The Student Room Group

repeatability or validity?

Each time the seeds are tested for viability (in seed bank), 50 seeds are used. State why 50 seeds are used.

Is this for repeatability or validity?
(edited 2 years ago)
An experiment that is repeatable i.e. for which similar results can be obtained by repeating THE WHOLE EXPERIMENT has a high repeatability.

Using a large[r] number of samples OR a larger representative sample from the population in Q IN A SINGLE EXPERIMENT will give more valid results.

@Jpw1097 - what do you think? You are good at medical statistics - ta!
Original post by macpatgh-Sheldon
An experiment that is repeatable i.e. for which similar results can be obtained by repeating THE WHOLE EXPERIMENT has a high repeatability.

Using a large[r] number of samples OR a larger representative sample from the population in Q IN A SINGLE EXPERIMENT will give more valid results.

@Jpw1097 - what do you think? You are good at medical statistics - ta!

Thanks.

In questions on clinical trials, previous edexcel marks schemes expect "(sample size is large) for reproducibility / repeatable / reliable". I have seen this in more than one marks scheme. Why is it that taking a large sample of people improves repeatability and taking a larger sample of peas improve validity?

Edexcel definition of validity is that you are measuring what you are supposed to be measuring and this depends on both method and instrument. As far as I understand validity is when we control some variable. Is the large sample here for controlling for differences among peas/people? like in epidemiological studies? Is that why the answer is validity and not repeatability?
(edited 2 years ago)
Original post by thomas.rhett
Thanks.

In questions on clinical trials, previous edexcel marks schemes expect "(sample size is large) for reproducibility / repeatable / reliable". I have seen this in more than one marks scheme. Why is it that taking a large sample of people improves repeatability and taking a larger sample of peas improve validity?

Edexcel definition of validity is that you are measuring what you are supposed to be measuring and this depends on both method and instrument. As far as I understand validity is when we control some variable. Is the large sample here for controlling for differences among peas/people? like in epidemiological studies? Is that why the answer is validity and not repeatability?

Why is it that taking a large sample of people improves repeatability and taking a larger sample of peas improve validity?
This should NOT be the case - I am very sceptical with Qs/MSs of Edexcel: often they are downright incorrect - sorry that your school has selected this far from fair/optimal board.

Is the large sample here for controlling for differences among peas/people? like in epidemiological studies? Is that why the answer is validity and not repeatability?
I would have thought so:-
If you look at the maths of e.g. the student's t test, the significance figure [p value] is likely to be smaller [greater significance [similar to validity??]] when the sample is larger e.g. if comparing a treatment group with a placebo group to see if a drug is effective, with a null hypothesis that states no difference between drug and placebo groups, if the p value comes out at, let us say p < 0.001, [v small]

THEN:
there is a 1/1000 probability that any difference between the 2 groups has occurred by chance i.e. there is a 99.9% chance that the two sets of data ARE DIFFERENT i.e. that the drug works.

Sorry I am not a statistician, this is only from my knowledge of medical research.

M
Reply 4
Original post by thomas.rhett
Each time the seeds are tested for viability (in seed bank), 50 seeds are used. State why 50 seeds are used.

Is this for repeatability or validity?

Tell me more about the question? What’s the study?

Repeatability (which is the same as reproducibility and reliability) means that a similar experiment under similar conditions will produce similar results.

Internal validity is the extent to which an experiment measures what it is supposed to measure. External validity is the same as generalisability, which is to what extent the findings of an experiment are applicable to the general population, or a different population.
Original post by Jpw1097
Tell me more about the question? What’s the study?

Repeatability (which is the same as reproducibility and reliability) means that a similar experiment under similar conditions will produce similar results.

Internal validity is the extent to which an experiment measures what it is supposed to measure. External validity is the same as generalisability, which is to what extent the findings of an experiment are applicable to the general population, or a different population.



This particular question is from edexcel Biology B 2017 paper 3 Q9. This is not from a scientific study. The question is about testing seeds for viability in a seed bank. The expected answer in the marks scheme is "to produce valid (measure of seed viability / results / data)". So if I say this is for repeatability I won't get any marks.

But in questions about design of scientific studies (e.g. atherosclerosis) the marks schemes always consider large sample size is for repeatability and variables are controlled for validity and if I say large sample size is for validity I lose marks again!

Quick Reply

Latest