CogitoErgoCogitoSum
New member
distribution of sample means? The books and sites I read use confusing language... mean sample mean of sample means blah blah. WTF?
There the the population... which presumably we know nothing about. (I have asked questions before about inferring about a population from a sample, no answers)
From that population you randomly pick a sample. Sample size n. This sample has its own mean and its own standard deviation. But we generally dont care about that, do we? Resources often speak of the sample but Im pretty sure they're talking about the sample distribution... which is formed by many repetitions of sampling.
If the samples mean and standard deviation were computed... and the samples were taken many many times... we can plot the distribution formed by sample means and the distribution formed by sample standard deviations. The latter we dont seem to concern ourselves with.
We now we have a distribution formed by many repeated samplings, where we plot the mean of those samples as our random variable.
This distribution of sample means has its own mean and its own standard deviation (third distinct of each Ive mentioned thus far). According to the books Ive read, the mean of the distribution of sample means should be approximately equal to the population mean (which we wouldnt know). How does the standard deviation of the distribution of sample means relate to the population?
I keep seeing the equation s = ?/?n
This seems right to me. EXCEPT. Books tell me explicitly that this equation is used to find the standard error, not the population sd, and we find the standard error from the population sd which we wouldnt yet know.
Does anyone get my confusion? There dont seem to be any reliable online resources for statistics.
There the the population... which presumably we know nothing about. (I have asked questions before about inferring about a population from a sample, no answers)
From that population you randomly pick a sample. Sample size n. This sample has its own mean and its own standard deviation. But we generally dont care about that, do we? Resources often speak of the sample but Im pretty sure they're talking about the sample distribution... which is formed by many repetitions of sampling.
If the samples mean and standard deviation were computed... and the samples were taken many many times... we can plot the distribution formed by sample means and the distribution formed by sample standard deviations. The latter we dont seem to concern ourselves with.
We now we have a distribution formed by many repeated samplings, where we plot the mean of those samples as our random variable.
This distribution of sample means has its own mean and its own standard deviation (third distinct of each Ive mentioned thus far). According to the books Ive read, the mean of the distribution of sample means should be approximately equal to the population mean (which we wouldnt know). How does the standard deviation of the distribution of sample means relate to the population?
I keep seeing the equation s = ?/?n
This seems right to me. EXCEPT. Books tell me explicitly that this equation is used to find the standard error, not the population sd, and we find the standard error from the population sd which we wouldnt yet know.
Does anyone get my confusion? There dont seem to be any reliable online resources for statistics.