Edexcel S3 - Wednesday 25th May AM 2016

Original post by economicss

Thank you

Can I ask you which paper is that one? I never saw that question :smile:

Reply 882

username1599987

12

Original post by Nikhilm

Which paper?

2012, doesn't matter though because all of them are like that lol

Reply 883

ArafH

1

Guys this does not make any sense part A And B

Reply 884

username1162972

18

Original post by ArafH

Guys this does not make any sense part A And B

+-0.6mm so the modulus of the difference.
This is very known and comes up tonnes go over it in the book.

Reply 885

Music With Rocks

9

Is there ever a time in S3 when you would use the σ² rather than S² formula?

Reply 886

Armpits

Original post by Ayman!

How come we're noy allowed to use the sample mean in 6(a) but we have to use it in 6(b) on June 13R?

@Zacin

Not sure if it's already been answered.

But the distribution of xbar is N(pop mean i.e. e(xbar), pop var/n)

so you'd use the pop mean not the sample mean across 50 observations

Reply 887

Krollo

17

Does an S3 june 2015 R paper exist? Can't find it anywhere

Posted from TSR Mobile

Reply 888

Armpits

Original post by Music With Rocks

Is there ever a time in S3 when you would use the σ² rather than S² formula?

when it's appropriate

Reply 889

username1162972

18

Original post by Krollo

Does an S3 june 2015 R paper exist? Can't find it anywhere

Posted from TSR Mobile

na, 2015 onwards there re no R ppers due to IAL.

Reply 890

justalawyer1998

5

Original post by physicsmaths

na, 2015 onwards there re no R ppers due to IAL.

Do you know if OCR/ICCEA or whatever has a similar S3 paper to edexcel? Or any links to hard questions/papers made up by people? Thanks

Reply 891

Original post by Krollo

Does an S3 june 2015 R paper exist? Can't find it anywhere

Posted from TSR Mobile

Nah, regional papers were replaced by IAL papers from 2014 onwards (2014 being the transition year where both (R) and IAL papers were available).

Original post by gagafacea1

Capture dâ€™eÌcran 2016-05-24 aÌ€ 12.23.04.png

This was in the mark scheme of a chi squared test. Can someone why there is no followthrough?? It makes ZERO sense! Like what if I did a calculation mistake? What does my calculation mistake have to do with my ability to deduce/write a conclusion to a test? WHAT THE HELL IS WRONG WITH EXAM BOARDS??? Like this isn't the first time where they punish you fifteen million marks for small mistakes or mistakes that are unrelated.

Yeah, that is annoying. At the same time, I can see them having to do it because it's an accuracy mark. You do et a mark for making a "reject/accept H0" depending on your calculations regardless of whether it's correct but they can't award you the accuracy mark for something that's incorrect. But annoying and overly penalising anyways, I agree.

Original post by economicss

Thank you

No problem.

Original post by physicsmaths

cheers, q4d on the june 13r paper, what is this? it aint contingency **** so what is it?just seeing if it fits something? i got it right but i just guessed it the whole way lol

It technically is a contingency table or perhaps might be better viewed as a goodness of fit. It's just using the thing you're testing as a way to compute the expected values.

Original post by Geraer1

Part d of Question 3, Paper 2015,Could someone explain me the answer from the mark scheme:The variation within each stratum is quite small (o.e.) B1The difference in the means will be quite large, (so variations from theoverall mean will be large giving a larger overall s.e.)Is standard error similar to standard deviation?

It just means that the adults and childrens weight are vastly different from one another so if you put them in the same category then the variance is much larger because the the range is so huge. So since the variance is larger, then so is the standard deviation. But since

\text{s.e} = \frac{\sigma}{\sqrt{n}}

then the standard deviation being larger means that the standard error is larger as well.

And in the same paper, Q6 part (c) doesnt specify the significance level, so I guessed that I need to look at all the values of x^2 in the row n=3 and as every value is bigger than 6.188 leads to the conclusion that its not significant. I am not sure if this is right or not because in the mark scheme they dont specify anything. :smile:

Yes, that's correct.

Reply 892

username1210827

12

Was there an s3 paper in January?

Reply 893

Original post by Ayman!

How come we're noy allowed to use the sample mean in 6(a) but we have to use it in 6(b) on June 13R?

@Zacin

The CLT states that if your sample size is large enough then the sample mean is normally distributed with the mean of the population and the variance of the population divided by the sample size.

i.e: the sample mean may be 17.2, it doesn't mean the sample mean is distributed with mean 17.2.

In part (b), this is precisely what you are dong, you are working out the confidence interval for the mean.

Reply 894

Original post by the-anonymous-me

Was there an s3 paper in January?

No.

Reply 895

username1599987

12

Capture dâ€™eÌcran 2016-05-24 aÌ€ 14.07.00.png

In a question like this, should I calculate 'b' by adding all the expected frequencies and subtracting that from the total, which is 100 btw. Or should I use the formula, just like I would do 'a'. Cuz I remember doing a question where in the mark scheme they did the total minus other expectancies thing, but then in this questions mark scheme, they used the formula. So they got 'a' and 'b' the same way.

Reply 896

Original post by gagafacea1

In a question like this, should I calculate 'b' by adding all the expected frequencies and subtracting that from the total, which is 100 btw. Or should I use the formula, just like I would do 'a'. Cuz I remember doing a question where in the mark scheme they did the total minus other expectancies thing, but then in this questions mark scheme, they used the formula. So they got 'a' and 'b' the same way.

Generally, if it's not too much trouble, I would do it both ways to make sure my answer is correct. (they amount to the same thing, although a bit of care has to be taken w.r.t decimal places rounding) If it was an annoying calculation where the class was >= 6 and I'd need to look up cumulative crap in the tables, then I'd just add up the previous ones and do total - (sum of previous).

Reply 897

Rkai01

6

Could we actually be asked to calculate c in a pdf and those type questions?

I can't remember much of those s2 things

Reply 898

Geraer1