This web page contains a live and functional spreadsheet created with
SpreadsheetConverter
. Please enable JavaScript in your web browser for the live calculations to work properly.
Sample size calculator for estimating a proportion & with a detectable change to a known proportion
Sample size for detetcting a change
Known proportion P
_{0}
This is the known proportion,
known from other sources than from a sample survey
by Ruilin Ren, PhD
Type I error (α)
This is the type one error for a two-tailed test; the default value is set to 0.05 if not provided
ruilin.ren@icf.com
This is the power of the test; the default value is set to 80% if not provided
This is the net sample size needed for detecting the difference between P and P0 if the desired sample size is not specified; otherwise, it gives the power of the test.
Current estimated proportion P
This cell is for the reference indicator which must be a decimal number.
Specified
size
This cell is for the desired sample size which is in priority.
Target population size
This cell is for the target population size when it is small.
Estimated Deft
This cell is for design effect Deft which is the square root of Deff. It is necessary for a cluster or multi-stage sampling.
Sample Size
95% confidence limits of P
Individual
Lower-limit
Upper-limit
You can specify the desired relative standard error here other than the above given ones
* Only the light green colored cell (for current P) must be filled. Input for other dark green colored cells are optional.
* The sample size calculated/you specified are net sample size without taking non-response into account.
* The sample size calculated for detecting a change is based on a two-sided test with
α=
5% and β=80%.
Print All Information
Please host the converted page in server to use Print All.
Update
Reset
Print All