Huge NNAT scoring variability

[Anonymous]

Child was administered the test in kindergarten last year and scored in 98th percentile.

Same child was recently administered the test again this year and scored in 52nd percentile.

Just trying to process and understand this. Tests were administered by school (Alexandria City Public).

We haven't done any independent testing, but have considered it. Child reads about two grades above reading level, shows pretty advanced artistic capabilities, and is generally very intellectually curious.

Thoughts or suggestions? Trying to reconcile the huge discrepancy in scores on the same test.

[Anonymous]

Our DD had that too. Just a function of teacher not explaining the test the second time. Teacher was fired by the next year.

[Anonymous]

Okay, interesting. I wonder if I should talk to anyone at the school about it. To be honest, I'm not even sure who exactly administered the test. I could see some variation in score but this is obviously quite significant. Just wondered what some of the possible explanations were and whether we need to do anything to address it.

[Anonymous]

Anonymous wrote:Okay, interesting. I wonder if I should talk to anyone at the school about it. To be honest, I'm not even sure who exactly administered the test. I could see some variation in score but this is obviously quite significant. Just wondered what some of the possible explanations were and whether we need to do anything to address it.

It's a 30 minute timed test. I'm guessing he skipped a page or two accidentally. Or he was sick on the day it was administered.

[Anonymous]

I've heard it's not the most reliable test. Lots of false highs and false lows, but reliable in the middle.

[Anonymous]

This happened to my child on the cogat, only his scores went up, not down. I think there is the reason that test publishers don't do studies on score reliability over time - they aren't super reliable. I'm not sure if it is a problem with the test or the fact that young children are taking them.

[Anonymous]

This is like saying the doctor measured your kid's height at 4 years old and he was at the 98th percentile.

Then the same doctor measured your kid's height at 5 years old and he is now at the 52nd percentile.

There could be many reasons for it.

[Anonymous]

It's a short test (<50 questions) and the swing from median (50th percentile) to 2 standard deviations (98th percentile) is only 5 questions (last years numbers, raw number of questions could vary somewhat based on population of participants) so scores can fluctuate wildly based on a few more or a few less missed questions. Could be anything, poor instructions, skipped questions, a butterfly flying around.

[Anonymous]

Anonymous wrote:It's a short test (<50 questions) and the swing from median (50th percentile) to 2 standard deviations (98th percentile) is only 5 questions (last years numbers, raw number of questions could vary somewhat based on population of participants) so scores can fluctuate wildly based on a few more or a few less missed questions. Could be anything, poor instructions, skipped questions, a butterfly flying around.

This is exactly why using these tests to track kids for 6 years (3rd-8th grade) in FCPS is ridiculous.

[Anonymous]

Why is ACPS administering this test in kindergarten? What's the purpose?

[Anonymous]

Anonymous wrote:

Anonymous wrote:It's a short test (<50 questions) and the swing from median (50th percentile) to 2 standard deviations (98th percentile) is only 5 questions (last years numbers, raw number of questions could vary somewhat based on population of participants) so scores can fluctuate wildly based on a few more or a few less missed questions. Could be anything, poor instructions, skipped questions, a butterfly flying around.

This is exactly why using these tests to track kids for 6 years (3rd-8th grade) in FCPS is ridiculous.

They aren't. FCPS uses them as one piece of a referral packet. The teacher observations weigh heavily as do the work samples. I'm an administrator in FCPS, and we spend a lot of time discussing which kids demonstrate that they would be successful in a Level IV classroom. It is not just about the number. I know plenty of kids who came up with the "right" score but didn't have the classroom performance to match it. And conversely, I've seen plenty of kids who did not have the cut score, but demonstrated curiosity, strong work habits and/or creativity that supported a move to a Level IV classroom.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:It's a short test (<50 questions) and the swing from median (50th percentile) to 2 standard deviations (98th percentile) is only 5 questions (last years numbers, raw number of questions could vary somewhat based on population of participants) so scores can fluctuate wildly based on a few more or a few less missed questions. Could be anything, poor instructions, skipped questions, a butterfly flying around.

This is exactly why using these tests to track kids for 6 years (3rd-8th grade) in FCPS is ridiculous.

They aren't. FCPS uses them as one piece of a referral packet. The teacher observations weigh heavily as do the work samples. I'm an administrator in FCPS, and we spend a lot of time discussing which kids demonstrate that they would be successful in a Level IV classroom. It is not just about the number. I know plenty of kids who came up with the "right" score but didn't have the classroom performance to match it. And conversely, I've seen plenty of kids who did not have the cut score, but demonstrated curiosity, strong work habits and/or creativity that supported a move to a Level IV classroom.

Thank you so much for reiterating that!
My DC didn't have the best test scores but is now in 4th grade AAP and is doing incredibly well. It's not all about that number.

[Anonymous]

Anonymous wrote:Why is ACPS administering this test in kindergarten? What's the purpose?

They aren't. They are doing them in 1st. My guess is OPs kid did the test elsewhere in K

Since ACPS teachers and admins have little experience administering the tests, I too would question the results. This is only the 2nd year they are testing kids.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:It's a short test (<50 questions) and the swing from median (50th percentile) to 2 standard deviations (98th percentile) is only 5 questions (last years numbers, raw number of questions could vary somewhat based on population of participants) so scores can fluctuate wildly based on a few more or a few less missed questions. Could be anything, poor instructions, skipped questions, a butterfly flying around.

This is exactly why using these tests to track kids for 6 years (3rd-8th grade) in FCPS is ridiculous.

They aren't. FCPS uses them as one piece of a referral packet. The teacher observations weigh heavily as do the work samples. I'm an administrator in FCPS, and we spend a lot of time discussing which kids demonstrate that they would be successful in a Level IV classroom. It is not just about the number. I know plenty of kids who came up with the "right" score but didn't have the classroom performance to match it. And conversely, I've seen plenty of kids who did not have the cut score, but demonstrated curiosity, strong work habits and/or creativity that supported a move to a Level IV classroom.

How many proactive teacher referrals of these low score high performing kids have you done? If the scores don't matter, you should get rid of them and save taxpayers a lot of money. The vast majority of AAP admittance is based on test scores, no matter how much thinking administrator do about other factors.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:It's a short test (<50 questions) and the swing from median (50th percentile) to 2 standard deviations (98th percentile) is only 5 questions (last years numbers, raw number of questions could vary somewhat based on population of participants) so scores can fluctuate wildly based on a few more or a few less missed questions. Could be anything, poor instructions, skipped questions, a butterfly flying around.

This is exactly why using these tests to track kids for 6 years (3rd-8th grade) in FCPS is ridiculous.

They aren't. FCPS uses them as one piece of a referral packet. The teacher observations weigh heavily as do the work samples. I'm an administrator in FCPS, and we spend a lot of time discussing which kids demonstrate that they would be successful in a Level IV classroom. It is not just about the number. I know plenty of kids who came up with the "right" score but didn't have the classroom performance to match it. And conversely, I've seen plenty of kids who did not have the cut score, but demonstrated curiosity, strong work habits and/or creativity that supported a move to a Level IV classroom.

How many proactive teacher referrals of these low score high performing kids have you done? If the scores don't matter, you should get rid of them and save taxpayers a lot of money. The vast majority of AAP admittance is based on test scores, no matter how much thinking administrator do about other factors.

Perhaps you need some reading comprehension lessons. The first poster claimed that the scores were the sole basis for tracking students. The second poster responded by saying that they are one piece and not the sole determinant. They do matter, and I don't think the administrator was claiming they didn't.