Sampling Distribution

Objectives
• Identify the reason/”Why” for conducting random sampling
• Distinguish parameter and statistic
What is the use of “Statistic/s”
• Decision making
• Q: Why is decision making so difficult?
• A: Because we are never truly certain
• “Nakakatakot ang bagay na hindi tayo sigurado kasi alam nating importante
ito sa atin”
• In every decision we make, we let go of other possibilities/opportunities
• Thus, with information we have, we have to decide, but then again,
we’re bound by constraints- the data we have
• With what we have, can we say that’s enough for us to
conclude/decide about something?
Inferential Statistics
• We deal with decision making problems
• We decide/conclude about traits/characteristics of our population based on
data that we’re able to collect from a sample, a subset of population
• Sometimes, kailangan mo rin “mag-assume”
• With sample or information, you have to reach a conclusion that extends
beyond your collected data
• However, it’s impossible to be 100 % certain if your conclusions about your
population are correct or not
Sampling Distribution
• Equips us with tools to quantify the accuracy of our conclusions by
understanding uncertainty.
• Kahit imposibleng maging sigurado, posibleng masabi kung “gaano
kasigurado”
Sampling Distribution: Basic Definitions
• Population- totality of items, things or people under consideration;
relative(i.e. HUMMs students, this class, ASHS students)
• Sample- subset of population, part taken from population(i.e. Male
students in this class, this class(from ASHS classes) )

• Parameter- any measurable characteristic of the population

• Statistic- any measurable characteristic of the sample(i.e. Pulse Asia,
SWS, Nielsen)
Why make use of samples?
1. Saves time and money
2. Physically impossible to get the entire population
3. Some tests are destructive in nature
4. Sample results are sufficient already
Sampling Distribution
• Consider a small population of 20 employees in a certain company.
The amount of money each employee spent on a certain day is shown
in the table below:
Sampling Distribution
• For this population, the mean 𝜇 = 374.60 pesos.
• Suppose that we do not now this value of the population mean and
ത
we just want to estimate it by computing for the sample mean, 𝑋,
that is the mean of the sample.
• For instance, we get a sample of size 5 and call it as SAMPLE 1.
Sampling Distribution

ത for Sample 1 is 419.40 pesos.

• The sample mean, 𝑋,
• This is much bigger than the population mean, 𝜇 = 374.60
• Is this difference typical or is the sample mean for Sample 1 just
unusually far from the population mean, 𝜇?
Sampling Distribution
• We note the following:
• The values of the sample means vary from sample to sample.
• Some sample means are larger than the population mean while
some are smaller.
• Some sample means are closer to the population mean while others
are farther
Sampling Distribution
• We can investigate further by getting more samples, and for each
ത
sample we get their respective means, 𝑋.
Sampling Distribution
• But when we get the mean (or average) of all the 50 sample means, it
is 376.6 which is very close to the population mean of 374.6

• If we get all sample of size 5, not just 50 samples, the mean or

average of all the possible samples is the value of the population
mean.
Sampling Distribution
• The sample mean is actually a function of the random variable X =
amount spent by an employee on a particular day.

• So the sample mean, which is a statistic or measure, being a function

of a random variable in in itself a random variable.

• This implies that the sample mean, being a random variable also has
a probability distribution.
Population vs. Statistic
• Population includes all the elements from a set of data.

• Sample consists of one or more observations from the population.

• A measurable characteristic of a population, such as a mean or

standard deviation, is called a parameter while for a sample, we call
the measurable characteristic as a statistic.
Parameter vs. Statistic
• The population mean is denoted by the symbol 𝝁, but the sample
ഥ.
mean is denoted by 𝑿

• The standard deviation of a population is denoted by the symbol 𝝈;

but the standard deviation of a sample is denoted by the symbol s.

• The formula for the standard deviation of a population is DIFFERENT

from the formula for the standard deviation of a sample.
Statistic
• Suppose (𝑋1 , 𝑋2 , … , 𝑋𝑛 ) is a random sample. A statistic is a random
variable that is a function of 𝑋1 , 𝑋2 , … , 𝑋𝑛 .
• Example:
a. The sample mean is a random variable that is a function of
𝑋1 , 𝑋2 , … , 𝑋𝑛 . Thus, it is a statistic.
• The formula for the sample mean, 𝑿 ഥ is given by:
n

åX i
X= i=1
n
Statistic
• Suppose (𝑋1 , 𝑋2 , … , 𝑋𝑛 ) is a random sample. A statistic is a random
variable that is a function of 𝑋1 , 𝑋2 , … , 𝑋𝑛 .
• Example:
a. The sample mean is a random variable that is a function of
𝑋1 , 𝑋2 , … , 𝑋𝑛 . Thus, it is a statistic.
• The formula for the sample mean, 𝑿 ഥ is given by:

å( X )
n 2

i
-X
S2 = i=1
n -1
Statistic as Random Variable
• This definition of a statistic is consistent with our previous definition
of “statistic” as a measure describing the sample. Now, however, we
are viewing the statistic as a random variable.

• As a random variable, the statistic is a function whose value depends

on the outcome of a random experiment.

• The random experiment here is the drawing of a sample.

• The value of a statistic will then vary from sample to sample.

Sampling Distribution: Remarks
• We can only determine the realized value of a statistic once we have
actually selected our sample. But, it is impossible to predict with certainty
what its realized value will be.

• Since the statistic is random variable, then it must have a probability

distribution.

• This probability distribution is what we call a sampling distribution.

• Even if we cannot predict with certainty what the value of a statistic will
be, we can use its sampling distribution to understand its behavior in
probabilistic terms.
Sampling Distribution
• The probability distribution function of a statistic is called its sampling
distribution.

