For a standardized test... the DAT isn't very standardized, anyone else agree?

[DentalStudent9]

---

Members do not see this ad.

Anyone else agree?

 

[The Anhedonia]

what are you comparing it to, the SAT? give us a reference point to go from.

i've taken many standardized test before (including the Fundamentals of Engineering/EIT exam - an 8 hour multiple choice test), and they all share this common theme of being standardized. one doesn't necessarily stand out to me as "more standardized" than the others.

 

[Ryltar]

Yes and no. In order to be considered a standardized, the test must be administered and scored in a consistent manner. The scoring is consistent, but some argue that the way that it is administered varies too much because there are different versions of the test that are quite a bit harder than the others. For example, we are all familiar with the infamous FRAPs or Piano version of the DAT.

Your scores are calculated based on percentile comparison with everyone else who took the test recently, which is an average of ALL three versions of the DAT.

 

[The Anhedonia]

ah now i see what you meant.

 

[Ryltar]

The SAT seems to be graded differently. From the collegeboard.com, "Your raw score is then converted to a scaled score (reported on a 200-800 scale) by a statistical process called equating. Equating ensures that the different forms of the test or the level of ability of the students with whom you are tested do not affect your score. Equating makes it possible to make comparisons among test takers who take different editions of the test across different administrations."

The different versions of the DAT on the other hand can have a significant impact on the final score.

 

[DentalStudent9]

Yes, I was refering to the different versions. Looking through people's scores, I can almost pinpoint which versions or range of versions people took, although only among those in the higher ranges. The DAT grading doesn't seem to standardize your score against others who took the same version as yours, they seem to clump everyone together, even though there seem to be obvious difficulty differences between the versions. I've only taken it once, so this is not based on personal experience or any sort of fact, but by reading through other people's breakdowns, I can tell there are very obvous differences. On the other hand, it also seems that a version with a "harder" section is compensated with an "easier" section in another subject.


But then again... I never paid attention to other people's SAT and ACT breakdowns and maybe there were differen't versions that were harder or easier than others too. Maybe I just spend too much time on this forum...

 

[americanpierg]

Do schools see which version you took?

 

[doc3232]

No, it is standardized. If you do bad on the DAT, please don't blame the version...just blame yourself. I'm sorry, but that's just the way it is.

 

[lazybummy]

No, it is standardized. If you do bad on the DAT, please don't blame the version...just blame yourself. I'm sorry, but that's just the way it is.

+1. I always hate it when people complain about "omg i got a 13 in blah blah section cuz i got this particular version!" Really? look at the percentile.. if u're in the 20 something percentile... it's not the test.

 

[Ryltar]

Well yeah, if someone scores that low it's obviously their problem. I'm just saying that a different test version can mean the difference between a 17 and an 18-19, a 19 and a 20-21, etc.

 

[Tommy43087]

Well yeah, if someone scores that low it's obviously their problem. I'm just saying that a different test version can mean the difference between a 17 and an 18-19, a 19 and a 20-21, etc.

👍 agreed

some tests are "easier" because when you see a consistent trend of high scores on a certain test version(crocodile) or consistent low scores on another(piano, frap) it simply does state the fact that there is a difference and can be "unfair" in some ways...With that being said it is still really stupid to blame the test version for your low scores, because nothing in real life is completely "fair"

I had the piano version and i can say it was not that bad at all, do i think i woulda done better on the croc version?...maybe.But it doesnt matter, and if i studied longer and harder I wouldve done alot better on the test version i did have, so at the end of the day complaining about test versions is an even more stupid argument than the difficulty of recieving A's at certain schools...

At the end of the day if you work hard, it shows!

 

[ZzZzSleepyhead]

Uhhhhhhhh most of the scores on SDN aren't representative of DAT scores in general... If these scores are the ones you're looking at and using to make your claim, you're pretty wrong.

SDN seems to breed DAT killers and who knows what version ppl might get?

 

[americanpierg]

Hey guys, not to put anyone down or anything, but different test versions DO differ in difficulty, and in some versions the difference can be substantial.

For example, look up "microarray" or "microarrays" in the "DAT Discussions" forum with the search feature.

