kitchen table math, the sequel: I'm confused

## Thursday, February 23, 2012

### I'm confused

What Do the NAEP Tests Really Measure?by David Klein | January 2011

Or is there some aspect of this item I don't understand?

Maybe there's a right triangle hiding in there somewhere. Speaking as a veteran of SAT prep.

TerriW said...

Maybe it (the question) exists as a sanity check to see how many people mark that they used their calculator on that question. Maybe?

TerriW said...

Oh my, I just realized this is a grade *4* question.

God help us all.

Anonymous said...

It might be grade four with a calculator.

I'm thinking sixth grade without :-)

-Mark Roulo

Anonymous said...

I think it assumes a lot that just because Jan left her bike between Michael's hous and Hilda's house the green bicycle is hers. Hasn't anyone else ever had their bike stolen? Or even just borrowed by an annoying sibling?

SteveH said...

The key to answering this is to see "grade 4" and "NAEP". My son would get problems wrong on tests because he would over-think them. This is NOT the SAT. Even now, he seems to do better on tests where the median is low rather than high.

I like to tell people to go online and look at some of the sample 4th grade NAEP math questions and the raw percent correct scores. Then they can wonder what schools do in math for an hour each day. One education writer told me once that the NAEP is the "gold standard".

This is something we haven't talked much about on KTM; the expectation divide between different parents and different teachers. If someone calls the NAEP a gold standard, I know that I can't have a reasonable discussion of education with that person. When I talk with other parents, there seems to be clear dividing lines of thought and expectations.

Anonymous said...

"I think it assumes a lot that just because Jan left her bike between Michael's hous and Hilda's house the green bicycle is hers."

Yes, it makes some assumptions.

I can add that it also it assumes that the picture is of the state *after* the bicycle was left rather than before.

And that *this* diagram is related to the question.

And that the diagram is correct.

And that there are no *other* bicycles not covered by the diagram.

And that the bicycles have not been painted in the meantime (or, as you point out, stolen).

And that the author of the question is telling the truth.

I'm sure I could go on.

But most 4th graders who can read the question and comprehend it will get the correct answer, no? Since the point seems to be to nail down if the student knows what the word "between" means.

-Mark Roulo

Anonymous said...

Considering that the questions seem to be so easy, it seems to me that we should be more, not less, worried by poor results. A lot of preschoolers could answer the bicycle question. But I do agree that the test should actually based on the kinds of math that is actually taught in schools.

Obi-Wandreas, The Funky Viking said...

IQ Test in the year 2505, from the movie "Idiocracy"

"If you have one bucket that can hold 2 gallons, and one bucket that can hold 5 gallons, how many buckets do you have?"

Jen said...

This question especially in 4th grade immediately tells you which kids can not only read the words, but can actually understand them. And yes, the word that they have to understand is "between."

I have seen the word "between" on SAT geometry problems, therefore this is clearly a perfect problem. (kidding, I'm kidding)

I know that at least 2 of 3 of my kids would absolutely want to say they used their calculator. At least one of the three would have done it.

TerriW said...

RE: test scores and grade levels and whatnot, that reminds me of our experience with doing standardized tests as homeschoolers.

My kids are still pretty young (8 & 5), so we do the oral Peabody test at our kitchen table (takes about an hour) with the tester.

The scores confused me at first -- we'd get a percentile for their current grade level, but also a grade level equivalency score, like, say, last year's "4 years, 9 months" for math.

It took me awhile to wrap my head around what that *really* meant. First there's the flush of pride of "Wow, my 7 year old scored like a 4th grader in the 9th month of 4th grade!" ... but then reality sets in.

Because when I originally envisioned what that means, I imagined *my* child (product of an engineer and a former unix admin, white, middle class, nice suburb, house with 10,000 books, math is huge priority), what she is going to be able to know and do during the 4th year, 9th month of school.

But that ain't it, is it? It's what the 50th percentile student can do during that time frame. And that number includes all students tested, from all over the country, all SES backgrounds, good and crappy schools alike.

I have some friends who are radical unschoolers who were happy that their child scored in the 50th percentile for their grade level, and I was thinking to myself, "Wow, if my child scored that level, I would have completely failed -- she'd be doing far worse than she would in regular school."

TerriW said...

I think what I really want from my standardized tests is not that they be normed against the whole of US students ... but I want them normed against, say, the other first world countries, too.

Meaning, okay, say, my kid tests 99th percentile against US students ... do they test 75th percentile in Singapore, etc, etc?

Hmmm.

AmyP said...

I've seen test scores for my daughter's private school where they show how she scores relative to the nation, relative to suburban schools and relative to private schools. That's helpful, I think.

Anonymous said...

Actually, there's a far better example considering that getting this question wrong demonstrates an achievement level below the basic cutoff score and is meant to be one of the simplest questions. Getting it wrong demonstrates an achievement level at the kindergarten or first grade level using the rule of thumb that an increase of 11 points represents a gain of one grade level. The example I am talking about is the one at eight grade where students had to determine which shape was used to construct a square if four of them were used. While a problem similar to this one but in three dimensions was proposed in Hilbert's list of 23 unsolved problems and the general case is interesting in its own right, this version is so simple that a second grader should be able to do this one. The PISA actually has a similar issue as pointed out by the Fordham Institute and Finnish mathematicians, who have actually called their victory pyrrhic because of the decreasing pass rates on matriculation despite simplification along with serious weaknesses in basic skills such as the use of fractions. The PISA has several problems listed and it turns out that a good portion of the problems are at the 5-7 grade levels. Some are actually at kindergarten to first. The Dice problem is one of these. The url is here if you want to see them: http://www.oecd.org/dataoecd/14/10/38709418.pdf.

Catherine Johnson said...

Oh my, I just realized this is a grade *4* question.

God help us all.

Well, I'm glad to hear that!

I was seriously worried that there was something important in this question that I was completely blind to ----

Catherine Johnson said...

After all this time, I had NO idea this was the kind of thing appearing on NAEP.

Catherine Johnson said...

Terri-

I'm laughing.

Chris and I, a few years back, were reveling in the fact that he had finally managed to get .... maybe some kind of high B in his middle school math class....

We were walking along with the dogs, celebrating the great news, when one of us brought up the kids in Singapore. We decided that, compared to the kids in Singapore, Chris had just managed to rise to the level of Singaporean students with developmental disabilities.

TerriW said...

As my dad was fond of saying: "In the Valley of the Blind, the one-eyed man is king."

Scoring 99th percentile math in America feels like that.

Catherine Johnson said...

oh lord, don't say that!

Catherine Johnson said...

A student in my class the other day said she scored a 400 on the SAT --- on all 3 scales, I think.

That's not possible, is it?

TerriW said...

I know I shouldn't say that, but ... you know, every standardized test I took in school, I always scored 99th percentile. Now, I am no rocket scientist and I never have been, and even as a kid I had some cognitive dissonance there because I knew the kids who were _really_ smart, and I was not one of them.

So, to this day, I still kind of roll my eyes a little when I see that 99th percentile because ... I guess I expected more from what 99th percentile should mean than *me*, you know?

TerriW said...

I guess what I'm trying to say was that I'm not too impressed by a test that doesn't distinguish between me and the really smart kids. I suppose that's why we have the SAT and all that folderol we talked about all those months.

ChemProf said...

All these tests always remind me of the old CTBS. We had to take it every couple of years when I was in school, and what I remember was that after some grade, they never changed the test. It wasn't a bad exam, but I remember in senior year, we all got to a passage we remembered (about a camel) and the whole class cracked up since it was the fourth time we'd read it.