Competition - post a photo you've taken of nature >>

Edexcel S4 - Friday 24th June 2016 AM

Scroll to see replies

Well, just starting to recap this module alongside my FP2 preparations - In terms of difficulty the module isn't too bad but is anyone finding it a lot more boring than S3?

I mean, S3 was reasonably varied, this module seems to just be confidence intervals and hypothesis tests.

Yeah, I see what you mean. It's not very varied.

At least you get to draw a pretty graph in the exam sometimes. :teehee:

Reply 61

electrifeye

Could someone help me with June 2012 q2: http://puu.sh/phJ44/2a7ee2f093.png

For a), is there a reason you couldn't do an S3 test using the normal distribution, seeing as you are given that the populations are normal, meaning the small sample size doesn't matter as we don't need to use CLT?

In the mark scheme, they use a t-test, which I guess makes sense seeing as it's S4. The bit I don't understand though is why in b) i) the critical values in the show that are derived from the normal distribution (+/- 1.96). I expected this to be using a t-distribution, which meant critical values of +/- 2.12 which gives an incorrect answer. So why are they using the t-distribution in part a) but normal in b) i)? the same goes for b) ii), I used normal for this and got the right answer.

Reply 62

tazza ma razza

qn 6 june 11

can someone do a complete answer and method for this as i dont get the mark scheme at all. I would be much obliged

Reply 63

Euclidean

Does anyone know where the potential tricks are in this module?

Not studied it at all yet but I have a feeling I'm gonna need to do it from scratch very quickly so wanna know which parts to allocate most time to understanding :tongue:

Reply 64

username1162972

Original post by Euclidean

cmon mate, its stats. no understanding needed.

Reply 65

Kevin De Bruyne

Original post by electrifeye

In part a), the variance is not known so you can't use the normal distribution. In part b, the variance is given to be 0.9 so you can then use the normal.

Original post by tazza ma razza

qn 6 june 11

can someone do a complete answer and method for this as i dont get the mark scheme at all. I would be much obliged

Please link the paper if you can :tongue:

Reply 66

Kevin De Bruyne

Original post by Euclidean

Powers of tests are probably the most confusing thing (and power functions, if they ever appear). Apart from that.. nothing particularly special.

Reply 67

electrifeye

Original post by SeanFM

In part a), the variance is not known so you can't use the normal distribution. In part b, the variance is given to be 0.9 so you can then use the normal.

Thanks, that helps :smile:

Reply 68

tazza ma razza

anyone got links to the specimen s4 paper? Ii am about to do the mock paper (listed on maths and physics tutor) but they don't list the specimen. couldn't find it on the edexcel webste either

Reply 69

130398

I take it there are no notes for S4? I have found a sample but who is willing to pay for notes. Having looked at the 2014 paper (both normal and R variant) I have come to realize that the module isn't as easy as first expected (especially when compared to the earlier papers). Any tips on how to study for this? Is the book really that useful or should I just go for memorizing and half understanding the mark schemes?

Spoiler

Reply 70

tazza ma razza

for the difference tests using mu=D

how do we know which way to work out the difference? I always make mistakes on this qn

Reply 71

tazza ma razza

How do we go about evaluating a test and whether it is good or not?

Reply 72

Kevin De Bruyne

Original post by 130398

Spoiler

The textbook is enough - trust me on that one. :lol:

Reply 73

Kevin De Bruyne

Original post by tazza ma razza

for the difference tests using mu=D

how do we know which way to work out the difference? I always make mistakes on this qn

Can't say I remember those types of questions, but I imagine it'd be to do with the context and deducing what the hypothesis test should be.

Original post by tazza ma razza

How do we go about evaluating a test and whether it is good or not?

IIRC, it's testing whether it's unbiased or not as well as studying its variance. (maybe letting n tend to 0, I'm not sure).

Reply 74

tazza ma razza

Original post by SeanFM

Can't say I remember those types of questions, but I imagine it'd be to do with the context and deducing what the hypothesis test should be.

IIRC, it's testing whether it's unbiased or not as well as studying its variance. (maybe letting n tend to 0, I'm not sure).

1) chapter 3 stuff - i figured out how to do it; it basically comes with reading the qn carefully :s-smilie:

2) i meant by a part f style qn where it would come after doing a test comparison of different powers and sizes etc and then you have to figure which test is better / if it is a good test (chapter 1)

Reply 75

Kevin De Bruyne

Original post by tazza ma razza

1) chapter 3 stuff - i figured out how to do it; it basically comes with reading the qn carefully :s-smilie:

2) i meant by a part f style qn where it would come after doing a test comparison of different powers and sizes etc and then you have to figure which test is better / if it is a good test (chapter 1)

Yeah.. I think you look for the one(s) that are unbiased, first and formost, then whether or not the variance tends to 0 as n tends to infinity. If not.. it should be in your textbook or in the markscheme :tongue:

Reply 76

Med3

Are you guys Yr 12?