Sampling and Sampling

Techniques
 Census vs Sample Survey

Census or Complete Enumeration

• the process of gathering information from every
unit in the population
Sample Survey
• the process of obtaining information from a
subset unit in the population
 Two Types of Sample Surveys:

1. Descriptive
The main objective of a descriptive sample survey is simply to
obtain certain information about large population groups like
number of men and women who have finished college.
2. Analytical
Comparisons are made between subgroups of the population to
find out if there are differences that exist among them and to
form or verify hypotheses about the reasons for these
differences.
 Examples of Sample Surveys

• Quarterly Labor Force Survey

• Family Income and Expenditures Survey
• National Demographic and Health Survey
 Role of Sampling in Survey Work:

 The main purpose of conducting sample surveys is generally to

estimate certain population parameters

Examples: average income of families

average age of the population
participation rate of female workers in factories
most common occupation of employed persons
average production
fertility rate
infant mortality rate
 Role of Sampling in Survey Work:

 Information is obtained or desired from a sample of statistical

population, but inferences are required on characteristics of the
whole statistical population.

 A second possible purpose may be to test a statistical hypothesis

about the statistical population
 Roles of Sampling in Survey Work

Example: the hypothesis that at least 40% of the households in

Metro Manila have T.V. sets. To do this, a sample of households
in Metro Manila is selected and the proportion of households
possessing TV sets will be calculated.
 Appropriate statistical test will be used to prove whether the
hypothesis is supported by the survey findings.
 Advantages of Sampling

1. Less expensive
 The use of sampling in the conduct of a survey is less
expensive compared with a complete coverage of the
population like a census when absolute precision is not
necessary.

 Also when funding is limited, it is still possible to collect

information through sampling with certain degree of precision
 Advantages of Sampling

2. Results are obtained faster with better quality of data

 Enumeration of a sample rather than a census will

definitely save time in data gathering as well as
processing of data aside from saving money.
 Advantages of Sampling

 Example: Release of the 1970 Census of Population and

Housing in a shorter span of time was made possible through
the use of sampling. A sample of 5% of census returns was
selected for advance processing and tabulation to generate
socio-economic characteristics of the population in less than one
year from the conduct of the census. Normally, it takes more
than 2 years to come out with the socio-economic
characteristics of the population from a census if all the
questionnaires will be processed.
 Advantages of Sampling

3. Gather more detailed information

 When gathering very detailed data, it is not feasible to consider
them on a complete enumeration basis except through a
sample.
Examples: family income and expenditures studies;
health and demographic studies on the
population at large, such as fertility,
morbidity, mortality, mother and child care
and other health conditions and other social
concerns studies.
 Advantages of Sampling

4. Establish time series

 Use of sampling in the establishment of time series data is the
most practical and least expensive way of obtaining data
promptly.

Examples of time series data: Consumer Price Index (CPI), Labor

Force statistics, and Family Income and Expenditures statistics
including poverty statistics.
 Advantages of Sampling

 The frequency of the series depends upon the type of survey

conducted.

Examples: The survey that gathers data used as inputs to the

computation of the CPI is done every month; the labor
force survey is conducted every quarter; the Family Income
and Expenditures Survey is conducted every 3 years; the
National Health and Demographic Survey is usually conducted
every 3 years; and the Crop and Livestock Survey is also
conducted every quarter.
 Advantages of Sampling
5. Use of sampling to control non-sampling errors
 A survey is a chain of operations that extends from
data collection to publication of survey results.
Varying degrees of control can be exerted on the
quality of each link in the chain.
 Definitions of Terms:

Example: In a certain community, an opinion poll was conducted

to determine public sentiment toward a bond issue in an
upcoming election. The objective of the survey was to estimate
the proportion of voters in the community who favored the
bond issue.
1. Element
- is an object on which a measurement is taken.

Example: In the given example, an element is a registered

voter in the community.
 Definitions of Terms:

2. Target Population
- the population about which information is desired.

Example: The target population in the given example is the

collection of voters in the community.

