A Guide to Using Statistical Thinking in Science Fair Projects

from the Cleveland Chapter of the American Statistical Association with contributors: Gerry Beck, Steve Richardson, John Schollenberger, and Gary Skerl

Northeastern Ohio Science and Engineering Fair Special Awards from the Cleveland Chapter of the American Statistical Association

Each year, the Cleveland Chapter of the American Statistical Association awards a number of prizes for the best uses of statistics in a science fair project at the Northeastern Ohio Science and Engineering Fair (NEOSEF). Our aim is to improve the scientific reasoning used by the students to reach their conclusions through proper use of statistical thinking. We believe that statistical thinking is inseparable from the scientific method. For example, every science project begins with a hypothesis. Statistical thinking helps define the hypothesis and provides the means for testing the hypothesis.

Reproducibility - One of the basic tenants of the scientific method is the idea of reproducibility: will the same results be seen when the experiments are repeated? It is for this reason that scientists replicate their experiments. That is, they run the same experiments a number of times to see if the results are reproducible. However, there will always be some variation whenever experiments are repeated, due to experimental error. Statistical thinking helps determine whether differences observed between groups (e.g., control and treatment) are real or are just due to experimental error.

Criteria for Awards

Statistical thinking does not necessarily require the use of statistical formulas or analysis - it only requires that the student recognizes and plans for the presence of variability. Because statistical thinking is useful in all fields of science, every science fair entry is evaluated by our panel of judges regardless of which category it is entered in.

There are two ways of using statistics in a science fair project:

• Descriptive statistics are ways of using statistics to describe data, through plots of data and calculating statistics like average and standard deviation.
• Inferetial statistics are ways of inferring conclusions from the data. This includes "hypothesis tests." Inferential statistics and hypothesis tests are very powerful ways students can use to improve the quality of their science fair projects.

Descriptive statistics and inferential statistics are both important tools, and can be used in nearly any science fair project. At a minimum, the Cleveland Chapter of ASA would like to see appropriate descriptive statistics used in all science fair entries. The use and understanding of appropriate inferential statistics would make stronger projects.

Prizes

Prizes awarded by the Cleveland Chapter of ASA have included Certificates of Award, cash prizes ranging from $25 up to $100, and books such as "Statistics: A Guide to the Unknown" by Tanur or "Exploring Data" by Landwehr and Watkins.

Who We Are

Our judges are members of the Cleveland Chapter of the ASA who volunteer their time to judge the science fair. They all either use statistics in their scientific work in Cleveland area industry and health care institutions, or teach in area colleges.

For More Information

Contact: Gerald Beck, NEOSEF Special Awards Judging Organizer for ASA Cleveland Chapter, 216-444-9927, gbeck@bio.ri.ccf.org

Gary Skerl, 440-683-8481, gary_skerl@progressive.com

If you wish to arrange a visit to your class from a statistician, contact Jerry Moreno, 216-397-4681,moreno@jcu.edu

References

• Anderson, Mark J. and Anderson, Hank P., "Applying DOE to Microwave Popcorn," PI Quality, July-August 1993, pp. 30-32.
• Bochinski, Julianne Blair, "The Complete Handbook of Science Fair Projects," Wiley, 1991.
• Briscoe, Mary Helen, "Preparing Scientific Illustrations: A Guide to Better Posters, Presentations, and Publications," Springer-Verlag, 1996.
• Cohn, Victor, "News and Numbers: A Guide to Reporting Statistical Claims and Controversies in Health and Other Fields," Iowa State University Press, 1989.
• Cleveland, William S. "The Elements of Graphing Data," Wadsworth Advanced Books and Software, 1985.
• Hunter, W., "Some Ideas About Teaching Design of Experiments, with 25 Examples of Experiments Conducted by Students," The American Statistician, 31, February 1977, pp. 12-17.
• Iritz, Maxine Haren, "Science Fair: Developing a Successful and Fun Project," TAB Books (division of McGraw-Hill), 1987.
• Krieger, Melanie Jacob, "Means and Probabilities: Using Statistics in Science Projects," Franklin Watts, 1996.
• Landwehr, James, and Watkins, Ann, "Exploring Data," Dale Seymour Publications, 1996, Teacher's Edition, Revised.
• Tanur, J., Mosteller, F., Kruskal, W., Lehmann, E., Pieters, R., and Rising, G.,"Statistics: A Guide to the Unknown," Wadsworth and Brooks, 1989, 3rd edition.

Your Science Fair Storyboard

by John Schollenberger

Let's start from the basic assumption that of course statistics are an integral part of any data- based science project: statistical methods (and thinking) are necessary for handling variation in the data, for reducing and summarizing data, for interpreting data and drawing appropriate conclusions, etc. But, how and where does one present these results? Where is in the report of course, but also on the storyboard. Further, especially on the storyboard, the presentation of the results should include more visuals (information graphics) and less text. Information graphics include statistical graphics plus flow charts, illustrations, maps, diagrams, photographs, and tables.

The purpose of the storyboard is to provide an easily-followed and quickly-comprehended overview-and, to attract interest and attention. One of the best ways to increase the speed and comprehension of the message is to replace long detailed paragraphs with graphics that communicate at a glance. The end result will be improved storyboards for the students, an easier-to-follow presentation of their research, and an easier job for science fair judges to judge the projects.

