Field research trials are important to improving agronomic practices 

By Dan Quinn, Purdue Extension Corn Specialist 

Field research trials are an important part of understanding how specific agronomic practices can improve farm productivity. In upcoming growing seasons like 2025, when corn prices are lower and margins become tighter, it becomes even more important to assess, question, and look at reputable data prior to making input or management decisions. 

Universities such as Purdue use both research station and on-farm research trials across the state to help drive our recommendations and provide management information for Indiana farmers. However, some of our research practices and conclusions may differ from various private-sector research trials and potentially what you may see on your own farm. 

For example, questions may arise, such as: Why did they set up the research trial that way? The yield numbers look different, but why are they are telling me they are not different? And, why does it seem the university never sees any yield responses from various products? 

Therefore, it is important to understand how we approach field research trials, the steps we take to determine our conclusions, and how understanding these approaches can help you better understand and test practices more accurately on your own farm. 

Crop management and assessing the practices/products that work or don’t work is challenging and often debatable. Why? Because we work, manage and assess practices/inputs in a highly variable and biologically active environment. 

Many variables to consider 

Product and practice responses can change from hybrid to hybrid, field to field, and year to year. Just because some new “5-plus bushel guaranteed” product or technology worked one year and 100 miles from your farm, doesn’t mean it will work for you. Therefore, when choosing to use a various product or practice, it is important to look at what data is available, how was it acquired, where was it acquired, and how might it help my current operation. 

The first two questions I often ask people when discussing research is: 

• Do you have a yield monitor in your combine? 

• When traveling across the field during harvest, do those yield values stay the same? 

The answer I receive 100 percent of the time is “no.” (If yes, you may need to consider a new monitor.)The reason that yield values don’t stay the same is largely due to the variability throughout the field caused by soil type differences, elevation differences, etc. Therefore, when setting up field research trials we often designate a treatment – or new product – and compare that to a non-treated control, or a business-as-usual approach. 

Two of the most important questions we ask after harvest is: 

• Was the yield difference observed truly caused by the product we applied? 

• Was the yield difference only due to the treated areas being in a more productive part of the field? 

For example, in Figure 1, if I split a field in half and apply my treatment on one half of the field and don’t apply my treatment on the other half of the field, I may find a yield difference of 15 bushels per acre and think to myself, “I should apply this product on all of my acres.” 

However, when you look closer, it is easy to see that the treated area of the field encompassed a larger portion of one soil type, whereas the non-treated area encompassed a larger portion of another soil type. Therefore, it is difficult to differentiate if the yield response was due to the product applied, or was it just due to the treated area being in a more productive area of the field. 

An approach to testing a treatment 

In our university research trials, we approach testing a treatment within a field using randomization, replication (repetition of an experiment in similar conditions), and statistics (Figure 2 and Table 1). For example, if you compare Figure 1 and Figure 2, Figure 2 highlights how we typically set up one of our research trials using replication and randomization of the treated and non-treated passes to account for field differences. 

Each of these practices help us improve the reliability of our conclusions, account for random error, such as field variability, and determine the true causes of yield differences observed. Furthermore, it is also important for us to perform these research trials across multiple locations and multiple years to determine how treatment responses may differ in different fields and different environments. 

We also use statistical models to help determine our conclusions (Table 1). Using statistics helps us determine if the differences we detect are due to random error, or due to the treatment we tested. For example, if you have ever seen university data presented, you have probably seen data presented similar to Table 1. At first glance, after we randomized and replicated our treatments (Figure 2), the treated areas seem to have increased corn yield by 4 bushels per acre (Table 1). 

However, our conclusions were that no yield differences were observed. Therefore, through the research steps we implemented, it was determined that the yield difference was due to random error, such as field variability, and not due to the product or management practice tested. 

The letters next to the yield values help us highlight where statistical (yield differences due to treatments) differences were observed. 

In conclusion, when testing a new product or practice on your own farm, it is important to think about how to design and set up a trial to accurately test the new product or practice. Just because a yield difference is observed, doesn’t always mean the new product or practice you tested is the reason for this difference. 

Also, when trying out a new practice or product, don’t just jump in and apply it across the whole farm. Leave a check-strip or turn the planter/sprayer off a couple times in the field to get an idea how something is performing. At Purdue, it is our goal to accurately assess new products and practices to determine whether or not these are truly the reason behind observed yield differences. 

In addition, as you sit in on various meetings, presentations, and examine research results, ask yourself how did they design and set up this research trial? Did they use randomization, replication and statistics; and if not, is the yield differences being discussed truly due to the product applied? And, how many different environments and years was this product tested? 

Understanding and asking these questions can help determine the best products and management practices to implement and improve your operation.