Data Checking Functions

There are a number of things you should do to check your data for clear errors. One is looking at the minimum and maximum data points (either by eye, by sorting the data, or using a function in a program like Excel). If these are very different than what you expected, or there is a data point that is greatly different than the rest of the range, you should double check in case this is an error (typographical, measuring error, etc.). 

Trait histograms like the examples below are a good quick way to look for outliers. In the first histogram, note that there is one, and only one, plant that scores 8.5 for brix (a measure of soluble solids) (red arrow). This is highly unusual and could be an error (Illustration 6). It could be better to remove this from the data set, or at least keep it in mind when you analyze results.

Figure 7/Illustration 6: Checking data for errors Data from Theresa Fulton, using the software QGene (