The new Charlotte-Mecklenburg school progress reports have helped me to better understand North Carolina's formula for academic growth -- and I'm a bit taken aback.
The state's "expected growth" calculation is key to state ratings and other measures CMS uses to gauge how well its schools and officials are performing. It has been explained to me as roughly translating to an average of a year's academic gain per student, based on their performance on state exams (read the state's description on pages 4-5 here).
So I was looking at West Charlotte High's progress report and scratching my head. If, as the CMS document reports, only 64 percent of last year's students demonstrated at least one year's progress, why did the school get a "high growth" rating?
The answer lies in CMS's explanation, the clearest one I've seen yet: "Each student is expected to perform as well (or better) on the End-of-Course assessment as he or she did, on average, during the previous two years. Average Growth for a school is calculated by comparing actual performance to expected performance and then averaging the difference across all students and all subjects. ... Across the state, about half of students typically meet or exceed this growth expectation. To earn the high growth designation, a school must meet the average-growth standard described above and also have more than 60% of its students make expected growth."
So a student who has performed poorly in the past will go in with a relatively low projected performance. And up to 40 percent of students at any given school can fall even further behind, making less than the gains expected based on their past performance, while the school still gets a "high growth" rating. That helps explain why CMS and the state have so many "high growth" schools (just over 81 percent statewide met or exceeded the growth standard last year) while proficiency levels remain persistently low at some of those same schools.
It's also worth noting that Ardrey Kell and Providence high schools, two of CMS' top performers, had 65 percent of students making the expected gains last year, a number barely above West Charlotte's. In cases like that, a significant number of strong students who have little trouble clearing the "grade level" bar on exams are still falling short on the progress they're expected to make. Districtwide, about 40 percent of students logged less than a year's gains.
Kudos to CMS for including a measure that gives a clearer picture of what lies behind the averages -- and how many students are continuing to fall behind. The real challenge, of course, lies in figuring out what to do about it.
Wednesday, January 18, 2012
N.C. academic growth: Many left behind
Subscribe to:
Post Comments (Atom)
Posts
36 comments:
...and just think of how much that $55 million from Project LIFT will do for those high achieving students at West Charlotte!!
Ann, you've already predicted great success and they have yet to spend a dime.
You already know what the data will say.
Perhaps we should spend that money at Ardrey Kell?
"Pay no attention to that man behind the curtain!"......
"The real challenge, of course, lies in figuring out what to do about it."
Simple. Throw more money at it, as usual.
CMS, want to improve high school achievement? Change the high school start time to 8:15am. Sleep will do wonders for growing teens' behavior, academics and health.
I would agree with the sleep comment but many of these kids would just stay up later. There is no one doing any parenting at home teaching good habits so that they can be prepared to learn at school.
The CMS schools are pushing these kids way too far, way too fast, to except them to succeed. In addition, they have teachers just out of college teaching middle school classes, and its obvious that they are full of themselves and have no interpersonal skills with which to deal with what these teenagers have going on. These teachers are young enough that they should remember the difficulties of that age....but no....they act as if it never happened to them!
ooops...."to expect them to succeed".....sorry.
Growth is like interest in your banking account.
If you start with nothing, even spectacular growth won't give you much in the end.
You have to bring something to the bank, so to speak, to get real returns.
Same is true for schools.
Quick comment on the above sleep comment, studies (lots of them!) suggest that teenagers will go to sleep at the same time everynight and get an additional hour of sleep. I know that's true in my house, my teenagers drop at 10-10:30pm in exhaustion (but we are also involved in sport activities).
Just my thoughts, Yes on the later high school start time. I wish my kids were able to get the required 9 hours of sleep per night.
My kids go to AK so I am concerned with Ann's growth, or lack of growth, at that school. Lots of new teachers there too. My son's math teacher didn't know what she was doing and is just starting to get in the groove (and his course is over now). Do "mature" teachers mentor the new ones?
If the schools were like banks, CMS would give higher interest on the lower deposits and lower interest on the higher deposits.
If a bank did this, they would lose all their wealthier customers and would only be left with those who can only make small deposits.
Of course, a bank would eventually collapse with such policy.
That's why banks usually reward those with the most to deposit with better rates and services.
Schools just can't seem to get that simple concept right.
Baix...
How dare you try to use a business analogy to make a point about public education!..
You know schools can't be run "like" a business!
*laughs....
Even though they loooove to push that ROI thingy.
These grades will continue to fall as long as CMS continues to put mediocre teachers in schools. They did it this year so they could pay them less and work them longer. The general make up of our teachers has fallen in step with the grades. You get what you pay for plus a Broad Award ! Title one champ baby !!
What will be interesting to see is how sanitized the Observer will be during the DNC.
