Statistically Valid Sample Size
There is an art as well as a science to determining the proper size of a sample. To do this, you need to set several key factors. The confidence level determines the probability you can accurately judge the difference in response. "Do I need to be correct 95% of the time or is 90% accuracy acceptable?" Then you need to estimate what you think the response rate would be if you marketed all available names. The acceptable error factor is the range of difference that would cause you to change your marketing tactics in the future. In other words, the challenger would have to beat the champion by X% to make it worthwhile. If X = 10%, you would enter an amount equal to: Response Rate x 10% as the acceptable error.
|
