Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

What is my mistake when calculating the sample size for my ACBC-study?


I have used the Sample Size Calculator to determine the sample size for my ACBC-study (https://www.sawtoothsoftware.com/support/knowledge-base/general-issues/1544-sample-size-calculator-2).

The values I entered are:
Marging of error = 5%
Confidence level=95%
Population size = 20000
response distribution=30%

According to the Sample Size Calculator the recommended sample size should be 318.

However, when I do the calculation I obtain another recommended sample size which is 147. My calculation is as follows:

x= Z*(c/100)^2 * r * (100-r) = 1.96 * (95/100)^2 * 30 * (100-30)=3714.69

n = (N*x) / ( (N-1) * E^2  + x) = (20000*3714.69) / ((19999)*5^2+ 3714.69) = 147.499

Therefore, may I ask where I did wrong in my calculation?


asked May 17, 2019 by anonymous

1 Answer

0 votes
The formula on the page you cited is NOT applicable for conjoint studies.  It is a general formula for the size of standard errors around a proportion and does not apply to the regression or logit models used on conjoint and choice experiments.  

See the note on that same page:  "If you are conducting a conjoint or discrete-choice survey, this tool may not be relevant. Please see the following article for other things you need to consider when determining sample size: Sample Size Issues for Conjoint Analysis Studies (2009)."
answered May 17, 2019 by Keith Chrzan Platinum Sawtooth Software, Inc. (92,075 points)