This article is behind a paywall. Previous HN commenters have pointed out that you can get the full text of the article if Google is your referrer. (I.E. you google the article title and click on the result).
Standard WSJ hit piece. It's all the fault of unions.
In spite of the fact that out of 178 people the superintendent and 34 principals were in on it. People very much NOT in the teacher's union and generally hostile to it.
But, hey, an actual conversation about education, socioeconomics, and achievement? Can't have that--people might actually figure out that testing is a gigantic Republican boondoggle.
Just blame the unions--it's what our readership want.
Okay, let's have that "actual conversation" -- if testing is just a giant boondoggle, what method(s) do you favor for measuring education and achievement in a socioeconomic-neutral fashion?
I have never seen or heard of a test that is SES-neutral. Even simple "G-Loaded" activities like counting backwards shows both racial and SES bias in younger children.
I personally think giving principals more power over hiring and firing and salaries of teachers would go a lot further than testing to improve academic achievement at the school level.
The problem isn't testing. Testing is fine and is actually necessary. Even testing with socioeconomic bias is okay since you can norm to historic values to see trends.
The problem occurs at two points:
1) when you then use the results of those tests more broadly than applicable.
Comparing results in the same school year on year is generally valid (although you have to watch out for events like a big employer closing). Comparing two different schools close in location requires a bit of finesse. Comparing schools in completely different locations with completely different demographics is generally useless--yet NCLB effectively tries to lump all schools together with the same tests.
2) when you apply those tests to provide punitive measures
Why don't people get that doing this is going distort the system? Software people wail continuously about "you can't judge my craft with objective measures." Practically any measure in software I can dream up: number of bugs fixed, number of bugs found, number of commits, number of lines of code written, etc. will be met with cries of "that doesn't measure my ability" AND "if you promote on those measures, people will start gaming the measures". And they will be right.
Why can't people see that the exact same thing will occur in teaching?
The standard complaint is that only programmers can judge programmers and that non-programmer managers can't judge correctly. Yet, with respect to teaching, we magically think that any moron can judge teaching ability and that non-teaching managers are well equipped to judge teachers.
This is one of my hot buttons. Tech folks are so very quick to defend their own area with "you can't objectively judge and punish me" and so quick to apply "but I can objectively judge and punish you."
And, even when a school is doing things right, it's really hard to keep it going. Positive results take years to show up unequivocally while negative results can appear almost immediately.
The bit about how quickly things dropped one year when they didn't have parents come in 3-times-a-year to review report cards is particularly enlightening. Achievement is getting a lot of little things right, and only the teachers can see when they're right and when they're wrong.
By contrast, it took 10 years to move from 15% passing reading and math to 66%. That's roughly 5% a year, and I suspect the first years were slower than that. 5% is statistical noise to a test--which means that all of the things that were occurring in the first years would have been filtered out by a test. Not the result you wanted.
So, given that achievement at that school is now dropping, what does the test tell us to do? Where does the test say: "Whoops. The teachers are the same, so A) either the children got worse or B) we need to unroll every single change made by administrators in the last year." Where is THAT application of the test?
> Practically any measure in software I can dream up ... will be met with cries of "that doesn't measure my ability" AND "if you promote on those measures, people will start gaming the measures". And they will be right.
Sure, any measurement system will fail, but there are means to make them more robust. One way is to make it so that what is measured is what is desired: gaming the system results in more of the desired behavior. Consider a support desk measured not by time-to-resolution or call volume, but post-resolution surveys: in order to 'game' the measurement, the agents must actively seek to promote customer satisfaction.
Another, more controversial, means for more robust measurements might be an overt, active gamification. Using your example of LoC: instead of unilaterally praising large (or small) work efforts, LoC is used as the measurement in a bid war. I'm not envisioning a particular bid process (silent auction? Liar's dice style round-robin?) but anybody can call BS on the latest bid ... which then holds that bidder to implement. Beat your own estimate for praise, but for punishment you would have to exceed the prior (non-BS) wager. Couple a second bid value (deadline, for example) and the complexities of actively "gaming the system" becomes nigh on impossible.
Between this article about the Atlanta cheating scandal and the NPR story you linked, I wholeheartedly believe that the "standardized test" regimen is currently broken. However, I was hopeful that you might have some suggestions on alternative measurements or ways we could better apply the current ones.
Using tests for an "objective measure" is doomed on its face. That's gigantic sign that the people involved are more interested in blame transfer than in progress.
You can use tests for trends. That's a good use of tests. You can use tests for feedback to the individuals directly involved. That's also a good use of tests.
This corresponds to my "software metrics" points. If I suddenly see an uptick in average lines of code committed, I should take a look. If I start seeing a big increase in bugs filed, I should take a look and see what just happened. Most of the time, there's a good explanation. Sometimes, someone is doing something stupid and I need to go fix that. And, rarely, someone is doing something malicious and needs to be fired.
