|
Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
Standard Errors for Sampling Could you help me please! I am learning about standard errors for sampling and the book I have isn't very well written. Could you please explain it to me in plain English? Hope you can help - cheers!
READERS RESPOND: Re: Standard Errors for Sampling The standard error allows us to make probability statements about the true population mean. Assume that prior to the presidential election a pollster found that support for candidate Bush was 49% among a sample of 1000 voters in Florida. The true support for Bush in Florida, however, is not likely to be exactly 49%. To increase our confidence in our estimate for the true support of Bush we could establish confidence intervals about the 49% estimate. To do so, we use the standard error (mean +/- 1.96 standard error) and the 49% estimate to establish a 95% confidence interval. Assume that the confidence interval turned out to be 49% +/- 3% or 46% to 52%. Now we are more confident in our estimate of the true support for Bush in Florida. We can say that "There is a 95% probability that this confidence interval includes the true population mean." Another interpretation is that if we repeatedly obtained samples of 1000 from the state, 95% of the time the support for Bush would be between 46% and 52%.
Re: Standard Errors for Sampling 1. Within 1 SD of the mean, there are 66% of the population Basically anything that does not fit into these three statements show up standard error, so just follow the rules and you will be fine.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.
|
|||||||||
|