OKAPS are finally done

[ckyuen]

Whew! That is one tough exam even the second time around. Boy am I glad I didn't use the BCS series for optics and used Guyton instead.

---

Members don't see this ad.

 

[Andrew_Doan]

Whew! That is one tough exam even the second time around. Boy am I glad I didn't use the BCS series for optics and used Guyton instead.

I hope it went well for you.

I am so glad it's over too!!! They recycled some of the same questions from last year. I got stumped on reflective power of mirrors last year, but not this time!
I'll have to use Guytons next year. We have a low vision specialist (Dr. Wilkinson) who prepares at 40 page review of optics. His notes and lectures are excellent too. I felt much better about optics this year.

The problem with the OKAP is that they have to make the test hard, otherwise, all of us would ace it. So many questions are just a pathology slide or one sentence, and then you have to make a diagnosis; however, instead of asking what is the diagnosis, it'll be picking out one of the classic features associated with the disease.

Because I can't regurgitate questions, this article may be of interest to you!

http://www.ncbi.nlm.nih.gov/entrez/...ve&db=pubmed&dopt=Abstract&list_uids=12542994

If this is common knowledge to you... then I bow to your greatness!

BTW, are you going to the advocacy meeting in D.C. next week or ARVO? I'll be in D.C. but not ARVO.

 

[ckyuen]

Well that's another one I got wrong. Thanks for the reference. Who would've thought. I bet about one in four got that one right. I highly recommend using guyton next year. There are very few questions that appear on the exam that you would not get by going through guyton. maybe one or two. the example problems are great. first half of it is more applicable to the written okaps, second half to orals, but the second half also discusses some important principles to know also.

 

[Andrew_Doan]

Well that's another one I got wrong. Thanks for the reference. Who would've thought. I bet about one in four got that one right. I highly recommend using guyton next year. There are very few questions that appear on the exam that you would not get by going through guyton. maybe one or two. the example problems are great. first half of it is more applicable to the written okaps, second half to orals, but the second half also discusses some important principles to know also.

Anyone interested in writing up a question database for the OKAPs? We can offer it for free for future residents. Clearly, we can't copy the questions, but I think collectively, we could write a pretty good database of questions.

 

[ckyuen]

That's a great idea as long as we don't copy questions on the test I can definitely help out with optics, I must have done every type of mathematical optics question in preparation.

 

[ckyuen]

I have a theory on how they write okap questions, everyone will probably laugh. My theory is that tables and pictures especially color photos take a lot of time and are expensive to print. Also if I were writing test questions I would flip through BCS and ask about those b/c it's faster than reading the entire book again and much easier to ask a question about a picture. I think for embryology the only thing I looked at was the table in the comprehensive oph book and I found that was sufficient to answer their questions. I dont think it was necessary to read about a hundred pages on embryology. However I don't have my score yet, and I may have failed miserably.

 

[ckyuen]

Andrew,
I asked my vicechairman "the green bean" question and he gave the correct answer in about 2 seconds along with a detailed explanation why. I was blown away.

 

[Andrew_Doan]

I've met your vice chairman. He's a smart man.

Now we'll know how to answer the latex-fruit allergy question!

 

[Andrew_Doan]

Does anyone now the correlation between the raw score and scaled score?

There are 250 questions on the test. Two-thirds of residents will fall within 400-600 on the scaled score, and the average is 500 for all test takers.

I've estimated that one needs to get ~80% correct to be at the mean for all test takers. This was estimated by looking at the scaled score, percentage score, and question item feedback from several test takers.

For first year residents, I think ~65-70% (162-175/250) correct is the mean. For second year residents, I think ~75-80% correct (187-200/250) is the mean. I think ~80-85% (200-213/250) correct is the mean for third years and beyond.

 

[vwhan]

Now we'll know how to answer the latex-fruit allergy question!

My girlfriend, 2nd yr peds resident, diagnosed this in one of her kids about a month ago (with the help of a 3rd yr resident). Who would've thought that diagnosis would pop up again?

 

[GlaucomaMD]

Andrew,

I believe you'll find that the % of correct response needed to achieve the mean score is far below what you have calculated.