You will find that most people who took that version have 25+ RC scores and <18 QR scores. I know the sample size is small, like 8 people, but the similarity is definately strong enough to not be due solely to coincidence (considering 25+ in any section is so hard to get), and the scores are in such a small range (23-25). Now there ARE people who took this version and got 19s in both section, so it doesn't show that this version is an easy giveaway, but it does show that this version does seem to be easier in the RC department and harder in the QR than a lot of the other versions. (I could also be wrong and theres a lot more people who failed but didn't post their scores on this site)

This either means people who are good at reading comprehension are bad at math, or that exam had a very easy RC section and a very hard QR section. I guess every easier section is balanced with a harder section.

The real "fair" thing about this is that its totally random. Some people get lucky, some people don't. But everyone has an equal chance of getting one or the either (or one of however many versions there are) so I guess in a sense it IS fair.

 

[Atticus]

Hey guys, not to put anyone down or anything, but different test versions DO differ in difficulty, and in some versions the difference can be substantial.

For example, look up "microarray" in the "DAT Discussions" forum with the search feature.

You will find that most people who took that version have 25+ RC scores and

While I agree that tests might differ in difficulty, the way that they are curved causes standardization. I absolutely guarantee that the frap or piano dat versions are curved heavier than some other versions.

 

[Tommy43087]

While I agree that tests might differ in difficulty, the way that they are curved causes standardization. I absolutely guarantee that the frap or piano dat versions are curved heavier than some other versions.

??????

 

[THINKOFMYFUTURE]

...
You will find that most people who took that version have 25+ RC scores..

Objection! Where is the proof for this statement? Last time I checked, nobody could go back and see what questions there were on any real DAT 😱. I love it when people just assume that just because a score was low on a particular section this was inevitably due to that test being a 'tougher' version.

People, stop making excuses. Do you think the ADEA test makers were that stupid to overlook the cornerstone of the DAT - the standardized part?

The test is fair. If you have the ability it will show, so study smart! There is no section on it testing your ability to question its legitimacy!

 

[americanpierg]

Objection! Where is the proof for this statement? Last time I checked, nobody could go back and see what questions there were on any real DAT 😱. I love it when people just assume that just because a score was low on a particular section this was inevitably due to that test being a 'tougher' version.

People, stop making excuses. Do you think the ADEA test makers were that stupid to overlook the cornerstone of the DAT - the standardized part?

The test is fair. If you have the ability it will show, so study smart! There is no section on it testing your ability to question its legitimacy!

You must have gotten a bad RC score. I clearly said to use the search feature to see those scores. These are the results you get when you search for people taking the microarrays exam by searching "microarray" and "microarrays" with the search feature.

Bio ---------21 21 19 20 21 22 20 26 24 21 29
GChem------21 17 18 17 21 20 19 23 24 23 24
OChem -----22 21 22 20 30 22 21 23 23 20 23
PAT------- 18 20 19 19 18 20 21 21 20 21 21
RC----------25 24 18 26 19 23 22 24 22 25 21
QR--------- 18 13 19 13 20 14 16 24 17 16 15
TS--------- 22 20 20 19 23 22 20 24 24 22 26
AA ---------21 20 19 19 22 20 20 24 22 21 22

Now they're all obviously smart and worked hard for these scores, but the point of emphasis here are the QR and RC sections. More than half of these peolpe have RC as their highest subscore, and more than half of them also score LOWER than 50% of all test takers in QR, (17<). The AA all range within "normal" SDN range (19-22 ish), but notice the abnormally high numbers in the RC sections even for SDN (25, 24, 26, 23, 22, 24, 22, 25) and their corresponding QR numbers, respectively (18, 13, 13, 14, 16, 24 this guy has a 24 AA so hes a freak, 17, 16). Also notice the abnormally low PAT sections, no one got higher than 21 even with all these high scores. All of this could be coincidence, but it could also mean something else.

 

[Atticus]

While I agree that tests might differ in difficulty, the way that they are curved causes standardization. I absolutely guarantee that the frap or piano dat versions are curved heavier than some other versions.

??????

What is so difficult to understand? Regardless of the difficulty of the exam version, test taker's results still have to fit a standard Gaussian curve... So if the exam is ridiculously difficult, the majority of test takers will hypothetically score lower and thus less correct answers will yield a higher score.

