I think DAT is no good, because...

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

bor0000

Membership Revoked
Removed
10+ Year Member
15+ Year Member
Joined
Jun 5, 2005
Messages
190
Reaction score
0
I looked at the contents of the test. And here is how it is scaled:
30 ts: 0 wrong
25 ts: 4 wrong (out of 100)
23 ts: 7 wrong (out of 100)

now assuming you even did extremely well on this test and got 25 ts. and the school's average is 22. you think this will help you out really? but in fact, since the difference between a 22 and a 25 is so small, and most of the questions on the dat are not so tough..., your score will not mean nearly as much as your gpa.... i wish i could be wrong. but in college admissions, SAT scores did not mean too much to the elite schools. you could get rejected with 1550 or accepted with 1400. and there the difference between 1400 and 1550 was much much larger than 3 questions.
 
i guess i could be wrong but i'm pretty sure that the scale you are using is not correct.

a 25 is about 96%, but my understanding is that the % is based on other peoples scores who are taking the DAT, not a straight % of correct questions.
 
25 is supposedly 99%+ as compared to the rest of population(like 1600 on sat). but im talking about the raw score. i.e. if you get 1 question wrong in general chemistry you already get a 25 in that section. and my point is that in this case schools will not care for someone with a 3.4gpa and a 25 or even a 29 over someone with a 3.8gpa that got a 22-23. %-wise 22-23 is probably equivalent to 1500 on the sat. and everyone knows ivyleague schools wouldnt care if you got a 1500 or a 1600, even though the difference between the 2 is something like 10-15 questions(in raw scores)!
 
bor0000 said:
I looked at the contents of the test. And here is how it is scaled:
30 ts: 0 wrong
25 ts: 4 wrong (out of 100)
23 ts: 7 wrong (out of 100)

now assuming you even did extremely well on this test and got 25 ts. and the school's average is 22. you think this will help you out really? but in fact, since the difference between a 22 and a 25 is so small, and most of the questions on the dat are not so tough..., your score will not mean nearly as much as your gpa.... i wish i could be wrong. but in college admissions, SAT scores did not mean too much to the elite schools. you could get rejected with 1550 or accepted with 1400. and there the difference between 1400 and 1550 was much much larger than 3 questions.

well, personally, i don't think the DAT is a good test period. regardless of how it is scaled...

the DAT is purely a recall test, and anyone who puts in enough time to memorise the material can do well...as for the gpa, assuming we're talking about a respectable institution, most the testing going on there will be more than just recall, but rather more of integration and application..and being able to integrate knowledge is what is important, i think...
 
r0entgen said:
well, personally, i don't think the DAT is a good test period. regardless of how it is scaled...

the DAT is purely a recall test, and anyone who puts in enough time to memorise the material can do well...as for the gpa, assuming we're talking about a respectable institution, most the testing going on there will be more than just recall, but rather more of integration and application..and being able to integrate knowledge is what is important, i think...

For ONCE I can agree with you.

DAT was implimented as a mean to standardize everyone's undergraduate achievement. Now it is wrong to put more or even equal emphasis in a 4.5 hour test than 4 years of education when clearly not everyone is taking the same test.
 
this absolutely sucks. i wish that DAT were designed in such a way that the difference between a 25 and a 22 were more than 3 questions. I.e. if the questions themselves were harder... without this, there is no way for me to redeem myself if i got some bad grades. Of course it's 90% my fault, but also many people go to lesser schools or take bull**** classes. If im really an unworthy student, i shouldnt do well on the dat anyway. but in case i do well, it wont help me🙁
 
Comet208 said:
For ONCE I can agree with you.

DAT was implimented as a mean to standardize everyone's undergraduate achievement. Now it is wrong to put more or even equal emphasis in a 4.5 hour test than 4 years of education when clearly not everyone is taking the same test.