You also have to remember that there are some "pre-test" questions in the exam - questions that are asked to see if they are good questions to use in the future exams - I believe that is why some questions are asked several times in the same exam in a very slightly different way. The examiners are trying to look for questions that are good "discriminators."

Just my 2 cents...

 

[Andrew_Doan]

Andrew,

I believe you'll find that the % of correct response needed to achieve the mean score is far below what you have calculated.

You also have to remember that there are some "pre-test" questions in the exam - questions that are asked to see if they are good questions to use in the future exams - I believe that is why some questions are asked several times in the same exam in a very slightly different way. The examiners are trying to look for questions that are good "discriminators."

Just my 2 cents...

Good point. I forgot about the "pre-test" questions. Because we don't know how many are in the test, then it's difficult to calculate. Do you have any idea how the raw scores correlates?

Let's assume that 200 are scored questions.

First years: ~65-70% (130-140) correct is the mean.
Second years: ~75-80% correct (150-160) is the mean.
Third years: ~80-85% (160-170) is the mean.

 

[Eyesore]

Andrew,

When they send you the scores, they also send you a sheet with an itemized breakdown of what topics you need to work on. If I'm not mistaken, each one of those bullets is a question that you missed. So, you can pretty much figure out the number of questions you answered correctly. Correlate that with your raw and scaled score, then you have one data point. If your buddies tell you what they got, then now you have more info to work with. However, if none of you Iowa guys/gals get in the 50th percentile, then you still don't know how many questions need to be answered correctly to get the mean per class year. That is the price you pay for being a bunch of geniuses. But as you know, the percent correct correlated with each raw/scaled score will vary from year to year based on the difficulty of the exam.

What GlaucomaMD I believe was saying, is that you are overestimating the number of questions we need to get right in order to achieve the mean. To use your breakdown examples, you estimate that for second years, you need to answer 75-80% correct to achieve the mean. On last year's exam, if you answered 75% correctly, that would put you in the mid 80s percentile-wise.

Just my 2 cents....

 

[Andrew_Doan]

One caveat: the item list breakdown consists of questions representing one or several "concepts". This is stated in the OKAP manual.

For instance, one resident scored 50% with 70 items on the list. Several of these concepts are represented by 1 question.

The question is: how many of these concepts are linked to only one question? For instance, if 70 concepts were listed on the breakdown list and this represented 50 questions, then the percent correct is 150/200 = 75% for 200 scored questions. If there were 220 scored questions, then it's 77%.

If all 70 concepts correlated to 70 questions, then it's 65% for 200 scored question. However, I think many questions represented 2 concepts in the breakdown list. For example, an endophthalmitis after cataract surgery question represents concepts under Lens and Uveitis. Just my OCD getting the best of me!

 

[Eyesore]

It does say on the score report that one question may be represented in different sections with associated keywords. However, I do remember last year that for pretty much every (or almost every) keyword that I had listed, I could remember the question that was asked. I don't remember any of them overlapping. And if you look at the keywords, they are fairly broad. To use your example again, for endophthalmitis, it would be under endophthalmitis or post-surgical endophthalmitis for the lens and cataract section as well as the uveitis section. But look at your breakdown for keywords, how many of them actually are repeated. For me, there are no repeats. Just my 2 cents again...

 

[Andrew_Doan]

How many of them actually are repeated. For me, there are no repeats.

There are several repeats, but I think a question can have 2 concepts that are being tested. For instance, a path slide with pseudoexfoliation can be listed as lens disorders under pathology and pseudoexfoliation under lens/cataracts.

 

[JL_Nightwing]

just wondering how this year's OKAP Qs compared with last year's

 

[Andrew_Doan]

just wondering how this year's OKAP Qs compared with last year's

I always think the OKAPs are difficult. However, I didn't study this year. Now I know what I need to study for the real boards next year.

 

[dyk33]

what is guyton's optics? is this a book?

 

[ckyuen]

yes it's a spiral bound book cost about 18.00
this was a few years ago

David Guyton's book is called Ophthalmic Optics and Clinical Refraction. He is developing the course on a CD and should be available next year. I just spent 16 hours being lectured by him last week...he is a phenomenal teacher! His book is $17.50 (not sure how much now) sent to:
David L. Guyton, M.D.
Wilmer Institute, 233
Johns Hopkins Hospital
Baltimore, MD 21287-9028