Statistics Poster Competition Update - Putting It All Together
In the previous three issues of the Interchange, the Michigan Statistics Poster Competition (MSPC) was described. All K-12 Michigan students are eligible to enter the competition. Further information, the registration form, and a complete set of rules are available here. You may also contact Dan Frobish at firstname.lastname@example.org for additional information.
A statistics poster is a visual display that uses one or more related graphs to summarize data, discuss different points of view, and answer question(s) about the data. In the December 1999/January 2000 issue of Interchange, we offered suggestions for the display of statistical data. We reviewed the types of data; qualitative data - data that can be broken into categories, and quantitative data - numerical data. We further broke down quantitative data into discrete data - data that can only take on certain values within a range of numbers, and continuous data - data that can be any value within a range of numbers. Then, we discussed graphs that are appropriate for each type of data. In this article, we focus on how to most effectively communicate your message in a statistical poster.
Let us begin with one important rule: There is no one right way to make a statistical poster. However, there are common features that make for an effective poster. The most important feature of a successful statistical poster is that everything on the poster is correct. All the bells and whistles imaginable can not make up for incorrect information. The graphs must be an accurate description of the data. Titles, axis labels, legends, conclusions, etc. must be representative of and supported by the data. This does not mean that your graphs should "play it safe." On the contrary, the more bold you are in your conclusions the more likely that a reader of the poster will become engaged in the subject matter. Just be certain that you are drawing conclusions based on the evidence contained in the data and not on personal bias or preconceived notions.
Another feature of the successful statistical poster is that there is one unifying message throughout the poster. Decide what question is central to your study. The poster is your medium for delving into the issues surrounding this question. All material contained in the poster should have a connection to the question and that connection should be identifiable by the reader of the poster. If the reader asks herself the question: "What does this graph have to do with the study?"; then the poster has failed to convey your message. Titles of graphs should be descriptive enough that the reader can ascertain the relationship of the graph to the central question.
Now that we have made sure that each graph is focused on the central question, let us discuss some features of an effective graph. Effective graphs draw the reader's attention. They sell the poster to the reader. The use of color, catchy titles, unique graphical components, etc., adds to the visual appeal of the poster. For example, suppose you collect data on the rebound champion in the National Basketball Association. You might use a histogram to display this data. If so, you might fill in the bars of the histogram with cutout basketballs rather than just coloring in the bars. As another example, suppose you collect data on the fat grams and number of calories in common candy bars. You might use a scatterplot to display the data points. One potentially interesting question is: "What candy bar has the highest fat content?" You could just ask and answer this question directly, but a more interesting approach would be to create a little pocket out of construction paper that has the question written on it. Inside the pocket would be the wrapper of the candy bar that is the answer to the question. This draws the reader's attention and forces her to become an active participant in the poster. If you can get the reader of the poster to physically interact with the poster it is a big plus.
The overall placement of materials on the poster is important. Surely, we can all agree that there should be a logical placement of the components of the poster such that the reader can comprehend the intellectual path leading to the conclusions. The placement of the components should enhance not interfere with the message contained within the poster.
Lastly, be creative. The more uniquely you represent the information; the more likely the poster will draw the reader's attention. Ask yourself the question: "How can I make this more appealing to someone who is not familiar with my data?" Or better yet, ask a friend, family member, or schoolmate to describe what the poster says to them. Oftentimes you will find that what seemed obvious to you is not at all obvious to someone who is not familiar with your topic. Usually, if you can get a parent to understand what the poster is about, then you are in pretty good shape.
Remember that the most important rule is: HAVE FUN! If you're creative and willing to try various ideas, then you'll have a high quality statistical poster. Good luck in the competition and happy posterizing.
Page last modified November 30, 2010