Section
Section
Up to this point, the probabilities we have found have been based on individuals in a sample, but suppose we
want to find probabilities based on the mean of a sample. In order for us to find these probabilities we need
to know determine the sampling distribution of the sample mean. Knowing the sampling distribution of the
sample mean will not only allow us to find probabilities, but it is the underlying concept that allows us to
estimate the population mean and draw conclusions about the population mean which is what inferential
statistics is all about.
Sampling Error : The error resulting from using a sample to estimate a population characteristic.
For a variable x and a given sample size n, the distribution of the variable x
̅ (all possible sample means of size
n) is called the sampling distribution of the mean.
Note: The larger the sample size the smaller the sampling error tends to be in estimating a population mean,
, by a sample mean x
̅.
Mean of x
̅: denoted μx̅
Example:
The following data represent the ages of the winners (age, in years, at time of award given) of the Academy
Award for Best Actress for the years 2012 – 2017.
69
b) The following table consisting of all possible samples with size n = 2 and calculate their corresponding
means.
Sample Mean Sample Mean Sample Mean Sample Mean Sample Mean Sample Mean
62, 62 22, 62 44, 62 54, 62 26, 62 28, 62
62, 22 22, 22 44, 22 54, 22 26, 22 28, 22
62, 44 22, 44 44, 44 54, 44 26, 44 28, 44
62, 54 22, 54 44, 54 54, 54 26, 54 28, 54
62, 26 22, 26 44, 26 54, 26 26, 26 28, 26
62, 28 22, 28 44, 28 54, 28 26, 28 28, 28
c) Calculate the mean of the sampling distribution of the mean, μx̅ . _________________________
(i.e. calculate the mean of the sample means)
Note: The above example is exactly that an example, it is not a proof of μx̅ = μ.
Standard deviation of x
̅ denoted σx̅
For samples of size n, the standard deviation of the variable x
̅ equals the standard deviation of the
population under consideration divided by the square root of n.
σ
So, σx̅ = σx̅ is smaller than
√n
σx̅ is sometimes called the standard error of the mean.
b) Calculate the standard deviation of the sampling distribution of the mean, σx̅ . ______________________
(i.e. calculate the standard deviation of the sample means)
σ
c) Use the formula σx̅ = to calculate σx̅ . ____________________
√n
70
The Sampling Distribution of the Sample Mean for a Normally Distributed Variable
Suppose that a variable x of a population is normally distributed with a mean and a standard deviation .
̅ is also normally distributed and has mean and standard deviation
Then, for samples of size n, the variable x
𝛔
𝛔𝐱̅ = .
√𝐧
Of course, not all distributions are normal, but given certain conditions we can assume the variable x
̅ is
approximately normally distributed regardless of the distribution of x. This leads to one of the most important
theorems in statistics; the central limit theorem.
Central Limit Theorem (CLT) For a relatively large sample size (n > 30), the variable x
̅ is approximately
normally distributed, regardless of the distribution of the variable under consideration. The approximation
becomes better with increasing sample size.
Population Sample of n = 3
1.0
0.8
Probability
Probability
0.6
0.4
0.2
1 2 3 4 0.0
1.0 1.5 2.0 2.5 3.0 3.5 4.0
1.00 1.24 1.48 1.72 1.96 2.20 2.44 2.68 1.40 1.54 1.68 1.82 1.96 2.10 2.24
71
The Sampling Distribution of the Sample Mean
Suppose that a variable x of a population has mean, and standard deviation, . Then, for samples of size n,
2) The standard deviation of x̅ equals the population standard deviation divided by the
σ
square root of the sample size, in other words: σx̅ =
√n
3) If x is normally distributed, so is x
̅, regardless of sample size
σ
Therefore, we can say, x
̅ is normally distributed with parameters μx̅ and σx̅ , where μx̅ = μ and σx̅ = .
√n
Note: Since the sampling distribution of the sample mean is normally under certain conditions you can use
the normal approximation to find probabilities, therefore you need convert x
̅ to a z-score.
x̅− μx̅
Converting x
̅ to a z-score: z= σx̅
Examples:
1) The times that college students spend studying per week have a distribution that is right skewed with a
mean of 8.4 hours and a standard deviation of 2.7 hours. Suppose a random sample of 45 students is
selected.
a) What is the sampling distribution of the mean number of hours these 45 students spend studying per
week? (i.e. What is the sampling distribution of x
̅?)
n= _______
µ = ______
σ = ______
b) Find the probability that the mean time spent studying per week is between 8 and 9 hours.
c) Find the probability that the mean time spent studying per week is greater than 9.5 hours.
72
2) At an urban hospital the weights of newborn babies are normally distributed, with a mean of 7.2 pounds
and standard deviation of 1.2 pounds. Suppose a random sample of 30 is selected.
a) What is the sampling distribution of the mean weight of these newborn babies?
(i.e. What is the sampling distribution of x
̅?)
n= _______
µ = ______
σ = ______
b) Find the probability the mean weight is less than 6.9 pounds?
c) Find the probability the mean weight is between 6.5 and 7.5 pounds?
3) A battery manufacturer claims that the lifetime of a certain battery has a mean of 40 hours and a standard
deviation of 5 hours. A simple random sample of 100 batteries is selected.
b) What is the probability the mean life is less than 38.5? Would this be unusual?
73