Tests should be used as an indicator to a human to "look here!" However, test results are simply an indicator to a human to exercise judgment--not an objective measure to remove humans from the loop.
The problem with education is that people really don't want to use the tests as "indicators". As in the article I pointed out, that particular school now is having a decrease in results--the tests are indicating with a big, red klaxon that something is going wrong. People should be examining the situation with a magnifying glass to figure out the issues and fix them.
The problem is that "something going wrong" is very clear--a new principal and new directives from the head bureaucrats. Do you think that they will look at the test results and say "Oh, we need to undo these policies and get a new principal?" Unlikely.
Would you like a metric that only ever gets used against you but that your management gets to ignore? Of course not. It's why the complaints by programmers about management are legion.
And that's why teachers hate all this stuff. Somehow, it only ever gets used against the teachers. When the evidence points at the administrators, school board or politicians, it gets ignored.
This isn't even limited to teachers. Physicians oppose the collection of this kind of data about outcomes for the sames reasons--it will very quickly get turned into a ranking system that gets used against doctors rather than for improving outcomes.
>> If the environment that produced this horrific behavior in Atlanta is “toxic,” blame the people who control that environment, not the testing regime that attempts to hold those people accountable. The teachers unions that run our schools view public education, first and foremost, as a jobs program for adults. That is why the unions fight so hard to keep open failing schools and want seniority—rather than teaching ability—to determine layoffs.
I don't think the author of this piece manages to back up his argument very well. The teachers' union says that the testing regime is flawed, and that the toxic environment is created by it; the author says that the testing regime is "flawed" but not "fatally" so, without substantitively adressing that concern. The "people in charge" of this environment are in turn governed by No Child Left Behind and state/federal regulation and laws -- there's a chain of authority here, and the brokenness of the system can rightly be argued to flow down from the highest policy levels all the way to local middle-management.
>> Would you want your child taught by someone who flunked the certification test five times, let alone 10?
This is a silly argument. How many people take multiple tries to get a driving license? The test either ceritifies you or it doesn't.
>> And would that instructor be more or less likely to resort to changing student test scores to hide his own incompetence?
Absolutely no logical connection, other than a wild assertion by an author clearly hostile to the teachers' union, and it seems to me, teachers in general.
Like it or not, public school teachers in the US are vastly underpaid and overworked (especially in urban school districts with low budgets), and have their ability to teach severely hampered by testing requirements that may or may not be beneficial to their students. The whole system is broken, but demonizing the people that are attempting to educate in these conditions without trying to understand the pressures at play isn't very constructive.
First, the environment was toxic because the school administration was corrupt at the top and the school system had too much influence over the testing. To prevent further such abuse all testing should be done by accredited third party groups.
Second, get off this silly mantra that they are underpaid. In some areas yes, but in Atlanta that is not the case. A simple query of various counties here produces results and for the number of days worked their pay is actually quite good; that and anecdotal being friend with a 10 year teacher who makes good money and who berates the system(which includes the union) for the problems. They cannot fix what they are not allowed too.
> all testing should be done by accredited third party groups.
These 'accredited third party groups' would need to be a) accredited, b) be paid. And it's already happening here in the US[0]. What you have now is an overhead of bureaucracy that 'accredits' and outsources the testing to these private companies, who pay call center wages to mark high school exams based on shoddy metrics that are aimed at maximising profits. All the markers are temps/contract, and this accreditation system is essentially a cartel of companies that lobby to ensure their business remains. I cannot see the logic how this can end up benefiting the students.
Secondly, teachers being underpaid is not a mantra - it is a fact [1]. In the US, the worth society places on a job is strongly correlated to the pay. The accreditation and pay, resources and social respect that teachers (do not) receive I wager is directly correlated with the quality of the education students receive. The system is so broken as to rely on standardized testing as to further punish 'under-performing' schools by limiting availability to funding based on scores.
Neither source you give is reliable, both are heavily biased.
I prefer actual charts from counties that pay, given 190 days of work http://www.dekalb.k12.ga.us/human-resources/teacher-salary-s... in a respectable county in Georgia this isn't what I would called underpaid. This is before benefits. Its not hard to find these charts for most counties.
The areas that need attention is all the money spent on administrative people other than teachers. The US has spent increasing amounts of money on education with near zero results.
> To prevent further such abuse all testing should be done by accredited third party groups.
How do those groups earn money? By charging for their testing? So groups that produce better results are selected for, which risks groups that have easier tests or more relaxed scoring being selected for by schools.
This has happened in some parts of England for GCSE exams. (These exams are taken by schoolchildren aged 16. )
While this may reek of "ad hominem", I believe it's important to know that the Wall Street Journal is for conservative audiences by conservative authors.