3. Sampled Population
- the population from which a sample is actually taken
 Definitions of Terms:

Notes about target and sampled population:

a) Ideally, the target and the sampled populations should be the

same collection of elements.

b) The target population is the collection under study, that is, we

wish to infer on the characteristics of this population.
However, if we cannot get a sample from this population then
get another similar population, which we call the sampled
population, from where we obtain the sample.
 Definitions of Terms:

4. Sampling Units
- are nonoverlapping collections of elements from the
population that cover the entire population

Example: In the bond issue example, a sampling unit may be a

registered voter in the community.
 Example of Elementary Unit
and Sampling Unit

Suppose we select our sample from a list of

registered voters then the sampling unit is a
registered voter, or the element in the study.
However, if we select our sample from a list of
families in the barangay then the sampling unit
this time is a family, which may consist of one or
more elements (registered voter) in the study.
 Definitions of Terms:

Notes on Sampling Units:

a) As the definition states, sampling units should be
nonoverlapping. However, situations do arise in which the
nonoverlapping condition is virtually impossible to achieve.
Example of overlapping sampling units: Field plot samples
taken, in studies of animal habitat are often circular. The circular
pattern is a convenient one to lay out and has advantages in terms of
the amount of walking necessary to study the plot. Obviously,
circular plots cannot cover a field without some overlap.

b) The intent here is to suggest that the overlap should be as small

as possible for efficient sampling.
 Definitions of Terms:

5. Sampling Frame
- is a list of sampling units
Example: If we specify the individual voter as the sampling unit, a
list of all registered voters may serve as a frame for a public
opinion poll.
Notes on the Frame:
a) The frame may not include all the elements in the population
because updating the list daily is impossible.

Example: If we take the household as the sampling unit, then a

telephone directory, a city directory, or a list of household
heads obtained from census data can serve as a frame.
 Definitions of Terms:

Types of Frames:
1. list frame
2. area frame
3. telephone numbers for telephone surveys

Inadequacies of Frames:
1. missing elements or noncoverage
2. duplicate listings
 Definitions of Terms:

Notes on the frame:

b) Some sampling schemes may involve multiple frames.
Examples of Multiple Frames:
1. In sampling voters, one could start by sampling housing units
or city blocks then sampling voters within the selected housing
units or blocks. One frame, then, is a list of housing units or a
list of city blocks, and the second frame is a list of voters
within those larger units. The second frame may not be
available until the housing units or blocks are selected and
studied in some detail.

2. Estimation of crop yields in a state may involve sampling from a

list of growers to be interviewed and a list of fields to be
objectively measured.
 Definitions of Terms:

6. Sample
- is a collection of sampling units drawn from a frame or frames.

Example: Sampling unit - individual voter

Frame - list of registered voters
Sample - number of voters in the
public opinion poll
 Definitions of Terms:

7. Probability of Selection
- The probability of selection is the chance that each
unit in the population has of being included in the
sample. Probability values range from 0 to 1. Usually
the probability of selection is determined from the
elements of the sampling frame.
 2 Methods of Sampling

Probability Sampling
• procedure wherein every element of the population
is given a (known) nonzero chance of being
selected in the sample

Nonprobability Sampling
• procedure wherein not all the elements in the
population are given a chance of being included
in the sample
 Types of Sampling Methods

Samples

Non-Probability Samples Probability Samples

Haphazard Systematic
Quota
Convenience
Stratified

Volunteer Cluster
Purposive
Simple Random

Example: Phoned-in calls

 Methods of Nonprobability Sampling

 Probabilities of selection are not specified for the individual

units of the population in nonprobability sampling.

 The clear disadvantage of nonprobability sampling is that the

researcher cannot assert that the sample is representative of
the larger population because the probability that a person or
object will be chosen is not known.
 Methods of Nonprobability Sampling

Purposive Sampling
• sets out to make the sample agree with the population in
regard to certain characteristics
Quota Sampling
• a specific number of particular types of elements are selected
Convenience Sampling
• chooses units which come to hand or are convenient
Judgment Sampling
• selects sample in accordance with an expert’s judgment
 Cases Wherein
Nonprobability Sampling is Useful
• Only few are willing to be interviewed

• Extreme difficulties in locating or identifying

subjects

• Probability sampling is more expensive to

implement
 Methods of Probability Sampling

Probability Sampling is a method of selecting a sample

wherein each unit in the population (for example, a person)
has a known non-zero probability of being included in the
sample. This probability is used in computing estimates of
population characteristics.

