This discussion is now closed.

Check out other Related discussions

AQA Biol A2 EMPA Megathread

Just wanted to ask, what sort of conclusion would I write, if my confidence limits overlap. They did in a practice emap, june 2010 one, and it said use words like chance and probability, but I didnt have a clue what to write. An exact answer would be appreciated

If they overlapped:
There is no significant difference between the means;
The probability of results being due to chance is greater than 5%;
Accept the null hypothesis.

(you can't really accept a null hypothesis, but it's what the mark scheme wants, so ho hum)

Reply 101

thegodofgod

19

Original post by paul272

Just wanted to ask, what sort of conclusion would I write, if my confidence limits overlap. They did in a practice emap, june 2010 one, and it said use words like chance and probability, but I didnt have a clue what to write. An exact answer would be appreciated

I would say something along the lines of:

"Since the 95% confidence limits overlap, there is more than a 5% probability that the results are due to chance. Therefore, I accept the null hypothesis."

Reply 102

Revd. Mike

17

Personally I'd say 'cannot reject the null hypothesis' rather than accept it, but I'm not sure what the exam spec wants.

Reply 103

thegodofgod

19

Original post by Revd. Mike

Personally I'd say 'cannot reject the null hypothesis' rather than accept it, but I'm not sure what the exam spec wants.

The marking guidance is always:

'Correct interpretation of statistical test in terms of acceptance and rejection of null hypothesis;

'Interpretation involves appropriate reference to the probability of the results being due to chance;'

And the 'Comments' box says:

'Use candidate’s values even if calculation incorrect.

'Correct critical value needed for 1st mark.'

Reply 104

Revd. Mike

17

Original post by thegodofgod

The marking guidance is always:

'Correct interpretation of statistical test in terms of acceptance and rejection of null hypothesis;'

'Acceptance' of H0 isn't something you actually do in real science though. Take, for example, one of the projects I'm contributing to looking at the efficacy of gene therapy or stem cell transplants in the treatment of NCL (a childhood neurodegenerative disorder). For a given experiment, the H0 might be "There's no significant difference between gene therapy treated or vehicle treated knock-out mice". So you treat the animals, remove the brains, section them and stain them to look at a particular disease marker, for example, cortical thickness. You notice when you crunch the numbers that although there might be a trend, it doesn't reach statistical significance. At this point you wouldn't say "Well, I have to accept the H0, there's no difference, clearly gene therapy doesn't work". You'd say "I can't reject the H0, further investigation is required - maybe with a larger sample size"

Obviously, stick with what the markscheme says for the purposes of your exams and coursework. Just be aware that it's a simplification :smile:

Reply 105

Kidneyjean

7

Original post by Revd. Mike

Personally I'd say 'cannot reject the null hypothesis' rather than accept it, but I'm not sure what the exam spec wants.

Yeah, this really annoys me too :tongue:

But the AQA mark scheme says "accept the null hypothesis" every year. My bio teacher doesn't seem to get why that's not correct :s-smilie:

Reply 106

maxpark3r

2

So if confidence limits don't overlap:
There is a significant difference between the means and therefore the probability of results being due to chance is less than 5% and reject null hypothesis?

Reply 107

raveen789

9

Anyone else find task 3 a bit ambiguous?

Reply 108

Kidneyjean

7

Original post by maxpark3r

So if confidence limits don't overlap:
There is a significant difference between the means and therefore the probability of results being due to chance is less than 5% and reject null hypothesis?

Yup!

Reply 109

excell

I've been told to use n and not n-1 as that's what is stated in the AQA book

Reply 110

maxpark3r

2

Original post by excell

I've been told to use n and not n-1 as that's what is stated in the AQA book

When you're taking a sample of the population you use n-1 not n, because it's something to do with the fact that you're not measuring the whole population.

Reply 111

Kidneyjean

7

Original post by maxpark3r

When you're taking a sample of the population you use n-1 not n, because it's something to do with the fact that you're not measuring the whole population.

Even if that's correct, we still have to use n just because AQA say so... we're just expected to use the formulae on the sheet, :s-smilie:

Reply 112

Benzene123

How do we write the null hypothesis for Standard error? My teacher says you lose marks if you write it in the style of Spearman's rank, so we can't use the word 'effect' in the hypothesis. Could anyone give an example of a null hypothesis for standard error? :smile:

Reply 113

maxpark3r

2

Original post by Benzene123

How do we write the null hypothesis for Standard error? My teacher says you lose marks if you write it in the style of Spearman's rank, so we can't use the word 'effect' in the hypothesis. Could anyone give an example of a null hypothesis for standard error? :smile:

When you draw a graph;
If the confidence limits overlap, there is no significant difference between the two means, accept null hypothesis and the probability of the results being due to chance is greater than 5%.
If confident limits don't overlap, reject null hypothesis, there is a a significant difference between the two means and the probability of the results being due to chance is less than 5%.
"The height that a diver jumps off has no effect on the height of the splash" is a null hypothesis.
edit: We weren't taught not to use effect, did your teacher explain why?

(edited 11 years ago)

Reply 114

Benzene123

Original post by maxpark3r

When you draw a graph;
If the confidence limits overlap, there is no significant difference between the two means, accept null hypothesis and the probability of the results being due to chance is greater than 5%.
If confident limits don't overlap, reject null hypothesis, there is a a significant difference between the two means and the probability of the results being due to chance is less than 5%.
"The height that a diver jumps off has no effect on the height of the splash" is a null hypothesis.
edit: We weren't taught not to use effect, did your teacher explain why?

He said when you use standard error, you're looking for differences between means so you should use the word effect because you're not looking for correlation. I was just wondering how you could word it. Maybe 'there is no difference between the height of the splash at 25m and the height of the splash at 50m'?

Reply 115

handsome7654

Original post by Benzene123

How do we write the null hypothesis for Standard error? My teacher says you lose marks if you write it in the style of Spearman's rank, so we can't use the word 'effect' in the hypothesis. Could anyone give an example of a null hypothesis for standard error? :smile:

There is no SIGNIFICANT difference between the number of aphids on the upper surface and the lower surface of nettle leaves!

The key word is 'significant' difference.

Reply 116

E.J

Hey guys,
I have my task 3 on Tuesday, really not sure what to revise or go through :frown:

Any tips would be great :smile:

Reply 117

aria 11

Hii, I also have the EMPA on tuesday and I'm really worried about it as we have to do all three parts on the same day! All I know is about seeds! What should I revise please?
Also, is it hard?
xx

Reply 118

masterhr1

5

Why is it when setting up a transect, in one of the aqa past paper mark schemes, it says to lay the measuring tape perpendicular/right angle to the road?? Wouldn't you lay it in line with the road so you can see the intervals properly as you go along?

(edited 11 years ago)

Reply 119

raveen789

9

Original post by masterhr1

Why is it when setting up a transect, in one of the aqa past paper mark schemes, it says to lay the measuring tape perpendicular/right angle to the road?? Wouldn't you lay it in line with the road so you can see the intervals properly as you go along?

You lay it perpendicular because you are measuring the transition from a certain distance from the road :smile:

AQA Biol A2 EMPA Megathread

Related discussions

Latest

Trending

Trending

Articles for you