CHAPTER 7---- SAMPLING AND SAMPLING DISTRIBUTIONS

An element is the entity on which data are collected.

A population is a collection of all the elements of interest.
A sample is a subset of the population.
The sampled population is the population from which the sample is drawn.
A frame is a list of the elements that the sample will be selected from.

The reason we select a sample is to collect data to answer a research question about a population.
The sample results provide only estimates of the values of the population characteristics.
The reason is simply that the sample contains only a portion of the population.
With proper sampling methods, the sample results can provide “good” estimates of the population
characteristics.

SELECTING A SAMPLE

Sampling from a Finite Population

 Finite populations are often defined by lists such as:
o Organization membership roster
o Credit card account numbers
o Inventory product numbers
 A simple random sample of size n from a finite population of size N is a sample selected such that each
possible sample of size n has the same probability of being selected.
 Replacing each sampled element before selecting subsequent elements is called sampling with
replacement.
 Sampling without replacement is the procedure used most often.
 In large sampling projects, computer-generated random numbers are often used to automate the
sample selection process.

Example: National Baseball League teams

There are 16 teams that played in 2012

national baseball league. Suppose we
want to select a simple random sample of 5
teams to conduct in-depth interviews about
how they manage their minor legal
franchises.

Step 1: Assign a random number to

each of the 16 teams in the population.
The random numbers generated by Excel’s
RAND function follows a uniform
probability distribution between 0 and 1.

Step 2: Select the five teams

corresponding to the 5 smallest random
numbers as the sample.
Use Excel’s sort procedure to select the five teams corresponding to the five smallest random numbers.
Steps:

1. Select any cell in the range B2:B17

2. Click on Home tab on the ribbon
3. In the Editing group click sort and filter
4. Choose Sort smallest to largest

 Sometimes we want to select a sample, but find it is not possible to obtain a list of all elements in the
population.
 As a result, we cannot construct a frame for the population.
 Hence, we cannot use the random number selection procedure.
 Most often this situation occurs in infinite population cases.

Sampling from an Infinite Population

Populations are often generated by an ongoing process where there is no upper limit on the number of
units that can be generated.

Some examples of on-going processes, with infinite populations, are:

 Parts being manufactured on a production line
 Transactions occurring at a bank
 Telephone calls arriving at a technical help desk
 Customers entering a store

In the case of an infinite population, we must select a random sample in order to make valid statistical
inferences about the population from which the sample is taken.

A random sample from an infinite population is a sample selected such that the following conditions are
satisfied.
Each element selected comes from the population of interest.
Each element is selected independently.

Point Estimation

 Point estimation is a form of statistical inference.

 In point estimation we use the data from the sample to compute a value of a sample statistic that
serves as an estimate of a population parameter.
 We refer to x as the point estimator of the population mean m.
 s is the point estimator of the population standard deviation s.
 p is the point estimator of the population proportion p.

Example: EAI Employee Data

Out of a total number of 2500 employees, a

simple random sample of 30 employees and
corresponding data on annual salary and
management training program participation are
shown in the given table. x1, x2 ….. xn is used to
denote annual salary of the employees and the
participation in the management training program
is indicated by a yes / no.
To estimate the value of population parameter, we
can compute the corresponding characteristic of the
sample, referred to as sample statistic.

Note: Different random numbers will identify different

sample which would result different point estimates.

 x as Point Estimator of m

x=
∑ x i = 1,554,420 = $51,814
n 30

 s as Point Estimator of s
s= √∑ ¿¿ ¿ ¿
 pas Point Estimator of p
x 19
p = = =.6 3
n 30

Practical Advice
o The target population is the population we want to make inferences about.
o The sampled population is the population from which the sample is actually taken.
o Whenever a sample is used to make inferences about a population, we should make sure that the
targeted population and the sampled population are in close agreement.

 The sampling distribution of x is the probability distribution of all possible values of the sample mean x .
 Expected Value of x ------ E( x ) = m
where: m = the population mean
When the expected value of the point estimator equals the population parameter, we say the point estimator is
unbiased.

Standard Deviation of x
Finite Population Infinite Population


σ x=
√ ( )
N −n σ
N−1 √ n
σ x=
σ
√n
A finite population is treated as being infinite if n/N < .05.
 √( N −n)/( N −1) is the finite population correction factor.
 σ x is referred to as the standard error of the mean.

