The Precision dialog is shown below. Here, you enter the information that Maritz Stats needs to compute either the precision that a particular base size provides, or the base size needed to achieve a particular precision.
Show me how to use it! To see an example of how to use precision, click here.
Show me how it works! For more information on the formulas used for precision, click here.
Precision is the margin of error around the test statistic (e.g. mean, proportion). For example, if your mean is 8.5 and the precision is 0.5 at the 95% confidence level, you can be 95% confident that the actual mean (if you measured the entire population) would be between 8.0 and 9.0. In Maritz Stats, you can calculate the precision associated with a particular base size, or you can determine the base size needed to achieve a target precision. Depending on which option you select, you will be prompted to enter the known value (Base or Precision) below.
If you are trying to determine 'precision based on base size,' you will be required to provide the base size. To ensure that the base size is the valid n, ascertain that the base size entered is the same as the base size used in the calculation of the test statistic under consideration (mean or proportion).
If you are trying to determine the 'base size needed to achieve a precision,' you will be required to enter the precision.
In the case of means, you will need to provide the standard deviation of the sample. If this information is unavailable, estimates from prior similar studies may be used.
In the case of proportions, precision can be obtained for the results to a specific survey question summarized in percentages. If this information is unavailable, 50% should be entered. This percent provides the most conservative estimate for calculating precision.
The confidence level indicates the degree of confidence that the interval actually encompasses the unknown population parameter variable. For most applications, the default of 95% should be used. However, you have the option of being more stringent (using the 99% confidence level) or less stringent (90% or 80% confidence level) based on your needs.
The results will present to you the base size needed or the precision achieved, based on the information provided in the form.