Analysis will also consider whether type of vegetable impacts results. Journals will be examined for gardening care and experience, and results may be re-analyzed to exclude outliers or other unexpected events. What are the limitations to your experiment?

Vegetable yield can vary by a number of factors, of which fertilizer type is only one. The randomization of treatment type should minimize underlying differences between intervening variables such as volunteer efficacy, initial plant condition, initial soil condition, etc. Moreover, the analysis of results will use the weekly journals in order to control for outside factors. Still, we should use caution when attempting to generalize results from a limited sample over a single growing season.

This research design also relies heavily on self-reporting from participants. Thus, results are subject to measurement...

Again, the randomization of fertilizer type to volunteers should minimize the impact on overall conclusions.
What are the ethical concerns with your study?

Volunteer gardeners must agree to a fairly onerous amount of physical labor (planting, caring for, and harvesting vegetables) and paperwork (reporting) to participate in this study. In return, they receive free vegetable plants and (perhaps) fertilizer. This could open the study to charges of exploitation, if volunteers are low-income and in need of free plants. However, by targeting current volunteers with existing community gardens (who presumably freely labor in the gardens already), ethical concerns are minimized. Further, while the study encourages participation throughout the 20-week study period, no penalty occurs if a volunteer decides to withdraw before the end of the study.