Hello everyone,
We are conducting two questionnaire-based studies with almost 80 items and 50 items in the questionnaire. There are almost 500,000 people in one city and 1 million in the other. What is my sample size for each city? I could not find similar studies to our topic; our topic is a post-occupancy evaluation.
There are many sample size calculators on the internet, most of which determine the sample size according to the degree of precision you want in your estimates. Another alternative is to determine the "power" you need for a statistical estimate, such as a t-Test to compare your two sites, and the most widely used software for that purpose is G*power.
David L Morgan Can I use Morgan's table? and go with 384 as the sample size for 1mil?
A size of 384 for 1 million means that you are 95% likely to get a sample value that it is +/- 5% from the populations value. For example, if the population value for a yes or no question is 50%, then 95% of the time your sample estimate would be between 45% and 55%.
It is up to you to decide if this is an adequate level of precision.
David L Morgan I think that's a good precision for a post-occupancy evaluation. Thank you!
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size