'not everyone is taking the same test.' that's key here...

on my test, i got a 28 in QR, and that was 100th percentile...but someone else got a 30 (on a different test), and if we just look at the scores, it may look like i performed worse, when in fact, i just took a different test...

since the dental schools don't get the percentiles (or so i heard), this number really doesn't mean a DAMN thing...

i thought the point of the scaled score was to make it so dental schools could COMPARE the scores across the nation..

obviously, if one compared a 28 vs a 30, the 30 is better...but my 28 was also 100th percentile...

this system makes no sense...
 
roentgen, maybe you got 1 or 2 questions wrong? did your score report tell you if you answered 40/40 on the QR? if you got all questions right wouldnt it say 28-30? But even if thats not the case, 28 is obviously well above the 22-23 average that is at harvard. The bad news is that both 28 and 30 only beat 22-23 by several raw points. Which means that the schools may think you got 28 because you're lucky, not because youre smart.

p.s. i remember on sat1, 800math gave only 99th %ile. and the test had 60 questions instead of 40 where you had to get every one of them right, and it was of about the same difficulty. weird,eh?
 
Comet208 said:
Now it is wrong to put more or even equal emphasis in a 4.5 hour test than 4 years of education when clearly not everyone is taking the same test.

Yeah but even throughout 4 years of education, clearly not everyone has the same difficulty of classes, teachers, tests, etc. DAT is the best way to standardize it; even though everyone does not get the same questions, they are more or less equal in difficulty. Whereas, for classes, there can be a huge difference in difficulty level from school to school and professor to professor. If one has a high gpa, why wouldn't they be able to score high on the DAT if it is just a recall test? Also, going back to the above SAT example--not everyone got the same SAT test either, but a 1500 is always going to be better than a 1480, regardless.
 
no, schools consider any sat score over 1500 to be the same. then they look at your h.s. gpa and your essays/recs. the reason being that the difference between for example 1500 and 1550 is too small in terms of the raw score. But on dat the difference in raw scores is even smaller... i hope my whole reasoning is wrong. Because im looking towards the DAT as my final chance to redeem myself.
 
i think that most admission don't give a rats ass what happens after 23 on the DATs. so go get your 23 and be happy!
 
Gurl21 said:
Yeah but even throughout 4 years of education, clearly not everyone has the same difficulty of classes, teachers, tests, etc. DAT is the best way to standardize it; even though everyone does not get the same questions, they are more or less equal in difficulty. Whereas, for classes, there can be a huge difference in difficulty level from school to school and professor to professor. If one has a high gpa, why wouldn't they be able to score high on the DAT if it is just a recall test? Also, going back to the above SAT example--not everyone got the same SAT test either, but a 1500 is always going to be better than a 1480, regardless.

yes, we need a means of standardising..but the DAT does a horrible job doing so...

high ppl should get high dat score...
but low ppl can also get a high dat score..
so what's the point?

basically, given enough time and preparation, ANYONE can make a high dat score...
we do need a standardised test..but something else that requires more analytic skills...
 
r0entgen said:
yes, we need a means of standardising..but the DAT does a horrible job doing so...

high ppl should get high dat score...
but low ppl can also get a high dat score..
so what's the point?

basically, given enough time and preparation, ANYONE can make a high dat score...
we do need a standardised test..but something else that requires more analytic skills...

I agree.

I also like to add that since the same test is given at different times, it allows for cheating.

MCAT - Its possible for you to study your whole life and not do well.
DAT - Study hard and you'll get a high score.

In dental school, you are given a limited amount of time to study a whole lot of information. IMO, SOME people with low GPA's (relative to the students at their school) who do well on the DAT aren't necessarily prepared to handle this.
 
QuietGuy said:
I agree.

I also like to add that since the same test is given at different times, it allows for cheating.

MCAT - Its possible for you to study your whole life and not do well.
DAT - Study hard and you'll get a high score.

In dental school, you are given a limited amount of time to study a whole lot of information. IMO, SOME people with low GPA's (relative to the students at their school) who do well on the DAT aren't necessarily prepared to handle this.

yep..very good point...we see a lot of this on sdn..people posting questions about what was on their tests...not that i care, but i'm just trying to make it a point that the DAT really doesn't measure much....and i'm not sure why people don't realise this, and continue to use it...

i'm not saying we should do away with standardised testing..we just need a better one that actually measures some degree of aptitude...

i agree with your comment about mcat...that test is about applying your analytic schools, and background information one has learned in university...and to apply CONCEPTS to passages and readings...

one can memorise all they want, but if he doesn't know the concepts behind what he is memorising, that will serve him no good on the mcat...

for the dat, no concept is needed....you barely even have to think much on it...you either know it, or you dont....and if you know it, you mark your answer...if you don't, then you eliminate a few answers and guess...and it's as simple as that...i just don't understand how this exam is supposed to measure one's potential success in dental school..
 
