The NCTM Standards state that: Probability, the measure of the likelihood of an event, can be determined theoretically or experimentally. Students in the middle grades must actively participate in experiments with probability so that they develop an understanding of the relationship between the numerical expression of a probability and the events that give rise to these numbers (e.g., 2/5 as it relates to the probability of choosing a red marble from a hat)…. Students have many misconceptions and poor intuitions about probabilistic situations. In order to bring these ill-formed notions to the conscious level so that they can be confronted, students should be asked to guess what will happen next or what the result of the experiment will show. An unexpected result has a much greater potential to cause students to rethink their basic assumptions if they have articulated their ideas before their experiment or analysis of the situation

With this statement in mind, educators in the Windsor Public Schools have designed the following experiment for their 5^th grade students.

Students will predict the number of times heads will show in repeated flips (10, 100 or more) of a United States penny (participating classrooms from outside of the United States will use their own local coin if U.S pennies are not available) and will compare their results with their predictions. They discuss the fact that flipping a head and flipping a tail is equally likely (each has the probability of ½). Students will see that although the theoretical probability of flipping a head is ½, in actual trials the experimental probability may be very different. They will investigate the probability of larger amounts of flips by gathering data from around the world via the Internet: a penny will be flipped 100 times and participating classrooms will pool the collected data so that comparisons and better predictions can be made. By combining all the data and having a very large number of flips, students will then understand that probability predicts the behavior of random systems over the long run. Also, as we increase our pool of data through use of the Internet, we will get closer to the theoretical outcome of ½.

Technology will allow students from classrooms around the world to submit, share and analyze data in a timely manner via the Internet at this web site. In addition this site will also allow participants from around the world to learn about other cultures throughout the United States and the world. E-mail will provide for the interaction between classrooms in the sharing of outcomes of the data collected.