Instructions
Applet Overview: The large ‘jar’ contains 10,000 blue and orange balls. What proportion of the balls are blue?
To know the true proportion of blue balls, it would be necessary to count all the blue balls in the large jar. A large population like this one is often inaccessible due to constraints of time and other resources. Instead, we choose a (much smaller) sample from the jar, compute the proportion of blue balls in the sample, and use that to estimate the proportion of blue balls in the whole population.
Buttons:
- Sample – Use this button to choose a single sample from the large jar. The sample will appear in the small jar and the proportion of blue
balls in the sample is indicated on the plot below. Repeat this process to draw more samples.
- Sample 100 – This button will cause the applet to choose 100 samples from the large jar in rapid succession. Only the last of the 100 will
be shown in the sample jar. The proportions of blue balls observed the samples are plotted on the graph below.
- Clear plot – This button erases the graph and empties the sample jar but does not change the population. Use this button when you want to
look at the sampling distribution based on a new sample size.
- New Population – This button clears the plot, empties the sample jar, and fills the large jar with a new population of balls.
Other components:
- Show p check box – Use this box to show the value of the parameter, that is, the true proportion of blue balls in the population (large jar).
The proportion is displayed beside the check box and is indicated in the plot.
- Sample size radio buttons – Use the radio buttons to set the size of the sample: 50, 100, or 250 balls.
- Plot Type radio buttons – Use these to determine whether to display a histogram or a dot plot. A dot plot is preferable when starting out as it
will enable you to see the specific values of the proportion computed for the various samples. The histogram is most useful when you have chosen
many samples (the dots in the dot plot will eventually reach too high to be contained in the plot).