|
0 members (),
80
guests, and
241
robots. |
|
Key:
Admin,
Global Mod,
Mod
|
|
|
S |
M |
T |
W |
T |
F |
S |
|
|
|
|
|
1
|
2
|
3
|
|
4
|
5
|
6
|
7
|
8
|
9
|
10
|
|
11
|
12
|
13
|
14
|
15
|
16
|
17
|
|
18
|
19
|
20
|
21
|
22
|
23
|
24
|
|
25
|
26
|
27
|
28
|
29
|
30
|
31
|
|
|
|
Joined: Feb 2012
Posts: 181
Member
|
OP
Member
Joined: Feb 2012
Posts: 181 |
I got the following blurb from the CTY SCAT website": [/quote]What is a percentile?
PERCENTILE: The percentile score allows you to see the percentage of students who scored lower than your child on the SCAT. It is extremely important for you to understand that your child was not compared to students in his/her same grade. We compare your child's raw score to scores obtained by an older group of students. If your child scored in the 50th percentile, your child’s performance on the test was better than 50 percent of the children two (2nd-4th) or three (5th-6th) grades above your child who took the test. EXAMPLE: "63 (compared to Grade 4)" means your 2nd grade child scored better than 63% of a normative population of students in grade 4. [quote] I was under the impression that SCAT is used only for gifted testing. Given that how do they determine the normative population?
|
|
|
|
|
Joined: Oct 2014
Posts: 675
Member
|
|
Member
Joined: Oct 2014
Posts: 675 |
I believe when the test was created, it was validated with higher grade level populations, but that is a one-off development activity. Once finalized, though, it is pretty much only used by CTY on out-of-level kids.
|
|
|
|
|
Joined: Apr 2014
Posts: 4,088 Likes: 10
Member
|
|
Member
Joined: Apr 2014
Posts: 4,088 Likes: 10 |
It was originally developed by ETS for grade 4-12 students, but was purchased by JHU in, I think, 1996, which is why it is now used exclusively for GT testing of students in grades lower than the nominal test levels.
...pronounced like the long vowel and first letter of the alphabet...
|
|
|
|
|
Joined: Feb 2012
Posts: 181
Member
|
|
OP
Member
Joined: Feb 2012
Posts: 181 |
Platypus/aeh thank you. So given that these norms are more than 2 decades old is there any validity to the percentile? OR is it totally useless information? A percentile against current testers would seem to make more sense. But they do not provide it.
|
|
|
|
|
Joined: Apr 2014
Posts: 4,088 Likes: 10
Member
|
|
Member
Joined: Apr 2014
Posts: 4,088 Likes: 10 |
Ideally, yes, one would prefer norms under 20 years old.  They likely do not provide norms against current examinees because they do not have the resources to do a nationally-representative re-norming like ETS did, and prefer not to use the convenience sample for norming, because it's self-selected, and not representative of the US population. CTY does have some experience using relatively aged out-of-level testing norms with GT populations, as a similar issue was involved with the old SATs, prior to the re-centering in 1995. I wouldn't say it's totally useless information. You just have to understand the limits of the data.
...pronounced like the long vowel and first letter of the alphabet...
|
|
|
|
|
Joined: Oct 2014
Posts: 675
Member
|
|
Member
Joined: Oct 2014
Posts: 675 |
If you're registered with CTY, you do have access to data tables for the last year's testers, but I think they may not be available via the public site. DS did SCAT (and STB) a couple of years ago, and I have tables which provide a breakdown by grade and sex, frequency distribution of scores/ percentiles, etc. For the out-of-level numbers, I just have raw score and a percentile against grade 8 norms (he was in grade 5).
|
|
|
|
|
Joined: Apr 2014
Posts: 4,088 Likes: 10
Member
|
|
Member
Joined: Apr 2014
Posts: 4,088 Likes: 10 |
Actually, all data tables for CTY administrations of SCAT through 2014 are available here: http://cty.jhu.edu/talent/after/about/#SCAT_test_dataFWIW, this still doesn't tell you what the "true" standing is against current out-of-level groups is (say, hypothetical current 8th grade norms). Just against the test population of self-selected examinees, who are, of course, not representative of the overall population, or even necessarily of the top 5-10% of the population (cognitively). Some of the distributions are noticeably less normal than others.
...pronounced like the long vowel and first letter of the alphabet...
|
|
|
|
|
Joined: Feb 2012
Posts: 181
Member
|
|
OP
Member
Joined: Feb 2012
Posts: 181 |
aeh Did look at these. I also found it weird that they break it down by sex. Have not seen that in any other assessment. I guess you can combine them and do the math.
|
|
|
|
|
Joined: Oct 2014
Posts: 675
Member
|
|
Member
Joined: Oct 2014
Posts: 675 |
Ah - thanks aeh! I only seemed to be able to find that page after I logged in - but then I often have a hard time finding stuff on CTY even when I know I've been there before!
|
|
|
|
|
Joined: Feb 2011
Posts: 1,432
Member
|
|
Member
Joined: Feb 2011
Posts: 1,432 |
VR00, I would definitely stick with the stats for other talent search participants even though it still may not be representative of the top 5 to 10%. I think that the above-level stats may be particularly problematic for the quantitative scores due to elementary math curriculum changes in the last two decades - lots of topics are pushed to earlier grades. Even though SCAT is supposed to measure ability, I think it only does so accurately for kids who have not been exposed to higher curriculum. Remember, you don't need exact answers on many of the math problems, which can be solved either intuitively or procedurely.
|
|
|
|
|