Instead of focusing on various statistical methods that might be useful for a science project, focus on improving storyboards (and reports) by using insightful statistical graphics-graphs that are easily comprehensible, that draw interest, and that quickly convince. Presenters too often try to post their entire report on a board, page by page. The details, however, should stay in the report. Only the main ideas, results, and conclusions, should be highlighted on the storyboard. Further, the easiest way to convey an idea quickly is with a visual: if it's real, a photograph; if it's abstract, an illustration or diagram; and if it's measurable, a chart or graph. Paragraphs of text should serve mainly as links between the information graphs, or to provide some secondary information.

Graphs vs. tables: Graphs don't convey information with the same precision as tables. Viewers of graphs, however, are more likely to be able to grasp the big picture, which is the intent of the storyboard. Graphs are more easily interpreted, more easily remembered, and more likely to be believed than less-visual alternatives for presenting the same information.

The storyboard cannot, and should not, attempt to provide all of the details-these should stay in the report. The purpose of the storyboard is to attract attention, and to provide an overview of the science project and the results of the research and study. It is not just an enlarged version of the report. However, since a storyboard viewer may wish to follow-up something in your project, relate your storyboard visuals to their place in the report.

Show, rather than tell. One of the best ways to increase the speed and comprehension of your message is to replace long detailed paragraphs with information graphics that communicate at a glance. Use paragraphs of text mainly as links to connect your information graphics, or to provide some additional detail about a preceding or forthcoming graphic. Always consider, however, whether the detail is necessary. It may be needed in the report, but is it necessary on the storyboard?

Flow charts can simplify even the most complicated procedures and sequences (e.g., the process flow of your own research). Flow charts permit you to visually indicate sequences, as well as important landmarks-points where decisions have to be made and the consequences of those decisions. In addition to adding visual interest, flow charts greatly reduce the amount of text required. Even for the report, if you are bogged down trying to describe a process or sequence, flow chart it and save the text for adding explanatory detail.

Use organization charts to describe relationships.

Use time lines to describe "history". Here history refers to the stages of the project or of the sequence of events or outcomes which occur as part of the experiment or research. That is, the project may be carried out in different stages or with measurements being taken at specific times, or there may be changes in what is being observed (e.g., life cycle of a virus of plant). History can also show both internal and external events. For example, it might show both the life of a virus plus the corresponding changes that take place in the body as it responds to infection.

Charts and graphs help readers and viewers quickly interpret the importance of numbers. A paragraph of numbers, no matter how well written, can rarely communicate relationships or trends as effectively as a good chart or graph.

Eliminate unnecessary detail from your charts and graphs. Often, for example, software programs' defaults add unnecessary background vertical and horizontal lines. Remember, exact numeric amounts and relationships are less important when your point is to emphasize trends, for instance.

Color: don't scatter it about; use it to highlight important information; use it to link sections together.

Steps in producing a storyboard:

1. Plan your storyboard. Inventory the text and graphic elements from the report that might be included. Identify the goals (the main message) and, based on that, try to envision the viewer's needs and expectations-what will the viewer need or expect to be able to understand your message or conclusions?
2. Experiment with different layouts. Does it all fit? If not, which parts are needed to convey the main message, to provide the big picture-and which provide secondary information that can be left to the report? Once you have a set of text and visuals that fit, check the flow. Are there any gaps? Review the text portions again. Can some parts be replaced by a table or graphic? Review the graphics. Do they need to be simplified or changed for the storyboard? For example, for the storyboard, you might replace a stem-and-leaf plot in the report with a histogram.
3. Produce a full-size version. Do the pieces fit in a large enough size to be easily readable from four to six feet? Some secondary details may not need to be legible at that distance, but the important points should stand out-even from a distance. Indeed, make sure that your main message is the most prominent feature. Try to give your storyboard "aisle" appeal. Use visuals that will grab the viewers' attention and make them want to learn more about your project.
4. Revise, revise, revise. Again, experiment with alternative layouts. It is unlikely that the first version will be your final choice. When you see your storyboard at full size, you can usually identify many areas for improvement. Also, simplify as much as possible. As you review the text and graphics, constantly ask yourself:
   • Is this word (or sentence) absolutely necessary?
   • Can this paragraph be replaced by a list or table or a graphic?
   • Can this title be rewritten to communicate the same idea in fewer words?
   • Can this legend be removed and the parts labeled directly?
5. Review the goals and expectations. When you are satisfied with your revisions, check your storyboard in terms of your goals and viewer expectations from Step 1. If necessary, start again.
6. Refine and fine tune. Revise as necessary to enhance impact, to show the main conclusions and their justification,and to help the viewer move through the project.

The graphics on a storyboard serve a different purpose than those in a report. Storyboard graphics are mainly for communicating the results and conclusions of an analysis, not for performing the analysis. Thus, they should be more like presentation graphics-intended to grab attention and, in the interest of simplicity and ease of comprehension, perhaps giving up some detail in order to get it.

Taking this discussion a step further, consider the audience. The person passing by a storyboard or poster is perhaps not yet interested in the topic; for them the graphics need to have an impact in order to get their attention. The person reading the report, however, is already interested and is looking for details. Thus, exploratory graphs may be appropriate in the report, but probably not on the storyboard. For science projects the audience may be a little more technical or professional, but, at least for the storyboards, the message should be accessible to almost everyone. Again, it is the message that is important-if a simple graph does the job, then that should be sufficient.