Which is better? -- 384 completed
surveys from an organization with 1 million members, or
384 from a membership of 2,000 members.
Believe it or not, both will produce a
95% confidence level with a margin of error of 5%
(assuming a random sample and a reasonable response
rate).
Associations often feel the need to
survey the entire membership at huge expense, and that
becomes the reason why they survey infrequently.
Surveying a sample of the audience can
produce just as meaningful results, at a lower cost,
thus allowing the organization to survey their members
more often.
The budget for a survey can be
calculated based on the acceptable confidence level and
margin of error, and the expected response rate.
Using those numbers, the organization can calculate how
many surveys that must be sent out (randomly).
Knowing the number of surveys distributed and the number
expected to be returned will allow the organization to
calculate the anticipated cost of the project.
After the survey is completed, the
confidence level and margin of error can be calculated
using the actual distribution and response numbers.
If you don't know the equations to do
these calculations yourself, try this on-line calculator.
