Tuesday, September 15, 2009

Praxis II Trivia

Interestingly, according to this PDF from ETS, of states that accept or want the Praxis II test that I took ("Mathematics Content"), Colorado's required score of 156 is the highest. (A score of 165 apparently qualifies you for a "recognition of excellence" from ETS.) The median required score to pass, among states that use this test, is 136.

In 2004-2005 at least, apparently the median score on the exam was 143, and what Praxis calls the "average range" was 127-156. I'm not sure what an "average range" is. (The possible scores on this test go from 100 to 200.)

Anyway, here is a map showing the required passing scores for the various states. I left out Guam, but for the record you need a 124 to pass this in Guam.

Click to embiggen.

I suspect the reason that Colorado wants such a "high" Praxis II score is that they'd prefer you to take the Colorado-specific PLACE test instead, which is more specific (I guess) to their needs. So if you insist on having the Praxis instead, they want a really good one.


Sally said...

Since they give a score of 165+ as being a category that corresponds to "excellent" (or whatever), perhaps average range means the range of scores that correspond to the category "average" rather than "you can only teach math in Guam" or whatever other categories they may have...? This seems unlikely, I admit, but I struggle to interpret "average range" as meaning jack squat, so this is the best I've got.

Tam said...

In the place I saw it, it was near some other statistics that also didn't seem entirely useful. I'm guessing it might mean an area within some multiple of a standard deviation of the mean.

rvman said...

According to the "Understanding your Praxis score" pdf's glossary, "Average Range" is ETS-speak for the 25th to 75th percentile ranks.

Sally said...

Rvman, ah, I wondered later last night (as I lie in bed and ponder statistics) whether "average range" was supposed to be the same as interquartile range. Hah! (I had never heard of interquartile range either until it came up in the class I'm TAing.) I think it's classic that they used some made-up term that is meaningless to both the population at large AND the kind of people who would recognize & understand the usual technical term.

rvman said...

They also noted in their descriptions that, for the scores which show a range of 100 to 200, a score of 200 isn't actually available on some of the tests. It seems a perfect score doesn't actually rate a 200 for some tests. I didn't notice whether tha was also true of the 200-990 scales, but I do remember that the top scores were also unavailable in some of the GRE Subject exams when I took the Econ one, lo so many years ago.

Sally said...

I think if you have to make up a new term for interquartile range for this audience, something like "Typical Range" sounds a lot better than "Average Range." But ETS doesn't listen to me.