STATISTICAL PROJECT

	DEFINITION AND PURPOSE A statistical project is an extremely powerful teaching tool that integrates numerous concepts and skills presented in the classroom. This article gives some basic guidelines for developing a statistical project. A statistical project is the process of answering a research question using statistical technique and presenting the work in a written report. The research question may arise from any field of scientific endeavor, such as athletics, advertising, aerodynamics, nutrition. It differs from a statistical poster in that a written report is used to present the findings.   A team of 3 – 6 people can be used to investigate the question(s). DATA-BASED PROBLEM SOLVING The process of developing a statistical project should demonstrate the scientific method of solving a problem: 1.    Pose a focused question or questions 2.       Collect appropriate data 3.       Analyze the data intelligently 4.       Draw correct conclusions Because you are continually asking questions about things that touch your life, you should have little trouble generating questions about yourself, your school, your family, your neighborhood, or interesting phenomena in the world. Once a question is proposed, you should examine it. First, is it a question that can be answered ? ("Is there intelligent life in the universe that does not come from Earth?" is an extremely interesting question, but not one that is likely to be answered in a short-term project.) Second, can you collect data to answer the question ? Or has someone else already collected data that could be used to find the answer ?

DATA-BASED PROBLEM SOLVING

The process of developing a statistical project should demonstrate the scientific method of solving a problem:

1.       Pose a focused question or questions

2.       Collect appropriate data

3.       Analyze the data intelligently

4.       Draw correct conclusions

Because you are continually asking questions about things that touch your life, you should have little trouble generating questions about yourself, your school, your family, your neighborhood, or interesting phenomena in the world.

Once a question is proposed, you should examine it. First, is it a question that can be answered ? ("Is there intelligent life in the universe that does not come from Earth?" is an extremely interesting question, but not one that is likely to be answered in a short-term project.) Second, can you collect data to answer the question ? Or has someone else already collected data that could be used to find the answer ?

Once the question is chosen, data must be collected. If published data are used, you must  write down where they were obtained and record the source.

You will collect your own data. Time should be spent deciding how to collect it. If a survey is used, how are the people chosen to answer the questionnaire ? If two treatments (treatments could be models, grades, genders, etc.) are to be compared, how can comparisons be made fairly? How will the data be recorded? After the details have been worked out, then you are ready to take the data. Great care should be exercised at every stage of data collection. Remember that careless measurement or recording of data cannot be remedied in the analysis phase of the project.

Intelligent analysis of the data may take many forms and should be guided by the question and how the data were collected. Usually it is best to begin by graphing the data. Can graphs be used to give the answer to the question or questions? Graphics is one method of data analysis and use  some  statistical methods such as a chi-squared test or a t-test may be used. Regression has been used occasionally. Sometimes, estimation is most appropriate, and hypothesis testing is not needed. Other methods may be used depending on the question and the data.

Once the analysis is complete the question should be answered correctly. The data may not be able to provide a conclusive answer. For example, one treatment may appear to be better than another, but the difference was not significant. If the question has a definitive answer, then that should be presented. A check should be made at this point to make certain that the answer matches the question. It is easy to get "caught up" in the analysis phase and obtain many answers, none of which addresses the research question.

Finally, consider the strengths and weaknesses of the project. What went right? What went wrong? What would be changed if it was done again?

THE WRITTEN REPORT or WEB PAGES

Great latitude may be taken in developing the written report or web pages. You may use movies, power point presentations, and computer software such as fathom.  Plan how to communicate your work effectively. The longest report does not necessarily represent the best project. A web page or Power Point presentation will be worth more points than a written paper.  A movie is extra credit.  However, the report MUST do the following:

1.       Demonstrate how and why the particular topic was chosen.

2.       Show how the research was conducted.

3.       Delineate what conclusions were obtained.

4.       Include the collected data and analysis of the data.

5.       Discuss the strengths and weakness of the selected statistical methods.

A FINAL NOTE

The NCTM Standards for Curriculum and Evaluation in School Mathematics presents the vision that problem solving is a primary goal of mathematics instruction and recommends student involvement in statistical activities at all grade levels. According to the Standards, statistical thinking should begin in the primary grades with the creation of student data from class activities. In upper grades, collecting, organizing, summarizing, and interpreting data from other school disciplines such as the physical or the social sciences, as well as outside interests of the students, are emphasized. The statistical project is a powerful tool for attaining these goals while exercising essential communication skills.

Project requirements

A panel of classroom teachers of statistics and statisticians will assess each project according to six broad areas:

·         Question of interest
Is the question of interest focused, clearly stated and relevant?

·         Research design and data collection
Can the data, as collected, answer the question? Were the data collected in an appropriate manner? Were data collected directly by the students?

·         Analysis of data
Is the analysis appropriate for the design?

·         Conclusions
Are conclusions consistent with analysis? Has the question been answered?

·         Reflection on process
What went right? What went wrong? What would you do differently? Are there any suggestions for further study?

·         Final Presentations
Is the report well organized and presented? Does the project display creativity and/or originality? Are supporting graphs and charts carefully prepared?