It specifies rules and procedures for both sample selection

and estimation. Used if the main objective of the sample
survey is making inferences about the characteristics of the
population under study.
 Probability vs. Non-Probability
Sampling

ISSUE: Which is “better”?

 There is no straightforward answer to resolve this issue

outright.

 Non-probability sampling may be “cheaper” to implement.

 It is only with probability sampling that one is able to

derive objective measures of “errors”.

 The strength of one method is the weakness of the other.

However, the cost of doing a survey can be minimized. With this,
the use of probability sampling is recommended.
Methods of Probability Sampling

* Simple Random Sampling

* Stratified Sampling
* Systematic Sampling
* Cluster Sampling
* Multi-Stage Sampling
 Simple Random Sampling (SRS)
Description of the Design
Random sampling is the process of selecting a sample of size (n)
giving each sampling unit an equal chance of being included in
the sample. An SRS of n observations of the population is a
sample that is chosen in such a way that each subset of n
observations of the population has the same chance of being
selected.
Random sampling may be with replacement (SRSWR) or
Without replacement (SRSWOR). In SRSWR, a chosen
element is always replaced before the next selection is made, so
that an element may be chosen more than once.
Sample Selection Procedure

1. Make a list of the sampling units and number them

from 1 to N, where N is the population size.
2. Select n (distinct for SRSWOR, not necessarily distinct
for SRSWR) random numbers from 1 to N using some
random process; for example, the table of random
numbers or lottery. The sample consists of the units
corresponding to the selected random numbers.
 How To Use The Table of Random Numbers:
The table of random numbers consists of the digits 0 through 9
generated by the computer using a random process and
recorded randomly in rows and columns. To use the table, the
researcher can start with any row and any column. We list
down the steps on how to use the Table of Random Numbers:

1. Determine how many digits there are in the population size N.

Assume that there are M digits in N.
2. Using any m columns of the table, read the random numbers
down the column.
3. Choose the elements whose serial numbers are the same as
those of the random number read.
 How To Use The Table of Random Numbers:

4. Ignore the random numbers greater than N, ignore also those

that have already been selected.
5. If the random number of the first m columns have been
exhausted, continue reading the next m columns. Stop when n
elements have been chosen.

Example: Let the size of the population of interest be 300 students

(N = 300) and the size of the sample be 10 students (n = 10). We
list down the 300 students in a serial listing as 001, 002, 003, . . . ,
300. Following the steps above, the number of digits of the
population size is 3 since N = 300. :Let our starting point be the
3rd row and the 5th column of the table of random numbers.
We use m = 3 columns and the columns will be 5, 6, and 7.
How To Use The Table of Random Numbers:

Starting with row 3, we then read down the table using columns
5, 6, and 7. We get the following three-digit numbers: 694, 016,
983, 940, 000, 848, 641, 176, 830, 374, 528, 092, 759, 036, 124,
214, 773, 129, 371, 438, 868, 003, 011, 670, 131

Numbers greater than 300 are discarded and we stop as soon as

we have obtained the sample size of n = 10. The 10 sampling
units then are 016, 176, 092, 036, 124, 214, 129, 003, 011, and
131. The students corresponding to these serial numbers
comprise the sample. However, if we have a population size of
N = 10,000, the same steps will be performed but this time the
population list will consists of serial numbers 00001, 00002, …,
10000. The number of columns to be utilized will be 5 columns.
 Examples of Simple Random Sampling:

1. Employee records are recorded on their respective

201 file. The Personnel Manager has a list of all the
employees totaling to N = 15,000. If a sample size of n
= 75 employees is to be drawn, then the sample is
called a simple random sample if every possible
sample of n = 75 employees has the same chance of
being included in the sample.
 Examples of Simple Random Sampling

2. The Bureau of Internal Revenue Auditors may study

simple random samples of individual tax payments to
check for correctness of tax declared. Thus they may
wish to estimate the proportion of tax payments not
correctly filed by individual taxpayers.
Advantages
– The theory involved is much easier to
understand than the theory behind other
sampling designs.
– Estimation methods are simple and easy.
Disadvantages
– The sample chosen may be widely spread, thus
entailing high transportation costs.
– A population list, or frame, is needed.
– The sample chosen may not be truly typical of
the population if the population is heterogenous
with respect to the characteristic under study
 Stratified Sampling
Description of the Design
There are cases wherein the population consists of items
which are heterogeneous with respect to the
characteristic under study. In such situations the
population should be divided, or stratified, into more or
less homogeneous subpopulations or strata before
sampling is done.