As such, the editorial pages normally present the truth if and only if that truth conveniently intersects with their ideology. Not saying that liberal outlets don't have the same bias, more so that its opinion pages are not a good source for reasonable, effective analysis of, or solutions for, United States policy shortcomings.
This is not new, nor is it unexpected[1]. If you incentivize an outcome, people will try to get to that outcome. The problem with No Child Left Behind, is that they (foolishly) presumed that everyone would take the most difficult path to that outcome.
If they wanted to stay with the testing metric framework, they would have been much better off with external testers. However, that still would have teachers "teaching the test" instead of teaching the subject.
16 comments
[ 3.8 ms ] story [ 52.3 ms ] threadIn spite of the fact that out of 178 people the superintendent and 34 principals were in on it. People very much NOT in the teacher's union and generally hostile to it.
But, hey, an actual conversation about education, socioeconomics, and achievement? Can't have that--people might actually figure out that testing is a gigantic Republican boondoggle.
Just blame the unions--it's what our readership want.
I personally think giving principals more power over hiring and firing and salaries of teachers would go a lot further than testing to improve academic achievement at the school level.
The problem occurs at two points:
1) when you then use the results of those tests more broadly than applicable.
Comparing results in the same school year on year is generally valid (although you have to watch out for events like a big employer closing). Comparing two different schools close in location requires a bit of finesse. Comparing schools in completely different locations with completely different demographics is generally useless--yet NCLB effectively tries to lump all schools together with the same tests.
2) when you apply those tests to provide punitive measures
Why don't people get that doing this is going distort the system? Software people wail continuously about "you can't judge my craft with objective measures." Practically any measure in software I can dream up: number of bugs fixed, number of bugs found, number of commits, number of lines of code written, etc. will be met with cries of "that doesn't measure my ability" AND "if you promote on those measures, people will start gaming the measures". And they will be right.
Why can't people see that the exact same thing will occur in teaching?
The standard complaint is that only programmers can judge programmers and that non-programmer managers can't judge correctly. Yet, with respect to teaching, we magically think that any moron can judge teaching ability and that non-teaching managers are well equipped to judge teachers.
This is one of my hot buttons. Tech folks are so very quick to defend their own area with "you can't objectively judge and punish me" and so quick to apply "but I can objectively judge and punish you."
And, even when a school is doing things right, it's really hard to keep it going. Positive results take years to show up unequivocally while negative results can appear almost immediately.
See: http://www.thisamericanlife.org/radio-archives/episode/275/t...
The bit about how quickly things dropped one year when they didn't have parents come in 3-times-a-year to review report cards is particularly enlightening. Achievement is getting a lot of little things right, and only the teachers can see when they're right and when they're wrong.
By contrast, it took 10 years to move from 15% passing reading and math to 66%. That's roughly 5% a year, and I suspect the first years were slower than that. 5% is statistical noise to a test--which means that all of the things that were occurring in the first years would have been filtered out by a test. Not the result you wanted.
So, given that achievement at that school is now dropping, what does the test tell us to do? Where does the test say: "Whoops. The teachers are the same, so A) either the children got worse or B) we need to unroll every single change made by administrators in the last year." Where is THAT application of the test?
Now do you see the problem with testing?
Sure, any measurement system will fail, but there are means to make them more robust. One way is to make it so that what is measured is what is desired: gaming the system results in more of the desired behavior. Consider a support desk measured not by time-to-resolution or call volume, but post-resolution surveys: in order to 'game' the measurement, the agents must actively seek to promote customer satisfaction.
Another, more controversial, means for more robust measurements might be an overt, active gamification. Using your example of LoC: instead of unilaterally praising large (or small) work efforts, LoC is used as the measurement in a bid war. I'm not envisioning a particular bid process (silent auction? Liar's dice style round-robin?) but anybody can call BS on the latest bid ... which then holds that bidder to implement. Beat your own estimate for praise, but for punishment you would have to exceed the prior (non-BS) wager. Couple a second bid value (deadline, for example) and the complexities of actively "gaming the system" becomes nigh on impossible.
Between this article about the Atlanta cheating scandal and the NPR story you linked, I wholeheartedly believe that the "standardized test" regimen is currently broken. However, I was hopeful that you might have some suggestions on alternative measurements or ways we could better apply the current ones.
You can use tests for trends. That's a good use of tests. You can use tests for feedback to the individuals directly involved. That's also a good use of tests.
This corresponds to my "software metrics" points. If I suddenly see an uptick in average lines of code committed, I should take a look. If I start seeing a big increase in bugs filed, I should take a look and see what just happened. Most of the time, there's a good explanation. Sometimes, someone is doing something stupid and I need to go fix that. And, rarely, someone is doing something malicious and needs to be fired.