 When the population has a normal distribution, the sampling distribution of �𝑥� is normally distributed
for any sample size.
 In most applications, the sampling distribution of �𝑥� can be approximated by a normal distribution
whenever the sample is size 30 or more.
 In cases where the population is highly skewed or outliers are present, samples of size 50 may be
needed.
 The sampling distribution of �𝑥� can be used to provide probability information about how close the
sample mean �𝑥� is to the population mean m.
Central Limit Theorem
When the population from which we are selecting a random sample does not have a normal
distribution, the central limit theorem is helpful in identifying the shape of the sampling distribution of �𝑥�.

In selecting random samples of size n from a population, the sampling distribution of the sample mean
�𝑥� can be approximated by a normal distribution as the sample size becomes large.

Example: EAI Employee Data

Note: n/N = 30/2500 = .012

Because sample is less than 5% of the population
size we ignore the finite population correction factor.

Suppose the personnel director believes the sample

mean will be an acceptable estimate of population
mean if the sample mean is within $500 of the
population mean.

What is the probability that the sample mean

computed using a simple random sample of 30 employees will be within $500 of population mean?

Step 1: Calculate the z-value at the upper endpoint of the interval.

z = (52,300 – 51,800)/730.30 = .68

Step 2: Find the area under the curve to the left of the
upper endpoint.
P (z < .68) = .7517

Step 3: Calculate the z-value at the lower endpoint of the interval.

z = (51,300 – 51,800)/730.30 = -.68

Step 4: Find the area under the curve to the left of the lower endpoint.
P (z < -.68) = .2483

Step 5: Calculate the area under the curve

between the lower
and upper endpoints of the interval.
P (51,300 < �𝑥� < 52,300) =
P (z < .68) – P (z < -.68)
= .7517 - .2483
= .5034

Use of Excel to Compute Sampling Distribution

of �𝑥�

 Function used - NORM.DIST

 We do not have to make separate
computation of z value.
 Evaluating NORM.DIST function at each
end point of the interval provides cumulative probability at the specified end point of the interval.
 The result obtained using NORM.DIST is more accurate.
Relationship Between the Sample Size and the Sampling Distribution of �𝑥�

Example: EAI Employee Data

 Suppose we select a simple random sample

of 100 employees instead of the 30 originally
considered.
 E ( x ) = m regardless of the sample size. In
our example, E ( x ) remains at 51,800.
 Whenever the sample size is increased, the
standard error of the mean σ x is decreased.
With the increase in the sample size to n =
100, the standard error of the mean is
decreased from 730.3 to 400.
σ x=
( √σn ) ( 4√100
=
000
) = 400

Recall that when n = 30, P(51300 < �𝑥� < 52300) =

. 5034

We follow the same steps to solve for P(51300 <

�𝑥� < 52300) when n = 100 as we showed earlier
when n = 30.

Now, with n = 100, P(51300 < �𝑥� < 52300) = .7888

Because the sampling distribution with n = 100 has a

smaller standard error, the values of �𝑥� have less
variability and tend to be closer to the population
mean than the values of �𝑥� with n = 30.

 The sampling distribution of p is the probability

distribution of all possible values of the sample
proportion p.
 Expected Value of p
E( p ¿= p
where: p = the population proportion

Standard Deviation of p
Finite Population Infinite Population

σ p=
√ √
N −n p (1− p)
N −1 n
σ p=
√ p(1− p)
n

 σ p is referred to as the standard error of the proportion.

 √( N −n)/( N −1) is the finite population correction factor.
The sampling distribution of �𝑝� can be approximated by a normal distribution whenever the sample size is
large enough to satisfy the two conditions: np > 5 and n(1 – p) > 5

When these conditions are satisfied, the probability distribution of x in the sample proportion, �𝑝� = x/n, can
be approximated by a normal distribution (because n is a constant, the sampling distribution of �𝑝� can also
be approximated by a normal distribution).

Example: EAI Employee Data

For the EAI study we know that the population proportion of employees who participated in the management
training program is p = .6.

What is the probability that a simple random sample of 30 employees will provide an estimate of the
population proportion of employees attending management program that is within plus or minus .05 of the
actual population proportion?

