Ask Your Question
0

When I use a data range for error bars, is this the total error or the error up and down?

asked 2018-11-11 12:03:17 +0100

Calllack gravatar image

I have selected a range of values to use for y error bars. I have chosen the "same value for both" option when asked for separate values for the positive and negative error. My question is: say that the value from the range used for one data point is 0.3. When I ask Calc to use "same value for both", is it creating a positive error of 0.3 and a negative error of 0.3, or is it spreading the error across, giving a positive error of 0.15 and a negative error of 0.15 (adding up to a total error of 0.3)?

edit retag flag offensive close merge delete

Comments

1

Why not do an experiment? Put a value in your spreadsheet of, say 50, with an "error" of 10. Then plot it and see if the error bar goes from 45 to 55, or from 40 to 50. That should answer your question pretty definitively.

ve3oat gravatar imageve3oat ( 2018-11-12 02:46:26 +0100 )edit

Good idea. Originally the large spacing of gridlines on graphs on LibreOffice set me against this (it makes it hard to read where you are on a scale), but I could make a new graph just for this test and set the error to something big enough to be easy to read.

Calllack gravatar imageCalllack ( 2018-11-12 11:52:22 +0100 )edit
1

You can change the gridlines by formating the axis. Double-click on the chart to put it in Edit mode; left-click on the axis in question; then right-click to bring up a dialog box. Choose Format Axis. Under the Scale tab, choose a different (smaller?) Major Interval. When finished, click on a cell outside the chart to cancel Edit mode. Many things are possible. (Have fun!)

ve3oat gravatar imageve3oat ( 2018-11-12 16:29:47 +0100 )edit

Thanks, I didn't know this!

Calllack gravatar imageCalllack ( 2018-11-12 20:09:10 +0100 )edit

1 Answer

Sort by » oldest newest most voted
1

answered 2018-11-12 20:08:45 +0100

Calllack gravatar image

When the "same value for both" option is selected, the data range selected is used for values for the positive error and for the negative error, and not considered as "total error"; in other words, the data are used as the error up AND down, and not as total error.

It is easier to explain by example: if the data used for the error of one data point is 10 and the "same value for both" error bars option is selected, then a positive error of 10 will be created as well as a negative error of 10, resulting in a total error (or "uncertainty") of 20.

edit flag offensive delete link more

Comments

You have been able to answer your own question, a Good Thing by any criterion. Hint -- Since it was your question, if you believe the answer is satisfactory, you should click on the uncolored check-mark at the left of the answer, so that others will know that the answer is a good one.

ve3oat gravatar imageve3oat ( 2018-11-12 21:05:04 +0100 )edit

I would like to, but I haven't got enough karma.

Calllack gravatar imageCalllack ( 2018-11-12 21:06:24 +0100 )edit

Understood. Maybe one of the moderators will do it for you. :-)

ve3oat gravatar imageve3oat ( 2018-11-13 00:43:45 +0100 )edit
Login/Signup to Answer

Question Tools

1 follower

Stats

Asked: 2018-11-11 12:03:17 +0100

Seen: 107 times

Last updated: Nov 12 '18