Statistics: SD vs. SE/CI

[Function]

Hello everyone

 

Quick one; I don't seem to find what is preferred in reporting statistics: mean (SD) or mean (95 % CI)?

 

Basically one can be calculated from the other, so it shouldn't matter too much, but what is more commonly used or favourable?

 

Thanks.

 

F

[Prometheus]

SDs and CIs are indicators of subtly different things so it depends on what you want to report on.

 

If you simply want to report on how variable some measurements are around the mean go for SD. In this case you are not making a statement about the population mean, just describing your sample. But if you want to report on the uncertainty of the point estimate for the mean then go for CI: you are then making an inference on the population parameter (the 'true' mean).

[Function]

CI it is then. Thanks!