For our example, with n = 30 and p = .6, the normal distribution is an acceptable approximation because:
np = 30(.6) = 18 > 5 And n(1 - p) = 30(.4) = 12 > 5

σ p=
√ . 6 (1−. 6)
30
= .0894

Step 1: Calculate the z-value at the upper endpoint of the interval.

z = (.65 - .6)/.0894 = .56

Step 2: Find the area under the curve to the left of the upper endpoint.
P(z < .56) = .7123

Step 3: Calculate the z-value at the lower endpoint of the interval.

z = (.55 - .6)/.0894 = - .56

Step 4: Find the area under the curve to the left of the lower endpoint.
P(z < -.56) = . 2877

Step 5: Calculate the area under the curve between the lower and upper endpoints of the interval.
P(-.56 < z < .56) = P(z < .56) - P(z < -.56)
= .7123 - .2877
= .4246
Other Sampling Methods

Stratified Random Sampling

 The population is first divided into groups of elements called strata.
 Each element in the population belongs to one and only one stratum.
 Best results are obtained when the elements within each stratum are as much alike as possible (i.e. a
homogeneous group).
 A simple random sample is taken from each stratum.
 Formulas are available for combining the stratum sample results into one population parameter
estimate.
 Advantage: If strata are homogeneous, this method is as “precise” as simple random sampling but with
a smaller total sample size.
 Example: The basis for forming the strata might be department, location, age, industry type, and so on.

Cluster Sampling
 The population is first divided into separate groups of elements called clusters.
 Ideally, each cluster is a representative small-scale version of the population (i.e. heterogeneous
group).
 A simple random sample of the clusters is then taken.
 All elements within each sampled (chosen) cluster form the sample.
 Advantage: The close proximity of elements can be cost effective (i.e. many sample observations can
be obtained in a short time)
 Disadvantage: This method generally requires a larger total sample size than simple or stratified
random sampling.
 Example: A primary application is area sampling, where clusters are city blocks or other well-defined
areas.

Systematic Sampling
 If a sample size of n is desired from a population containing N elements, we might sample one element
for every n/N elements in the population.
 We randomly select one of the first n/N elements from the population list.
 We then select every n/Nth element that follows in the population list.
 This method has the properties of a simple random sample, especially if the list of the population
elements is a random ordering.
 Advantage: The sample usually will be easier to identify than it would be if simple random sampling
were used.
 Example: Selecting every 100th listing in a telephone book after the first randomly selected listing

Convenience Sampling
 It is a nonprobability sampling technique. Items are included in the sample without known probabilities
of being selected.
 The sample is identified primarily by convenience.
 Example: A professor conducting research might use student volunteers to constitute a sample.
 Advantage: Sample selection and data collection are relatively easy.
 Disadvantage: It is impossible to determine how representative of the population the sample is.

Judgment Sampling
 The person most knowledgeable on the subject of the study selects elements of the population that he
or she feels are most representative of the population.
 It is a nonprobability sampling technique.
 Example: A reporter might sample three or four senators, judging them as reflecting the general
opinion of the senate.
 Advantage: It is a relatively easy way of selecting a sample.
  Disadvantage: The quality of the sample results depends on the judgment of the person selecting the
sample.

Recommendation
o It is recommended that probability sampling methods (simple random, stratified, cluster, or systematic)
be used.
o For these methods, formulas are available for evaluating the “goodness” of the sample results in terms
of the closeness of the results to the population parameters being estimated.
o An evaluation of the goodness cannot be made with non-probability (convenience or judgment)
sampling methods.

Errors in Sampling
o The difference between the value of sample statistic and the corresponding value of the population
parameters is called the sampling error.
o Deviations of the sample from the population that occur for reasons other than random sampling are
referred to as non-sampling errors.
o Non-sampling error can occur in a sample or a census.

 Reasons for Nonsampling Errors

- Coverage error
- Non-response error
Interviewer error
Processing error
- Measurement error

Steps to minimise non-sampling Errors

 Carefully define the target population and design the data collection procedure.
 Carefully design the data collection process and train the data collectors
 Pre-test the data collection procedure
 Use stratified random sampling when population-level information about an important qualitative
characteristic is available.
 Use systematic sampling when population-level information about an important quantitative
characteristic is available.