Finally done with it! 1530 (800Q, 730V)

[Oldman]

Kickboxer, thanks for your contributions on the board. I've learned a lot from you posts!

 

Personally, I think the GRE measures how good you are at taking the GRE, and that's about it. I have seen numerous people score extremely high on the verbal section, yet they can barley compose a legible debrief. This makes me wonder how well the verbal section measures one's mastery of the English language.

[GRE_Kickboxer]

Oldman:

 

I think there are several reasons for that.

 

1) Some foreign students are cheating on the GRE. Also, some foreign students are good at analytical reasoning, but haven't developed practical English skills yet. (American students are also cheating.)

2) Some people with high ability are very analytical in some ways, and are able to reason through analogies, data and also have good enough memories. As a result, they can go through vocabulary lists, memorize them, and use the info on the test. These people, however, may lack verbal creativity and practical verbal skills. They may also lack social skills. Sometimes these people tend to be good at math and logic, poor writers and less than eloquent speakers.

3) Don't ignore the analytical writing score. (This test still has a lot of weaknesses too though.)

[Oldman]

Thanks for the analysis kickboxer.

[Eyohimself]

GRE_KICKBOXER:

 

Hmmm, I must commend you for your insightful analysis of issues on this forum and for your active participation. I am sure a lot of guys out there lurking on this forum have learnt a lot from your posts.

 

Frankly speaking, do you mean people cheat on the GRE? How? It's hard for me to believe this.

[sanjay21]

Gre_K & the rest of TM,

 

How you guys?. Long time.

 

Yes some do cheat. But I believe this cheating will help only for about 30-40 extra points. Nothing more.

 

College look at the entire score (Q,V,AWA), so it's very important to get a 4+ in AWA.

[Ancalagon The Black]

Umm, GRE_K: Just how do foreign students cheat? Heh, if I knew maybe, I would have cracked that verbal up to a 750 :D

[GRE_Kickboxer]

Hi Ancalagon, buddy. I can't answer that question specifically, but I know that they are sharing answers. That's why ETS stopped such answer sharing in China. Students were busted.

[GRE_Kickboxer]

No, MOST PEOPLE do not cheat. Moreover, this is the last post I am going to write about this subject. Erin (the moderator) doesn't like these posts and I can't say I blame her. ETS gives her a hard time when the subject of cheating arises because it gives people ideas and draws attention.

 

I agree with Sanjay. The stuff that some people are currently doing helps the score a little bit. For instance, it can boost your score 30 points. That's about it. Sometimes it won't even help that much.

 

After all, if you aren't very good at math, you won't get a high score because the database ETS uses is quite large.

 

The only exception to this is with students that were cheating in China awhile back, for instance. They had something going on that was entirely different in nature and resulted in perfect scores for VERY WEAK students.

 

They were busted, however. The type of cheating they were doing was entirely different and much more serious than what currently goes on.

 

Again, enough "talk" about cheating. Let's not give Erin (the moderator) a headache. Thanks.

[Jive Miguel]

GRE_Kickboxer wrote:

 

"I am not saying that IQ tests are good instruments. I am simply stating that the GRE was created as an IQ test. This isn't an opinion, this is a fact. The SAT was created as an IQ test too.

 

If you don't believe me then please go read about the history of the old SAT test and the correlation they claim it had with the WAIS IQ test. The GRE is essentially the same as the old SAT test.

 

Now, this doesn't mean that these tests do a good job of measuring intelligence. That's a completely different debate.

Vocabulary is used in almost all professional IQ tests including the Wechsler IQ test and the Stanford Binet which are considered the GOLD STANDARD IQ tests. Vocabulary measures "acquired intelligence" or "crystalized intelligence". Again, this is a construct they created, not one I necessarily take very seriously. General knowledge is also part of many IQ tests. They also have "fluid intelligence" tests which includes things like arranging blocks into patterns.

 

Again, I'm not arguing whether or not this is the right thing to do. I am simply stating that this is what IS DONE.

 

The makers of the SAT used to try and convince colleges that unless a person had a high level of ability to start with, prepping for the tests would not help. The same arguments used to be made about the GRE. Again, this is what they claimed, not what I claim.