Gurl21 said:
Yeah but even throughout 4 years of education, clearly not everyone has the same difficulty of classes, teachers, tests, etc. DAT is the best way to standardize it; even though everyone does not get the same questions, they are more or less equal in difficulty. Whereas, for classes, there can be a huge difference in difficulty level from school to school and professor to professor. If one has a high gpa, why wouldn't they be able to score high on the DAT if it is just a recall test? Also, going back to the above SAT example--not everyone got the same SAT test either, but a 1500 is always going to be better than a 1480, regardless.

SAT has proven to be linked to race and income. So it is another standardized test that needs to be revamped. how can you standardize something when your measuring tool varies from one case to another? how can you put 4+ years of education next to 4.5 hours? this is why they need to look at the total package and not just the scores. I dont appreciate it when your application is rejected during the first screening when they just look at your GPA and DAT scores. Dentistry is a lot more than grades.
 
bor0000 said:
no, schools consider any sat score over 1500 to be the same.
Actually, at some schools, scores >1350 (back in 1998 when I applied to college) were thrown in the same bucket as the 1480 and 1600s, according to the ADCOM person I had an impromptu interview with.

bor0000 -- my advice to you is if you want to score well on the DAT, hit the books and practice exams. Even if the DAT system is flawed, you're not going to change it so roll with the punches.
 
Comet208 said:
SAT has proven to be linked to race and income. So it is another standardized test that needs to be revamped. how can you standardize something when your measuring tool varies from one case to another? how can you put 4+ years of education next to 4.5 hours? this is why they need to look at the total package and not just the scores. I dont appreciate it when your application is rejected during the first screening when they just look at your GPA and DAT scores. Dentistry is a lot more than grades.
i would reply but dont want to create a controversy.
 
drat said:
Actually, at some schools, scores >1350 (back in 1998 when I applied to college) were thrown in the same bucket as the 1480 and 1600s, according to the ADCOM person I had an impromptu interview with.

bor0000 -- my advice to you is if you want to score well on the DAT, hit the books and practice exams. Even if the DAT system is flawed, you're not going to change it so roll with the punches.

drat, thanks for the response, thats exactly what i plan to do-study. but i am really bummed because even if i get 25+ on all the sections, the school wont care, because this test is only significant up to 22-23🙁
 
bor0000 said:
i would reply but dont want to create a controversy.
I think you should reply! Comet has a valid argument! Nuthin' wrong w/ a little debate.
 
drat, comet's argument is about politics(race), so he would probably want any test to become easier(while i want it to become harder), and i dont have any intent to turn a post into politics, i'll just say that i disagree with him.
 
bor0000 said:
drat, comet's argument is about politics(race), so he would probably want any test to become easier(while i want it to become harder), and i dont have any intent to turn a post into politics, i'll just say that i disagree with him.

I don't think Comet meant so much as making the test easier, but more reflective of various cultures within our population. And why is it that you want the test to be harder? The scoring for the DAT is done on a bell curve, so scores tend get pulled toward 15. This is not very different from the MCAT or the majority of classes I took in undergrad.
 
crazy_sherm said:
I don't think Comet meant so much as making the test easier, but more reflective of various cultures within our population.

Thank you. I am not concerned with the SAT being made harder or easier, if made harder, it is garder for everyone. But they need to eliminate questions that individuals of different races and socioeconomic backgrounds uncountiousely respond different to.


bor0000, I'd like to hear your comments. Please.


All i am sking is that ADCOMS should look at the total package and not eliminate/accept based on DAT/ GPA.
 
crazy_sherm said:
And why is it that you want the test to be harder? The scoring for the DAT is done on a bell curve, so scores tend get pulled toward 15. This is not very different from the MCAT or the majority of classes I took in undergrad.


