1/22/18
Introductory Statistics
ECON1005
Unit 1: Explaining Statistics
Session 1.2: Sampling
1
Session Objectives:
After completing this session, you will be able
to:
• Discuss why sampling is often favoured over
census;
• Differentiate between random and non-
random sampling methods;
• Explain simple random sampling, stratified
random sampling, systematic, cluster, quota
and convenience sampling;
• Define sampling frame and discuss possible
sources of error.
1
1/22/18
Required Reading
Refer to text: OpenStax College,
Introductory Statistics. OpenStax
College. 19 September 2013.
Retrieved from
http://cnx.org/content/col11562/latest/
Read Chapter 1: Sampling and Data;
pages 18 to 24.
Primary data refers to information
collected, or developed, by the
researcher exclusively for the current
research project. Gathering primary
data may be costly and timely.
Secondary data refers to information
previously collected by another
researcher for some other purpose
other than the current research
project.
4
2
1/22/18
Classify the output of each of the following according to primary or
secondary data:
(i) Census Primary/Secondary
(ii) Household survey Primary/Secondary
(iii) Telephone survey Primary/Secondary
(iv) Mail out survey Primary/Secondary
(v) Document search Primary/Secondary
(vi) Focus Group Primary/Secondary
(vii) Internet search Primary/Secondary
Classify the output of each of the following
according to primary or secondary data:
(i) Census Primary/Secondary
(ii) Household survey Primary/Secondary
(iii) Telephone survey Primary/Secondary
(iv) Mail out survey Primary/Secondary
(v) Document search Primary/Secondary
(vi) Focus Group Primary/Secondary
(vii) Internet search Primary/Secondary
3
1/22/18
Sampling
Surveys:
1. Census – data is
collected from all the
items of the population
2. Sample Survey – data
is collected from a subset
of the population
4
1/22/18
Sampling
Sampling is the technique
used to make conclusions
about a population by
testing a sample.
Reasons for sampling
•t o m a k e i n f e r e n c e s a b o u t a
population without having to
investigate the whole population, eg
vehicle crash test
• facilitates the impossibility of
accessing the population
•less expensive
•reduces time taken to complete
research
•increases the possibility of collecting
more information
•increases the reliability of results. 10
5
1/22/18
Sampling Methods
The sampling technique or method is the process of
selec=ng the members of the sample.
Sampling Methods
Random Sampling Non-Random Sampling
members of sample are members of sample are
obtained by chance inten(onally selected
& each member of the
popula;on has a chance of
being a part of the sample.
11
RANDOM NON- RANDOM
1. Simple
Random 1. Quota Sampling
Sampling
2. Convenience
Sampling
2. Stratified
Random
Sampling
3. Systematic
Sampling
4. Cluster
Sampling
12
6
1/22/18
Simple Random Sampling
Each member of the population is selected entirely by chance
and has an equal chance of being included in the sample.
Lottery technique, calculator or the table of random numbers
technique.
The sample is thus selected in a way that is completely
unrelated to any conceivable characteristic of the elements.
The sample selected is free from selection bias – that is, no
human prejudice, conscious or unconscious, has influenced
the selection procedure.
13
Simple Random Sampling
However, this procedure can be very costly, for instance,
in cases where the population being sampled is
distributed across a wide geographical area.
Taking a simple random sample does not automatically
ensure that the results obtained are reliable, that is, we
feel sure that the sample value is close to the population
value.
You should always remember that the possibility that a
sample selected through a simple random method may
prove to be a bad or an imprecise indicator of the
population from which it was taken, and thus may lead
to inaccurate estimates.
14
7
1/22/18
Stratified Random Sampling
The population is divided into mutually exclusive small
subgroups (strata), eg males and females. The members,
for the sample, are then simple random sampling is used
to select a sample from each group. The samples from
each group are put together to form the stratified random
sample. The number selected from each stratum should
be in proportion with the size of the strata.
One important consequence of stratifying the population
is that, now, each element of the population no longer has
an equal chance of being drawn. However, the chance of
any individual being drawn is still measurable and all
possible samples of equal size still have the same chance
of selection.
15
Stratified Random Sampling
The choice of the stratification factor is
dependent on our knowledge of the population.
One uses stratified random sampling when the
variation within groups is relatively small
compared to the variation between groups.
Stratified random sampling can achieve quite
accurate results from smaller samples than
would be needed for simple random sampling,
and, therefore, it significantly reduces the cost-
effectiveness of sampling.
