tournedos Posted April 14, 2008 Report Share Posted April 14, 2008 I've found error bars extremely difficult, can someone explain the process to me? Reply Link to post Share on other sites More sharing options...
Dooga Posted April 15, 2008 Report Share Posted April 15, 2008 (edited) If you know the uncertainties, the error bars just mean that you make a point at the largest and smallest value possible within the uncertainties, and then connect those two points. Sounds confusing? Try drawing it. For example, if you have 4.5 grams of something, and the uncertainty is +/- 0.5 grams, then you would make a point at 5.0 and a point at 4.0. Simply connect those points in one line, and you have the uncertainty bars for that specific point. Remember to make the ends of the bar obvious with little dashes (I instead of l). Sorry I really suck at explaining something graphical with text. Here's an image I stole: [img]http://img215.imageshack.us/img215/3271/img3po1.gif[/img] Original image: [url="http://musr.physics.ubc.ca/~jess/hr/skept/Meas/node2.html"]http://musr.physics.ubc.ca/~jess/hr/skept/Meas/node2.html[/url] Edited April 15, 2008 by Dooga Reply Link to post Share on other sites More sharing options...
tournedos Posted April 15, 2008 Author Report Share Posted April 15, 2008 eh cool thanks! Reply Link to post Share on other sites More sharing options...
Rawrn3ss Posted April 20, 2008 Report Share Posted April 20, 2008 You need to use error bars when you're graphing averages too. You'll graph the average amounts, but you need to show the maximum and minimum value you got for that particular value. Ex. {24,25,26} The average of that group is 25, so you'll graph 25. Then, you'll have to graph the error bars. The top bar will go to 26 and the bottom will go to 24. Excel isn't very nice when you're graphing error bars for averages because your bars will always be different amounts, but excel only lets you graph error bars for an entire series together. You have to do all your error bars separately or draw them on if you're doing averages :x Reply Link to post Share on other sites More sharing options...
IBMan9000 Posted April 20, 2008 Report Share Posted April 20, 2008 The functions in excel make the error bars not so bad for averages. If you're going to be typing the entire data table for the lab report anyways, you can just make a function column to find the averages and two others to find the min/max of each dataset row. Excel then allows you to use a data series as the error bars' min/max. At least, it did it somehow last time I had an IA for physics due, and it wasn't too much of a hassle to set up. If you're not good with excel then you're probably better off drawing the error bars on the graph by hand. PM me if you want me to look up exactly how I did it in excel, but I remember google helping to solve it in under a few minutes. Reply Link to post Share on other sites More sharing options...
hellogoodbye Posted June 1, 2008 Report Share Posted June 1, 2008 just a question. how do you know how large or small to make the error bars? Reply Link to post Share on other sites More sharing options...
Rawrn3ss Posted June 4, 2008 Report Share Posted June 4, 2008 For uncertainties, you just make the error bars the value of the uncertainty, as Dooga mentioned with his diagram. Following his example, the line above the circle would be 0.5 as would the line below the circle. For averages, you have to make the lowest value of the uncertainty the lowest value among the averages and the same for the highest value. Excel isn't good for this because the averages might not be consistent throughout the entire function. Reply Link to post Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.