# Error Bars.

Could anyone help me out by giving me a step-by-step guide as to how to work out error bars for graphs and then how to apply and analyse them? (The second part isn't really nessasary I just need to know how to work them out but all help is good!)

Theres rep for anyone who helps.

Thanks.
Cyst
Could anyone help me out by giving me a step-by-step guide as to how to work out error bars for graphs and then how to apply and analyse them? (The second part isn't really nessasary I just need to know how to work them out but all help is good!)

Theres rep for anyone who helps.

Thanks.

Ok, error bars are used whenever you have taken repeats, and have worked out an average. You plot the graph of the average results. Then at each poit, you plat the highest and lowest values which you got. e.g. average 20, highest 21, lowest 19. Then you draw a line from the highest value to the lowest value. This is an error bar. Then measure each error bar. The longer the error bar,the more variet in the results. The range of results is greater and the spread of data is greater. the shorter the error bars, the more accurate the results.
Hope this helps.
Cyst
Could anyone help me out by giving me a step-by-step guide as to how to work out error bars for graphs and then how to apply and analyse them? (The second part isn't really nessasary I just need to know how to work them out but all help is good!)

Theres rep for anyone who helps.

Thanks.

Once you've plotted a graph you need to decide on the apparent error you were working with.

So, say for example you were plotting a force extension diagram for a spring: The extension would have an error of +/- 0.5 mm (presuming the ruler you were using had a sensitivity of 1mm).

The force error would be accountable to the potential error of F=ma. So, say you were using masses, and had weighed them with an electronic balance with a sensitivity of two decimal places, your accuracy would be 9.81ms^-2 x 0.005g, which is +/- 0.04905N.

Looking at your graph, plot the equivalent to 0.5mm (according to your scale) either side of your point either vertically or horizontally (depending on the orientation of your axis), and 0.04905N either side of the original point (again either horizontally or vertically).

Now, draw a cross from these points through the original plotted point, and draw small bars on the end of the lines to the error points.

Incidentally, MS Excel can do this for you.
Xenon
Ok, error bars are used whenever you have taken repeats, and have worked out an average. You plot the graph of the average results. Then at each poit, you plat the highest and lowest values which you got. e.g. average 20, highest 21, lowest 19. Then you draw a line from the highest value to the lowest value. This is an error bar. Then measure each error bar. The longer the error bar,the more variet in the results. The range of results is greater and the spread of data is greater. the shorter the error bars, the more accurate the results.
Hope this helps.

Thanks, this is really helpful. Just one more question, how do I work out the highest and lowest values for a single point?
Cyst
Thanks, this is really helpful. Just one more question, how do I work out the highest and lowest values for a single point?

I covered that in my response.
Alec
I covered that in my response.

So did I!
Xenon
So did I!

So you did, sorry - I didn't read it.
So you both did. Sorry, I did read both replies but I've been up since 6am doing biology coursework and Im not really in the sharpest of states so forgive my absent mindedness.

I'll give you both some rep over the next two days when I get some. Haven't got any today.

Thanks again.
Cyst
So you both did. Sorry, I did read both replies but I've been up since 6am doing biology coursework and Im not really in the sharpest of states so forgive my absent mindedness.

I'll give you both some rep over the next two days when I get some. Haven't got any today.

Thanks again.

No probs. Glad to be of help.