Lecture 8
Lecture 8
LECTURE88
P I C 3 0 0 1 RSEPS111
E A RP C
SHY CMHEOTLHOOGDI C
S AI LN REEASRELY
A RCCHHI LMDEHTOHOODD E1D U C AT I O N
OUTLINE
2
POPULATION
Definition
3
Generalize :
4
Not generalize : The sample
selected is not an accurate
representation of the
population.
5
IMPORTANT TERMS:
6
POPULATION VS. CENSUS
7
POPULATION PARAMETER VS. STATISTIC
8
SAMPLING FRAME VS UNIT OF ANALYSIS
9
SAMPLING FRAME
10
E.G., SAMPLING FRAME
11
SAMPLES
Definition :
Sample is a subset of the population.
12
POPULATION
Sample
Sample
13
WHY SAMPLE?
14
Types of sampling
1. Probability sampling
2. Non probability sampling
15
Sampling Techniques
Nonprobability Probability
Sampling Techniques Sampling Techniques
16
PROBABILITY SAMPLING
17
NON-PROBABILITY SAMPLING
Population parameters not of interests.
Adequacy of sample unknown.
Cheaper, easier, quicker to carry out.
Cant generalized findings.
Non-representative.
18
PROBABILITY SAMPLING
A type of sampling where the likelihood of
any one member of the population being
selected is known.
19
NON-PROBABILITY
20
TYPES OF PROBABILITY SAMPLING
21
1. Simple Random Sampling
22
http://www.google.com.my/search?q=cluster+sampling+design+ppt&ie=utf-8&oe=utf-
8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
23
Adv:
o Ensures a high degree of
representativeness
Disadv:
o Time consuming and tedious
24
PS- SIMPLE RANDOM SAMPLING
25
26
27
HOW TO USE A RANDOM NUMBER TABLE?
28
We close our eyes and randomly point to a spot on
the chart. For this example, we will assume that
we selected 20631 in the first column.
29
As we work down the column, we find that the first
number to match our population is 100 (actually
10005 on the chart). Student number 100 would be
in our sample. Continuing down the chart, we see
that the other four subjects in our sample would be
students 049, 082, 153, and 005.
http://www.google.com/imgres?imgurl=http://www.gifted.uconn.edu/sieg
le/research/Samples/RANTBLE.JPG&imgrefurl
30
2. Systematic Sampling
31
http://www.google.com.my/search?q=cluster+sampling+design+ppt&ie=utf-8&oe=utf-
8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
32
Population = 27 people
Sample size = 9 people for your study.
k = population/ sample size, so 27/9 = 3
Start with any random number (1st Respondent),
then select every kth
33
Adv :
Ensures a high degree of
representativeness; no need to use a
table of random numbers.
Disadv :
Less truly random than simple
random sampling
34
3. Stratified Random Sampling
35
http://www.google.com.my/search?q=cluster+sampling+design+ppt&ie=utf-8&oe=utf-
8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
36
37
Adv :
Ensures a high degree of
38
TWO TYPES OF STRATIFIED RANDOM
SAMPLING (SRM)
ProportionateSRM
Non-Proportionate SRM
39
PROPORTIONATE SRM
40
EXAMPLE: PSRM
Population = 100
Layer 1 = 40 males
Layer 2 = 60 females
For a sample size of 10, you will take 4 males + 6
females.
41
NON-PROPORTIONATE SRM
42
E.G., NPSRM
Population = 100
Layer 1 = 40 males
Layer 2 = 60 females
For a sample size of 10, you will take 5 males + 5
females.
43
4. Cluster Sampling
When the population consist of units rather
than individuals.
44
http://www.google.com.my/search?q=cluster+sampling+design+ppt&ie=utf-8&oe=utf-
8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
45
A TWO-STEP AREA CLUSTER SAMPLE (SAMPLING SEVERAL CLUSTERS) IS
PREFERABLE TO A ONE-STEP (SELECTING ONLY ONE CLUSTER) SAMPLE
UNLESS THE CLUSTERS ARE HOMOGENEOUS
http://www.google.com.my/search?q=cluster+sampling+design+ppt&ie=utf-8&oe=utf-
8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
46
Adv :
Easy and convenient
Disadv :
Possibility that members of units are
different from one another,
decreasing the sampling’s
effectiveness
47
TYPES OF NON-PROBABILITY SAMPLING
1. Convenience Sampling
2. Quota sampling
3. Purposive Sampling
4. Snowball sampling
48
1. Convenience Sampling
When the sample is captive.
Adv :
convenient and inexpensive
Disadv :
results in questionable representativeness.
49
2. Quota sampling
When strata are present, and stratified,
sampling is not possible
Adv :
Ensures some degree of representativeness of all the strata
in the population
Disadv :
Results in questionable representativeness
50
3. Purposive Sampling
51
4. Snowball sampling
52
SAMPLING ERROR
Lack
of fit between the sample and the
population.
53
Reducing sampling error is the major
goal of any selection technique.
54
SAMPLE SIZE
How big?
Depends on type of research design.
Desired confidence level of results.
Amount of accuracy wanted.
Characteristics of population of interest.
55
SAMPLE SIZE
56
General Rule of Thumb
30 participants/ respondents in
each group.
57
GENERAL GUIDE FOR SAMPLE SIZE:
58