24/04/2025

• Creation of sample set

• Retains the accuracy in bringing out the
correct statistical information
• Population → Sample → Population

Sampling is the process of

selecting a small number of
elements (sample) from a
larger defined target group of
elements (population) such
that the information gathered
from the small group will allow
judgments to be made about
the larger groups.

1
24/04/2025

2
 24/04/2025

• Sampling error is any type of bias 1. Define the Population of Interest

that is attributable to mistakes in 2. Identify a Sampling Frame (if possible)
either drawing a sample or 3. Select a Sampling Method
4. Determine Sample Size
determining the sample size 5. Execute the Sampling Plan

3
 24/04/2025

Population of interest is entirely Some Bases for Defining Population:

dependent on
a) Management Problem ◦ Geographic Area
b) Research Problems and; ◦ Demographics
c) Research Design ◦ Usage/Lifestyle
◦ Awareness

For example: If you are interested in Which population of interest would best fit your
research requirement of gathering data from
the average time a person between the those considered “Poor” in the table below?
age of 30-35 takes to recover from a
COVID-19 after undergoing a particular
type of medication, the population of
interest is all persons between 30-35
y.o with COVID-19 undergoing the
medication.

4
 24/04/2025

1. Define the Population of Interest • A list of population elements (people, companies,

2. Identify a Sampling Frame (if possible) houses, cities, etc.) from which units to be sampled can
3. Select a Sampling Method be selected.
4. Determine Sample Size • Difficult to get an accurate list.
5. Execute the Sampling Plan • Sample frame error occurs when certain elements of
the population are accidentally omitted or not included
on the list.

Population: Suppose a researcher is conducting a Population: Consider a study investigating the

survey on customer satisfaction with a particular prevalence of a rare medical condition in a
brand of smartphones. specific region.

Sampling Frame: Consist of a database Sampling Frame: The sampling frame could be a
containing contact information or purchase list of patient records from hospitals, clinics, or
records of all customers who have bought the medical registries within the region.
brand's smartphones from authorized retailers
or online stores.

5
 24/04/2025

NOTE: YOU CAN’T JUST USE ANY LIST YOU A good sample frame for a project on living
COME ACROSS! Care must be taken to make conditions would:
sure your sampling frame is adequate for
your needs. • Include all individuals in the target
population.
• Exclude all individuals not in the target
population.
• Includes accurate information that can be
used to contact selected individuals.

1. Define the Population of Interest 1. Probability Sampling

2. Identify a Sampling Frame (if possible) 2. Non-probability Sampling
3. Select a Sampling Method
4. Determine Sample Size
5. Execute the Sampling Plan

6
 24/04/2025

• Probability sampling is defined as a sampling Types of Probability Sampling

technique in which the researcher chooses
• Simple Random Sampling
samples from a larger population using a • Systematic Random Sampling
method based on the theory of probability • Stratified Random Sampling
• Cluster Sampling
• Random Selection.

SIMPLE RANDOM SAMPLING SYSTEMATIC RANDOM SAMPLING

• A method of probability sampling in which every • a method of probability sampling in which the
unit has an equal nonzero chance of being defined target population is ordered and the
selected. sample is selected according to position using a
skip interval.
Example: Suppose a researcher wants to study the
reading habits of students at a university. They
assign a unique number to each student and then
use a random number generator to select a sample
of 100 students from the entire student population.

7
 24/04/2025

SYSTEMATIC RANDOM SAMPLING STRATIFIED RANDOM SAMPLING

Example: In a study on customer satisfaction at a • a method of probability sampling in which the population
shopping mall, researchers choose to survey every is divided into different subgroups and samples are
10th shopper exiting the mall after selecting a selected from each.
random shopper as the starting point. If the random
start selects the 5th shopper, then the 15th, 25th, Example: A researcher wants to assess the job satisfaction of
35th, and so on will be surveyed. employees in a large company. They stratify employees by
departments (e.g., sales, marketing, HR) and then randomly
select a sample of employees from each department
proportional to its size.

• Steps in Drawing a stratified random sampling:

1. Divide the target population into homogeneous CLUSTER SAMPLING
subgroups or strata
2. Draw random samples from each stratum
3. Combine the samples from each stratum into a single • A cluster sample has all
sample of the target population members from randomly
selected segments of a
population. This is used
when the population falls
into naturally occurring
subgroups.

8
 24/04/2025

CLUSTER SAMPLING
1. Probability Sampling
Example: A public health researcher wants to
study the prevalence of a disease in a city. 2. Non-probability Sampling
They divide the city into neighborhoods
(clusters) and randomly select several
neighborhoods. Within each selected
neighborhood, they test all residents for the
disease.