 

The book "the Bell Curve" based much of its research into IQ on the SAT and the GRE."

 

This is an excellent post. You do a good job in clarifying the issue people tend to get confused on of whether the GRE predicts IQ well, or whether it is predictive of more amorphous and qualitative concepts of intellect and aptitude.

 

From a statistical standpoint, the GRE correlates well with IQ tests and the reason is, as you point out, it was designed as an IQ test. The SAT and the GRE and Miller Analogies are surrogates for an IQ test and they're considered "surrogates" mainly for political reasons. They are basically de facto IQ tests.

 

A few people are dismissive of vocabulary as a measure of anything substantial and tend to dismiss the GRE verbal as a measure of vocabulary, not anything else. But bear in mind, If this were true, and the GRE measured just vocabulary, then the correlation to a full scale IQ would be extremely high (around .8).

 

This seems counterintuitive, because many people assume reading comprehension and analogies are indeed 'harder' than learning words. On the Wechsler, for instance, the vocabulary section is considered the most "g loaded" (a statistical word that means a test loads on general intelligence, what they're out to measure). Another subtest on the Wechsler the verbal abstract reasoning test (a subtest similar to analogies) is a good measure of IQ in and of itself but not quite as high as vocabulary. Taken together, the vocabulary and analogies on the GRE is a good verbal test -- this, without reading comp and sentence completion (neither of which measure G quite as well).

 

The quant portion is a good measure of non-verbal IQ; it is set up with ETS testing more concepts rather than math skills - it gets at the G factor (rather than the s factor - specific skills). The quant, as studies have shown, has a good correlation with the performance portion of the Wechsler, tapping as it does non verbal reasoning. The problem is the ceiling is a little low for those who are extremely proficient in math and many people who are good in math ace it without much of a problem. But, then, if the GRE quant were taken by the general population, the ceiling of the test would be quite high).

 

My primary hesitation with IQ as a unitary construct is that verbal and non verbal portions are weakly correlated. I tend to think of math and verbal as separate abilities but arguably connected enough with a big enough swath of people so you could say that a G factor exists and ergo "IQ." That being said, I do think of verbal and non verbal reasoning as different enough in kind that one could almost say two types of intelleginces exist, or if not intelligence, two types of reaoning that have very little to do with one another, both of which require significant mental power for their manifestation.

 

I'm not a fan of the multiple intelligence scheme - that confuses talent with intelligences to make those people who lack high intelligence feel as if they have it. I think Gardner's work is somewhat interesting, but he's out to make everyone happy.

 

I more or less agree with the conventional ETS stand regarding prepping, though they are understating what can be done by all. I think it's hard to prep beyond ones ability but the catch here is that many people tend to understimate their own ability, forgetting that they need to be "in shape" for the exam.

 

The analogy to an athelets applies here. "I used to be very good athelete" doesn't mean much if you've put on 100 pounds and have turned into a couch potato. Someone who can learn and adapt "tricks" and raise their scores is a high IQ individual, probably. Someone who spends over 6 months to a year learning words and math is benefitting from an actual education. Remember, even the most avowed determinists allow 20-30% of intelligence to one's environment and education helps improve intelligence. (Btw, I'm not a psych major, but I have done some of my own research on intelligence debate, b/c it's such a hot topic. I'd like to get this as an AWA topic);.

[GRE_Kickboxer]

Jive_Miguel:

 

I agree with you, for the most part. I would add that there are some important aspects of intelligence that exist, and are used almost every day of our lives, but are very hard to test. As a result, these types of abilities are not as easy to subject to factor analysis even though they may or may not correlate very well with "G". In fact, these mental abilities may even be as "general" as "G", poorly correlated with "G", and "generally difficult to test". (Sorry about the funny wording, there.)