• If the statistic is discrete random variable, then its sampling

distribution is its probability mass function.
• If the statistic is a continuous random variable, then its sampling
distribution is its probability density function.
Sampling Distribution
• Sampling distribution of a statistic depends on various factors. These
are the:
• Sample size, n
• Method of choosing the random sample
• Population under study

• We need to consider these factors whenever we evaluate the

performance of the statistic we will be using in making our inferences.
Sampling Distribution
• The standard deviation of the sampling distribution is called the
standard error of the statistic. It tells us the extent to which we
expect the values of the statistic to vary from different possible
samples.

• The probability distribution of the sample mean is called the sampling

distribution of the mean.
Standard of Error
• A small standard error indicates that the computed values of our
statistic in the different samples generated are close to one another.
• So that even if we know that the value of a statistic varies from one
sample to another, a small standard error gives us an assurance that
at least the variation among their values is not too large.
• A small standard error assures us that the different realized values of
the statistic under repeated sampling are very close to the average
value of the statistic.
Sampling Distribution: Population
• Consider 4 observations making up the population values of a
random variable X having the probability distribution.

• Compute for E(X)

• Compute for Var(X)
Sampling Distribution: w/replacement
2. Suppose we have four numbers, which are 0, 1, 2 and 3.
We list all possible samples of size 2, with replacement, and for
each sample compute for the value of the sample mean, 𝑋.ത
No. Sample ഥ
𝑿 No. Sample ഥ
𝑿
1 0, 0 9 2, 0
2 0, 1 10 2, 1
3 0, 2 11 2, 2
4 0, 3 12 2, 3
5 1, 0 13 3, 0
6 1, 1 14 3, 1
7 1, 2 15 3, 2
8 1, 3 16 3, 3
Sampling Distribution: w/replacement
2. Suppose we have four numbers, which are 0, 1, 2 and 3.
We list all possible samples of size 2, with replacement, and for
each sample compute for the value of the sample mean, 𝑋.ത
No. Sample ഥ
𝑿 No. Sample ഥ
𝑿
1 0, 0 0.0 9 2, 0 1.0
2 0, 1 0.5 10 2, 1 1.5
3 0, 2 1.0 11 2, 2 2.0
4 0, 3 1.5 12 2, 3 2.5
5 1, 0 0.5 13 3, 0 1.5
6 1, 1 1.0 14 3, 1 2.0
7 1, 2 1.5 15 3, 2 2.5
8 1, 3 2.0 16 3, 3 3.0
Sampling Distribution of sample mean:
w/replacement
ഥ
𝑿 0 0.5 1 1.5 2 2.5 3

ഥ)
𝒇(𝑿 1/16 2/16 3/16 4/16 3/16 2/16 1/16
Sampling Distribution of sample mean:
w/replacement
ഥ
𝑿 0 0.5 1 1.5 2 2.5 3

ഥ)
𝒇(𝑿 1/16 2/16 3/16 4/16 3/16 2/16 1/16
Sampling Distribution: w/o replacement
3. Suppose we have four numbers, which are 0, 1, 2 and 3.
• We list all possible samples of size 2, without replacement, and
ത
for each sample compute for the value of the sample mean, 𝑋.
No. Sample ഥ
𝑿 No. Sample ഥ
𝑿
1 0, 1 7 2, 0
2 0, 2 8 2, 1
3 0, 3 9 2, 3
4 1, 0 10 3, 0
5 1, 2 11 3, 1
6 1, 3 12 3, 2
Sampling Distribution: w/o replacement
3. Suppose we have four numbers, which are 0, 1, 2 and 3.
• We list all possible samples of size 2, without replacement, and
ത
for each sample compute for the value of the sample mean, 𝑋.
No. Sample ഥ
𝑿 No. Sample ഥ
𝑿
1 0, 1 0.5 7 2, 0 1.0
2 0, 2 1.0 8 2, 1 1.5
3 0, 3 1.5 9 2, 3 2.5
4 1, 0 0.5 10 3, 0 1.5
5 1, 2 1.5 11 3, 1 2.0
6 1, 3 2.0 12 3, 2 2.5
Sampling Distribution of sample mean: w/o
replacement
ഥ
𝑿 0.5 1 1.5 2 2.5

ഥ)
𝒇(𝑿 2/12 2/12 4/12 2/12 2/12

ത
• We now solve for E(𝑋)
Sampling Distribution of sample mean: w/o
replacement
ഥ
𝑿 0.5 1 1.5 2 2.5

ഥ)
𝒇(𝑿 2/12 2/12 4/12 2/12 2/12

ത
• We now solve for Var(𝑋)
Sampling Distribution of Sample Mean
• In summary,
• For the original random variable X,
• 𝐸(𝑋)=3/2
• 𝑉𝑎𝑟(𝑋)=5/4
• Taking samples of size 2 with replacement,
• 𝐸(𝑋)=3/2
• 𝑉𝑎𝑟(𝑋)=5/8
• Taking samples of size 2 without replacement,
• 𝐸(𝑋)=3/2
• 𝑉𝑎𝑟(𝑋)=5/12
Theorem 1
• If all possible random samples of size n are drawn with replacement
from a finite population of size N with mean μ and standard deviation
σ, then the sample mean 𝑿 ഥ will have mean and variance given by:

• Taking samples of size 2 with replacement,

• 𝐸(𝑋)=3/2
• 𝑉𝑎𝑟(𝑋)=5/8
Theorem 2
• If all possible random samples of size n are drawn without replacement from a
finite population of size N with mean μ and standard deviation σ, then the
sample mean 𝑿 ഥ will have mean and variance given by:

• Taking samples of size 2 without replacement,

• 𝐸(𝑋)=3/2
• 𝑉𝑎𝑟(𝑋)=5/12