• Non-probability sampling is defined as a Types of Nonprobability Sampling

sampling technique in which the researcher
selects samples based on the subjective • Convenience Sampling
judgment of the researcher rather than • Judgment Sampling
random selection. • Quota Sampling
• Snowball Sampling

9
 24/04/2025

CONVENIENCE SAMPLING
CONVENIENCE SAMPLING
• A convenience sample consists only of
available members of the population. Example: You are doing a study to determine
the number of years of education each
Example: A researcher conducting a survey on teacher at your college has. You select only
campus satisfaction stands outside a the teachers you currently have this
university library and surveys students as semester.
they exit, selecting those who are
conveniently located for data collection.

JUDGMENT SAMPLING JUDGMENT SAMPLING

• Sampling that relies upon belief that • For example: If you are to create a study of
participants fit certain characteristics who are most likely to be accepted in the
audition of Miss Universe Philippines, then
upon checking the characteristics required,
you can create a judgment on who can
audition for the part.

10
 24/04/2025

QUOTA SAMPLING QUOTA SAMPLING

• a non-probability sampling method that • For example, if your population consists of

involves dividing the population into 40% female and 60% male, your sample
subgroups based on predetermined should reflect those percentages.
characteristics (quotas) and then selecting
participants from each subgroup until the
quota is filled.

SNOWBALL SAMPLING
SNOWBALL SAMPLING
• For example, if you are studying the level
• Snowball sampling or chain-referral
of customer satisfaction among the
sampling involves recruiting participants
members of an elite country club, you will
through referrals from initial participants
find it extremely difficult to collect primary
or through social networks, creating a
data sources unless a member of the club
"snowball" effect as the sample size grows.
agrees to have a direct conversation with
you and provides the contact details of the
other members of the club.

11
 24/04/2025

SNOWBALL SAMPLING
1. Define the Population of Interest
2. Identify a Sampling Frame (if possible)
3. Select a Sampling Method
4. Determine Sample Size
5. Execute the Sampling Plan

1. The sample size should be carefully fixed so that it will Level of Precision
be adequate to draw valid and generalized conclusions.
• Also known as sampling error
2. To determine the appropriate sample size: the level of • the range in which the true value of the population is
precision, the level of confidence or risk, and the degree estimated to be.
of variability in the attributes being measured are to be • Often expressed in percentage points(ex. ±5%)
determined. • Precision is determined by a statistical method called a
standard deviation.

12
 24/04/2025

Level of Precision Level of Confidence

• High standard deviations indicate low precision, low • The risk level of confidence level is based on ideas of
standard deviations indicate high precision Central Limit Theorem.
• The key idea in the Central Limit Theorem is that when a
population is repeatedly sampled, the average value of
the attribute obtained by those samples is equal to the
true population value.

Level of Confidence Confidence Interval

• In a normal distribution, approximately 95% of the • a range of values we are fairly sure our true value lies in.
sample values are within two standard deviations of the
true population value. In solving for the confidence interval, the sample mean,
• In other words, this means that if a 95% confidence level sample standard deviation, confidence level value, and
is selected, 95 out of 100 samples will have the true sample size are important.
population value within the range of precision specified.

13
 24/04/2025

Confidence Interval Confidence Interval Example:

• We measure the heights of 40 randomly chosen men, and
get a mean height of 175cm. We also know the standard
deviation of men's heights is 20cm.
Step 1: Determine the following:
𝒙 (𝒔𝒂𝒎𝒑𝒍𝒆 𝒎𝒆𝒂𝒏) = 𝟏𝟕𝟓 𝒄𝒎
𝒔 (𝒔𝒕𝒂𝒏𝒅𝒂𝒓𝒅 𝒅𝒆𝒗𝒊𝒂𝒕𝒊𝒐𝒏) = 𝟐𝟎 𝒄𝒎
𝒏 𝒏𝒐 𝒐𝒇 𝒐𝒃𝒔𝒆𝒓𝒗𝒂𝒕𝒊𝒐𝒏 = 𝟒𝟎 𝒎𝒆𝒏

Step 2: Determine the confidence level value (z-value). • Step 3: Use the z-value to compute for C.I.
Assume: Given:
x = 175 cm
Confidence level = 95% s = 20 cm
n = 40 sample
Therefore, z=1.95996 z = 1.960

𝟐𝟎
𝑪𝑰 = 𝟏𝟕𝟓 ± 𝟏. 𝟗𝟔𝟎
𝟒𝟎

𝑪𝑰 = 𝟏𝟕𝟓 ± 𝟔. 𝟏𝟗𝟖

14
 24/04/2025

1. Using census for small populations

Degree of Variability(Proportion, p)
2. Using a sample size of a similar study
• refers to the distribution of attributes in the population
3. Using published tables
• The variables with more homogeneous population, the
4. Using formulas to calculate a sample size
smaller the sample size required. If the more
heterogeneous population, the larger the sample size
required to obtain a given level of precision.
• When assuming the degree, always use 50% for maximum
variability, p.