Tests should be used as an indicator to a human to "look here!" However, test results are simply an indicator to a human to exercise judgment--not an objective measure to remove humans from the loop.
The problem with education is that people really don't want to use the tests as "indicators". As in the article I pointed out, that particular school now is having a decrease in results--the tests are indicating with a big, red klaxon that something is going wrong. People should be examining the situation with a magnifying glass to figure out the issues and fix them.
The problem is that "something going wrong" is very clear--a new principal and new directives from the head bureaucrats. Do you think that they will look at the test results and say "Oh, we need to undo these policies and get a new principal?" Unlikely.
Would you like a metric that only ever gets used against you but that your management gets to ignore? Of course not. It's why the complaints by programmers about management are legion.
And that's why teachers hate all this stuff. Somehow, it only ever gets used against the teachers. When the evidence points at the administrators, school board or politicians, it gets ignored.
This isn't even limited to teachers. Physicians oppose the collection of this kind of data about outcomes for the sames reasons--it will very quickly get turned into a ranking system that gets used against doctors rather than for improving outcomes.
I don't think the author of this piece manages to back up his argument very well. The teachers' union says that the testing regime is flawed, and that the toxic environment is created by it; the author says that the testing regime is "flawed" but not "fatally" so, without substantitively adressing that concern. The "people in charge" of this environment are in turn governed by No Child Left Behind and state/federal regulation and laws -- there's a chain of authority here, and the brokenness of the system can rightly be argued to flow down from the highest policy levels all the way to local middle-management.
>> Would you want your child taught by someone who flunked the certification test five times, let alone 10?
This is a silly argument. How many people take multiple tries to get a driving license? The test either ceritifies you or it doesn't.
>> And would that instructor be more or less likely to resort to changing student test scores to hide his own incompetence?
Absolutely no logical connection, other than a wild assertion by an author clearly hostile to the teachers' union, and it seems to me, teachers in general.
Like it or not, public school teachers in the US are vastly underpaid and overworked (especially in urban school districts with low budgets), and have their ability to teach severely hampered by testing requirements that may or may not be beneficial to their students. The whole system is broken, but demonizing the people that are attempting to educate in these conditions without trying to understand the pressures at play isn't very constructive.
Further reading: http://www.thisamericanlife.org/radio-archives/episode/275/t...
Second, get off this silly mantra that they are underpaid. In some areas yes, but in Atlanta that is not the case. A simple query of various counties here produces results and for the number of days worked their pay is actually quite good; that and anecdotal being friend with a 10 year teacher who makes good money and who berates the system(which includes the union) for the problems. They cannot fix what they are not allowed too.
These 'accredited third party groups' would need to be a) accredited, b) be paid. And it's already happening here in the US[0]. What you have now is an overhead of bureaucracy that 'accredits' and outsources the testing to these private companies, who pay call center wages to mark high school exams based on shoddy metrics that are aimed at maximising profits. All the markers are temps/contract, and this accreditation system is essentially a cartel of companies that lobby to ensure their business remains. I cannot see the logic how this can end up benefiting the students.
Secondly, teachers being underpaid is not a mantra - it is a fact [1]. In the US, the worth society places on a job is strongly correlated to the pay. The accreditation and pay, resources and social respect that teachers (do not) receive I wager is directly correlated with the quality of the education students receive. The system is so broken as to rely on standardized testing as to further punish 'under-performing' schools by limiting availability to funding based on scores.
[0] http://www.theatlantic.com/features/archive/2014/07/why-poor...
[1] http://www.nea.org/home/2012-2013-average-starting-teacher-s...
I prefer actual charts from counties that pay, given 190 days of work http://www.dekalb.k12.ga.us/human-resources/teacher-salary-s... in a respectable county in Georgia this isn't what I would called underpaid. This is before benefits. Its not hard to find these charts for most counties.
The areas that need attention is all the money spent on administrative people other than teachers. The US has spent increasing amounts of money on education with near zero results.
How do those groups earn money? By charging for their testing? So groups that produce better results are selected for, which risks groups that have easier tests or more relaxed scoring being selected for by schools.
This has happened in some parts of England for GCSE exams. (These exams are taken by schoolchildren aged 16. )
http://en.wikipedia.org/wiki/The_Wall_Street_Journal#Politic....
As such, the editorial pages normally present the truth if and only if that truth conveniently intersects with their ideology. Not saying that liberal outlets don't have the same bias, more so that its opinion pages are not a good source for reasonable, effective analysis of, or solutions for, United States policy shortcomings.
If they wanted to stay with the testing metric framework, they would have been much better off with external testers. However, that still would have teachers "teaching the test" instead of teaching the subject.
http://www.nydailynews.com/new-york/report-shows-cheating-te...