1

WHAT IS SAMPLING?
Sampling is the process of choosing a representative portion of a population.

A sample is a subset of individuals from which data are collected to make inferences
about the characteristics of the population.

Ideally, the sample corresponds to the larger population on the characteristic(s) of

interest. In that case, the researcher's conclusions from the sample are probably
applicable to the entire population.

This type of correspondence between the sample and the larger population is most
important when a researcher wants to know what proportion of the population has a
certain characteristic – like a particular opinion or a demographic feature.

2
WHY SAMPLE?
We know what the target population is, why not simply use the whole
population?

In some cases, where the target population is relatively small, and where
the need to generalize beyond the sample isn't a requirement, then
the whole population could indeed be used.

In reality there are a number of reasons for not using the whole
population:

- Time and funding constraints.

- Sampling beyond a certain point is wasteful because nothing else
will be gained.
3
BASIC SAMPLING CONCEPTS
 Population
- It is the collection of ALL the items about which we want to know
some characteristics.
- Examples of populations are hospital patients, road accidents, pet
owners, unoccupied property or bridges.
 Target Population
- It is the entire group a researcher is interested in; the group about
which the researcher wishes to draw conclusions.
- Suppose we take a group of men aged 35-40 who have suffered an
initial heart attack. The purpose of this study could be to compare the
effectiveness of two drug regimes for delaying or preventing further
attacks. The target population here would be all men meeting the
4
same general conditions as those actually included in the study.
BASIC SAMPLING CONCEPTS
 Sampling Population
- It refers to the population from which sample will actually be taken as
determined by the sampling frame.
 Sampling Frame
- It is a list of all elements in the population of interest (e.g., names of
individuals, telephone numbers, house addresses, census tracts).
- The sampling frame operationally defines the target population from
which the sample is drawn and to which the sample data will be
generalized.
 Elementary Units
- It refers to the units that make up a population. 5
BASIC SAMPLING CONCEPTS
 Sampling Units
- It refers to the units that make up a population.
 Independent Samples
- These are samples selected from the same population, or different
populations, which have no effect on one another. That is, no
correlation exists between the samples.
 Matched or Related Samples
- Those samples in which the same attribute, or variable, is measured
twice on each subject, under different circumstances. Commonly
called repeated measures. Examples include the times of a group of
athletes for 1500m before and after a week of special training; or the
milk yields of cows before and after being fed a particular diet. 6
BASIC SAMPLING CONCEPTS
 Sampling Variability
- It refers to the different values which a given function of the data
takes when it is computed for two or more samples drawn from the
same population.
 Sampling Error
- It is the standard deviation of the values of a given function of the
data (parameter), over all possible samples of the same size.
 Bias
- It refers to how far the average statistic lies from the parameter it is
estimating, that is, the error which arises when estimating a quantity.
7
BASIC SAMPLING CONCEPTS
 Probability Sampling
- A general approach to sampling where all elements (e.g., persons,
households) in the population have the same opportunity of being
included in the sample, and the mathematical probability that any
one of them will be selected can be calculated.
- Samples are obtained using some form of random sampling.
 Non-probability Sampling
- It is a sampling strategy where population elements are selected on
the basis of their availability (e.g., because they volunteered) or
because of the researcher's personal judgment that they are
representative.
8
- Random sampling is not used.
PROBABILITY SAMPLING MAY NOT
BE USED WHEN. . . . . .
A scientist can demonstrate that a particular trait occurs in a population
by documenting a single instance.
For example, the assertion that all lesbians are mentally ill can be
refuted by documenting the existence of even one lesbian who is free
from psychopathology.

A researcher wishes to describe a particular group in an exploratory way.

For example, interviewing 25 people with AIDS (PWAs) about their
experiences with HIV could provide valuable insights about stress and
coping, even though it would not yield data about the proportion of
PWAs in the general population who share those experiences.

9
FACTORS IN SELECTING WHICH
SAMPLING TECHNIQUES
 Availability of and access to the individuals in the representative
group.

 Availability of resources to use in the selection of the sample.

 Technical expertise of those involved in the data collection.

