I’m trying to find the true latent value of r which is the average proportion of a population that was attending/viewing at any given second of an event. The measurement of r is derived from the (sum of the viewing seconds / duration of event) / population size or r = (sum_viewing_seconds / t) / p. I don’t know how they come up with sum_viewing_seconds as it’s dependent on each sampled person’s weight and actual attendance/viewing. Note that the weight of the person is the number of people that person is meant to represent. The only thing I know is the number of people that are used in the sample.
Since there are events measured that use much less people in the sample, I’m skeptical of the measured sum_viewing_seconds and therefore the measured r. Further, I know generally how r should behave based off of the interests of the population and the event in question. I was hoping there would be some standard way of incorporating the sample size into the uncertainty but it doesn’t sound like that is the case.
Hopefully this additional information is helpful.
I’ll try and see if there are any papers in ecology that measure population proportions of events using samples that I can borrow from. Thanks!