I have a question regarding the roof of the AFQT.

I'm not allowed to post direct URL's, but the data is in "about dot com". Just Google "AFQT composite scores" to see this crudely pasted chart.

The left of each column is composite score and the right is relative percentile.


Standard Score Percentile (AFQT) Standard Score Percentile (AFQT)
80-120 1 204 50
121-124 2 205 51
125-127 3 206 52
128-131 4 207-208 53
132-134 5 209 54
135-137 6 210 55
138-139 7 211 56
140-142 8 212 57
143-144 9 213 58
145-146 10 214 59
147-148 11 215 61
149-150 12 216 62
151-153 13 217 63
154 14 218 64
155-156 15 219 65
157-158 16 220 66
159-160 17 221 67
161-162 18 222 68
163-164 19 223 69
165 20 224 70
166-167 21 225 71
168-169 22 226 72
170-171 23 227 73
172 24 228 74
173-174 25 229 75
175 26 230 76
176-177 27 231 77
178 28 232 78
179-180 29 233 79
181 30 234 80
182 31 235 81
183-184 32 236 82
185 33 237 83
186 34 238-239 84
187-188 35 240 85
189 36 241 86
190 37 242 87
191 38 243 88
192 39 244 89
193 40 245 90
194 41 246 91
195-196 42 247 92
197 43 248 93
198 44 249 94
199 45 250 95
200 46 251 96
201 47 252 97
202 48 253 98
203 49 254-320 99

Based upon this, it would seem that the absolute roof of the AFQT (in composite score, of course) is significantly higher than the top one percentile of the population. Am I right in this deduction, or thinking nonsense?

The different between the top 1 percentile and the maximum composite score would be 66 points, this is higher than the difference between the median and the top percentile.

That can't be right, could someone explain?

I realize the mathematical scores are weighted, therefore perhaps you could argue that these disproportionately heavy scores would be making up the difference and that the roof isn't massively higher than the highest listed percentile.

But then you must examine how each composite score and percentile score (relative to a particular score) is almost evenly distributed (with some exception, but these exceptions do not severely deviate from the trend).

For example, a 4 point difference below 1% in composite score would indicate a score in the top 5% and so on.

Could anyone help explain?

Max score on the AFQT is 99. It is not a percentile score but a standard score. Without the conversion charts (raw to ss) you cannot determine actual scores from the ASVAB.

>It is not a percentile score but a standard score

What? You mean that in reverse? A 99 would indicate 99th percentile, 98 would be 98th percentile and so on.

By "raw score", I don't mean the score that you were given upon completion of the test, I mean the amount you had correct that is transferred into what the chart considers a "Standard Score".

My question is if the raw score measures significantly higher than the top 1 percentile, as the chart suggests.

I'd like to know how much higher, if at all.

A 99 is not necessarily the max raw score one may obtain, only the max percentile one may be told they've scored.

Raw scores in "amount correct" are relevant as according to this website, they're considered for various other purposes.

No the AFQT is a standard score not a percentile. Raw scores are not used by any service for anything, only the standard scores. Each service has different standard scores that combine various raw scores. But the AFQT is the same for all services.

A 99 does not mean you did better than 99% of the people taking the test or obtained 99% correct. It is the max score possible on the AFQT.

You are getting way to involved in the scores. Just study and do your best. Unless you are taking or planning to take graduate level courses on test construction and scoring your questions don't really have anything to do with enlisting.

The score given at MEPS or the test site is the standard score for the AFQT because this is the score used by the services to determine enlistment eligibility.

I hope wikipedia articles are considered safe for usage on this website.

http://en.wikipedia.org/wiki/A...nal_Aptitude_Battery

"Arithmetic Reasoning (AR), Mathematics Knowledge (MK), Paragraph Comprehension (PC), and Word Knowledge (WK). AFQT scores are reported as percentiles between 1-99. An AFQT percentile score indicates the percentage of examinees in a reference group that scored at or below that particular score. For current AFQT scores, the reference group is a sample of 18 to 23 year old youth who took the ASVAB as part of a national norming study conducted in 1997. Thus, an AFQT score of 90 indicates that the examinee scored as well as or better than 90% of the nationally-representative sample of 18 to 23 year old youth. An AFQT score of 50 indicates that the examinee scored as well as or better than 50% of the nationally-representative sample."

This seems to concur with all other sources I've read, including some topics on this forum.

Wikipedia is not an acceptable link. And you seem to be caught up in a dicusssion that odes not matter. Lt William is former recruiter. There is a difference between a standardized score and a percentile. But it does not matter. There are scores you need to join and for certain jobs.

That is not a tough concept. If you want to focus on the basis for generating the scores, you are in the wrong place. You take the test. They give you scores. You do not see raw results. What do you not understand.

quote:

This seems to concur with all other sources I've read, including some topics on this forum.

Yes but none of these sources matter since they are inaccurate. Take a graduate level (or undergrad for that matter) stats class and you will understand the difference.

As Dave and I have told you this is a mute discussion. Study and score as best you can to qualify for the most jobs. End of discussion.

Eh, alright.

I was merely wondering if my score on the vocabulary section of the AFQT would correlate with my score I received on a verbal analogies based IQ test I had taken recently.

I scored around 1 in 215 (.5% and 140 on a standard deviation of 15) and I got 3 wrong on the AFQT practice test (vocabulary subtest) that is posted on the website.

I'm worried that my scores aren't correlating very well (within maybe 2/3 a deviation of one another). I realize the AFQT isn't a proper IQ test, but it certainly correlates relatively well (not well enough to be a valid measure of general intelligence of course)due to it measuring aptitude in a couple major facets of cognition, them being numerical reasoning and vocabulary/verbal reasoning.

Is the practice test on this website an accurate prediction of AFQT performance?

I'm still going to stand by the score givem being a percentile score, as I've viewed dozens of sources that state this. Such as a 50 would be scoring better than 50% of the population, etc. If I remember correctly "amount correct" per subtest is data given in your score sheet (as I've read), but it is not the composite score.

Unless my data is very recently outdated?

I'd require a new source to be swayed.

EDIT:

I just saw an official example of a score sheet on "Asvabprogram dot com". It very explicitly states both standard score and percentile.

What is listed as your "Military Entrance Score" is a composite percentile. The poor girl being used as an example looks like she has a severe numerical learning disorder.

This message has been edited. Last edited by: 26206385,

It is not an IQ test, it is an aptitude test. I remember taking one of those decades ago in High School. I think you are way to caught up in the BS for your own good.

quote:

Originally posted by Dave_M:
It is not an IQ test, it is an aptitude test. I remember taking one of those decades ago in High School. I think you are way to caught up in the BS for your own good.

It obviously correlates to a relevant degree....composite scores in school achievement tests usually correlate high enough to give a rough approximation of one's general intelligence.

This test is like a reasoning test combined with an achievement test.

The arithmetic section is sort of similar to the problems seen in WAIS and Wonderlic.

I think the vocabulary section is the exact same thing as the vocabulary section on the written form of WAIS. The format at least.

this is VERY helpful!

the answer to all your alls problems!


http://www.education.com/study.../article/score-job1/