Session 9

Chapter Seven

Sampling and Sampling distribution of the

sample mean
 Basic Concepts
Most researchers come to a conclusion of their study
by studying a small sample from the huge population
or universe.

To draw conclusions about population from sample,

there are two major requirements for a sample:

i) sample size should be adequately large;

ii) sample has to be representative of the
population

Sampling Methods are concerned with the selection

of representative sample of adequate size, especially
for the purposes of statistical inference.
 Population vs. Sample

Population Sample
a b cd
b c
ef ghi jkl m n
g i n o
o pq rs t uv w r u y
x y z

Values calculated using Values computed from sample

population data are called data are called statistics
parameters
 Reasons for Sampling
• Reduced Cost
• Greater Speed (less time)
• Greater Accuracy: Measurement errors typically
can be controlled more effectively in a small
undertaking than in a large one
• Greater Scope (more issues can be addressed)
• When a test involves the destruction of an item
(complete enumeration = complete destruction)
• When a population is infinite, information
about it can be obtained only from a sample
 Types of Sampling
A sample can accurately reflect a huge population if
proper sampling methods are used.
Sampling techniques/methods can be grouped into
two categories
1. Probability sampling methods
Each member of the population has a specified
chance of being selected (random selection).
These methods involve probabilistic process.

2. Non-probability sampling methods

Sample selection is based on subjective
judgment rather than selection with
specified probability (non random selection)
 Sampling Frame

Sampling frame is a list of all elements in the target

population.
There is a risk of drawing wrong conclusion from the
survey if the sample has been selected from a sampling
frame that differs from the population.
The problems are:
1. Under-coverage:
Occurs if the target population contains elements that do
not have a counterpart in the sampling frame.
e.g. an online survey, where respondents are selected via the
Internet. In this case, there will be under-coverage due to
people having no Internet access.
2. Over-coverage: sampling frame contains elements that do
not belong to the target population. If such elements end up
in the sample and their data are used in the analysis,
estimates of population parameters may be affected.
 Probability Sampling Methods
• Simple random sampling (S.R.S)

• Stratified/cluster random sampling

• Systematic random sampling

• Multi-stage random sampling

 Non-Probability Sampling
• Judgment sampling

• Quota sampling

• Convenience sampling
Probability Sampling
 Simple Random Sampling (SRS)
In simple random sampling, there is an equal chance (probability) of
selecting each unit from the population being studied.
Steps in taking SRS:
✓ Define population
✓ Determine sample size
✓ List the population (Sampling Frame)
✓ Assign number to each the unit
✓ Get random numbers (Table/software (R; SPSS,)
✓ Select (take) sample

Problem in SRS: Difficulty in getting complete list (sampling frame) of Population:

 Example (SRS)
Selecting 5 students from 2nd Year Chemistry Department Students
to study the weights of students using SRS
✓ Define population: 34 students
✓ Determine sample size: 5 students
✓ List the population (Sampling Frame)
1. Abenezer G., 2. Abenezer Y., …. 34. Zelalem A.
✓ Assign number to each unit
1. Abenezer G., 2. Abenezer Y., …. 34. Zelalem A.
✓ Get random numbers (Table/software (R; SPSS, Excel))
Simple-random-example.xlsx
✓ Select (take) sample
Sample-1 taken: 23, 14, 18, 24, 22
 Stratified Random Sampling
• It is preferred when the population is heterogeneous with respect to
characteristic under study.
• In this method, the complete population is divided into homogenous
sub groups called "Strata" and then a stratified sample is obtained by
independently selecting a separate simple random sample from each
population stratum.
• Population is Heterogeneous, but Strata are Homogenous
• Some of the criteria for dividing a population into strata are: Sex
(male, female); Age (under 18, 18 to 28, 29 to 39); Regions (Region-1
to Region-12); Universities (1st Generation, 2nd Generation, 3rd
Generation)
• Random samples taken within a stratum will have much less
variability than a random sample taken across all strata. This is true
because sample units within each stratum tend to have
characteristics that are similar.
 Example (Stratified Sampling)
Selecting 5 students from 2nd Year Chemistry Department Students to
study the weights of students using stratification by Sex (4 male, 1 female)
✓ Define population: 34 students (22 male, 12 female)
✓ Determine sample size: 5 students (3 male, 2 female)
✓ List the population (Sampling Frame)
Separate List for Male: 1 to 22; female: 1 to 12
✓ Assign number to each unit
1. Abenezer G., 2. Abenezer Y., …. 21. Zelalem A.
1. Absera A. … 12. Sara A.
✓ Get random numbers (Table/software (R; SPSS, Excel))
Stratified-sample-example.xlsx
✓ Select (take) sample
Sample-Male: 22, 11, 18 Sample-Female: 2, 5
 Cluster Sampling
• Clusters are formed by grouping units on the basis of their
geographical locations. Thus, elements within a cluster are
heterogeneous.
• Population and Clusters share similar distribution of
characters
• It is obtained by selecting clusters from the population on
the basis of simple random sampling so that each and every
units in the selected clusters will be included in the sample.
(Complete enumeration will be conducted for selected
cluster)
• The advantage of cluster sampling is that sampling frame
for entire population is not required and in practice when
complete lists are rarely available, cluster sampling is
suitable.
 Example (Cluster Sampling)
Selecting 10 Woredas out of about 770 Woredas (Clusters) in Ethiopia to
study the size/magnitude of malnourished children under five..
Assumptions:
- Woredas (Clusters) are internally Heterogeneous
- Woredas (Clusters) have similar distribution of population charecters
- Clusters all in all cover entire population
- No overlaps between clusters
Random selection of 10 out of 770
List the population (Sampling Frame) and assign number to each unit
List of Woredas of all regions starting from Region-1 to Region ---
W-1, W-2, ….., W-770
✓ Get random numbers (Table/software (R; SPSS, Excel))
Cluster-sampling-example.xlsx
✓ Select (take) sample
Sample of woredas:
W411, W21, W50, W33, W22, W29, W15, W24, W53, W40
❖ For selected Woredas, take complete enumeration of children under five
 Systematic Random Sampling
• Systematic sampling is a commonly employed
technique, when complete and up to date list of
sampling units is available.
• A systematic random sample is obtained by selecting
one unit on a random basis and then choosing
additional units at evenly spaced intervals until the
desired number of sample size is obtained.
• Let N=population size; n=sample size and k is sampling
interval.
=> Then choose randomly a number between 1 and k.
 Example (Systematic Sampling)
Selecting 5 students from 2nd Year Chemistry Department Students to
study the weights of students using systematic sampling
✓ Define population: 34 students
✓ Determine sample size: 5 students
✓ List the population (Sampling Frame) 1 to 34
✓ Assign number to each unit
1. Abenezer G., 2. Abenezer Y., …. 34. Zelalem A.
✓ Divide students in to groups k (interval). Let k be five
✓ 1,2,,3,4,5;6,7,8,9,10; ….,26, 27, 28,29, 30; 31, 32, 33, 34, ….
✓ Get random numbers (Table/software (R; SPSS, Excel)
Randomly select a number between 1 and 5. Let the randomly selected
number be 4:
✓ Select (take) kth element until you get the desired sample size (5)
4th, 8th, 12th , 16th and 20th students
 Multistage Sampling
• In this method, the whole population is divided in first
stage sampling units from which a random sample is
selected.

