Reply to Iready

Forum Index » Fairfax County Public Schools (FCPS)

Reply to "Iready"

Subject:

Emoticons





More smilies

Text Color: Font: Close Marks

[quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous]I'm trying to see if my child might be placed in a better class.  He scored at 93% in I Ready math and 91% in IReady reading on the last report.  According to his teacher some students score 30 to 40 POINTS above the 99th percentile. Since these scores are way above the 99th percentile. am I being told the truth? I guess if there dozens of little geniuses at the school, he's just looked at as average. [/quote]
Across the entire county, there are maybe 2 or 3 kids per grade level scoring 100 points over the 99th percentile line.  There could easily be another 50-100 kids scoring 30-40 points over the 99th percentile line.  Keep in mind that kids who are taking math classes at AoPS or RSM could be far above the grade level being assessed in the iready.  [/quote]

Oh boy. Math is hard.  For example, there are 3.9 million fifth graders in the country according to the NCES.  Assuming Iready ain't lying and the 99th percentile is...well...the 99th percentile, then according to math, there are 39,000 fifth graders scoring at the fifth grade 99th percentile. We know that isn't the case cuz not every fifth grader takes an Iready assessment. But whatevs.

Of those 39,000 99th percentile kids, how many score ten points over the 99th percentile? Well, using the Iready percentile chart, scoring a 524 will get you the 99th percentile. Scoring a 520 gets one into the 98th percentile, or 78,000 kids score ABOVE this. The difference of 4 points knocked out 39,000 kids from getting into that 99th percentile. I guess it could be that America's little geniuses ALL score well above the 99th percentile; that is each kid scoring above the 99th percentile didn't just eke into that category.  But that kind of flies into the face of common sense.  So if scoring 520 to 524 loses 39,000 kids, one must assume that the rate of degradation will be the same and would be linear.  But let's pretend it isn't.  Let's pretend that for every 5 point score above 524 that we lop off half of that amount.  My trusty abacus says that there will be a loss of  19,000 kids between a 524 score to a 529.  Again, I am being generous.  If that is the case, scoring 5 points above the 99th percentile cut line leaves us will approximately 19,000 kids in America proudly still standing.  Let's do it again. Let's assume that an additional 5 point increase will cut another half of kids.  That'll leave 9,500 kids left.  Do it one more time and be only 15 points over the 90th percentile and you're left with...drum roll...4,750 kids in America scoring 15 points above the 99th percentile.  Yee haw. 'Merica. Keep at  it, you say.  5 more points and you're at 2,350. 5 more, 1,175. And finally 5 more for a total of 30 points above the 99th percentile and you're at 587 precious little snowflakes scoring 30 points above the 99th percentile.

Of those 587 kids, each region of the country will have their fair share.  Those tiger Moms out in the Bay area and Irvine say hi.  Let's give the West 20% of them.   Don't forget about those brilliant New Englanders.  Son of Chad and Buffy of Cos Cob checking in. The southeast and Floriduh gotta be good for 10%.  Fly over country around Iowa and Chicago maybe another 10%.  Texas 5%.  New York New York, yee of the magnet schools, with New Turnpike and Pennsylvania...You get the idea.  This is a big country outside our fair area.  Well, you get the idea.

Virginia is home to 8.6 million people.  Fairfax has 1.4 million peeps. I'm going to WAG that there are 10,000 students in Fairfax County, representing about .2% of America[s fifth grade student enrollment. We're told by the poster above that there are 100 kids scoring 30 points above the 99th percentile.  Out of the 587 kids in the entire country. Yeah, no.
[/quote]

Your understanding of how norming a nationally representative sample is very flawed.  You are bringing an arithmetic mindset to a probabilistic math problem.  They don't have data on all kids and the data they are missing is likely--like most norming systems-- systematically distorted to under-represent the extremes.[/quote]

Options

	Disable HTML in this message
	Disable BB Code in this message
	Disable smilies in this message

Review message

Search Recent Topics Hottest Topics