# Statistical tests? Watch

1. In what situation would you use standard error, and what exactly does it prove?
2. I too would like to know the answer to this question...
3. Its the standard deviation of the sampling distribution. It tells you how much variability there is in that statistic across samples from the same population. So if its a large value, it probably means that your sample is not an accurate representation of the population (that the sample came from).
4. (Original post by Callipygian)
Its the standard deviation of the sampling distribution. It tells you how much variability there is in that statistic across samples from the same population. So if its a large value, it probably means that your sample is not an accurate representation of the population (that the sample came from).
Ah okay. Thanks for the help.

(Original post by widgets)
You'd love it, wouldn't you?
Dirty little whore.
5. lol, okay, didnt realise you knew each other. Yeh sorry i cant help you with applications to biology since I dont study it. Part way through a stats coursework myself - actualy dieing. We have to do a multivariate linear regression and then a logistic regression - theres 14 variables but i only ended up using like 5. Its worth 50% of a module though so i have to do well
6. (Original post by CapnHooch)
In what situation would you use standard error, and what exactly does it prove?
It does not really 'prove' anything.

It is the standard deviation of your estimator from your sample. A large standard error (SE) suggests that your estimator is not accurate. So say you want to find an estimator of parameter x. Call this x'. The standard error of x' is the standard deviation of x' given your sample.

You can use the SE to prove significance. If you make assumptions about the distribution where your sample came from, then you can create test statistics e.g. a t-test that involve the SE and then you can calculate the distribution of this test statistic and use that distribution to see if x' is significant. you could also create a confidence interval for your estimator x'.
7. (Original post by CapnHooch)
In what situation would you use standard error, and what exactly does it prove?

(Original post by LordBlighty)
I too would like to know the answer to this question...
I know some people have already mentioned their version, I think they're all useful, but anyway here's my version.

Standard error is the standard deviation divided by the square root of n where n = the sample size. For standard error bars, you times this value by 2 and plot them above and below the mean.

the purpose of this? Standard error bars show you the range of which faults in your experiment which could have affected this. In other words, you may have done something wrong in your experiment which could have affected your results, but whatever happened, your slightly affected results will still lye in the standard error bars. Scientists use standard error bars to see whether there is a significant difference between two sets of data, this is called the null hypothesis. The null hypothesis tells us that any significant difference between two sets of data is due to chance alone. We carry out the experiments to see whether or not we can reject the null hypothesis(H0).

Say for example we have plotted 2 mean values and the standard error bars. If the error bars overlap eachother, then we cannot reject H0, and that it is possible that all the significant differences are due to chance alone. However, if the error bars do not overlap one another, we can reject H0 and we can say there is a significant difference between the two mean values.

Sorry for the essay guys, but here's my input, just hope this helps you understand better

I'm not saying my explanation is better than others, its just my input on this thread.
8. (Original post by danny111)
It does not really 'prove' anything.

It is the standard deviation of your estimator from your sample. A large standard error (SE) suggests that your estimator is not accurate. So say you want to find an estimator of parameter x. Call this x'. The standard error of x' is the standard deviation of x' given your sample.

You can use the SE to prove significance. If you make assumptions about the distribution where your sample came from, then you can create test statistics e.g. a t-test that involve the SE and then you can calculate the distribution of this test statistic and use that distribution to see if x' is significant. you could also create a confidence interval for your estimator x'.
I know some people have already mentioned their version, I think they're all useful, but anyway here's my version.

Standard error is the standard deviation divided by the square root of n where n = the sample size. For standard error bars, you times this value by 2 and plot them above and below the mean.

the purpose of this? Standard error bars show you the range of which faults in your experiment which could have affected this. In other words, you may have done something wrong in your experiment which could have affected your results, but whatever happened, your slightly affected results will still lye in the standard error bars. Scientists use standard error bars to see whether there is a significant difference between two sets of data, this is called the null hypothesis. The null hypothesis tells us that any significant difference between two sets of data is due to chance alone. We carry out the experiments to see whether or not we can reject the null hypothesis(H0).

Say for example we have plotted 2 mean values and the standard error bars. If the error bars overlap eachother, then we cannot reject H0, and that it is possible that all the significant differences are due to chance alone. However, if the error bars do not overlap one another, we can reject H0 and we can say there is a significant difference between the two mean values.

Sorry for the essay guys, but here's my input, just hope this helps you understand better

I'm not saying my explanation is better than others, its just my input on this thread.
Thank you!

TSR Support Team

We have a brilliant team of more than 60 Support Team members looking after discussions on The Student Room, helping to make it a fun, safe and useful place to hang out.

This forum is supported by:
Updated: December 8, 2010
Today on TSR

### Cambridge interview invitations

Has yours come through yet?

### Official Oxford interview invite list

Discussions on TSR

• Latest
• ## See more of what you like on The Student Room

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

• Poll
Useful resources
Discussions on TSR

• Latest
• ## See more of what you like on The Student Room

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

• The Student Room, Get Revising and Marked by Teachers are trading names of The Student Room Group Ltd.

Register Number: 04666380 (England and Wales), VAT No. 806 8067 22 Registered Office: International House, Queens Road, Brighton, BN1 3XE

Reputation gems: You get these gems as you gain rep from other members for making good contributions and giving helpful advice.