This is what FCPS is saying about CogAT custom form for grade 2.

[Anonymous]

Unfortunately, the school work is not based on date of birth.
I remember a very immature late birthday child who got into the GT program when it was based purely on the tests. No question, he was smart. But, he was functioning below his grade level peers. However, he did not succeed in the program--and did poorly in high school.

You all talk about wanting your child to be with his/her ability peers--yet you want your child who is functioning below the others to get in. Can't have it both ways.

With the cheating on testing and so much subjective observation of teachers, this program is a crock. FCPS should go back to the drawing board. Eliminate it until some integrity can be put into the program.

[Anonymous]

Anonymous wrote:Doesn't anyone else think the "insufficient numbers" excuse is bullshit on Melba toast?

13,000 second graders. That's almost exactly 250 birthdays per week, or 35.6 per day.

I'm not a statistician, but it seems that plotting score against day of birth would yield a pretty clear curve.

Why don't you contact the program, you make a good point. Though, It's hardly bull**** on Melba toast.

[Anonymous]

Anonymous wrote:Unfortunately, the school work is not based on date of birth.
I remember a very immature late birthday child who got into the GT program when it was based purely on the tests. No question, he was smart. But, he was functioning below his grade level peers. However, he did not succeed in the program--and did poorly in high school.

You all talk about wanting your child to be with his/her ability peers--yet you want your child who is functioning below the others to get in. Can't have it both ways.

+1

With the cheating on testing and so much subjective observation of teachers, this program is a crock. FCPS should go back to the drawing board. Eliminate it until some integrity can be put into the program.

-1, Here not so much... Prepping may or may not be cheating, both sides make thoughtful arguments (though I side with the non preps). Regarding the "subjective observation of teachers" I really disagree with this cynical comment. GBRS are a necessary element of the reveiw. Of course there's an element subjectivity, but they are professionals who work with our DCs every day and have a good understanding of ability. However it's only one element. Finally, to call the program a crock is just bitterness. Yes it's not perfect and I believe the board is constantly reviewing and trying to improve all FCPS programs---read the other threads regarinding Board meetings and review of all levels of services.

[Anonymous]

Anonymous wrote:Doesn't anyone else think the "insufficient numbers" excuse is bullshit on Melba toast?

13,000 second graders. That's almost exactly 250 birthdays per week, or 35.6 per day.

I'm not a statistician, but it seems that plotting score against day of birth would yield a pretty clear curve.

That assumes equal distribution throughout the year. I will bet my son, who has a late September birthday and it one of the youngest in the grade, does not have 34 other students who share his birthday in the grade across Ffx Cty because this area has such a huge rate of academic red shirting.

[Anonymous]

Anonymous wrote:Unfortunately, the school work is not based on date of birth.
I remember a very immature late birthday child who got into the GT program when it was based purely on the tests. No question, he was smart. But, he was functioning below his grade level peers. However, he did not succeed in the program--and did poorly in high school.

You all talk about wanting your child to be with his/her ability peers--yet you want your child who is functioning below the others to get in. Can't have it both ways.

With the cheating on testing and so much subjective observation of teachers, this program is a crock. FCPS should go back to the drawing board. Eliminate it until some integrity can be put into the program.

I'm calling foul.

Please. How many older kids functioned below their peers?

The question should be whether or not the kids can keep up until the magnitude and significance of age-driven differences begin to level off (usually, for the classroom, that starts around the 4th grade). But those differences are more pronounced in test results than they are in the classroom. Test conditions allow for greater differentiation at the top end. In the classroom, there will always be a ceiling that acts as a limit to the highest performers.

Think Doogie Howser. He'd ace the test, but no public classroom experience would come close to challenging his ceiling. So the younger bright 2nd graders that may have just missed the cut-off may well be better situated to keeping up with their peers than older second graders that just barely edged into the pool.

All in all though, I'm not so exercised over the issue. I've got a August-born 2nd grader that just missed the pool on both tests. But the teachers know what my kid can do and I'm confident the GBRS will make up for the shortfall.

