October 19th, 2022
Statistical analysis of trial data
Trials often have a central question that needs to be answered. It’s usually in the last step of the trial, the statistical analysis of the collected data, where answers to the question emerge. To help with this important step we’ve added some more statistical analysis capabilities to QuickTrials.
In a new “Analysis” tab in the trial data, we now show the common Analysis of Variance (ANOVA) for plot based Randomised Complete Block Design (RCBD). It allows users to easily perform these steps from within QuickTrials:
1. Select the trial and traits to analyse
2) View scatterplots to make sure the data is in a shape that warrants deeper analysis
3) Check the common statistics used for an ANOVA, in an automatically generated table. Multiple traits can be selected if desired.
4) View possible interactions between test subjects and sites
5) View simple barcharts of the averages
The statistical calculations and charts generated by QuickTrials provide an immediate insight into the trial and can point to areas worth further analysis.
To use the built-in ANOVA analysis for a trial, the following conditions should be considered:
- Continuous numerical traits. This could be a yield derived trait or a plant height for example.
- RCBD should be used for the experimental design of plot layouts. There need to be multiple repetitions in blocks. The environmental difference inside of a block should be minimal.
- Multiple sites. A part of the analysis is only relevant if the same trial has been implemented at different sites.
- Plot based traits. A trait has to be represented as a single value for a single plot. If there are multiple samples needed per plot, a derived trait can be used to aggregate the samples into a single plot value.
We trust you will find the additional statistical analysis helpful and look forward to hearing any feedback you have on how we might enhance these features further.
Retiring Legacy Reporting
Since we are now building out our analysis capabilities directly in the trial data section, we will slowly start retiring the similar feature set in the Report functionality (in the trial context menu). If there is some aspect of the earlier Report functionality that you use and feel is not covered with the new Analysis tools, please let us know.
QuickTrials is growing
We’ve been steadily growing the capabilities of the solution and our team has also increased this year. Here are the people who are working hard to make QuickTrials the best field trial management software out there.
We know that further improvements are needed and we have an exciting vision for the platform. We’re working systematically to make it all happen.
As always, we love to hear any feedback you might have on things we could do better, features that would make your life even easier or just questions whenever you are not sure of how something works. You can always reach out to us at firstname.lastname@example.org.