BIOL 300 Chapter Notes - Chapter 4: Sampling Distribution, Confidence Interval, Statistical Parameter
May 22nd, 2018 BIOL 300
Week 1 Reading
Chapter 4: Estimating with Uncertainty
- The saple ea ( is used to estiate the true ea of the populatio µ)
- The sample standard deviation (s) is used to estimate the population standard deviation
(õ)
- The sample proportion (ñ) is an estimate of the population proportion (p)
- For an estimate of a population parameter to be useful, we also need to quantify its
precision. We need a measure of how far the estimate is likely to be from the target
parameter being estimated. If precision is high, then our uncertainty is low. We can be
reasonably confident that our estimate is close to the truth. If instead precision is low,
then our uncertainty is high ad e’ll eed ore data to redue it.
4.1 The sampling distribution of an estimate
- Estimation is the process of inferring a population parameter from sample data
o The value of an estimate calculated from data is almost never exactly the same
as the value of the population parameter being estimated, because sampling is
influenced by chance
- We use the sampling distribution of the estimate, which is the probability distribution of
all the values for an estimate that we might have obtained when we sampled the
population
o The sampling distribution is the probability distribution of all values for an
estimate that we might obtain when we sample a population
o The saplig distriutio represets the populatio of alues for a estiate
▪ It is not a real population
o Rather, the sampling distribution is an imaginary population of values for an
estimate
- Increasing sample size reduces the spread of the sampling distribution of an estimate,
increasing precision
4.2 Measuring the uncertainty of an estimate
- The standard deviation if the sampling distribution of an estimate is called the standard
error
o Because it reflects the differences between an estimate and the target
parameter, the standard error reflects the precision of an estimate
o Estimates with smaller standard errors are more precise than those with larger
standard errors
o The standard error of a estiate is the stadard deiatio of the estiate’s
sampling distribution
- We represet the stadard error of the ea ith the sol ō( because of its
remarkal straightforard relatioship ith ō the populatio stadard deiatio of Y)
find more resources at oneclass.com
find more resources at oneclass.com
o Standard error decreases with increasing sample size, according to the equation
above
- The troule ith the forula for the stadard error of the ea ō() is that we almost
eer ko the alue of the populatio stadard deiatio ō, ad so e aot
calculate it
o The next best thing is to approximate the standard error of the mean by using
the saple stadard deiatio s as a estiate of ō
o To show that this is an approximation, we use SE(
o According to this relationship, all we need is one random sample to approximate
the spread of the etire saplig distriutio for (
o The standard error of the mean is estimated from data as the sample standard
deviation (s) divided by the square root of the sample size (n)
4.3 Confidence intervals
- The confidence interval is another common way to quantify uncertainty about the value
of a parameter
o It is a range of number calculated from the data that is likely to contain within its
span the unknown value of the target parameter
o A confidence interval is a range of value surrounding the sample estimate that is
likely to contain the population parameter
- An example is the 95% confidence interval for the mean
o This confidence interval is a range likely to contain the value of the true
population mean µ
o It is calculated from the data and extends above and below the sample estimate
(
- For the 95% confidence interval, you have a lover limit and an upper limit. This
alulatio allos us to sa e are 95% confident that the true mean lies between ____
ad _____
o All numbers falling between the lower and upper bounds of a confidence interval
can be regarded as the most plausible values for the parameter, given the data
sampled
- In general, the width of the 95% confidence interval is a good measure of our
uncertainty about the true value of the parameter
o If the confidence interval is broad, then uncertainty is high and the data are not
very informative about the location of the population parameter
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
The sa(cid:373)ple (cid:373)ea(cid:374) (cid:894)((cid:895) is used to esti(cid:373)ate the true (cid:373)ea(cid:374) of the populatio(cid:374) (cid:894) ) The sample standard deviation (s) is used to estimate the population standard deviation. The sample proportion ( ) is an estimate of the population proportion (p) For an estimate of a population parameter to be useful, we also need to quantify its precision. We need a measure of how far the estimate is likely to be from the target parameter being estimated. If precision is high, then our uncertainty is low. We can be reasonably confident that our estimate is close to the truth. If instead precision is low, then our uncertainty is high a(cid:374)d (cid:449)e"ll (cid:374)eed (cid:373)ore data to redu(cid:272)e it. It is not a real population: rather, the sampling distribution is an imaginary population of values for an estimate. Increasing sample size reduces the spread of the sampling distribution of an estimate, increasing precision.