Pecentile differences...

[jmh018]

---

Members don't see this ad.

I have been seeing people put up pictures of their score reports and the precentiles are different for the same scores. Someone got a 20 in Bio, same as me and theirs was a 92.8 percentile and mine says 86.5 percentile. I saw another person who got a 21 AA, same as me, and his percentile showed like 95.6 and mine says 91.3.

Anyone know why that is?

 

[FeralisExtremum]

These tests are based on a curve, and performance always varies between exams - more people likely achieved higher scores on your exam than the other person's, so despite getting a similar number of questions right/wrong, your performance relative to others that took the same test differs.

 

[412smb412]

I have been seeing people put up pictures of their score reports and the precentiles are different for the same scores. Someone got a 20 in Bio, same as me and theirs was a 92.8 percentile and mine says 86.5 percentile. I saw another person who got a 21 AA, same as me, and his percentile showed like 95.6 and mine says 91.3.

Anyone know why that is?

basically i big part of doing well in the dat is luck. u can prepare all u want but it depends who else has prepared for the exam and which version of the exam u get. sad to say, but thats what every1 including u and me have to deal with...

 

[scsc7211]

basically i big part of doing well in the dat is luck. u can prepare all u want but it depends who else has prepared for the exam and which version of the exam u get. sad to say, but thats what every1 including u and me have to deal with...

According to the ADA DAT User's Manual:
"Each of the tests used on the Dental Admission Test battery yields a raw score which is the sum of the examinee's correct answers. The raw score is converted to a standard score so that it is possible to compare an examinee's performance across tests on the battery and across different editions. Since the adoption of the Rasch psychometric model in 1988, each part of the DAT contains a set of anchor items, which has been used in previous administrations of the test. Difficulty parameters of these items are used to equate the test."

Considering the above, it doesn't necessarily matter who else has prepared for the exam that day (or how well they have prepared), because the difficulty parameters of the questions that will be used to compute the standard score have already been determined based on previous administrations of the test. The only measure that will be affected by the performance of other examinees on the same day is your percentile rank, which is not reported to AADSAS or schools.

In other words, let's say 100 people (assuming they are all gunner SDNers) across the country take the "version A" DAT on a particular day. If each of them prepares equally well (they're SDNers, so of course they have), then there is nothing stopping every one of them from scoring 21+. The percentiles will be significantly lower than what would be typical for a 21, but all that is reported at the end of the day is a 21, which was equally earned by each test taker.

 

[412smb412]

According to the ADA DAT User's Manual:
"Each of the tests used on the Dental Admission Test battery yields a raw score which is the sum of the examinee's correct answers. The raw score is converted to a standard score so that it is possible to compare an examinee's performance across tests on the battery and across different editions. Since the adoption of the Rasch psychometric model in 1988, each part of the DAT contains a set of anchor items, which has been used in previous administrations of the test. Difficulty parameters of these items are used to equate the test."

Considering the above, it doesn't necessarily matter who else has prepared for the exam that day (or how well they have prepared), because the difficulty parameters of the questions that will be used to compute the standard score have already been determined based on previous administrations of the test. The only measure that will be affected by the performance of other examinees on the same day is your percentile rank, which is not reported to AADSAS or schools.

In other words, let's say 100 people (assuming they are all gunner SDNers) across the country take the "version A" DAT on a particular day. If each of them prepares equally well (they're SDNers, so of course they have), then there is nothing stopping every one of them from scoring 21+. The percentiles will be significantly lower than what would be typical for a 21, but all that is reported at the end of the day is a 21, which was equally earned by each test taker.

if u take the dat and every1 does horrible then the score to get a 21 will be different than som1 who takes the dat and has a an avg difficulty, no 2 tests have the same difficulty and it depends on the distribution and number of ppl who achieved each percentile to actually come up with a scale of 1-30.

i see what ur saying but here's the thing: the only way to truly standardize is to have ONE TEST AT ONE TIME FOR EVERY1! until then its not standardized. my point is: somedays its easier to get a 21 than other days, if u wrote all the tests from the dat u would get different scores each time, this is variability and this should not exist. the thing is they don't present the raw scores so they may say one thing and do the other which im sure happens a lot. all in all i've come to the reality that the only thing u can do is pray, and study and make sure on d-day that there was nothing else u could of done.

 

[Dgeorg6]

The percentage doesn't matter so who cares? a 21 is a 21 no matter if it was 91% or 97%