Begin by asking your child to come up with ideas for surveys. Surveys often involve asking
people questions. They can also involve simple observations. Surveys are related to chances
because they involve only a sample. You ask a sample of people questions, but you don't ask
all people. Chance comes into play because chances are that a larger group will give the same
answers or results as the small group sampled. The match between groups is more likely when
the sample is large.
Any survey needs to have a limited number of possible responses. This is important to keep
in mind when coming up with survey questions and thinking about possible responses. For example,
a survey of the hair color of people walking down the street on a particular day at a specific
time has a limited number of responses: black, brown, blond, gray, white, red. However, there
could be the occasional purple or green hair. These colors could be put in a category of “other.”
It's best to begin with surveys that have only two possible responses, like rain and no
rain. Chances are easier to calculate that way. Here are some ideas for survey questions that
have just two responses. Some are to be asked; others can be observed in a public place.
 | |
|
| Do you like eggplant? (yes, no) |
| |
|
| Are you older or younger than 25? (older, younger) |
| |
|
| Are people wearing shoes or sandals? (shoes, sandals) |
| |
|
| Are people walking to the left or to the right? (left, right) |
As your child gathers data, by either asking people questions or just watching them in
a public place, she needs to keep the data organized. Making tally marks in two columns is
an easy and quick way to do it. Once your child gathers enough data (and you might want to
decide how much is enough beforehand — 30? 50? 100 people?), chances can then be calculated.
Calculating the chances can be done by dividing the number of people who answered each
way by the total number of people. (Remember that a result of 0.125 is the same as 12.5%.)
This calculation will give a percent for each answer, and the two percents should add up to
100% (all the people).
You can use the data to predict the chance of the next person’s response. For example,
if 28% of the people who passed by are wearing sandals, then there is a 28% chance that the
next person who walks by will be wearing sandals.
