The Student Room Group

Is there any evidence of correlation between UKCAT scores and BMAT scores?

I know the tests are very different but in general does doing well in UKCAT mean you are Likely to do well at BMAT too. What is the evidence?

Scroll to see replies

Original post by rugose
I know the tests are very different but in general does doing well in UKCAT mean you are Likely to do well at BMAT too. What is the evidence?


I don't think so. A lot of people have done badly in the UKCAT but better in the BMAT or vice versa. Some people have done badly in both or really well in both. It depends on the person I guess.
Well there's not a research paper that I know off that shows this or looks into this, however speaking from experience, I did awfully in the UKCAT but much better in the BMAT
Reply 3
Original post by Pinkberry_y
Well there's not a research paper that I know off that shows this or looks into this, however speaking from experience, I did awfully in the UKCAT but much better in the BMAT


Hi

Thanks. What do you put this down to? it seems to me you can revise for bmat, but just practice UKCAT.
The first section of the BMAT was the one I did worst in, maybe I'll just never be good at reasoning type UKCAT/ section 1 of BMAT questions. But the second section of BMAT was science which I enjoy and am good at and the third section was the essay. I like writing essays and the topics they give you are always quite interesting. Whereas I don't find reasoning questions interesting and just zone out so end up doing Less well in it. Also I had a greater urgency to do well in BMAT because of my UKCAT score
Reply 5
Original post by rugose
I know the tests are very different but in general does doing well in UKCAT mean you are Likely to do well at BMAT too. What is the evidence?


I think the reason some people choose to d othe BMAT is becuase their UKCAT score was low! so in that sense their BMAT should be better! But this is jsut going by my own deduction.

And anyway, your UKCAT score only really correlates to your preperation and luck of the draw, really a useless exam in my opinion. So just going by the one factor you can control, if one is very organised and work hard for one, chances they will do the same for the other, so both should be relatively high (if t hey didn't flunk the ukcat first and then worked extra hard for BMAT)
This person seems to have done some analysis on the applicant profiles here on TSR. He's not very quantitative about his results though!
Reply 7
i have a friend at imperial who flopped the ukcat hard but aced the bmat so I personally wouldn't think there's a big link IMO.
Reply 8
Original post by nexttime
This person seems to have done some analysis on the applicant profiles here on TSR. He's not very quantitative about his results though!


you're tell me... that analysis was terrible! then again, probably 16-17 years old, that hasn't done any statistics. But even then I find his results dubious, since a large number only sat thebmat due to low ukcat score

Original post by 4rsenal
i have a friend at imperial who flopped the ukcat hard but aced the bmat so I personally wouldn't think there's a big link IMO.


From most of the undergrad medics I talked to that seemed to be the case. They didn't intend to do the BMAT if they aced the UKCAT. I didn't do the BMAT since they wanted UKCAT for grads. but GAMSAT is analogous BMAT, I said I wouldn't do the GAMSAT if i did well on the UKCAT, and thank god for that!
Reply 9
Hi

Someone in another thread posted this thread which is really useful.

http://www.thestudentroom.co.uk/showthread.php?t=865511

If I had time I'd put it in a spreadsheet but even just glancing through the entries you do get a good sense of UKCAT to BMAT correlation. I think I might work on the Spreadsheet during tonights olympics!
Reply 10
Original post by rugose
Hi

Someone in another thread posted this thread which is really useful.

http://www.thestudentroom.co.uk/showthread.php?t=865511

If I had time I'd put it in a spreadsheet but even just glancing through the entries you do get a good sense of UKCAT to BMAT correlation. I think I might work on the Spreadsheet during tonights olympics!


Would be an interesting analysis, quite a lot of data in that thread.

At the very least I think it needs to be adjusted for year of entry and A-Level grades though.
Reply 11
image.jpg
Original post by Beska
Would be an interesting analysis, quite a lot of data in that thread.

At the very least I think it needs to be adjusted for year of entry and A-Level grades though.


I'm on it. Entered 70 candidates so far. Should be finished in a couple of hours! Current headings are entry number on the thread, ukcat score, bmat sections a b c. (I am using the oxford system of weighting the sections), % a*. Will start to label year of entry too.