16
8
1/22/18
Systematic Random Sampling
A systematic random sample is obtained by selecting
one unit, on a random basis, and choosing additional
units, at evenly spaced intervals, until the desired
number of units is obtained.
For example, there are 100 students in your class.
You want a sample of 20 from these 100 and you
have their names listed on a piece of paper, maybe in
an alphabetical order. If you choose to use systematic
random sample, divide 100 by 20, you will get 5.
Randomly select any number between 1 and 5.
Suppose the number you have picked is 4, that will
be your starting number. So student number 4 has
been selected. From there you will select every 5th
name until you reach the last one. You will end up
with 20 selected students.
17
Cluster Sampling
This involves dividing the populations into clusters.
Each cluster is a representation of the population. A
random process is then applied to select among
clusters. Elements from each of the selected clusters
are examined.
Cluster sampling does not share this characteristic.
The design of cluster sampling (as the term cluster
indicates) involves the division of the population into
clusters. A random process is then applied to select
among the clusters, with the members of the chosen
cluster(s) constituting the sample. In other words, the
population is divided into clusters which are then
selected for the sample.
18
9
1/22/18
Cluster Sampling
Cluster sampling may seem to resemble stratified
random sampling, in that both sample procedures
involve the grouping of the members of the population.
But the similarity stops there. When we stratify, every
stratum is sampled, but when we cluster, we select
among the clusters.
Each cluster therefore has an equal chance of being
selected. Remember that this is not so for the
individual members of the clusters, since the selection
of one member immediately means the selection of the
rest of the cluster.
19
Quota sampling
This method involves the identification of a quota
that the selected sample is to fulfill.
Here, the population is divided into groups, eg sex,
or income level. A specified number of persons in
each group are interviewed. There is no specific
instruction as to how to fulfil the quota. Quota
sampling is widely used in opinion polling and
market research. Interviewers are each given a
quota of subjects of specified type to attempt to
recruit.
One major advantage of quota sampling is that it is
less costly to carry out than the other sampling
procedures we have discussed so far. In addition,
it does not require the existence of a sampling
frame and, therefore, is useful in situations where
no such frames exist. 20
10
1/22/18
Convenience sampling
Here the selected of the sample is based on accessibility,
ease, speed, and low cost. The selected process is
unguided. Persons who volunteer may constitute a
convenience sample. This sampling technique does not
allow for any control on the characteristics of the sample,
example, age, gender, socio-economic status.
This method may be used when other sampling
techniques are not possible. Caution must however ne
taken to ensure that the results are not used to make
generalizations about the population.
21
Discussion Question
Which of the following selection techniques will result in
random samples? Why or why not?
(a) Population: Nursing lecturers in the USA
Sampling Technique: randomly choosing a Nursing
University in the USA at random and interviewing all
lecturers in that University.
(b) Population: UWI Mona students
Sampling Technique: selecting every 10th person
entering the main library between 9am to 12noon on a
Monday.
(c ) Population: cases of beer about to leave the
brewery
Sampling Technique: blindfolded, selecting one beer
from each case and checking it for quality.
22
11
1/22/18
Discussion Question
Which of the following selection techniques will result in
random samples? Why or why not?
(a) Population: Nursing lecturers in the USA
Sampling Technique: randomly choosing a Nursing
University in the USA at random and interviewing all
lecturers in that University.
(b) Population: UWI Mona students not every student will be in
cafeteria
Sampling Technique: selecting every 10th person
entering the main library between 9am to 12noon on a
Monday.
(c ) Population: cases of beer about to leave the
brewery
Sampling Technique: blindfolded, selecting one beer
from each case and checking it for quality.
23
Question:
In attempting to derive a sample that is
representative of the population, a
researcher is about to make a choice
between stratified random sampling and
cluster sampling.
Which of the two sampling techniques will
require greater care in the process of
subdividing the population
(i) stratified random sampling
(ii) cluster sampling
24
12
1/22/18
Question:
In attempting to derive a sample that is
representative of the population, a
researcher is about to make a choice
between stratified random sampling
and cluster sampling.
Which of the two sampling techniques
will require greater care in the process
of subdividing the population
(i) stratified random sampling
(ii) cluster sampling
25
Question
Which of the sampling approach/es below is/
are not random in nature?
(a) Simple random sampling
(b) Systematic sampling
(c) Quota sampling
(d) Stratified random sampling
(e) Cluster sampling
(f) Convenience sampling
26
13
1/22/18
Question
Which of the sampling approach/es below is/
are not random in nature?
