Sample Design
Christensen Associates will develop and implement a sampling plan to ensure that the survey responses can be used to make appropriate inferences about the population of interest.
The first step in designing the sample is to determine the population that is of interest. For example, you may want results that reflect the views of all residential customers in a specific region.
The next step is to establish a protocol to ensure that the sample is representative of the target population. This may require a stratification of the sample, in which we over-sample some groups that are of particular interest but contain a relatively small number of potential respondents compared to other groups.
We also consider the mode of the survey in developing the sampling plan. For example, because not everyone has access to the Internet, extra care must be taken to ensure that an online survey sample is representative.
The final step is to determine the appropriate sample size. There is a clear tradeoff between cost and quality in making this decision. It generally takes more resources and sometimes takes more time to obtain a larger number of survey responses. However, larger samples also produce better estimates of customer preferences. We work with the client to set the appropriate targets for each project.
For more information, contact Dan Hansen.
©2010 Laurits R. Christensen Associates