 

[bhind]

You must have gotten a bad RC score. I clearly said to use the search feature to see those scores. These are the results you get when you search for people taking the microarrays exam by searching "microarray" and "microarrays" with the search feature.

Bio ---------21 21 19 20 21 22 20 26 24 21 29
GChem------21 17 18 17 21 20 19 23 24 23 24
OChem -----22 21 22 20 30 22 21 23 23 20 23
PAT------- 18 20 19 19 18 20 21 21 20 21 21
RC----------25 24 18 26 19 23 22 24 22 25 21
QR--------- 18 13 19 13 20 14 16 24 17 16 15
TS--------- 22 20 20 19 23 22 20 24 24 22 26
AA ---------21 20 19 19 22 20 20 24 22 21 22

Now they're all obviously smart and worked hard for these scores, but the point of emphasis here are the QR and RC sections. More than half of these peolpe have RC as their highest subscore, and more than half of them also score LOWER than 50% of all test takers in QR, (17<). The science sections and AA all range within "normal" SDN range (19-22 ish), but notice the abnormal high numbers in the RC sections even for SDN (25, 24, 26, 23, 22, 24, 22, 25) and their corresponding QR numbers, respectively (18, 13, 13, 14, 16, 24 this guy has a 24 AA so hes a freak, 17, 16). Also notice the abnormally low PAT sections, no one got higher than 21 even with all these high scores. All of this could be coincidence, but it could also mean something else.

I had that version of the test and got a 23 on the PAT but a 17 on the QR. I don't think this version of the test is any easier. I think the RC section of this test has a challenging passage in it while the other two are simply search and destroy. Maybe people just like the topics in these RC sections because they are science related and most students taking the DAT enjoy reading about that stuff. I can attest that sections of RC that I enjoyed to read I scored much higher on ( I took the exam twice and got floating bridges the first time and got an 18). This was even consistent for practice passages as well. Either way you look at it I think each exam is weighted fairly and that each numerical score you get is consistent with other versions of the exam. You cant take SDN scores and try to create trends.

 

[Tommy43087]

What is so difficult to understand? Regardless of the difficulty of the exam version, test taker's results still have to fit a standard Gaussian curve... So if the exam is ridiculously difficult, the majority of test takers will hypothetically score lower and thus less correct answers will yield a higher score.

First, you need to relax. The difficulty in understanding is the statement you are making. You are stating that they curve particular versions of the DAT? That is stupid, if i am interpreting you wrong my bad....but for right now you sound stupid...


or maybe im stupid and i just dont understand your statement, i dont know....

 

[bigstix808]

You must have gotten a bad RC score. I clearly said to use the search feature to see those scores. These are the results you get when you search for people taking the microarrays exam by searching "microarray" and "microarrays" with the search feature.

Bio ---------21 21 19 20 21 22 20 26 24 21 29
GChem------21 17 18 17 21 20 19 23 24 23 24
OChem -----22 21 22 20 30 22 21 23 23 20 23
PAT------- 18 20 19 19 18 20 21 21 20 21 21
RC----------25 24 18 26 19 23 22 24 22 25 21
QR--------- 18 13 19 13 20 14 16 24 17 16 15
TS--------- 22 20 20 19 23 22 20 24 24 22 26
AA ---------21 20 19 19 22 20 20 24 22 21 22

Now they're all obviously smart and worked hard for these scores, but the point of emphasis here are the QR and RC sections. More than half of these peolpe have RC as their highest subscore, and more than half of them also score LOWER than 50% of all test takers in QR, (17<). The AA all range within "normal" SDN range (19-22 ish), but notice the abnormally high numbers in the RC sections even for SDN (25, 24, 26, 23, 22, 24, 22, 25) and their corresponding QR numbers, respectively (18, 13, 13, 14, 16, 24 this guy has a 24 AA so hes a freak, 17, 16). Also notice the abnormally low PAT sections, no one got higher than 21 even with all these high scores. All of this could be coincidence, but it could also mean something else.

i don't remember learning about SDN range in statistics...SDN is a LOWW % of people taking the test. there is NO way you can generalize an entire population with 10-15 examples. and if you're gonna try and compare the small data set you actually have to see if there is a difference or a similarity in their data sets....try running a ANCOVA or ANOVA (if you only wanna use RC and QR) -- wait you couldn't even do that because these numbers aren't even "powerful" enough to do that. all you deduce from these numbers is correlation...which proves NOTHING!

