Skip to main content
ARS Home » Pacific West Area » Pullman, Washington » WHGQ » Research » Publications at this Location » Publication #328393
Research Project: Wheat Quality, Functionality and Marketablility in the Western U.S.

Location: Wheat Health, Genetics, and Quality Research

Title: What are you trying to learn? Study designs and the appropriate analysis for your research question

Author
item Kiszonas, Alecia

Submitted to: American Association of Cereal Chemists Meetings
Publication Type: Abstract Only
Publication Acceptance Date: 9/8/2016
Publication Date: 10/26/2016
Citation: Kiszonas, A. 2016. What are you trying to learn? Study designs and the appropriate analysis for your research question. American Association of Cereal Chemists Meetings. http://www.aaccnet.org/meetings/Documents/2016Abstracts/aacc2016abs234.htm

Interpretive Summary:

Technical Abstract: One fundamental necessity in the entire process of a well-performed study is the experimental design. A well-designed study can help researchers understand and have confidence in their results and analyses, and additionally the agreement or disagreement with the stated hypothesis. This well-designed study also gives confidence to the scientific community of the objectivity and purity of findings. Often, however, the specific questions and objectives of a researcher’s study are not fully developed when designing the experiment. This misstep often leads to difficulty in properly analyzing studies, and in some cases, decreases the usefulness of the data that has already been collected. The purpose of this talk is to explain why deciding on the overarching question we are trying to answer in our research is the most important first step in conducting research. Once the question has been distilled, we can move on to the best study design to create an unbiased, objective, controlled experiment with adequate replication and without confounding. Some of the important factors to consider when designing an experiment are identifying which factors to control and which to vary, what specifically needs to be measured, the number of replications, the cost of the experiment, quantitative vs. qualitative factors, and how the study will be analyzed. Taking all of these factors into consideration helps to create a strong design with high statistical power and a straight-forward analysis.