Sample size determination for observational studies

Dear all,

Relatedly to this thread I was wondering if any of you could kindly direct me to articles / guidelines offering insights into SSD – whether (and preferably to me) methodological or applied – with observational studies informing causal inference. I’ve been looking on major stat journals but with no luck.

Thank you in advance for any pointer or opinion!

how much relevance does power estimation have for obs studies? isnt it mostly ‘convenience samples’? incidentally, the IMS has a new journal on obs studies, in case youre not aware of it already: Observational Studies Home

1 Like


You might want to review the following two threads:


as they cover this topic and related issues.

1 Like

Surely any attempt to estimate power for an observational study is doomed to failure.

1 Like

Check out section 5.6.5 in Frank’s Biostatistics for Biomedical Research (pdf) and use estimate precision instead. There are some references that will help justify a precision calculation when your audience is expecting power.

Also read the following thread for a rigorous observational study design:

1 Like

Many thanks to those who’ve contributed with various pointers and comments; I appreciate it.

You can check this ref.:

and there is also part two for binary outcome.