Afterall, President O will do the usual shout-out to Mayor Fixxit, other locals and proudly proclaim how great CMS is and how it is a shining beacon to the rest of the world because we "won" The Broad Prize.
My prediction is the Observer will do all it can to bury or refrain from having any negative articles during that time.
...we shall see.
The reverse is true as well. If a student is a high performer and scored in the 99% and then the next year they score in the 98%, they are considered not meeting growth. This actually happened to my son and his teacher had to come up with a plan for him to "improve". These reports lack a common sense element.
CMS so obviously needs some serious instruction in basic math and science.
They apparently have no concept of "noise" in their measurements nor of diminishing returns as you reach an asymptote of a curve.
To think that a kid who was formerly 99% is backsliding because they're now in the 98% is just plain moronic.
This could just be an "error" well within the range of their "confidence" in their testing.
Of course, they probably assume their test results are the perfect measurement. At least this sounds like they do.
Seriously moronic.
And these buffoons have the gall to claim to "teach" our youth.
Anyone throw in the fact that teacher salaries have been FROZEN for 5 to 7 years in many districts. No cost of living raises... gas, rent, food prices, everything goes through the roof and not even a 2-3% cost of living raise.
The reality is that schools are pumped full of new grads because the experienced, mentor teachers are getting out of the profession, sometimes forced into an early retirement because the school system doesn't have the money to pay them any more. Cutting costs everywhere is done by filtering in the cheap, first year teachers that make $20k-$30k less per year than the vets.
You can bash the teachers, but what would you do if Bank of America froze your salary for the next 5-7 years? Would you feel under-appreciated? Would you keep buying the company line of just stick it out for the team?
We're throwing money at the wrong places...and yes, this IS from a school teacher.
is the cms website (www.cms.k12.nc.us) refusing to open for anyone else out there?
ANon 9:53...
It's working for me.
I doubt CMS would be protesting against SOPA like many are doing today.
Figures dont lie but liars can figure.
I think it's about time to let those students who are going to fail, fail. All the money in the world thrown at them isn't going to do any good until either the students or their parents start to care about the student's academic life.
Spend the money and the effort paying teachers correctly, developing a proper curriculum, and buying needed supplies for every student, but quit going out of the way to develop all these programs for low-performers, only to continue for the same student to be low-performing. Let's go back to the old days of school--if you show an interest, you can get help. If you don't care, flunk out and see how your life turns out. Don't undercut students who are interested in learning.
This describes how the growth is determined, doesn't it? Good luck figuring it out.
http://www.ncpublicschools.org/docs/accountability/reporting/abc/2010-11/academicchange.pdf
Excuse me, ma'am, but I speak geek...
Regarding
http://www.ncpublicschools.org/docs/accountability/reporting/abc/2010-11/academicchange.pdf
Just a quick glance.
Well, it looks like they've described their calculation pretty much step by step.
One thing worth noting is their fudge factors of .92 and .82 which are there for the confidence they have in their testing.
This means that they have a built in mechanism which excuses some noise and backsliding (a.k.a, regression to the mean) that may be normally expected in tests like these.
Now, given that this is in their calculations, a kid who performed in the 99%, followed by a year at the 98% should NOT have failed those expectations.
Of course, you'd need to plug in actual scores to verify, but their approx 10-20% range should be enough to cover it.
Fellow geeks can verify this.
As always, I am open to correction...
Oh, and FWIW, you need to know those raw scores instead of percentiles because there may not be a linear relation between the two.
IOW, since they measure raw scores, when you convert to percentiles, a 1% difference between the 48% and 49% may not be the same as a 1% difference between the 98% and 99%.
In fact, I'd be willing to bet money that this is the case.
It's one reason, for example, the top 1% in a test like the SAT may span a hundred points, but those closer to the middle may span much less.
It's that asymptotic thing that most people forget about when you're measuring close to the ends of the scales.
I'd say this is a flaw in their calculations for change.
But, hey, what do I know...
Anyone else out there feel free to correct me, but I think their Expected Change calculations might really suck for high performers.
You'd really need to see the distribution curve for those EOG scores and how large the score gap is for the 1% at the extremes (both high and low) and those closer to the mean.
Personally, the more I think about it, I'd give their statisticians a D- for that equation just based on my suspicion about those EOG score distributions.
But, who knows maybe the distribution curve is flat.
Nyuck, nyuck, nyuck*
*Official Bwahahaha in Geek.
If they are getting left behind then I say leave them behind. Enough is enough. For five decades we’ve thrown trillions, TRILLIONS!, at the same demographic and the results are always the same. They only want what is handed to them, they only want a job where all the work is picked out and still get paid on Friday, they want it ALL for nothing.