(a) Simple random sampling
(b) Systematic sampling
(c) Quota sampling
(d) Stratified random sampling
(e) Cluster sampling
(f) Convenience sampling
27
Question
As part of a household survey comprising 325
households numbered 001 to 325 on a map, the
interviewer was instructed to select a sample of
36 households by utilizing a skipping interval of
every 9 houses and a ‘start’ house at the 5th
house. What kind of sampling technique was
employed?
(a) Simple random sampling
(b) Systematic sampling
(c) Quota sampling
(d) Stratified random sampling
(e) Cluster sampling
(f) Convenience sampling
28
14
1/22/18
Question
As part of a household survey comprising
325 households numbered 001 to 325 on a
map, the interviewer was instructed to select
a sample of 36 households by utilizing a
skipping interval of every 9 houses and a
‘start’ house at the 5th house. What kind of
sampling technique was employed?
(a) Simple random sampling
(b) Systematic sampling
(c) Quota sampling
(d) Stratified random sampling
(e) Cluster sampling
(f) Convenience sampling
29
Question
Sampling that is performed in such a
way that each item of a population has
a measurable chance of being
selected is called
(a) Simple random sampling
(b) Non-random sampling
(c) Random sampling
(d) Cluster sampling
30
15
1/22/18
Question
Sampling that is performed in such a
way that each item of a population has
a measurable chance of being
selected is called
(a) Simple random sampling
(b) Non-random sampling
(c) Random sampling
(d) Cluster sampling
31
A sampling frame is an exhaus=ve list of
all the elements/items/persons in the
popula=on.
Suppose your popula=on of interest is all
the students currently enrolled in
ECON1005, then your sampling frame is
a list of all the names (or ID numbers) of
all the students currently enrolled in
ECON1005.
32
16
1/22/18
Deficient sampling frames may themselves be
responsible for many of the inaccuracies of the
resultant sample, for the following reasons:
• the frame is incomplete (which will exclude
some members of the population from the
sample);
• the frame is inaccurate;
• the frame is out of date.
Notwithstanding these limitations, a sampling
frame is a critical requirement for the selection of a
sample. We should be aware that no actual frame
will possess all of the characteristics of an ideal
sampling frame, that is, complete, accurate and
up-to-date.
33
Question
Which of the sampling approaches
below require complete knowledge of
the sampling frame? (multiple
response).
(a) Simple random sampling
(b) Systematic sampling
(c) Quota sampling
(d) Stratified random sampling
(e) Cluster sampling
(f) Convenience sampling
34
17
1/22/18
Question
Which of the sampling approaches
below require complete knowledge of
the sampling frame? (multiple
response).
(a) Simple random sampling
(b) Systematic sampling
(c) Quota sampling
(d) Stratified random sampling
(e) Cluster sampling
(f) Convenience sampling
35
Sampling errors and Non-sampling errors
The difference between the statistic (an estimate of a population
parameter) and the true, but unknown, value of the population
parameter may be due to (1) sampling errors and (2) non-sampling
errors.
The actual process of sampling causes sampling errors. Sampling
errors may be due to the sample not being large enough, error in
the sampling frame, errors made in selecting the sample (eg
selecting only persons interested in participating) and non-
response (selected persons refusing the respond).
Factors not related to the sampling process cause nonsampling
errors. This may be due to inaccurate inputting of data into software
for analysis, incorrect information given by respondents,
inappropriate statistical analysis.
There is always some sampling error because a sample will never
exactly represent the population. The rule of thumb is, the larger
the sample, the smaller the sampling error.
36
18
1/22/18
In statistics, a sampling bias is
created when a sample is collected from a
population and some members of the
population are not as likely to be chosen as
others (remember, each member of the
population should have an equally likely
chance of being chosen). When a sampling
bias happens, there can be incorrect
conclusions drawn about the population
that is being studied.
37
Wrap up:
You should now be able to:
• Discuss why sampling is often favoured over
census;
• Differentiate between random and non-
random sampling methods;
• Explain simple random sampling, stratified
random sampling, systematic, cluster, quota
and convenience sampling;
• Define sampling frame and discuss possible
sources of error.
38
19
1/22/18
End of Unit 1: Session 1.2
Powerpoint template
Kingsoft Office @Kingsoft_Office
39
published by www.Kingsoftstore.com kingsoftstore
20
