# etc..

Watch
Announcements
#1
etc..
Last edited by snoopyx; 1 year ago
0
1 year ago
#2
Hi, a p value and confidence intervals are separate parameters for assessing the significance of results.

If you are specifying a p value < 0.05 as being considered to be significant, then yes, oc you do need to calculate/look up your p value. If this is then less than 0.05, this means that the probability that the null hypothesis (h0) is untrue by chance is <5% [<0.05], so you have disproven the null hypothesis with a certainty of 95%.

Look up confidence intervals - they are different, and need separate calculation.

If you are unable to locate this info, then pls place another post here to that effect and tag me. I shall try to explain.
M.
0
#3
etc..
Last edited by snoopyx; 1 year ago
0
1 year ago
#4
Are you using a one-sample t test OR a paired t-test?

You are confusing the meaning of the word confidence as used in statistics. There is no such thing as interpretation of the result of calculation of the value a t test that can be interpreted directly.

Your answer from the t-test will ALWAYS need to be looked up in a table to find the p-value. You do not estimate anything with a level of confidence. You set a target or break-off point of your p-value beforehand e.g. you might say that, as is most often done, you will take a p value less than 0.05 as being significant - if your p value comes out at, say, <0.001, then your results are v significant - most medical papers that hold water well have results with such p values.

A confidence interval and e.g. a t-test are not mutually exclusive (the word "confidence" alone is not really applied in statistics). Rather, a 95% confidence interval is a range within which there is a 95% chance that the actual mean of the population will lie. So a confidence interval gives a further insight into the significance of your results, in addition to the p value. (you can use p values [and confidence intervals] in tests other than t-tests, too e.g. in a chi-squared test; Pearson's correlation coefficient, etc.

So, once you have your t statistic, look up the p value in the table, and interpret it appropriately - I will upload one in a few minutes for you (you will need to work out your degrees of freedom).

Sorry, all this is quite difficult to grasp in a short time on TSR.
M
0
#5
(Original post by macpatgh-Sheldon)
Are you using a one-sample t test OR a paired t-test?

You are confusing the meaning of the word confidence as used in statistics. There is no such thing as interpretation of the result of calculation of the value a t test that can be interpreted directly.

Your answer from the t-test will ALWAYS need to be looked up in a table to find the p-value. You do not estimate anything with a level of confidence. You set a target or break-off point of your p-value beforehand e.g. you might say that, as is most often done, you will take a p value less than 0.05 as being significant - if your p value comes out at, say, <0.001, then your results are v significant - most medical papers that hold water well have results with such p values.

A confidence interval and e.g. a t-test are not mutually exclusive (the word "confidence" alone is not really applied in statistics). Rather, a 95% confidence interval is a range within which there is a 95% chance that the actual mean of the population will lie. So a confidence interval gives a further insight into the significance of your results, in addition to the p value. (you can use p values [and confidence intervals] in tests other than t-tests, too e.g. in a chi-squared test; Pearson's correlation coefficient, etc.

So, once you have your t statistic, look up the p value in the table, and interpret it appropriately - I will upload one in a few minutes for you (you will need to work out your degrees of freedom).

Sorry, all this is quite difficult to grasp in a short time on TSR.
M
That's great thanks!
0
1 year ago
#6
(Original post by macpatgh-Sheldon)
If you are specifying a p value < 0.05 as being considered to be significant, then yes, oc you do need to calculate/look up your p value. If this is then less than 0.05, this means that the probability that the null hypothesis (h0) is untrue by chance is <5% [<0.05], so you have disproven the null hypothesis with a certainty of 95%.
This isn't quite correct. When doing hypothesis testing, you calculate the conditional probability of the observed test statistic assuming the null hypothesis to be true P(test stat | H0). If you want to know P(H0 | test stat), you need Bayes' theorem and a prior on P(H0).
1
1 year ago
#7
(Original post by Gregorius)
This isn't quite correct. When doing hypothesis testing, you calculate the conditional probability of the observed test statistic assuming the null hypothesis to be true P(test stat | H0). If you want to know P(H0 | test stat), you need Bayes' theorem and a prior on P(H0).
Hi thank you for your input - my knowledge of statistics is only from an introduction to medical statistics as a medical student, and some ideas picked up through reading numerous research papers. Would it be possible for you to put things in a simpler format yet in sufficient detail for the benefit of OP? I think he/she would appreciate it greatly, and you are qualified as a mathematician.to help him/her.

M
0
#8
(Original post by macpatgh-Sheldon)
Hi thank you for your input - my knowledge of statistics is only from an introduction to medical statistics as a medical student, and some ideas picked up through reading numerous research papers. Would it be possible for you to put things in a simpler format yet in sufficient detail for the benefit of OP? I think he/she would appreciate it greatly, and you are qualified as a mathematician.to help him/her.

M
Don't worry I don't need this explained
0
#9
(Original post by macpatgh-Sheldon)
Are you using a one-sample t test OR a paired t-test?

You are confusing the meaning of the word confidence as used in statistics. There is no such thing as interpretation of the result of calculation of the value a t test that can be interpreted directly.

Your answer from the t-test will ALWAYS need to be looked up in a table to find the p-value. You do not estimate anything with a level of confidence. You set a target or break-off point of your p-value beforehand e.g. you might say that, as is most often done, you will take a p value less than 0.05 as being significant - if your p value comes out at, say, <0.001, then your results are v significant - most medical papers that hold water well have results with such p values.

A confidence interval and e.g. a t-test are not mutually exclusive (the word "confidence" alone is not really applied in statistics). Rather, a 95% confidence interval is a range within which there is a 95% chance that the actual mean of the population will lie. So a confidence interval gives a further insight into the significance of your results, in addition to the p value. (you can use p values [and confidence intervals] in tests other than t-tests, too e.g. in a chi-squared test; Pearson's correlation coefficient, etc.

So, once you have your t statistic, look up the p value in the table, and interpret it appropriately - I will upload one in a few minutes for you (you will need to work out your degrees of freedom).

Sorry, all this is quite difficult to grasp in a short time on TSR.
M
how would you look up the p value? thanks
0
1 year ago
#10
Look up the p value from the few options at the top of this table by checking the location of your t-test result in the body of the table that is in the line horizontally that relates to your no. of degrees of freedom on the leftmost column.

1
#11
(Original post by macpatgh-Sheldon)
Look up the p value from the few options at the top of this table by checking the location of your t-test result in the body of the table that is in the line horizontally that relates to your no. of degrees of freedom on the leftmost column.

thanks!
0
X

new posts
Back
to top
Latest
My Feed

### Oops, nobody has postedin the last few hours.

Why not re-start the conversation?

see more

### See more of what you like onThe Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

### Poll

Join the discussion

Yes (21)
30%
No (49)
70%