A harder test with more questions gives you a wider distribution of scores. It reduces the chances of "lucky" guessers from doing well. A bell curve can only give you a very limited amount of information if it is narrow and high.
 
yeah, i meant if the test is harder then if you get a high score, the schools will be really impressed. like if you answer 90% of the questions correctly. but in this test you are expected to answer 90% of the questions correctly to begin with, and if you answer 95%, it may just be attributed to some last minute studying or luck, it's not like a difference between 70% and 90%. of course i can do badly whether it's a hard test or an easy one, no one knows, but at least if it were hard, i could be really rewarded for doing well. it's like a difference in gpa. 3.8 vs 3.4 is huge. but there is no possible dat score that could make up for such gpa. 25 vs 22 is probably like a .1 difference in gpa.

as for sat, i disagree with you that the questions are out to get anybody(blacks). and i am not going to say why.

and this whole thread should be deleted. im just so full of it! i really should just go and study, because i whine about things that cant be changed. i just came from a dental office with a cavity, and my jaw is numb🙂
 
I agree. Let's make the test much harder...after I have taken it of course.
 
*sigh*

I wrote a long response but it will just fall on blind eyes.

If you want to take a really tough exam to show how smart you are, take the CPA exam w/o studying. 🙂
 
i didnt see the long response. but if it was something about whether you want a harder test or an easier test, then thats behind me, i dont care for anybody's opinions, i dont think i raised that topic to learn anything, but just to express my frustrations. and cpa wouldnt interest me because im not interested in being an actuary or an accountant, or whatever that is.
 
I think that a few of the schools look at the total package when looking over applicants. On paper, my DAT is the only thing that made me stick out and I am glad it did. If schools only looked at GPA's I have a feeling that my 3.1 would have been ****canned as soon as it left the envelope. Basically no standardized tests are perfect but we all have to jump through the hoops before we can become dentists.
 
TucsonDDS said:
I think that a few of the schools look at the total package when looking over applicants. On paper, my DAT is the only thing that made me stick out and I am glad it did. If schools only looked at GPA's I have a feeling that my 3.1 would have been ****canned as soon as it left the envelope. Basically no standardized tests are perfect but we all have to jump through the hoops before we can become dentists.
Very true. Many people complain about the DAT, especially since it has nothing to do with dentistry, but it's a test that everyone has to take. If you are pissed off about it and neglect it you will not do well and probably not get into d-school. If you take it seriously you should do well and get yourself a spot somewhere.
 
bor0000 said:
yeah, i meant if the test is harder then if you get a high score, the schools will be really impressed. like if you answer 90% of the questions correctly. but in this test you are expected to answer 90% of the questions correctly to begin with, and if you answer 95%, it may just be attributed to some last minute studying or luck, it's not like a difference between 70% and 90%. of course i can do badly whether it's a hard test or an easy one, no one knows, but at least if it were hard, i could be really rewarded for doing well. it's like a difference in gpa. 3.8 vs 3.4 is huge. but there is no possible dat score that could make up for such gpa. 25 vs 22 is probably like a .1 difference in gpa.

as for sat, i disagree with you that the questions are out to get anybody(blacks). and i am not going to say why.

sorry about your cavity, and i dont mean to be argumentative, but i just want to comment on some things.

first, there is tons out there in the literature, internet, etc concerning the racial bias of the sat. it's not an opinion of one person, but a general concern of large groups of people, which is just tough to find a simple solution. if u search online u'll find tons of info on this.

so yes, sat is not perfect. but it's the most "efficient" way to normalize students' abilities out there right now. same w/ the dat. it can be made harder, true, to broaden the bell curve, but i guess the people upstairs decided that in order to most quickly evaluate students' study skills, they must test simple material regurgitation. so, if youre good at that, and score high on this test, that's what they're looking for. if they have info from a test that equates you to john nash, it complicates their whole admissions process.

i do pseudo agree w/ u about luck playing a part in separating 22-30, since extremely few questions separate these scores. and i hope it's true that schools dont care if u got 22 or 29. however i truly believe there is a significant knowledge gap b/w high teens, then low 20's, then 22 and up. at least personally, it took a little time to differentiate my scores through these levels.

just remember as easy as it is for people to get lucky into the upper 20's, people can do the same in the other direction.
 
netsn06 said:
so yes, sat is not perfect. but it's the most "efficient" way to normalize students' abilities out there right now. same w/ the dat. it can be made harder, true, to broaden the bell curve, but i guess the people upstairs decided that in order to most quickly evaluate students' study skills, they must test simple material regurgitation. so, if youre good at that, and score high on this test, that's what they're looking for. if they have info from a test that equates you to john nash, it complicates their whole admissions process.