GREAT teachers is the solution! Sadly, so many tenured teachers aren't great and getting rid of them is a BIG challenge.
Getting a great public education is difficult to come by no matter where you live in this country. Watch a few documentaries "Waiting for Superman" or "The Lottery" it will open your eyes as to how the United States is failing it's youth when it comes to education.
Where the CMS statisticians failed Stat 101...
http://www.socialresearchmethods.net/kb/regrmean.php
"The more extreme the sample group, the greater the regression to the mean."
"But if your sample is very extreme relative to the population (e.g., the lowest or highest x%), their mean is further from the population's and has more room to regress."
Let me change their grade to an F...
Fail.
For using those fixed values .82 and .92 to compensate for regression.
The sad thing is that the smart kids who get abused by this are probably smart enough to know it's BS as well.
Wiley, I'd be delighted to see and report great success from Project LIFT, but I surely don't remember predicting it. I generally wait to report what actually happens on various efforts, which isn't always success. Did I have a crystal-ball episode I've forgotten about?
Looking at the District Report, the graduation rate for CMS in 2009/2010 was 77.6%. The rate in 2010/2011 went DOWN to 73.6% or 4 percentage points.
CMS is still projecting a 90% graduation rate by 2014.
That means CMS has to meet a 16.4 percentage point increase (22.3%) for the next three graduating classes to get to 90%.
What magic will CMS pull out of their hat going forward, since last year they regressed?
Ann,
My comment was sarcasm related to the third paragraph in your story:
So I was looking at West Charlotte High's progress report and scratching my head. If, as the CMS document reports, only 64 percent of last year's students demonstrated at least one year's progress, why did the school get a "high growth" rating?
Your "prediction" is reporting "high growth" even though the way it is calculated is suspect.
Project LIFT will be calculated in some related manner to show "it worked" to justify spending $55 million dollars.
You're just that good Ann!
On the grad rate, more kerfuffling to me is the number on the progress report showing 98.2 percent of all CMS high school students are "anticipated to graduate on time." The data folks are re-running that one, because it's just not consistent with the fact that only 74 percent of them actually do. If it's not an actual error, it's a meaningless number, best I can tell.
CMS is great a throwing meaningless numbers around.
That is why the PfP or Teacher Effectiveness Project or whatever the hell they want to call it next week WONT WORK. The stats never match up ! How do you compare AK and West Charlotte students? How do you compare socioeconomic data with half class sizes and twice the spending? The public nor teachers TRUST the DATA !!! Please hire more "communications" employees to sell this worhless crap.
Here is an actual solution. TRADE SCHOOLS. Not ALL students have the desire or intellectual aptitude to go to college. Many would or do have the ambition to start a paycheck as soon as possible. Give the student the option of general studies for 2 hours of the day and trade and skills training the rest. This could also be combined with apprentice and job shadow time during and after school. Trillions wasted without measurable results is RIGHT. Start now and train our students for a better tomorrow.
I agree about the PfP plan to the extent which it depends on those bogus numbers they use to measure student progress.
Bad stuff. I just started reading an explanation of that (with their "justification" for the formulas they used) and it reeked.
Unfortunately, I'm not paid to analyze their analysis, so it's not an efficient use of my time to do so when I realize that it will most likely change nothing.
But, if I were a teacher whose pay depended on this crap, I'd get a group together and pay someone to write a solid critique, and then use it to embarrass educrats at every public forum possible.
I still think one of the biggest holes they have is in their analysis of individual student performance.
Their comparisons between entire schools isn't as bad and seems to fit with normal statistics.
I've noticed that the same techniques are used in other school systems in California for instance with a better explanation of how they compare schools, but none fot how they justify their calculations of change for individuals.
I just want to be everyone who reads these blogs understands where most of the posters are coming from in this debate. While a few have been forthright in their pessimism of the operation of CMS and of public schools in general, most of us truly believe in public education. For example, a number of posters have been critical of FRL. They do not mean they do not belive in the purpose and use of the program. They know and totally support the notion that there are poor families out there and they need this program. They criticze the program because of its lack of accountability and apparent unwillingness by too many to deal with the fraud. Additionally, its fraudulent numbers have expanded into too many reaches of spending education dollars.
We (well most of us) believe an informed and knowledgable citizenry is vital to our democracy and way of life. But as we have seen, our society has digressed into the battle of who can confiscate hard earned money of other people and exert thier power to the point the producers have lost their voice at the ballot box.
Humans are born with "free will". You can spout all these "Be kool, stay in school" cliches you want but some people have to learn by the "school of hard knocks". That is why CPCC is there for when someone decides to get their act together and get with the program. There is a way for them to do so.
So I do not believe there is much "left behind" but rather they choose and the family unit chooses to not "get with the program".
Post a Comment