Any maths / statistics whizzes here?

[SimonMac]

I am looking to see if I can extrapolate some numbers based on the information set I have.

If a sample is 100% and each week 0.4% "do a thing", how long would it take on average for all (or at least a statistically significant number) the people to do "the thing", I assume it's not as simple as 100 / 0.4 = X number of weeks

[ladymuck]

Not quite.

Week 1 would be 0.4. Week 2 would be 0.4 from week 1 plus 0.4 from week 2 so 0.8. week 3 would be 1.2 and so on.

Determining statistical significance requires knowing how big the sample size is

I'm assuming that people only need to do the thing once and so are counted towards the cumulative total of people doing the thing

[Old Greg]

I am looking to see if I can extrapolate some numbers based on the information set I have.

If a sample is 100% and each week 0.4% "do a thing", how long would it take on average for all (or at least a statistically significant number) the people to do "the thing", I assume it's not as simple as 100 / 0.4 = X number of weeks

Is a member of the sample group as likely to do the thing in any week if they have done it in a previous week?

[northernladuk]

42

HTH

[DimPrawn]

Who are you doing "the thing" to, have they consented, and have all legal requirements been met?

[northernladuk]

I am looking to see if I can extrapolate some numbers based on the information set I have.

If a sample is 100% and each week 0.4% "do a thing", how long would it take on average for all (or at least a statistically significant number) the people to do "the thing", I assume it's not as simple as 100 / 0.4 = X number of weeks

You'd need to take in to account if they are lazy feckless southerners doing the thing in which case most won't or can't but if it was some pure bred northern hard men it would be twice as fast with half the people and be done properly.

If it's going off shore then it's going to take three times as long with ten times more people doing it and for some obscure reason would never finish.

[WTFH]

Are you saying that if it was 1000 people,
Week 1: 4/1000
Week 2: because 4 were involved last week, they are no longer counted, so it's 4/996
Week 3: 4/994
...etc

Or are you saying that it's always a random 4/1000 that do it. It could be the same 4 people, or 4 different ones.

[scooterscot]

Assuming the number of people ‘do a thing’ decrease each week as your sample size is reduced.


week 1: population * 0.4 = n


week 2: (population - week 1 n) * 0.4 and so on… until you solve for 0

[DimPrawn]

This thread is going to run and run with all sorts of idiotic "experts" I can feel it already.

[AtW]

This thread is going to run and run with all sorts of idiotic "experts" I can feel it already.

Is this a gut feel or your expert opinion?