um...if this DAT is supposed to measure one's success in dental school, then it certainly doesn't do that. do you think dental school is pure regurgitation? do you think the career of dentistry itself is regurtiation? no, it's not. it requires application of your knowledge...and if the DAT measures one's ability to regurgitate (which it clearly does), then how is that a surrogate for one's success in dental school?

that's why there are many people with high DAT scores, but low gpas...because in university level classes, the tests often involve application of concepts...not merely regurtitation...

if anything, one's gpa (assuming they are from a respectable institution), will measure one's success in dental school more than the DAT would...because i'm sure written testing in undergrad and testing in dental school will remain similar...so one's undergrad gpa will reflect their written testing abilities....

i don't know about your university, but my university's exams are NOT EVEN CLOSE to a recall exam....it's all conceptual..
 
r0entgen said:
um...if this DAT is supposed to measure one's success in dental school, then it certainly doesn't do that. do you think dental school is pure regurgitation? do you think the career of dentistry itself is regurtiation? no, it's not. it requires application of your knowledge...and if the DAT measures one's ability to regurgitate (which it clearly does), then how is that a surrogate for one's success in dental school?

that's why there are many people with high DAT scores, but low gpas...because in university level classes, the tests often involve application of concepts...not merely regurtitation...

if anything, one's gpa (assuming they are from a respectable institution), will measure one's success in dental school more than the DAT would...because i'm sure written testing in undergrad and testing in dental school will remain similar...so one's undergrad gpa will reflect their written testing abilities....

i don't know about your university, but my university's exams are NOT EVEN CLOSE to a recall exam....it's all conceptual..



wow, so fiesty and defensive.

personal attacks on my opinions or university aside, i do agree w/ some of what youre saying. application of knowledge i think is much more useful than recollection abilities. but i never argued that dat's are direct gauges of university success. i said dat's (of course combined w/ gpa and other app parameters) are their best method of selecting dental students, given the position specific ad coms are in.
think of it this way, the dental gods billions of years ago i'm sure put a little more thought, analysis, and planning into this whole process than our obssessing selves have done these past few years. shortcomings included, this dat format is what they decided works best (balancing results with application processing time and ad com resources on their ends), so there should exist a lot of reasoning that we havent thought of.

your arguments make sense in an ideal world where dental schools have endless funds and an overflowing staff to handle app cycles.
 
Quit pussying around.

Just suck it up and do it.
 
netsn06 said:
wow, so fiesty and defensive.

personal attacks on my opinions or university aside, i do agree w/ some of what youre saying. application of knowledge i think is much more useful than recollection abilities. but i never argued that dat's are direct gauges of university success. i said dat's (of course combined w/ gpa and other app parameters) are their best method of selecting dental students, given the position specific ad coms are in.
think of it this way, the dental gods billions of years ago i'm sure put a little more thought, analysis, and planning into this whole process than our obssessing selves have done these past few years. shortcomings included, this dat format is what they decided works best (balancing results with application processing time and ad com resources on their ends), so there should exist a lot of reasoning that we havent thought of.

your arguments make sense in an ideal world where dental schools have endless funds and an overflowing staff to handle app cycles.

i'm not getting defensive...it's not like i made the DAT and people are attacking it...i'm just trying to state my point of view..

anyway, i don't see how this can work best..maybe if it were designed like the MCAT and required analytic skills, yes, it would work well...the point a lot of us are arguing is that using this DAT to assess students doesn't really give accurate results...

as for pikeyman, no one is pussying around...we're just here to state our opinions on the DAT...i've taken my DAT already..and i'm done with it so there's nothing for me to suck up.
 
According to many dental school ad coms, one's performance on DAT is is a direct predictive of their performance on board 1 exam. How they say this is beyond me, but if board 1 is as stupid as DAT, it should also be revamped.
 
Sorry for the long post, but my undergraduate degree is in psychology. I spent a lot of time in classes about statistics and about measures of human performance, and I have often thought about the problems of using tests like this for admission.

I see three problems.

