OCR MEI - S1 - 20th May 2015

http://www.mei.org.uk/files/papers/2011_Jan_s1.pdf

Can someone help me with 7v on this, i dont know how the variation changes with the addition of the equation?

yeah by timing something it changes the variation (makes the s.d. more varied) only adding makes no difference

M



I dont understand the mark scheme it says you need to square the sd and the 1.15? why?

times the standard deviation by 1.15,
then square this to get the correct variance
try to memorise that

does it have any reference to ex^2-ex? = var(x)

As someone above mentioned how do people approach sig figs with this? I usually do 4 for any binomial calculations and then 3 for the rest. But in marking a past paper just now there seems to be no consistency in how many the mark scheme uses.

I struggled with two tailed just two days ago but I hacked it haha
Okay so firstly, you need to know hypothesi testing for significance levels & critical region
One tail involes (lets say prob is 0.5)
Ho : p = 0.5
H1 : p > 0.5 OR p < 0.5 (depending on ques, as you probs know)
and then you either do sig level version:
In the ques you get x value (say its x = 14)
if p > 0.5 then P(X>/14)
if p < 0.5 then P(X</14) that means less than equal to btw haha
Now whichever it is, you compare to sig level (has to be below it)

or do critical region version:
if p > 0.5 then compare to high tail of critical region (trial & error)
if p < 0.5 then compare to low tail of critical region (both have to be below it)

For two tail testing, it involves:
Ho : p = 0.5
H1 : p DOESNT EQUAL 0.5
and then you do same thing but for sig level version:
-you HALF sig level
e.g. so say x = 12 and its two tail and sig level is 5% (0.05)
it'll be P(X</12) compared to 0.025

for CR version:
-you HALF sig level
-you do for both p > 0.5 AND p < 0.5
e.g. say sig level is 5% (0.05) and X~B(20, 0.5)
-check on tables for biggest n value for where P(X</ something) is less than 0.025
-check on tables for biggest n value for where P(X>/ something) is less than 0.025

Hope that made sense, if not, send a question & I'll happily do a question for you explaining all
Good luck

My dad works in the printing department for OCR, he told me there's no point revising two tailed tests...

Hello,

Can anyone explain how to do Q5 ii, from the Jan 2013 exam?

Do the chance of him not throwing a 6 10 times and take this from 1:
1-(5/6) to the power 10 gives 1-0.1615 so the answer is 0.8385.


Posted from TSR Mobile

Everyone ready?

Born ready.

Oh, apart from the combinations and permutations. Some of those are really difficult to get your head around

My dad works in the printing department for OCR, he told me there's no point revising two tailed tests...