Improving education is a "national imperative" in which the federal government can have a productive role
I agree wholeheartedly and without reservation to 1, 3, and 5. I suspect that one would have a hard time finding anyone in the educational world (even the non-"reformers") who disagree with 1 and 3. Even 5 would get you a pretty good degree of agreement. It's the even numbers up there that present some potential sticky points.
As I've written on this site over and over again, "accountability" is a great word that it's hard to argue against, but doesn't really mean anything. Does it mean testing? High stakes testing? High stakes for students? Teachers? Schools? What does it actually mean? It's well and good to say that people should be accountable for teaching and learning. In theory, I'm for that. I just haven't seen a good system yet for making that happen in a way that doesn't lead to the problems that even Petrilli outlines.
As for school choice, again, Petrilli put his finger right on the issue: even if we give everyone a choice of where they'll be going to school, there aren't enough good schools for every child to be in one. If there were, then every child would already be in a good school. It's just basic logic on this. So instead of starting with the idea of school choice, let's start with the idea of making the schools better. Then everyone benefits. And I don't think anyone would disagree with that goal.
Wednesday, January 7, 2009
Who Watches the Watchers?
That seems to be the question surrounding education systems with mayoral control recently. Or, perhaps more accurately (though less fluently), who holds the people who say everyone should be held accountable accountable?
One of the big arguments for mayoral control is that it makes the system accountable for its results because you have one person in charge and you know where the buck stops. In turn, the buck stoppers have tended to try to implement accountability measures within the schools for students, teachers, and entire schools. The irony of all this is that despite all the hard data being collected, there's still a lot of debate as to what it means and whether or not it's working. Check out the Eduwonkette and Public School Parents blogs for probably hundreds of posts on this subject.
The problem with mayoral control is that all the authority for the district and for analyzing its results ends up resting in the hands of the very people being analyzed. This leads to the kind of phony accountability measures we've seen in New York, like the school progress reports.
Now, DC is getting on the bandwagon as their city council is pushing to have independent oversight over Michelle Rhee and her educational reform agenda. They Mayor there would like the oversight to be done by people who are on the record supporting Rhee and mayoral control. (Do you think he'd let kids choose friends to write high stakes test questions for them?) The Council hasn't made its wishes known on the matter yet, but they'll probably be looking for someone more skeptical.
New York opened their own independent research center just recently, though we haven't seen any results from it yet and it took most of Mayor Bloomberg's two terms (so far) in office to get it open. The irony here is that for all the talk from the administration about how important accountability is for the system, we see very little real accountability. Kids are held "accountable" entirely based on standardized test scores. Teachers are then (sort-of) held accountable for how the kids do on the tests. Schools are given grades based on how the kids do. But the system itself has never been looked at from an outside, independent, analytical perspective and judged based on it's performance. That's true in New York and it's true in DC. And that's what we're calling accountability?
One of the big arguments for mayoral control is that it makes the system accountable for its results because you have one person in charge and you know where the buck stops. In turn, the buck stoppers have tended to try to implement accountability measures within the schools for students, teachers, and entire schools. The irony of all this is that despite all the hard data being collected, there's still a lot of debate as to what it means and whether or not it's working. Check out the Eduwonkette and Public School Parents blogs for probably hundreds of posts on this subject.
The problem with mayoral control is that all the authority for the district and for analyzing its results ends up resting in the hands of the very people being analyzed. This leads to the kind of phony accountability measures we've seen in New York, like the school progress reports.
Now, DC is getting on the bandwagon as their city council is pushing to have independent oversight over Michelle Rhee and her educational reform agenda. They Mayor there would like the oversight to be done by people who are on the record supporting Rhee and mayoral control. (Do you think he'd let kids choose friends to write high stakes test questions for them?) The Council hasn't made its wishes known on the matter yet, but they'll probably be looking for someone more skeptical.
New York opened their own independent research center just recently, though we haven't seen any results from it yet and it took most of Mayor Bloomberg's two terms (so far) in office to get it open. The irony here is that for all the talk from the administration about how important accountability is for the system, we see very little real accountability. Kids are held "accountable" entirely based on standardized test scores. Teachers are then (sort-of) held accountable for how the kids do on the tests. Schools are given grades based on how the kids do. But the system itself has never been looked at from an outside, independent, analytical perspective and judged based on it's performance. That's true in New York and it's true in DC. And that's what we're calling accountability?
Subscribe to:
Posts (Atom)