In general, I think that standardized test scores are often misused (be it for the DAT, the SAT, or the MCAT).
A standardized test involves scaling raw scores to scaled scores so that they fit a standard distribution. In most-cases (including the DAT, SAT, and MCAT) the distribution is the standard, normal curve (or the "bell" curve).

Basically this means that the curve has a single mode (which is also the mean and the median) and the rest of the scores are symetrically distributed on either side of it. Unfortunately the very assumption that DAT scores should be distributed on the "bell" curve is probably invalid. It is likely that the population of people that take the DAT are more intelligent than average and that the correct distribution should be skewed to the right (the side where the higher scores would be). Therefore the first problem is that trying to force the right-skewed distribution onto the "bell" curve tends to exaggerate small differences on the right side and mask large differences on the left side.

The second problem is the fixation that people (including admissions committee members) have with scaled scores. The scaled scores are really meaningless outside of their distribution. What people should focus on is the percentile score. It gives much more useful information. The percentile shows how a test-taker ranks compared to others. This is really what the admissions people want to know anyway. To illustrate the problem, consider that a few points difference in scaled score does not always mean the same difference in percentile. For example, on some subsections there is no percentile difference between a 27 and a 30 but more than twenty percent difference between a 17 and a 20. Also a given scaled score may correspond to different percentiles for each subsection.

The final problem is measurement error. Which is statistospeak for the fact a test never precisely measures what it is supposed to; there is a little variation each time. This is the concept of lucking your way into a high score or jinxing your way into a low score. Theoretically, if a person took the DAT say thirty times, that person's scores would be distributed on a "bell" curve with the mean/center representing their actual potential. It would be most fair to administer the test multiple times and to use each person's average score. Unfortunately, each person would need to take many (probably at least 15) unique DAT's on multiple occasions and this would not be feasible (not to mention it would be brutal).
 
Amen to all that!
 
ou_jay said:
Sorry for the long post, but my undergraduate degree is in psychology. I spent a lot of time in classes about statistics and about measures of human performance, and I have often thought about the problems of using tests like this for admission.

I see three problems.

In general, I think that standardized test scores are often misused (be it for the DAT, the SAT, or the MCAT).
A standardized test involves scaling raw scores to scaled scores so that they fit a standard distribution. In most-cases (including the DAT, SAT, and MCAT) the distribution is the standard, normal curve (or the "bell" curve).

Basically this means that the curve has a single mode (which is also the mean and the median) and the rest of the scores are symetrically distributed on either side of it. Unfortunately the very assumption that DAT scores should be distributed on the "bell" curve is probably invalid. It is likely that the population of people that take the DAT are more intelligent than average and that the correct distribution should be skewed to the right (the side where the higher scores would be). Therefore the first problem is that trying to force the right-skewed distribution onto the "bell" curve tends to exaggerate small differences on the right side and mask large differences on the left side.

The second problem is the fixation that people (including admissions committee members) have with scaled scores. The scaled scores are really meaningless outside of their distribution. What people should focus on is the percentile score. It gives much more useful information. The percentile shows how a test-taker ranks compared to others. This is really what the admissions people want to know anyway. To illustrate the problem, consider that a few points difference in scaled score does not always mean the same difference in percentile. For example, on some subsections there is no percentile difference between a 27 and a 30 but more than twenty percent difference between a 17 and a 20. Also a given scaled score may correspond to different percentiles for each subsection.

The final problem is measurement error. Which is statistospeak for the fact a test never precisely measures what it is supposed to; there is a little variation each time. This is the concept of lucking your way into a high score or jinxing your way into a low score. Theoretically, if a person took the DAT say thirty times, that person's scores would be distributed on a "bell" curve with the mean/center representing their actual potential. It would be most fair to administer the test multiple times and to use each person's average score. Unfortunately, each person would need to take many (probably at least 15) unique DAT's on multiple occasions and this would not be feasible (not to mention it would be brutal).
Ok.
 
Haha, sorry but I just wanted to revive this thread b/c I am in a posting mood today and b/c I've never read this thread before tonight 🙂 Is it really true that most dental schools consider scores b/t 22-30 pretty much the same? I mean, do the dental school admissions committee really realize how few questions separate a 22/23 from a 25/26??? I know this sounds silly, but I often find myself thinking that only if I hadn't made a couple mistakes (1-3 problems) I could have ended up with a higher score... 😳
 
Top