 

An example might be the ability to select strategies or modes of behavior from numerous options available in daily life. The ability to multi-task throughout your day might be a specific factor, here. The little research done on this has shown that when tested, these skills don't necessarily correlate very well with "G". The reason that IQ tests tend to correlate very well is that the easiest things to test tend to be those that emphasize "long term memory", "working memory" and rigid factual data. Hence if the tests are all similar in terms of skills they tap into, then obviously the scores on the different tests will correlate well. As you pointed out, associated verbal memory skills and visual/performance intelligence skills for these abilities correlate, but not as strongly as you might expect.

 

Nonetheless, anyone that argues that "IQ" measures nothing is fooling him or herself and/or is somewhat misinformed. IQ tests do measure something that is important, although you need to look at scores on a case by case basis for meaningful analysis.

 

Ramin

[Jive Miguel]

Right, my position is I accept the G-factor, but the overall measurement has, as you point out, holes in it. I don't subscribe much to Gardner's work, but I do think the work Thurstone did with discovering primary mental abilities -- (the breakdown of intelligence into 7 different factors) has as much or even more value than the full scale number...especially when one considers gifted people.

 

The threshold argument for intellectual/creative achivement to occur that Jensen posits (and even Gardner concedes), of an IQ needing to be in the neighborhood of about 1.5 standard deviation above the mean, seems very likely. Obviously, IQ matters, but other intangibles do as well ...the necessary but insufficient condition argument applies.

 

As for mental abililites that slip under the G factor radar -- definitely true. I never read any studies about multi-tasking (something I personally suck at) but that's a good example. Another one is social intelligence, which is very amorphous and hard to test, but clearly some people have it and others do not. This type of intelligence is clearly a mental ability but doesn't seem to correlate with "G" because it's nearly impossible to get at emprically.

[GRE_Kickboxer]

Your post is very eloquent and well-reasoned. You get a 6 on the essay.

 

Ramin

[Jive Miguel]

Cool deal. Do me a favor and email ETS and let them know my score.

[pratikashar]

Interesting.

[Guest seokool]

sanjay thats a great score........can you tell me which college you are think in of??

[Dr.VanderBildt]

I have to say I was rather amused by the somewhat obvious logical flaw in GRE_kickboxers post. The fact that that the SAT started out as an IQ test doesn't mean that the current version of it is one. It has undergone major changes since it was first introduced. Neither the current SAT, nor the current GRE is accepted as a valid IQ test with any of the high IQ clubs, such as Mensa or Triple Nine. They are now considered "achievement tests", rather than tests of innate ability.

 

It is no secret that the verbal section of the GRE, for example, correlates highly with IQ tests such as the WAIS. So I agree with that statement. It has to be said though that, intuitively, vocabulary is a measure of achievement. A person without any major memory impairment, can vastly increase their vocabulary through practice, over the course of months or years. However, as said before, the verbal part of the GRE parallels mainstream IQ tests very extensively.

 

The mathematical portion, on the other hand, has no real counterpart on the WAIS, except for the arithmetic part. And as it turns out, the quantitative score correlates, on average, poorly with overall IQ. The correlation between one of the best non-verbal measures of intellect, the Raven Progressive Matrices test, and the math GRE is, on average, lower still. I think it's funny how all these people swallow the salespitch from ETS whole. Fancy words fly around like "quantitative reasoning ability", etc. I seem to recall something about it "touching on concepts rather than math skills" from one particular poster here. Are you being serious? Are you saying that being able to work through a set of algebraic inequalities "touches on concepts rather than math skills"? Or finding the mode of a set of integers? Or finding percentage increases or averages? In what kind of a deluded world do you live in?

 

The GRE is sold to universities as somekind of a mysterious psychometric tell-all yardstick. ETS needs people like you to believe that it is uncrackable and that it predicts success or failure, so that it will be taken seriously.

 

The ETS refuses to call it an IQ test however. And as said before, the high IQ societies no longer consider them IQ tests either. It has been shown that students with a high IQ care, on average, more about their academic future than students with a low IQ. It therefore goes without saying that these students will invest more time in studying for the SAT and the GRE. It also goes without saying that a person with a high IQ is, on average, going to benefit more from every hour spent on studying than a person with a low IQ. This is science and common sense. And this is how the correlation gets upped. Not because the problems on the GRE are IQ problems, but because people with a high IQ come better prepared to take them. So you might hold onto the fact that these tests are correlated, and so in effect one has some predictive value for the other, but the fact still remains that the GRE is an achievement test, not an IQ test, and that the correlation between the two ranges from .445 to .8 depending on who you ask, and that there are students out there with a mediocre IQ that studied for these tests and aced them, and students with an exceptional IQ that either didn't bother to study or had learning disabilities and bombed. I know of personal examples in all these cases.

