Design and analysis are central tenets of any decent statistical course. Although it is tempting to skip right to the nuts and bolts of how you type the syntax into the program, it is critical that you understand what it is you are doing. This means gaining an awareness of how experimental design affects analysis as well as an understanding of the basic theory underlying the procedures themselves.

All too often, design is overlooked when running statistical analyses. This becomes an issue when it comes time to analyze the output. For example, people may use correlational or multivariate statistical analysis and then proceed to interpret the results in terms of one factor causing another. However, causation (as any stats professor will tell you) is a matter of study design, not analysis. Without an experimental manipulation, no causal conclusions can be drawn.

Comprehensive Design and Analysis Instruction
Of course, understanding the basics of design is insufficient in and of itself. You must also know a thing or two about analysis. Clearly, this will vary depending on the particular course; however, it is essential that instruction in data analysis impart some understanding of the principals that drive the techniques. All of this is necessary so that you can know which techniques are appropriate for your particular analysis and so that you can interpret your findings accurately.

Design and analysis are often viewed as dry, tedious subjects. However, they come to life in the context of your own project. Even if you have never been mathematically minded, consider educating yourself about new analytical design methods as a means of enhancing your next project. It is easier than you might think and well worth the investment of time.