Segregation

From ICISWiki

Segregation

In gathering data from field trials and transferring it over to Workbooks we came across the problem of segregation. Within the same plots we were having pants scoring differently for the same measurement.

For example, using Stripe Rust in wheat: in a single plot labelled A there were 10 plants. Six of these plants may have scored 7, while 4 scored between 2 and 4. This left us with a problem of how to represent this within the Workbooks as they were prepared for uploading into the database.

We developed a system in which the score that occurs most frequently, in the given example, 7, is recorded and then a zero is placed next to it and the score that occurs least frequently, in the given example, let’s say 2 is then recorded. The Workbook entry would look like this:

• 702

This same kind of issue could arise for traits such as maturity and/or height where would also need to use two digit representations such as:

• 11/15

Traditionally here in Australia this type of data would be scored with the use of a slash between the two different values hence it would appear like this:

• 7/2 or 11/15.

We wanted to keep the numeric value, hence for that reason we chose to use numeric characters over non-numeric characters. The reason for wanting to keep the numeric characters was for querying on these values once they were entered into the database.

We were wondering how others have dealt with this type of situation. It would be great to know as it may give us all insight into a better way to enter this data into Workbooks.

Thanks.

David McMeekin