10
ADVANTAGES OF USING A SAMPLE
It is faster and cheaper.
It can yield more comprehensive information because a small sample can be
thoroughly investigated whereas the cost of such thoroughness would be prohibitive
for a large population.
It is more accurate because a small highly skilled group of workers is likely to make
fewer errors in the collection and handling of data than a large census would.
Because of the savings it permits in time and money, the sample survey makes
possible the study of much larger and much more varied populations than would be
possible for the same expenditure if one were making a complete enumeration

11
SAMPLE SIZE AND SAMPLING ERROR
In most cases, there is no prior knowledge about the variance of the population. In
such a case, the best thing that can be done is simply to take a certain
percentage of the whole population, say 2, 5, 10, 20, or 50 percent of N. Some
pointers maybe used as guides in selecting the percentage to be used:

(a) When the population (N) is large, a small percentage is

recommended.
(b) When the population (N) is small, a large percentage is
recommended.
(c) The sample size should be preferably not smaller than 30. An
n=30 enables one to assume that the characteristics are
normally distributed. But even if it is not normally distributed,
for as long as you have n=30 or more the statistics will behave
12
like normal.
SAMPLE SIZE AND SAMPLING ERROR
The rule to follow is: use as large samples as possible. This is because the
larger the sample, the safer is the assumption that the sample is
representative of the population.

By “representative sample” we mean that the sample has approximately the

characteristics of the population relevant to the research in question.

Although we can never be completely sure that a random sample is

representative of the population, the larger the sample, the safer is the
assumption of representativeness.

13
SAMPLE SIZE AND SAMPLING ERROR

Large

ERROR

Small Large
SIZE OF SAMPLE
14
SAMPLE SIZE AND SAMPLING ERROR
 Whenever a mean, a percentage, or other statistics is calculated
from a sample, a population value is being estimated. The question
is: “How much error is there likely to be in the statistics calculated
from samples of different sizes?”
 The curve in the figure roughly expresses the relations between
sample size and error, error meaning deviation from population
values. The curve says that “the smaller the sample, the larger the
error and the larger the sample, the smaller the error”. This is
because statistics calculated from large samples are more
accurate, other things being equal, than those calculated from small
samples.

15
SAMPLE SIZE AND SAMPLING ERROR
All other things being equal, smaller samples (e.g., those with fewer than
1,000 respondents) have greater sampling error than larger
samples.
The sampling error is a number that describes the precision of an
estimate from any one of those samples. It is usually expressed as
a margin of error associated with a statistical level of confidence.
For example, a presidential preference poll may report that the
incumbent is favored by 51% of the voters, with a margin of error of
plus-or-minus 3 points at a confidence level of 95%. This means
that if the same survey were conducted with 100 different samples
of voters, 95 of them would be expected to show the incumbent
favored by between 48% and 54% of the voters (51% ± 3%).
16
SAMPLE SIZE AND SAMPLING ERROR
 If you increase your n, you decrease standard error (SE) resulting
to higher statistics.
 This results to the rejection of the null hypothesis (Ho). Including
Design Efficiency (Deff) factor in your sampling and survey design
will correct this. In most cases, without Deff, you overestimate
statistics that result to outright rejection of the Ho.

17
PROBABILITY SAMPLING TECHNIQUES
Simple Random Sampling
Systematic Sampling (with a random start)
Stratified Sampling
Cluster Sampling
Multi-stage Sampling

18
SIMPLE RANDOM SAMPLING
It is a sampling technique wherein every element in the sampled
population is given an equal non-zero chance of entering the sample.
It is sampling wherein all possible samples of fixed size n have the same
probability of being selected
A sample drawn at random is unbiased in the sense that no member of
the population has any more chance of being selected than any other
member. Random methods of selection do not allow our own biases
or any other systematic selection factors to operate.
May be either: (a) with replacement or (b) without replacement.
Sampling is done in a single stage with each element selected
independently.
19
ADVANTAGES OF SIMPLE RANDOM SAMPLING
 The sample mean is an unbiased estimate of the population mean
(samples are used to estimate population parameters).
 Estimation methods are simple and easy.

20
DISADVANTAGES OF SIMPLE
RANDOM SAMPLING

The sample chosen may be widely spread, thus, entailing high

transportation cost (especially if the respondents are geographically
dispersed.

A population frame or list is needed.

Hard to achieve in practice.

21
WHEN TO USE SIMPLE RANDOM SAMPLING
 If the population is not widely spread geographically.
 If the population is more or less homogenous with respect to the
characteristics under study (Use stratified sampling if individuals are
heterogenous).

22
SYSTEMATIC SAMPLING
 It is a method of selecting a sample by taking every kth unit from an
ordered population, the first unit being selected at random.

k refers to the sampling interval

n/N refers to the sampling fraction.
 How do you do systematic sampling?