[Anonymous]

I wouldn't think that you would need exact same birthdays to norm against. Using even scores of kids born in the same quarter would be better than no age adjustment at all. I thought that NNAT was normed by quarter but could be wrong.

[Anonymous]

Anonymous wrote:

Anonymous wrote:Doesn't anyone else think the "insufficient numbers" excuse is bullshit on Melba toast?

13,000 second graders. That's almost exactly 250 birthdays per week, or 35.6 per day.

I'm not a statistician, but it seems that plotting score against day of birth would yield a pretty clear curve.

That assumes equal distribution throughout the year. I will bet my son, who has a late September birthday and it one of the youngest in the grade, does not have 34 other students who share his birthday in the grade across Ffx Cty because this area has such a huge rate of academic red shirting.

I'm certain you're right. But even distribution isn't required to normalize. You draw the curve and extrapolate across relative "dead zones." With 34,000 data points, it's inconceivable to me that there wouldn't be enough data to construct a statistically sound normalization. I'm happy to be corrected, but I don't think I will be.

[Anonymous]

Anonymous wrote:Unfortunately, the school work is not based on date of birth.
I remember a very immature late birthday child who got into the GT program when it was based purely on the tests. No question, he was smart. But, he was functioning below his grade level peers. However, he did not succeed in the program--and did poorly in high school.

You all talk about wanting your child to be with his/her ability peers--yet you want your child who is functioning below the others to get in. Can't have it both ways.

With the cheating on testing and so much subjective observation of teachers, this program is a crock. FCPS should go back to the drawing board. Eliminate it until some integrity can be put into the program.

I would agree with the above if the specific tests we are discussing were testing expected knowledge and class-related subjects. But they are not. Nobody said that younger kids cannot keep up with peers. Just the fact that they are even being considered for AAP by their parents (some of them like mine are already in the pool based on their test scores) means that they are doing great at school. However, this doesn't change the fact that CogAT questions are of the kind that is affected by age (as all such tests are, worldwide) and the question is whether age-normalization can (and should) be done or not. And we are saying that the county's argument that because of insufficient data they cannot be age-adjusted this year is not valid.

FCPS also stated that the scores will be age-adjusted again in the future when enough new data are collected. This by itself is the answer to the age-normalization issue that people have been debating for weeks now. If they are not supposed to be adjusted, why are they going to be adjusted again in the future? And if they are supposed to be adjusted, how can they be used this year without any adjustment?

And by the way, I was the youngest kid in my class and I did better than anybody else. Not that it proves anything but I thought I should mention it.

[Anonymous]

For the last poster, that's great! But I do think there is a "growth curve" so to speak for cognitive and perceptual ability for young children. It probably does plateau out around the 4th grade. But it could be an important (not the only factor) factor in assessing cognitive and perceptual ability of children younger than 4th grade.

[Anonymous]

"The question should be whether or not the kids can keep up until the magnitude and significance of age-driven differences begin to level off (usually, for the classroom, that starts around the 4th grade). But those differences are more pronounced in test results than they are in the classroom. Test conditions allow for greater differentiation at the top end. In the classroom, there will always be a ceiling that acts as a limit to the highest performers.

Think Doogie Howser. He'd ace the test, but no public classroom experience would come close to challenging his ceiling. So the younger bright 2nd graders that may have just missed the cut-off may well be better situated to keeping up with their peers than older second graders that just barely edged into the pool. "

Actually, some of the biggest problems with the age differences occur in high school.

[Anonymous]

Anonymous wrote:

Anonymous wrote:Doesn't anyone else think the "insufficient numbers" excuse is bullshit on Melba toast?

13,000 second graders. That's almost exactly 250 birthdays per week, or 35.6 per day.

I'm not a statistician, but it seems that plotting score against day of birth would yield a pretty clear curve.

Why don't you contact the program, you make a good point. Though, It's hardly bull**** on Melba toast.

What makes it seem to be "BS on Melba Toast" is the statement that there is not enough data to age norm the sample, but then stating that in future years they hope to age normalize the results. Where is the additional data going to come from in future years? Will they combine results from this year to next to garner more sample points? That seems unlikely if they are going to use a new test. There certainly will not be a larger sample size. WTF? Smells very fishy.