The following is a checklist to assist in deciding whether the claims made in promotional material are valid, whether the data is accurate and reliable, and whether the claims are applicable to your situation.
- Is the data published from Australian trials?
- Were the trials from a range of sites and a range of farming conditions throughout dairy regions in Australia?
- Were the trials exposed to livestock grazing in order to simulate what occurs in a farming situation?
- Were the varieties used in the trials quality, proven cultivars which are commercially available today?
- Is there information available about performance of the cultivar given in terms of energy production (ME or MJME/ha), disease resistance, seasonal growth rates or is it simply presented as total drymatter production?
- Does the trial data presented cover a number of years? Does the trial data presented cover a whole year? If not, why only selected seasons?
- Is the heading date of the cultivar known?
- What was the endophyte status of the cultivar and those used for comparison? If they were different why and what effect might this have had? If it was a grazing and/or persistence trial, this can be very important There are three key measures which should be presented along with all trial data; these are Coefficient of Variation (CV), Trial Mean and Least Significant Difference (LSD).
- The CV is a term used to define the accuracy of a trial and can be seen as a measure of how well the trials were run. The CV% should be less than 20, otherwise there is a lot of unexplained variation in the trial.
- The Mean is the average of the values recorded for the trial. Often the data will be 100ised that is the mean of the trial set at 100. The performance of all cultivars is presented relative to this. A reading of 132 would mean that the reading was 32% greater than the mean.
- Least Significant Difference (LSD) is a value necessary to determine whether an actual difference between two values exists. If two numbers are different by less then the LSD, the two numbers are not significantly different. The difference that does exist could be explained by random chance.
There are a number of pieces of information which should be presented in order to make an accurate decision about pasture cultivar trial data. The checklist here outlines the most important of these which should enable a consumer to make an informed decision. These questions should ensure that the information provided is valid, reliable, and applicable.