What are the conditions of your experiment? If you took 30 readings under identical conditions, you can take average values of G' and G'' and take a ratio between them. But be careful, you might have to ignore some points if your sample took a while to equilibrate, or was changing with time.
But you say "should I calculate it before the yield stress?", which implies you are changing conditions through the experiment - whether stress, or strain, or strain rate, or possibly temperature. In this case it makes no sense to average the 30 points. Just calculate the ratio G'/G'' at each point and plot it against your independent variable.