- Number the units in the population from 1 to N

- Decide on the n (sample size) that you want or need
- Determine sampling interval (k = N/n)
- Randomly select an integer (or sample) between 1 to k
23
- Then take every kth unit
SYSTEMATIC SAMPLING
 Let's assume that we have a population that only has N=100 people in
it and that you want to take a sample of n=20. To use systematic
sampling, the population must be listed in a random order.
 The sampling fraction would be f = 20/100 = 20%. In this case, the
interval size, k, is equal to N/n = 100/20 = 5. Now, select a random
integer from 1 to 5.
 In our example, imagine that you chose 4. Now, to select the sample,
start with the 4th unit in the list and take every k-th unit (every 5th,
because k=5).
 You would be sampling units 4, 9, 14, 19, and so on to 100 and you
would wind up with 20 units in your sample. 24
SYSTEMATIC SAMPLING: AN ILLUSTRATION

25
ADVANTAGES OF SYSTEMATIC SAMPLING
 Selection of the sample is administratively easier, quicker, and
cheaper than simple random sampling.

 It is possible to select a sample in the field without a sampling frame.

 It may also be more precise than simple random sampling.

 There is simply no easier way to do random sampling in some

situations.

26
DISADVANTAGES OF SYSTEMATIC SAMPLING
 A systematic sample may give poor precision when unsuspected
periodicity is present in the population (e.g. every 25th household in
the sample or population is rich).

 If the population is not in random order, one cannot validly estimate

the parameters of the population.

 It may also be more precise than simple random sampling.

 There is simply no easier way to do random sampling in some

situations.

27
STRATIFIED SAMPLING
 There may often be factors which divide up the population into sub-
populations (groups/strata) and we may expect the measurement of
interest to vary among the different sub-populations.

 This has to be accounted for when we select a sample from the

population in order that we obtain a sample that is representative of
the population. This is achieved by stratified sampling.

 A stratified sample is obtained by taking samples from each stratum or

sub-group of a population.

28
WHEN TO USE STRATIFIED SAMPLING
 When the population is heterogeneous, or dissimilar.

 Where certain homogeneous, or similar, sub-populations can be

isolated (strata).

 When we want to reduce the cost per observation in the survey.

 When estimates of the population parameters may be wanted for

each sub-population.

 When increased accuracy at given cost is desired.

29
ADVANTAGES OF STRATIFIED SAMPLING
 It is more efficient than simple random sampling.

 It allows for more comprehensive data analysis since information may

be provided for each stratum or subpopulation.

 It is administratively convenient.

30
DISADVANTAGES OF STRATIFIED SAMPLING
 Stratification of the population may mean the nee for additional prior
information about the population and its subpopulations.

 A separate frame is needed is needed for each stratum.

 Difficulty in identifying appropriate strata.

 More complex to organize and analyze results.

31
HOW TO DO STRATIFIED SAMPLING
 Stratified sampling may be done thru:
(a) Proportional Allocation
(b) Quota/Non-proportional Allocation
 In proportional allocation, we determine the size of each stratum and
the sample in each stratum is determined in terms of their percentage
to the total sample size.
Example: Suppose that in a company there are the following staff: (A)
male, full time=90, (B) male, part time=18, (C) female, full time=9, and
(D) female, part time=63. You want to have a sample size of n=80.
A= 90/180 x 80 = 40 C= 9/180 x 80 =4
B= 18/180 x 80 = 8 D= 63/180 x 80 =28
32
CLUSTER SAMPLING
Cluster sampling is a sampling technique where the entire population is
divided into groups, or clusters, and a random sample of these
clusters are selected. All observations in the selected clusters are
included in the sample.
Cluster sampling is typically used when the researcher cannot get a
complete list of the members of a population they wish to study but
can get a complete list of groups or 'clusters' of the population.
It is also used when a random sample would produce a list of subjects
so widely scattered that surveying them would prove to be far too
expensive.
This sampling technique may well be more practical and/or economical
than simple random sampling or stratified sampling. 33
CLUSTER SAMPLING
Example 1
Suppose we would like to investigate the use of pesticides by farmers in a
barangay. A cluster sample could be taken by identifying the different sitio or
purok as clusters. A sample of these sitio or purok (clusters) would then be
chosen at random, so all farmers in those sitio/purok will be included in the
sample. It can be seen here then that it is easier to visit several farmers in
the same purok/sitio than it is to travel to each farm in a random sample to
observe the use of pesticides.
Example 2
In conducting a poll of voter preferences for a nationwide election, we can
randomly select congressional districts (or some other applicable grouping of
voters), and then conduct the poll among the people in the chosen
congressional district. Many voter polls that utilize cluster sampling would
carefully choose their clusters so that they best represent the voter
preferences.
34
CLUSTER SAMPLING
 Similar to to strata in stratified random sampling, clusters are mutually
exclusive subpopulations which together comprise the entire
population. Unlike strata, however, clusters are comprised of
heterogenous elements so that each cluster is typical of the
population.