1. Using census for small populations 1. Using census for small populations
2. Using a sample size of a similar study
• Eliminates sampling error 3. Using published tables
• Use the entire population as the sample. Although cost 4. Using formulas to calculate a sample size
considerations make this impossible for large populations,
a census is more attractive for small populations.

15
 24/04/2025

2. Using a sample size of a similar study 1. Using census for small populations
2. Using a sample size of a similar study
• Without reviewing the methods used in these studies 3. Using published tables
may run the risk of repeating errors that were made in 4. Using formulas to calculate a sample size
determining the sample size for another study.

3. Using published tables

• A third way to determine sample size is to rely on

published tables, which provide the sample size for a
given set of criteria.

16
 24/04/2025

1. Using census for small populations

2. Using a sample size of a similar study
3. Using published tables
4. Using formulas to calculate a sample size

Using formulas to calculate a sample size Cochran Formula

• Cochran Formula • allows you to calculate an ideal sample size given a

• Taro Yamane Sampling Formula desired level of precision, desired confidence level, and
the estimated proportion of the attribute present in the
population.

17
 24/04/2025

Cochran Formula Cochran Formula

• Suppose we are doing a study on the inhabitants of a
• Cochran’s formula is considered especially appropriate in large town and we want to find out how many
situations with large populations. households serve breakfast in the mornings.
• We don’t have much information on the subject to
begin with.
𝒛𝟐 𝒑𝒒 • We’re going to assume that half of the families serve
• Formula: 𝒏𝒐 = 𝟐 breakfast: this gives us maximum variability. So p = 0.5.
𝒆 • Now let’s say we want 95% confidence, and at least 5
• Where e is the desired level of precision (i.e. the margin percent—plus or minus—precision.
of error), p is the (estimated) proportion of the • A 95 % confidence level gives us Z values of 1.96, per
the normal table
population which has the attribute in question, q is 1 – p.

Cochran Formula Modified Cochran Formula

𝒛𝟐 𝒑𝒒
• Using the formula, 𝒏𝒐 = 𝒏𝒐
𝒆𝟐 • Using the formula, n= 𝒏𝒐 𝟏
𝟏
𝑵
• Here n0 is Cochran’s sample size recommendation, N is the population
𝟏.𝟗𝟔 𝟐(𝟎.𝟓)(𝟏 𝟎.𝟓) size, and n is the new, adjusted sample size. In our earlier example, if
𝒏𝒐 =
𝟎.𝟎𝟓𝟐 there were just 1000 households in the target population, we would
𝒏𝒐 = 𝟑𝟖𝟒. 𝟏𝟔 calculate

𝟑𝟖𝟓
𝒏𝒐 = 𝟑𝟖𝟓 𝒔𝒂𝒎𝒑𝒍𝒆 𝒔𝒊𝒛𝒆 𝒏= 𝟑𝟖𝟓 𝟏 = 𝟐𝟕𝟖 𝐬𝐚𝐦𝐩𝐥𝐞 𝐬𝐢𝐳𝐞
𝟏
𝟏𝟎𝟎𝟎

18
 24/04/2025

Taro Yamane Formula Taro Yamane Formula / Slovin’s Formula

• First of all, you should be working with a finite population and if the • Utilizing the previous example:
population size is known, the Yamane formula for determining the • if there were just 1000 households in the target population, we
sample size is given by: would calculate

𝟏𝟎𝟎𝟎
𝒏= = 𝟐𝟖𝟓. 𝟕𝟏 ≈ 𝟐𝟖𝟔 𝒔𝒂𝒎𝒑𝒍𝒆 𝒔𝒊𝒛𝒆
𝟏 + (𝟏𝟎𝟎𝟎)(𝟎. 𝟎𝟓𝟐 )
• Where n= corrected sample size, N = population size, and e =
Margin of error (MoE), e = 0.05 based on the research condition.

1. Define the Population of Interest

2. Identify a Sampling Frame (if possible)
3. Select a Sampling Method
4. Determine Sample Size
5. Execute the Sampling Plan

19