How to estimate the total error?

• Sep 14th 2010, 04:31 AM
susibom
How to estimate the total error?
I have a list of (systematic) errors obtained by calculating the percentage error of two sets of values. now from that list i don't know how to say what my total error was. these are the values:

-5.04
-3.09
-2.63
-0.20
0.00
1.40
3.88

I don't think that saying that the error was -5.04 to 3.88 is correct. I think I have to calculate the standard deviation or the standard error but I am not sure which. I know the formula for the standard deviation but for the standard error i'm not sure if it is the square root of the sum of the squares. What do you suggest I do to obtain the total error?
• Sep 14th 2010, 06:04 AM
CaptainBlack
Quote:

Originally Posted by susibom
I have a list of (systematic) errors obtained by calculating the percentage error of two sets of values. now from that list i don't know how to say what my total error was. these are the values:

-5.04
-3.09
-2.63
-0.20
0.00
1.40
3.88

I don't think that saying that the error was -5.04 to 3.88 is correct. I think I have to calculate the standard deviation or the standard error but I am not sure which. I know the formula for the standard deviation but for the standard error i'm not sure if it is the square root of the sum of the squares. What do you suggest I do to obtain the total error?

Try explaining what your figures denote more clearly please.

CB
• Sep 14th 2010, 08:04 AM
susibom
I made a comparison between the results I obtained through two methods.
One method has shown to be more accurate than the other. So when I made the comparison of the values I calculated the difference of the values with the accurate method with those of the tested method. With this value I calculated the percentage error assuming the accurate method was the theoretical value and the other was the experimental value. After this I obtained a list the list of values I showed before in the thread. So I was told that I could use those values as an indication of the error I get when using the tested method. But I think that I can't simply say that the error was -5.04 to 3.88%. So I thought maybe if I obtained a mean value of error from those I listed I could say my error is say +/-7.7%. And I used the formula of the square root of the sum of the squares. But then I read that that formula is only used when comparing means. So if my values ar not means then I suppose the alternative is the Standard deviation. Am I correct?