Table of contents | |
Population vs Sample | |
What kinds of numbers are parameters and statistics? | |
Statistical Notation | |
Estimating Parameters from Statistics |
You want to identify the level of support for the death penalty among US residents. Since the population you’re interested in is all US residents, it’s not practical to collect data from the whole population. Instead, you use random sampling to survey a sample of 2000 participants.
Using inferential statistics, you can estimate population parameters from sample statistics. To make unbiased estimates, your sample should ideally be representative of your population and/or randomly selected.
There are two important types of estimates you can make about the population parameter: point estimates and interval estimates.
Both types of estimates are important for gathering a clear idea of where a parameter is likely to lie.
Estimating a population parameter from a sample statistic
In your study on support for the death penalty among US residents, you find that 61% of participants in your sample support the death penalty. To estimate the population parameter, you calculate a point estimate and an interval estimate from your sample statistic.
Your point estimate is your sample statistic – you estimate that 61% of all US residents support the death penalty.
To find the interval estimate, you construct a 95% confidence interval that tells you where the population parameter is expected to lie most of the time. With random sampling, there is a 0.95 probability that the true population parameter for support for the death penalty among US residents lies between 57% and 65%.
58 videos|101 docs|63 tests
|
|
Explore Courses for Commerce exam
|