Four Tips for Better Academic Research Presentations

Have you ever wondered how professors and other researchers are so good at making convincing presentations out of relatively insignificant data? Last semester I was in charge of a research study that yielded seemingly insignificant results. With my tail between my legs I presented the results to the professor that I work under. In about twenty minutes he re-ran the SPSS statistical reports and had signed me up to present at the ACSM regional conference. As it turns out I had been approaching the data from the wrong angle. I am currently in my final semester of graduate school and am finally beginning to pick up on a few “tricks of the trade”.

1. Don’t Be Afraid to Turn a Large Study Into Several Smaller “Cohort Studies”

For example, one could set up a product validation study with 100 random participants. If the results are classified as insignificant you could break up the data into cohorts like: post menopausal women, college aged men, college aged obese females etc. You may find that there is a strong correlation in one of these group.

2. Save the Pies for Dessert

Another presentation technique I have learned is the art of using graphs. Great researchers ALLWAYS use the proper graph. Just because Microsoft Excel allows you to turn one group of data into any type of graph doesn’t mean proper graph selection is unimportant. Steven Phew is considered by many to be the world’s leading expert on graph selection and he has written numerous publications on graph types and color scheme optimization. In his most famous publication, “Save the Pies for Dessert” Steven cautions readers to not use pie graphs. Additionally, he suggests that 3D bar graphs are confusing and misleading.

3. Thoughtfully Label your Graphs

First, you must carefully determine how you will label your axes to convey your point to the audience. Generally the independent variable is assigned to the x-axis and the dependent variable is assigned to the y-axis. After you have labeled your axes you must determine which units you will use on each axis. If an SI unit is assigned to what you are measuring it’s best to stick with that. Then you must decide how you will scale your axes. Improper scaling can mislead the casual onlooker so this step is extremely important.

4. No Significance can be Extremely Significant

The purpose of most intervention studies is to “prove” that a given stimuli either positively or negatively effects a person. Generally when data that has no statistical significant is frowned upon. Don’t down play the fact that you have just “proven” that a stimulus has no measurable effect on a person. These results are still valid and could be useful. Great presenters are able to turn nominally important data into home run presentations.

Comments are closed.