 

And that's all there is to it.

[wanderlust]

I'm so glad it's over.

 

The Quant was similar in difficulty to that of powerprep. In fact I had enough time to double or even triple check all the answers. The harder tests like catprep and BigCD are good if you have to score a 800, or it's just excess preparation. GRE_Kickboxer's tests are really really useful. Many are saying the real GRE has increased in difficulty because the questions are getting more wordier and tricker. Also I noticed that *all* questions appear to be of the same difficult, so if you get a harder question in the beginning dont panic.

 

As for the verbal part, the questions are very very similar to that of big book. I seriously doubt any other test prep companies (like princeton) has this similarity. In fact, after doing many big book RC's you can easily eliminate many of the answer choices without even looking at the passage.

 

Moreover, I think with enough practice anyone can score a 700+ in verbal. I'm generally not very good in English and neither do I read novels. Add to this the fact that I'm an engineer and have spent most of the last four years with numbers. Also, I'm a non-native English speaker.

 

Materials that have really helped me

1) Barrons word-list- This is the definitive word list. I think it's hard for non-native English speakers to get a decent verbal score without knowing barrons.

2) TestMagic: Ofcourse, how can I forget this forum. Really helped me with SC's and Analogies and I have learnt a lot how people eliminate answer choices in them. The quant forum is also great. Lot of questions and exams (espl by GRE_K).

3) BigBook: I ignored the quant part of bigbook, just did the verbal. For verbal practice this is the ultimate book. 27 real tests and real questions

4) 800score, BigCD, Catprep: These have higher quant standard and are good for guys who are aiming a perfect 800

5) ScoreItNow - I'm not very good at writing and this really helped me refine my writing skills. I practiced about 3 essays have finally ended up getting a score of 5. Now after I get my AWA score I'll come to know if it's evaluation is accurate

 

As for real standard of quant, the only tests that are close to it are powerprep, kaplan and barrons. Other tests are either too hard or too easy. The princeton review tests for example are too easy and give a false sense of confidence.

 

Thanks TM!

:tup:wow....superb....do let me know...from whr to get the books tht u ve mentioned above...

second how much of preparation time is sufficient?

[eflect]

Congrats Sanjay and thanks for sharing many useful resources.

[tania_saha]

Congrats for great score. I have a doubt. If you need to do long calculations does that mean you are in wrong direction in Quant?? Need ur opinion on this plz. Thanks. Tania.

[munuchwa01]

Excellent score !!

congratulations !!

[monkeydluffy]

congrats, would you say its worth it to just rely on the barron's word list and ignore the other ones (compilations, other companies words list, etc)?

[GRE_Kickboxer]

In my previous post, I was simply explaining that the SAT was originally used as a quasi IQ test, and that until recently it was accepted as an IQ test by many people who were hiding behind closed doors. Please note that I do not approve of its use as an IQ test!

 

While the GRE is no longer accepted by many high IQ societies, it was accepted by them as such as recently as the 1990s. The same thing holds true for the ACT and SAT. In addition, similar tests such as the GMAT and LSAT are still accepted as IQ tests by high IQ societies.

 

I totally agree with Dr. V that the SAT will serve as a very poor measure of intelligence for many people! It is also true that its correlation with professional IQ tests such as the WAIS will vary significantly depending upon the life history of the test taker. In fact, some of the statements made by Dr. V can be seen on the Prometheus Society web site. Finally, the inconsistent correlations between the SAT and IQ mentioned by Dr. V were actually findings from Dr. Thomas Bouchard, whom I took a course from at the University of Minnesota.

Edited December 9, 2012 by GRE_Kickboxer