Statistical Components of a Good Science Fair Project and Statistical Judging Criteria
Gathering, analyzing and learning from data is an important part of a complete science fair assessment and report. The following are some essential statistical components of a good science project. Keeping these in mind will help you to start thinking statistically!
- Objective & hypothesis: Except in rare instances, the project must have a clearly defined goal and prediction as to what will happen as the project develops. These can be formed by reviewing previous research and making conclusions based on the results.
- Project design: Where appropriate, the project should have appropriate randomization of samples to ensure accurate data collection. This is a critical piece of statistical science that is applied to everything from polls to medical studies.
- Data collection procedure: Without data, a project provides nothing from which to draw conclusions. The procedure used to collect the data will help determine its accuracy. Statisticians spend a great deal of time determining the best way to collect and report data.
- Data summary: Use descriptive statistics to recap the project’s results.
- Analysis: What do the data say? Is the data collection complete? Should additional data points have been collected? The answers to these questions require applying statistical methods, including formal hypothesis testing procedures.
- Visualization: Bring the data to life with graphics that help tell the story of the project and demonstrate the conclusions. This is an important part of many statisticians’ work.
- Conclusions: Once statistical analysis and visualization help make the conclusions clear, be sure to detail them in the report. This is where your statistical thinking can really shine!
- Presentation: Clear communication is important in the science of statistics. The presentation of the process and findings of your project, through oral explanation and visuals, is important. How well you understand and can explain the statistical procedures used is essential.
While the judging criteria for science fair projects varies across regions, the following criteria are representative of those used for statistical considerations. All of these criteria are anchored in statistical thinking:
- Design and Data Quality
How were the data collected? Was randomness used correctly and appropriately in sampling, assignment, and/or simulation? Is the data collection documentation (often contained in a lab book) complete and appropriate? - Descriptive Statistics, Graphs and Visualization
Are the data summarized using appropriate descriptive statistics in clear and understandable tables and displays? Are appropriate graphs used and are they understandable on their own? Does the data visualization support the design and results of the study, and the conclusions? - Inference/Modeling
Were appropriate statistical techniques were used to construct statistical tests, estimates and/or models? Were the methods applied in correct and appropriate ways? Were diagnostics considered to explore statistical conditions and the quality of the fit of models? - Conclusion
Is the statistical evidence substantial enough to warrant mention in the conclusions to the project? Are the statistical elements of the conclusions communicated correctly?