Difference between revisions of "Data processing"
From Chemical Engineering @ UP wiki
(Link to dynamite plot article.) |
|||
| (One intermediate revision by the same user not shown) | |||
| Line 1: | Line 1: | ||
| − | + | ===Data normalisation=== | |
| + | You will make your life a lot easier by using a [https://en.wikipedia.org/wiki/Database_normalization normalised] structure for your data, whether you are using a spreadsheet or an actual database for storing them. | ||
| + | |||
| + | [https://www.tandfonline.com/doi/abs/10.1080/00031305.2017.1375989 This article] contains some excellent rules for data in spreadsheets. | ||
| + | |||
| + | |||
| + | ===Error bars=== | ||
| + | It is common to see error bars and averages on experimental results. When you have a small number of data, rather plot them. [http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2064100/ this article explains the use of error bars]. Points to note: | ||
| + | * '''Rule 1''' when showing error bars, always describe in the figure legends what they are. | ||
| + | * "... for such low values of n [as n=3], it is better to simply plot the data points rather than showing error bars, as we have done here for illustrative purposes" | ||
| + | |||
| + | [http://emdbolker.wikidot.com/blog:dynamite This article] expands on the problems of plotting means and error bars instead of the data. | ||
Latest revision as of 03:48, 29 March 2019
Data normalisation
You will make your life a lot easier by using a normalised structure for your data, whether you are using a spreadsheet or an actual database for storing them.
This article contains some excellent rules for data in spreadsheets.
Error bars
It is common to see error bars and averages on experimental results. When you have a small number of data, rather plot them. this article explains the use of error bars. Points to note:
- Rule 1 when showing error bars, always describe in the figure legends what they are.
- "... for such low values of n [as n=3], it is better to simply plot the data points rather than showing error bars, as we have done here for illustrative purposes"
This article expands on the problems of plotting means and error bars instead of the data.
