2017 BMAT analysis v 2016 BMAT. Good news I think Watch

rugose
Badges: 8
Rep:
?
#1
Report Thread starter 2 years ago
#1
I posted this on the Oxford Medicine thread. Its findings are relevant to all BMAT students.

Essentially I feel there is evidence that this year has been a very tough BMAT and that scores that would in previous years not entertain interviews / offers, will this year.

Evidence
1. Posted bmat scores on TSR this year are lower than last year.
2. UCL have released the average scores of applicants for this year as 4.4, 4.5, 3.1A. Last year it was 4.7, 4.8, 3.1A, and the year previously 4.8, 4.9, 3. So a significant drop in scores.
If one studies the summary of results on the BMAT website http://www.admissionstestingservice....g-and-results/
3. last year if you got 5.4 for section 1 you were just in the top 26% of bmat scores. This year you are in the top 10%.
4. last year if you got above 6 for section 2 you were in the top 10%. This year this puts you in the top 8%.

Caveat - I am not a statistician! I could do with someone checking this.

The way I derived the figures was by measuring (yes with a ruler!) the various bars on the bar graphs.

Have fun!
Attached files
2
reply
Kmalamp
Badges: 5
Rep:
?
#2
Report 2 years ago
#2
I don't understand how a score of 5.4 in s1 would be in the top 10% this year?
0
reply
rugose
Badges: 8
Rep:
?
#3
Report Thread starter 2 years ago
#3
(Original post by Kmalamp)
I don't understand how a score of 5.4 in s1 would be in the top 10% this year?
Measure the length of each of the bars. The total length of those above 5.5 is less than 10% of all the bars.
http://www.admissionstestingservice....of-results.pdf
0
reply
rugose
Badges: 8
Rep:
?
#4
Report Thread starter 2 years ago
#4
(Original post by Kmalamp)
It shows that around 18% of applicants scored between 5-5.5 and 12% between 5.5-6.0. So I still don't get how scoring 5.4 would put you in the top 10% of applicants overall.
5.5 is the top value . So 6 means 5.6 - 6.. count 6 and above and extrapolate a bit from from 5.5 and you get 10%. The graph on the spreadsheet shows it best.
0
reply
nexttime
Badges: 22
Rep:
?
#5
Report 2 years ago
#5
(Original post by rugose)
I posted this on the Oxford Medicine thread. Its findings are relevant to all BMAT students.
Well its not really relevant to oxbridge students is it - interview invites are either already released or about to be released in just 2 days time! It couldn't possibly make any difference! :p:

Possibly some relevance to other BMAT unis where being able to predict interview invites is a little useful, but given the uncertainty in your method (manually measuring) and original data (wide result ranges), and given the relatively small differences, it probably has very few applications.

But if you like crunching numbers for the sake of crunching numbers then its beautiful well done. I'm jealous of how much free time you have :p:
0
reply
rugose
Badges: 8
Rep:
?
#6
Report Thread starter 2 years ago
#6
(Original post by nexttime)
Well its not really relevant to oxbridge students is it - interview invites are either already released or about to be released in just 2 days time! It couldn't possibly make any difference! :p:

Possibly some relevance to other BMAT unis where being able to predict interview invites is a little useful, but given the uncertainty in your method (manually measuring) and original data (wide result ranges), and given the relatively small differences, it probably has very few applications.

But if you like crunching numbers for the sake of crunching numbers then its beautiful well done. I'm jealous of how much free time you have :p:
Just imagine hiw much time we would both have if I didn’t do numbers and you weren’t on here!😀
0
reply
Arroyo
Badges: 3
Rep:
?
#7
Report 2 years ago
#7
(Original post by rugose)
5.5 is the top value . So 6 means 5.6 - 6.. count 6 and above and extrapolate a bit from from 5.5 and you get 10%. The graph on the spreadsheet shows it best.
I was wondering, and maybe this is a stupid question, but where did you find that 6 represented the top value for that particular bar, rather than the bottom value (representing 6-6.5)? Back when I used to make histograms for statistics class, I put the bars in between the values so it was easy to find proportional intervals, but I'm clueless on how to read these.
0
reply
rugose
Badges: 8
Rep:
?
#8
Report Thread starter 2 years ago
#8
(Original post by Arroyo)
I was wondering, and maybe this is a stupid question, but where did you find that 6 represented the top value for that particular bar, rather than the bottom value (representing 6-6.5)? Back when I used to make histograms for statistics class, I put the bars in between the values so it was easy to find proportional intervals, but I'm clueless on how to read these.
Hi.
No such thing as a stupid question.
I was of the same opinion as you until i read on another thread about the number representing the top. When you examine the bars you can see that in previous years each score was given its own bar. This year they have but into broader bars. The highest possible score is a 9. Therefore in this case 9 must represent up to 9. It cannot include scores of 9.4 or 9.2 as you cannot get those scores.
0
reply
Arroyo
Badges: 3
Rep:
?
#9
Report 2 years ago
#9
(Original post by rugose)
Hi.
No such thing as a stupid question.
I was of the same opinion as you until i read on another thread about the number representing the top. When you examine the bars you can see that in previous years each score was given its own bar. This year they have but into broader bars. The highest possible score is a 9. Therefore in this case 9 must represent up to 9. It cannot include scores of 9.4 or 9.2 as you cannot get those scores.
I was considering this too, but wouldn't the same logic apply to the other argument, as a 1 is the lowest score so one couldn't get scores of 0.5-1 as would be the case with the number being at the top of the interval? It seems a little more likely to me that ~3% of test takers would get a score of 1-1.5 in s1 and next to nobody with exactly 9 than 3% with exactly a score of 1 and next to nobody with a score of 8.5-9
Maybe that's just my inherent pessimism though xD
0
reply
rugose
Badges: 8
Rep:
?
#10
Report Thread starter 2 years ago
#10
(Original post by Arroyo)
I was considering this too, but wouldn't the same logic apply to the other argument, as a 1 is the lowest score so one couldn't get scores of 0.5-1 as would be the case with the number being at the top of the interval? It seems a little more likely to me that ~3% of test takers would get a score of 1-1.5 in s1 and next to nobody with exactly 9 than 3% with exactly a score of 1 and next to nobody with a score of 8.5-9
Maybe that's just my inherent pessimism though xD
True that you can apply the same logic at 1 end. I think it is explained by reading the charts of previous years where each possible score had its own bar. Also I think it more likely that more people will get a 9 than a 1. As supported by the 2016 results btw. Also consider for example in 2016 s1 that to get a bmat score of 1 you had to get between 0 & 3 marks but you get a score of 9 for 31-35 marks. You would bave to have a touch of genius (and masochism) to select answers deliberately avoiding all the correct answers😄
Good luck on Tuesday (if applying to oxford) though pessimists tend to have a more accurate view of the world lets us hope for the best ! 😀
0
reply
X

Quick Reply

Attached files
Write a reply...
Reply
new posts
Back
to top
Latest
My Feed

See more of what you like on
The Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

Personalise

University open days

  • University of East Anglia
    PGCE Open day Postgraduate
    Tue, 3 Mar '20
  • University of Bradford
    Postgraduate Open day/Evening Postgraduate
    Tue, 3 Mar '20
  • Queen's University Belfast
    Postgraduate LIVE Masters & PhD Study Fair Postgraduate
    Wed, 4 Mar '20

Do you get study leave?

Yes- I like it (523)
60.05%
Yes- I don't like it (44)
5.05%
No- I want it (245)
28.13%
No- I don't want it (59)
6.77%

Watched Threads

View All