 The sample clusters maybe chosen by random sampling or by

systematic sampling with a random start.

 Clusters are either of equal or unequal size.

 In the field, one basic criteria employed is cluster of houses.

35
ADVANTAGES OF CLUSTER SAMPLING
 There is no need to construct a list of elements in the population as
one must do in random or stratified random sampling; the frame for
cluster sampling is simply a list of the clusters.

 It is cheaper because field cost would be minimized by the elements’

being physically closer together than elements selected by random or
stratified sampling.

36
DISADVANTAGES OF CLUSTER SAMPLING
 It is not as effective as random or stratified random sampling in
ensuring representativeness. In actual field applications, for example,
adjacent households tend to have more similar characteristics than do
households distantly apart. This affects the representativeness of the
sample.

 Larger sampling error than simple random sampling.

37
MULTI-STAGE SAMPLING
It In multistage sampling, we combine several techniques of sampling
into two or more phases of selection.

For example, in a household survey, you wish to select a sample of 30

households from a certain town. Suppose further that the town can
be divided into N=10 blocks of M=10 households per block. The units
need not have the same number of elements. A sample of N=5 blocks
is selected either by simple random sampling or by systematic
sampling with a random start. Then, from each selected block, a
sample of m=6 households is selected. Note that sampling is done at
each stage. This is an example of two-stage sampling with blocks as
first stage or primary units, and households as second stage or
38
secondary units.
ADVANTAGES OF MULTI-STAGE SAMPLING
 It is more efficient and flexible than single stage sampling.

 Except for the first stage units, a sampling frame is required only for
those units selected in order to sample the subunits.

 Transportation costs are greatly reduced especially when first stage

units are geographically distant from one another.

39
DISADVANTAGES OF MULTI-STAGE SAMPLING
The major disadvantage of multistage sampling is its complexity in
theory, which maybe difficult to apply in the field. The estimation
procedures are difficult for non-statisticians to follow.

40
NON-PROBABILITY SAMPLING TECHNIQUES
 Quota Sampling
 Purposive Sampling
 Accidental Sampling
 Snowball Sampling

41
QUOTA SAMPLING
This is usually done in surveys where the sampling frame is not used at
all and the choice of the actual sample units to be interviewed is left to
the discretion of the interviewer.
The interviewer is given quota controls. A control maybe socioeconomic
variables such as age, income, sex, occupation, place of residence,
etc.
It is widely used in opinion polling and market research. Interviewers are
each given a quota of subjects of specified type to attempt to recruit
for example, an interviewer might be told to go out and select 20 adult
men and 20 adult women, 10 teenage girls and 10 teenage boys so
that they could interview them about their television viewing.
It suffers from a number of methodological flaws, the most basic of
which is that the sample is not a random sample and therefore the
sampling distributions of any statistics are unknown.
42
PURPOSIVE SAMPLING
 This is characterized by the use of judgment by the researcher and a
deliberate effort to obtain representative samples by including
presumably typical areas or groups in the sample. Use this method
with extreme care.

 Maybe used in scientific research in combination with probability

sampling techniques.

43
ACCIDENTAL OR CONVENIENCE SAMPLING
 This a form of non-probability sampling technique in which one takes
available samples at hand. This form of sampling should not be used
at all.
 Because some members of the population have no chance of being
sampled, the extent to which a convenience sample – regardless of its
size – actually represents the entire population cannot be known.

44
SNOWBALL SAMPLING
Snowball sampling may simply be defined as a technique for finding research
subjects. One subject gives the researcher the name of another subject, who
in turn provides the name of a third, and so on (Vogt, 1999).

Snowball sampling contradicts many of the assumptions underpinning

conventional notions of sampling but has a number of advantages for
sampling populations such as the deprived, the socially stigmatised and
elites.

This strategy can be viewed as a response to overcoming the problems

associated with sampling concealed populations such as the criminal and the
isolated (Faugier and Sargeant, 1997).

45
SNOWBALL SAMPLING
 Although they violate the principles of sampling, the use of snowball
strategies provides a means of accessing vulnerable and more
impenetrable social groupings.

 This process is based on the assumption that a ‘bond’ or ‘link’ exists

between the initial sample and others in the same target population,
allowing a series of referrals to be made within a circle of
acquaintance (Berg, 1988).

46
DEFICIENCIES OF SNOWBALL SAMPLING
 Problems of representativeness and sampling principles.

 Finding respondents and initiating ‘chain referral’.

 Engaging respondents as informal research assistants.

47