## No relationship between CVinter and CVintra [Power / Sample Size]

Hi Sridhar,

» can any one help how i convert inter subject CV to intra subject CV for sample size estimation.

What do you mean by ‘convert’? It’s like saying ‘convert an apple to an orange’. Both are approximately spherical fruit, more or less sweet and of similar weight. However, they are exclusively distinct.

If you have a crossover study, both

$$\begin{matrix}

s_\textrm{inter}^2=\log_{e}(CV_\textrm{inter}^2+1)\\

s_\textrm{intra}^2=\log_{e}(CV_\textrm{intra}^2+1)\\

CV_\textrm{total}=\sqrt{\exp\left(s_\textrm{inter}^2+s_\textrm{intra}^2-1\right)}

\end{matrix}$$

It’s like asking a pupil

» can any one help how i convert inter subject CV to intra subject CV for sample size estimation.

What do you mean by ‘convert’? It’s like saying ‘convert an apple to an orange’. Both are approximately spherical fruit, more or less sweet and of similar weight. However, they are exclusively distinct.

If you have a crossover study, both

*CV*s are accessible. If you have a parallel design, you get only the total (aka pooled)*CV*(see this article). There is an infinite number of combinations of*CV*_{inter}and*CV*_{intra}which will give the same*CV*_{total}.$$\begin{matrix}

s_\textrm{inter}^2=\log_{e}(CV_\textrm{inter}^2+1)\\

s_\textrm{intra}^2=\log_{e}(CV_\textrm{intra}^2+1)\\

CV_\textrm{total}=\sqrt{\exp\left(s_\textrm{inter}^2+s_\textrm{intra}^2-1\right)}

\end{matrix}$$

It’s like asking a pupil

*“We added two numbers and their sum was five. What were the two numbers?”*

—

Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮

Science Quotes

*Dif-tor heh smusma*🖖Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮

Science Quotes

### Complete thread:

- inter subject CV to intra subject CV Sridhar.E 2021-04-27 14:49 [Power / Sample Size]
- No relationship between CVinter and CVintraHelmut 2021-04-27 15:14