Statistical Significance for a single patient

39 Views Asked by At

I have the mean values for a blood test from only 1 patient before and after treatment. My supervisor asked me to find statistical significance. I have the mean value of 4 parameters. Would it make sense to find statistical significance of a single patient? If so, what kind of test should I perform?

1

There are 1 best solutions below

0
On

Since you have only $1$ measurement it is meaningless to ask about statistical significance. However, if you have a model of the population distribution of those parameters (for instance, suppose one knows that members of the general population have their glucose level approximately normally distributed with some known mean and standard deviation) then you can compare the absolute value of the before/after difference to the population standard deviation as a proxy for statistical significance. A better way to phrase your question involves formulating a null hypothesis (that there is no change from before/after). Only if the difference is significantly larger than the population standard deviation could one gain evidence to reject this null hypothesis.