Stratified random sampling then consists of selecting an

SRS from each of the strata into which the population
has been divided.
Sample Selection Procedure

1. Stratify the population into strata so that ideally

each stratum will consist of more or less
homogeneous units.

2. After the population has been stratified, an SRS is

selected from each stratum.
 Stratified Sampling

Example: If we are making a health study of male students in a

college, we might recognize such strata as
- those who do or do not live at home;
- those who are totally, partially or not at all self-supporting;
- those who do or do not take regular exercise;
- those who do or do not smoke; and so forth.

However, there are other strata, which clearly have no bearing on

the problem.
Advantages
• Stratification may bring about a gain in
precision of the estimates of characteristics of
the population.
• It allows for more comprehensive data analysis
since information is provided for each stratum.
• It is administratively convenient.
Disadvantages
• A listing of the population for each stratum is
needed.
• The stratification of the population may
require additional prior information about the
population and its strata.
 (1-in-k) Systematic Sampling

Description of the Design

Systematic sampling with a random start is a
method of selecting a sample by taking every kth
unit from an ordered population, the first unit
being selected at random. Here k is called the
sampling interval and 1/k the sampling fraction.
Sample Selection Procedure
1. Number the units of the population consecutively
from 1 to N.
2. Determine k by the formula
k=N/n
= population size/sample size
3. Select the random start r (between 1 and k). The unit
corresponding to r is the first unit of the sample.
4. The other units of the sample correspond to r + k, r + 2k, ...

Example of Systematic Sampling

A medical investigator is interested in obtaining information about
the average number of times N = 15,000 specialists prescribed a
certain drug in the previous year. To obtain a sample of n = 1600
specialists, we could select one specialist at random from the first
k = 9 names appearing on the list and then select every ninth name
thereafter until a sample of size 1600 is selected.
Advantages

• Drawing of the sample is administratively easy.

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

Disadvantage

• If periodic regularities are found in the list, a systematic

sample may consist only of similar types.
 Cluster Sampling

Description of the Design

Cluster sampling is a method of selecting a sample of distinct
groups, or clusters, of smaller units called elements. The
sample clusters may be chosen by SRS or by systematic
sampling. Similar to strata in stratified sampling, clusters are
mutually exclusive subpopulations which together comprise
the entire population. Unlike strata, however, clusters are
preferably formed with heterogeneous elements so that each
cluster will be typical of the population.
The number of clusters C in the population is called the size
of the population of clusters. Clusters may be of equal or
unequal sizes.
Sample Selection Procedure
1. List the clusters and number them from 1 to C.
2. Randomly select n numbers from 1 to C. The clusters
corresponding to the selected numbers form the sample of
clusters.
3. Observe all the elements in each sample cluster.

Example of Cluster Sampling

A forester wishes to estimate the average height of
coconut trees on a plantation. The plantation is
divided into 386 plots. An SRS of 20 is selected
and all trees on the sampled plots are measured.
Advantages

• A population list of elements is not needed.

• Listing cost is reduced.

Disadvantages

• The costs and problems of statistical analysis are

greater.
• Estimation procedures are difficult.
 Multistage Sampling
Description of the Design
In multistage sampling the selection of the sample is
accomplished in two or more steps. The population
is first divided into a number of first-stage or
primary units, from which a sample is drawn.
Within the sampled first-stage units, a sample of
second-stage or secondary units is drawn. If desired,
one might add further stages, dividing the population
into a hierarchy of sampling units corresponding to
the different sampling stages.
 Example of Multi-stage Sampling

The following is one possible setup of nationwide

surveys.

