File "IA 0602 on Vimeo.vtt"

Path: /PaperPlainz/Internal Assessment Guide/Step 6/IA 0602 on Vimeovtt
File size: 1.94 KB
MIME-type: text/plain
Charset: utf-8

 
Open Back 
WEBVTT

00:00.000 --> 00:05.520
In this video we will discuss how to justify error bars and add them to your graph.

00:05.520 --> 00:10.880
We will talk about justifying whether you include or omit error bars and see how error

00:10.880 --> 00:13.520
bars look like on a graph.

00:13.520 --> 00:17.720
Error bars are considered under the analysis criterion.

00:17.720 --> 00:23.920
Error bars always use the absolute and not the percentage uncertainty in the given value.

00:23.920 --> 00:28.600
You do not have to write a lot about error bars, you just need a brief comment to justify

00:28.600 --> 00:29.600
them.

00:29.600 --> 00:32.600
Let's see how this is done in the example IA.

00:32.600 --> 00:38.920
First, I explain that I will include horizontal error bars to show an uncertainty of plus

00:38.920 --> 00:46.200
minus 0.2 centimeters in H. Then I add that from my process data table it is clear that

00:46.200 --> 00:52.720
uncertainty in the frequency is negligible, so I will not include vertical error bars.

00:52.720 --> 00:56.400
You can see my error bars in the graph above.

00:56.400 --> 01:00.840
There is no clear rule when you should or should not omit error bars, I usually add

01:00.840 --> 01:07.040
them in the graphing program that I am using and omit them if they are not clearly visible.

01:07.040 --> 01:09.440
The key is to justify your decision.

01:09.440 --> 01:14.880
Finally, the two best programs to use for your graphs and error bars are Excel and if

01:14.880 --> 01:17.880
you are using a Mac probably numbers.

01:17.880 --> 01:23.720
In theory you can also draw your graphs by hand, although students rarely do this.

01:23.720 --> 01:29.480
This completes step 6 action 2, now you know how to justify error bars and add them to

01:29.480 --> 01:30.880
your graph.

01:30.880 --> 01:34.880
In the next video we will discuss how to fit a trend line to your data.