Replicated field trial
- Also known as
- CRO trial, small-plot trial, randomized complete block design
A replicated field trial is an agricultural experiment in which each treatment, such as a product or practice, and an untreated check are applied to several plots arranged randomly across a field. Replication and randomization allow real treatment effects to be separated from natural field variability.
What is a replicated field trial?
Fields are not uniform. Soil, drainage, fertility and pest pressure vary from one spot to the next, so a yield difference between two strips may reflect where they were placed rather than what was applied. A replicated field trial deals with that by repeating each treatment in several plots, called replications, and by assigning treatments to plots randomly. Comparing the average of the replications, and the variation among them, shows whether a difference is likely to be real.
How it works
- Replication. Each treatment appears several times, commonly four or more, so that one unusual plot does not decide the result.
- Randomization. Treatments are assigned to plots by chance, so no treatment consistently gets better ground.
- Blocking. In the widely used randomized complete block design, the field is divided into blocks of similar conditions, with every treatment appearing once in each block.
- Untreated check. An untreated control, or a grower standard, provides the comparison.
- Statistical analysis. Methods such as analysis of variance test whether differences among treatments are larger than expected from chance, often reported as a p-value or a least significant difference.
CRO trials
A contract research organization (CRO) is an independent company that conducts field trials for product manufacturers under standardized protocols. CRO trials are valued because the organization running them has no stake in the outcome. Results across many locations and years show how consistently a product performs.
Replicated trials vs. strip trials
On-farm strip trials compare treatments in long field-scale strips, usually with fewer replications. They are practical with farm equipment and reflect real conditions, but are more affected by field variability. Replicated small-plot trials give more statistical power per site. Many evaluation programs use both.
Why it matters for biologicals
The effects of biological products often vary with soil, weather and management. Replicated, multi-location, multi-year trials are the most reliable way to judge how often, and by how much, a product pays off.
Replicated trials at Locus
Locus states that its agricultural products are validated by third-party CRO trials and publishes yield results from those trials.
Frequently asked questions
It means the difference between treatments is larger than would be expected from natural field variability alone, at a stated level of confidence, often 90 or 95 percent.
Because field variability can easily create or hide a difference. Without replication and randomization, there is no way to tell whether a result reflects the treatment or the location.
Sources
- On-farm research and field trial design guides, Land-grant university extension
- Statistical methods for agricultural experiments, Standard agronomy references