Stage Sample

Primary Region

Secondary Province

Third-Stage Municipality/City

Fourth-Stage Barangay

Fifth-Stage Household
Advantages

• Listing cost is reduced.

• Transportation cost is reduced.

Disadvantages

• Estimation procedures are difficult, especially when

the first-stage units are not of the same size.
• The sampling procedure entails much planning
before selection is done.
 Sampling Design

Determination of sample size involves answering a

series of questions on:
• precision or budget?
• margin of error that the researcher is willing to
accept
• probability that the desired margin of error is
achieved ( usually 90%, 95%, 99%)
 Which Sampling Plan to Use?

• When deciding which sampling plan to use, the investigator

must consider the efficiency of the scheme.
• It has already been noted that a stratified sample yields more
reliable results (that is, its sampling error is smaller) than does a
random sample of the same size.
• Cluster sampling may be expected to yield a less reliable result
than random sampling for random samples of the same size.
• The efficiency of a sample scheme refers to the reliability in
relation to unit cost.
 Which Sampling Plan to Use?

Example:

A geographic cluster sample with groups of units in, say

20 locations in a large city may have a lower cost per sampling
unit than a random sample of the same size with the units
scattered here and there about the city. The difference in unit
cost may be so great that the cluster sample will yield more
reliable results than could be had from a random sample for the
same expenditure.
 Margin of Error

• The “margin of error” is a common summary of sampling error that

quantifies uncertainty about a survey result.
• Estimate +/- margin of error  such intervals are sometimes called 95%
confidence intervals and would be expected to contain the true value of
the target quantity (in the absence of nonsampling errors) at least 95% of
the time.
• An important factor in determining the margin of error is the size of the
sample.
• Larger samples are more likely to yield results close to the target
population quantity thus have smaller margins of error than more
modest-sized samples.
 Example of Margin of Error
• In the case of the mayoral poll in which 55 of 100 sampled
individuals support Mr. Atienza, the sample estimate would be that
55 percent support Mr. Atienza – however, there is a margin of
error of 10 percent. Therefore a 95 percent confidence interval for
the percentage supporting Mr. Atienza would be
(55%-10%) to (55% +10%)
or (45 percent, 65 percent), suggesting that in the broader
community the support for Mr. Atienza could plausibly range from
45 percent to 65 percent.
• This implies – because of the small sample size – considerable
uncertainty about whether a majority of the people in the city
actually support Mr. Atienza.
 Example of Margin of Error

• However, if there had been a survey of 1,000 people, 550

of whom support Mr. Atienza, the sample estimate would
again be 55 percent, but now the margin of error for Mr.
Atienza’s support would only be about 3 percent. A 95
percent confidence interval for the proportion supporting
Mr. Atienza would thus be (55%-3%) to (55%+3%)
or (52 percent, 58 percent), which provides much greater
assurance that a majority of the city’s voters support Mr.
Atienza.
 Activity

1. Identify the sample selection procedure used in each of the following cases:
a) A survey obtained a sample of laborers by first classifying the different areas
as either rural area or urban area. After which, a sample of laborers is taken
from each area.
b) To select a sample of households in a province, a sample of provinces were
selected, then a sample of municipalities were chosen from each of the
selected provinces, then a sample of barangays were chosen from each of the
selected municipality, and all households in the selected barangays were
included.
c) For a survey, a sample of municipalities were selected from every province in
the country and included all child laborers in the selected municipalities.
d) In the game of lotto, 6 balls are selected from a container with 42 balls.
e) A car manufacturer conducts quality checking on every 20th car in the
production line.
 Activity

2. Determine the sampling procedure to be used and explain why.

Determine the population under consideration and the
sampling unit.
a) a study is to be conducted to determine the
housing condition of rice farmers in Region III.
b) The President of a University in Bulacan is
interested
in the employee’s opinion on his administrative policies.
c) If you wanted to determine what proportion of novels published
in the Philippines last year had themes based on covid 19 cases,
how might you get a random sample to answer your question?
 Activity

3. Give at least one way of stratifying the following:

a) height of school children enrolled in an elementary
b) expenditure of household for food annual earnings of
BulSU graduates since 2015
c) radio listening habits of radio owners in the Bulacan
Thank you.