Increase your sample size. The sample size is the number of data points you have collected during your investigation. A larger sample size is usually better because the more times a test is replicated, the more reliable the results will be. In other words, if an event happens once, it may or may not happen again. If it happens three times, it is more likely to be a recurring event. If it happens 3,000 times, it is very likely to be a reliably recurring event.
If you have already found a simple correlation, test for causation. You may have found from previous tests that two or more events are related. For example, you may have observed that plants that grow in areas where there is plenty of water are also large. But why are the plants large? Is it because of the access to water, or is it because this species of plant is naturally large? Correlation does not prove causation. You would have to test for causation by using many specimens of that plant species and growing them in experimental conditions (some with plenty of water, others with less water).
Look at additional variables that may affect the outcome of your experiment. You may be conducting a simple experiment that tests how access to water affects plant growth. This is a sound experiment; however, there are many factors that are likely to affect plant growth other than access to water. You could increase the breadth of your investigation by including other variables (such as access to sunlight, or access to different types of soils) that may affect plant growth, and even looking for interactions between the variables. (What effect does a lot of sunlight and a little water have? How about no sunlight, but plenty of water and fertilized soil?)
Look for variables that may affect the variables you initially were investigating. Perhaps you have found that increased water access does indeed increase plant size. Now look at variables that affect water access. Possibilities could include annual rainfall, landscape features, nearby aquifers, soil type, environmental disturbances, etc.