To begin, I suggest creating a purposive sample group, consisting of sixty elementary school students, thirty middle school students, and fourty high school students respectively assuming this is happening in a school district where elementary grades go from Kindergarten through Fifth; middle school is Sixth through Eighth; and High school from Ninth to Twelve grade. The purpose of this is to amass 10 representatives form each grade, all with different math abilities. A diagnostic will be given in schools to test math abilities, and the experiment directors would choose three students who performed exceptionally, three who performed poorly, and four who performed mediocre in effort to create a diverse sample pool, representing all elements of math students from every grade.
Once the final group is selected, they will be split up by grade and will be administrated another diagnostic test, similar to the original ones they originally took. This is the first time the kids will take it out of thirteen trials, making their scores the dependent variable. The independent variable will be the amount of soda consumed by the children.
The ten kids per grade will be split up into groups of two groups of five. Each group will consist of one student, who has exemplified mastery of math, and in some cases two (the groups with fifth, eighth, and high school seniors will have two rather than one) For the three students who performed poorly will be randomly placed in groups either having two or one. Two mediocre students will be given to each group. At the end of this, there should be twenty-four groups of five students. At first everyone will take the above-mentioned diagnostic and it will get graded. After that twelve of