 

[Atticus]

What is so difficult to understand? Regardless of the difficulty of the exam version, test taker's results still have to fit a standard Gaussian curve... So if the exam is ridiculously difficult, the majority of test takers will hypothetically score lower and thus less correct answers will yield a higher score.

First, you need to relax. The difficulty in understanding is the statement you are making. You are stating that they curve particular versions of the DAT? That is stupid, if i am interpreting you wrong my bad....but for right now you sound stupid...


or maybe im stupid and i just dont understand your statement, i dont know....

Rofl what is even the slightest bit agressive about my post that would justify requesting me to relax? Fifteen question marks with no other explanation in your post makes me think something was difficult to understand, and your post certainly didn't clarify what it was. Yes, each test version of the DAT is scaled slightly differently, just like many standardized tests ( like the MCAT)

 

[Tommy43087]

Rofl what is even the slightest bit agressive about my post that would justify requesting me to relax? Fifteen question marks with no other explanation in your post makes me think something was difficult to understand, and your post certainly didn't clarify what it was. Yes, each test version of the DAT is scaled slightly differently, just like many standardized tests ( like the MCAT)

Maybe i read it too fast and felt it was aggressive so RELAX, lol j/k...

But i really dont know where you are getting this nonsense from, its complete BS....the grading scale is the same for all versions of the test there is NO curve for particular versions....The MCAT is a completely different exam and is graded entirely different...

 

[americanpierg]

If a tree falls in the forest and no one hears it, does it really fall?

 

[UCB05]

You must have gotten a bad RC score. I clearly said to use the search feature to see those scores. These are the results you get when you search for people taking the microarrays exam by searching "microarray" and "microarrays" with the search feature.

Bio ---------21 21 19 20 21 22 20 26 24 21 29
GChem------21 17 18 17 21 20 19 23 24 23 24
OChem -----22 21 22 20 30 22 21 23 23 20 23
PAT------- 18 20 19 19 18 20 21 21 20 21 21
RC----------25 24 18 26 19 23 22 24 22 25 21
QR--------- 18 13 19 13 20 14 16 24 17 16 15
TS--------- 22 20 20 19 23 22 20 24 24 22 26
AA ---------21 20 19 19 22 20 20 24 22 21 22

Now they're all obviously smart and worked hard for these scores, but the point of emphasis here are the QR and RC sections. More than half of these peolpe have RC as their highest subscore, and more than half of them also score LOWER than 50% of all test takers in QR, (17<). The AA all range within "normal" SDN range (19-22 ish), but notice the abnormally high numbers in the RC sections even for SDN (25, 24, 26, 23, 22, 24, 22, 25) and their corresponding QR numbers, respectively (18, 13, 13, 14, 16, 24 this guy has a 24 AA so hes a freak, 17, 16). Also notice the abnormally low PAT sections, no one got higher than 21 even with all these high scores. All of this could be coincidence, but it could also mean something else.

Your data is strikingly convincing. by the numbers, all testtakers who got the microarray version can and will score over 20 on ochem and bio (with the exception of one who got a 19 bio so he's a freak) and total science except for one (a different freak). Notice the values of those abnormally high OChem scores. As a result, more than half of them scored 22 or higher on TS. The AA range is above normal compared to the national standard. Therefore I can only assume their exam is unfairly skewed to be extremely easy, and as a corrolary, there is a mass conspiracy to toy around with a large chunk of testtakers by feeding them outrageously high TS/AA scores but terrible QR sections. My outrage knows no bounds and we should all appeal to the ADA to immediately void all scores for this academic year and grant every testtaker a 23/23/23.

On second thought, I recant some of my argument. Instead, we should look at SDN scores. All the numbers reported show SDNers being in the top percentiles of the country. There is a startling trend in SDN presence and placement. Perhaps the ADA is doing a bad job at standardizing tests for this subset of people. I agree that the DAT is unfairly nonstandardized in favor of SDN members. The numbers shown here prove it. The only recourse is REVOLUTION. Remember remember the 3rd of October.