Currently the equation for the line of best fit is

Bmat score= 0.0773 x ukcat mean- 0.03. Rsquared is 0.28.

so there is a trend but really scattered to give everone hope ! Or worry:smile:

Oxford weighting of bmat is i believe 4x (section A +B) + section C. Remember to convert section c letter into a number eg 3a = 8. A=5.
(edited 7 years ago)
Reply 12
Original post by rugose
image.jpg

I'm on it. Entered 70 candidates so far. Should be finished in a couple of hours! Current headings are entry number on the thread, ukcat score, bmat sections a b c. (I am using the oxford system of weighting the sections), % a*. Will start to label year of entry too.

Currently the equation for the line of best fit is

Bmat score= 0.0773 x ukcat mean- 0.03. Rsquared is 0.28.

so there is a trend but really scattered to give everone hope ! Or worry:smile:

Oxford weighting of bmat is i believe 4x (section A +B) + section C. Remember to convert section c letter into a number eg 3a = 8. A=5.


That's interesting.

Is that regression you've done overall UKAT score versus this Oxford weighting of BMAT (to clarify)?

There's a fair bit of scatter but there seems to be a trend. I think it'll likely smooth out with more data points.

I don't think Excel can give any tests of significance, can it? That's the next (and most interesting) step....
Reply 13
Original post by Beska
That's interesting.

Is that regression you've done overall UKAT score versus this Oxford weighting of BMAT (to clarify)?

yes

Original post by Beska
There's a fair bit of scatter but there seems to be a trend. I think it'll likely smooth out with more data points.

plus it is zoomed in a bit. I have about 50 more data points to input.

Original post by Beska
I don't think Excel can give any tests of significance, can it? That's the next (and most interesting) step....


Beyond my skills set I'm afraid.

It's been useful exercise anyway - especially the advice from students. Same consistent messages coming through
Reply 14
Original post by rugose
yes


plus it is zoomed in a bit. I have about 50 more data points to input.



Beyond my skills set I'm afraid.

It's been useful exercise anyway - especially the advice from students. Same consistent messages coming through


Once you've collected it all I'm happy to run some multiple regressions and I can send you the results to do with whatever you want. Think it will be interesting...
Reply 15
Original post by Beska
Once you've collected it all I'm happy to run some multiple regressions and I can send you the results to do with whatever you want. Think it will be interesting...


Will do. Getting quite fatigued with it now so won't be with you for a couple of days.
Reply 16
Disaster! Laptop just turned off. Lost most of the input from today and last night. Schoolboy error.

Only got 11 data points saved. I got up to entry 630 - Amelia x

What I'll do is just enter data from this point onwards - It would wear me down to go over those points again.

I'll send you the spreadsheet once I've got to the end.
Original post by rugose
I know the tests are very different but in general does doing well in UKCAT mean you are Likely to do well at BMAT too. What is the evidence?


I don't think it's been studied, but I presume there would be some sort of correlation, but probably not a very strong one.

EDIT: Lol, just read more of the posts, it appears there are now much better-informed posts than mine. :p:
(edited 7 years ago)
Reply 18
Here is my spreadsheet of UKCAT v BMAT. It was bigger but lost 80 data points dammit - however all pre 2014 so not as relevant.

It looks to me like a weak correlation but I'd be interested to hear other opinions.

EDIT - thanks to nextitme. There is a consistent error in my calculation of the bmat score. Will update later today
(edited 7 years ago)
Original post by rugose
Here is my spreadsheet of UKCAT v BMAT. It was bigger but lost 80 data points dammit - however all pre 2014 so not as relevant.

It looks to me like a weak correlation but I'd be interested to hear other opinions.


There's something wrong with how you calculated the BMAT scores. They're lower than they should be if you're wanting to use the Oxford system.

If its a consistent error it might not matter, but if its not or it fails to weigh the sections correctly it might make a difference.

Edit: This formula (for row 35) gives the right answer: =((E35+F35-2)*5+(G35*2+H35)*4/3)
(edited 7 years ago)

Quick Reply

Latest

Trending

Trending