• The selected first stage is then subdivided into second

stage units from which another sample is selected.
Third and fourth stage sampling is done in the same
manner if necessary.

• e.g. Studying Hepatitis prevalence in Ethiopia

Region => Zone => Woreda => Kebele
 Non-Probability Sampling

• Judgment sampling

• Quota sampling

• Convenience sampling
 Convenience sampling

• In convenience sampling, we select individuals

into our sample based on their availability to the
investigators rather than selecting subjects at
random from the entire population

• The extent to which the sample is

representative of the target population is not
known.
 Quota Sampling
• We determine a specific number of individuals to
select into our sample in each of several specific
groups
• Similar to stratified sampling in that we develop
non-overlapping groups and sample a
predetermined number of individuals within each
 Example: Quota Sampling
Suppose we want to conduct a survey on the prevalence of
malaria in given region of residents by age group and our desired
sample size is n=500.
We wish to ensure that the distribution of subjects' ages in the
sample is similar to that in the population.
Prior Information:
We know from census data that approximately 30% of the
population are under age 20; 40% are between 20 and 49; and
30% are 50 years of age and older.
We would then sample
n1=150 persons under age 20,
n2=200 between the ages of 20 and 49 and
n3=150 who are 50 years of age and older.
 Judgment Sampling
• Samples selected according to the opinion of an
expert.
• Use when you want a quick sample and you believe
you are able to select a sufficiently representative
sample for your purposes
• It is a biased method that is useful when some
members of a population make better subjects than
others.
• Judgment sampling is often a last-resort method that
may be used when there is no time to do a proper
study.
 Sampling distribution of the sample mean
• If we consider all the samples of size n that can be
drawn from a population, we can compute sample
statistic such as mean or variance of each sample.
• Value of sample statistic will vary from sample to
sample
• Theoretical distribution that relates the possible
values of the sample mean to the probability of all
possible samples of size n is called the sampling
distribution of the mean
 Example: Sampling distribution of the sample mean
Suppose that all possible samples of size two are
drawn from a population, having the following N=4
elements: 2, 5, 7, 10
Sampling distribution of the mean.
Sample Mean
X =x P( X = x )
2, 5 3.5 3.5 1/6
2, 7 4.5 4.5 1/6
2, 10 6 6 2/6
5, 7 6 7.5 1/6
5, 10 7.5 8.5 1/6
7, 10 8.5 Sum 1
 Central Limit Theorem
Let the parent/original population is not
normally distributed and we a sufficiently
large sample and population such that:

n  30
N  2n

Then the distribution of the means of the

samples still tends to be normally
distributed.
 Central Limit Theorem (inequalities)
The mean and standard deviations of the
sample means will be:
x =

x =
n
2
X ~ N (, )
n
x1 −  x2 − 
P ( x1  X  x 2 ) = P ( Z )
/ n / n
 Survey Errors
• Sampling Error – Who are you sampling?
• Coverage Error – Does your list include
everyone/every unit required?
• Measurement Error – Does everyone answer a
question the same way?

• Non-response Error – Why did respondent not

answer:
- Instrument (whole questionnaire not returned)
- Item (question not answered)