MODULE 3: DATA
COLLECTION
Objectives
At the end of the lesson, the students should be able to:
1.enumerate and differentiate the methods of data collection;
2.define sampling;
3.compute for the sample size, margin of error, and population
size;
4.differentiate probability and non-probability sampling;
5.enumerate the different types of non-probability sampling;
6.enumerate the different types of probability sampling.
Methods of Data
Collection
The method to be used
depends on:
a. researcher’s design of the
study
b. type of data
c. time allotment to complete
the study
d. researcher’s financial
capacity
Methods of Data
Collection
Objective Method
Collects data by actual observation of the unit or process under the
study which may require the use of measuring instrument or a counting
device.
Subjective Method
Collects data or information provided by the respondents. Data are
usually collected by using a questionnaire.
Use of Existing Records
Data from this method were previously collected by another person or
institution for some other purpose. A researcher who makes use of data
from existing records must properly acknowledge the source(s) of data
Methods of Data
Collection
Remarks:
Primary data are those collected directly from the source, i.e. data
collected either by objective or subjective methods
secondary data are those which have been acquired through the use of
existing records.
Sampling
It is the process of selecting a portion of the population.
Reasons for Taking a Sample
1. The cost of collecting and processing data is obviously lower.
2. Complete survey is sometimes physically impossible as when the
number of units is infinitely large or when some of them are totally
inaccessible.
3. Complete survey is senseless whenever the acquisition of the
desired information destroys the elementary units of interest.
4. Complete survey is senseless whenever it produces information
that comes too late.
5. For a given cost, sampling can provide more detailed information
than a complete enumeration (total population).
6. Sampling can provide more accurate data than a complete survey.
Sample Size Determination
Slovin’s Formula (finding the sample size)
n= N where: n = sample size
1 + Ne2 N = population size
e = margin of error
finding the finding the margin of
population size error
√
𝑁 −𝑛
𝑛
𝑁=
1− 𝑛 𝑒
2
𝑒= ∗100
𝑁𝑛
Example:
A group of researchers will conduct a survey to find out the opinion of
residents of a particular community regarding the Homeowner’s
Association Project. If there are 20,000 residents in the community, what
should the sample size be using:
𝟏𝟎 % 5 1
Example:
Compute for the margin of error to be used if 800 sample units are
required from a population of 2,400.
Example:
A researcher plans to get 588 sample units from a population N using a 4%
margin of error. What is the value of the population size N?
Sampling Method
Probability Sampling states that every unit has a “chance” of being
selected.
Nonprobability Sampling is based on the assumption that there is an
even distribution of characteristics within the population and any
sample would be a representative, and hence, the results will be
accurate. There is no assurance that each item has a chance of being
included.
Types of Non-Probability
Sampling
Purposive Sampling
The respondents of the study will be chosen based on their knowledge of the
information required by the researcher.
Quota Sampling
This technique is commonly used in opinion polls. After meeting the required
number of sample points, the researcher is through with his collection of data.
Convenience Sampling
This technique is resorted to by the researchers who need information the
fastest way possible. For example, the telephone can be used to interview the
respondents about their opinions on a certain issue. This method may be fast
but it is also biased because those who have no telephones do not have a
chance to be included in the study.
SAMPLING FRAME - A complete listing of
Probability Sampling all the individuals or entities in the universe.
THREE RANDOMIZATION TECHNIQUES
Chips-in-the-box method
N identically-sized chips numbered from 1 to N are placed in a box. The chips will be thoroughly
mixed to ensure that each has equal probability of being selected. Desired number of chips will be
drawn and the number indicated in the chips will be recorded.
Random number generator key in the calculator
Using a calculator, a random number is generated by pressing the RAN# key. This will be multiplied
by N and rounded-off to the nearest whole number. The process is repeated, recording each
rounded-off number, until the preferred number of samples is obtained. Duplicates will be ignored
and replaced.
Table of Random Numbers
A starting point in the table will be randomly selected by dropping a pen vertically on the page.
The number where the pen landed will be the starting point. The number of digits in N will be the
number of digit-columns that will be considered. Move in any direction and the numbers less than or
equal to N will be recorded until the preferred number of samples is completed. If the last number
Types of Probability Sampling
Simple Random Sampling
This is the most basic sampling technique.
A simple random sampling is a sampling technique in which every
element of the population has the same chance of being selected for
inclusion in the sample.
A simple random sample maybe obtained using any of the
randomization techniques.
Types of Probability Sampling
Systematic Sampling
This is a random sampling technique in which every kth element of the
population is selected until the desired number of elements in the
sample is obtained.
The value of k is calculated by dividing the number of elements in the
population by the number of elements in the desired sample. The value
of k is the sample interval.
𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑒𝑙𝑒𝑚𝑒𝑛𝑡𝑠𝑖𝑛 𝑡h𝑒 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛
𝑘=
𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑒𝑙𝑒𝑚𝑒𝑛𝑡𝑠𝑖𝑛 𝑡h𝑒 𝑠𝑎𝑚𝑝𝑙𝑒
𝑁 k = sampling interval
𝑘= N = population size
𝑛
n = sample size
Types of Probability Sampling
Stratified Sampling
The population is partitioned into several subgroups called strata
based on some characteristics like year level, gender, age, etc.
This is a random sampling technique in which the population is first
divided into strata and then samples are randomly selected separately
from each stratum
Equal allocation – the sample size n to be taken from each stratum
will be determined by where L is the number of strata or subgroups.
Proportional allocation – the sample size n to be taken from each
stratum is given by the formula:
Example:
Select 120 grade 11 students in your school using stratified sampling with equal
allocation to determine their opinion on the new mode of learning.
Section Number Number
of of
Student Samples
s
ABM 54
Humss 21
Stem 1 44
Stem 2 46
Stem 3 47
Stem 4 45
Example:
Select 120 grade 11 students in your school using stratified sampling with
proportional allocation to determine their opinion on the new mode of learning.
Section Number Number
of of
Student Samples
s
ABM 54
Humss 21
Stem 1 44
Stem 2 46
Stem 3 47
Stem 4 45
Types of Probability Sampling
Cluster or Area Sampling
Population is divided into clusters. From these clusters, a random
sample of clusters will be drawn.
All the elements from the sampled clusters will make up the sample.
Sometimes, clusters are too large and there is a need for a second set
of smaller clusters to be taken from the original clusters.
For example, a researcher could divide the province into
towns/municipalities. A sample of towns will be selected using SRS.
She/he could then divide the towns into barangay/barrio. From these
towns, a sample of barrios will be selected at random. From these
barrios, a sample of houses will be identified. This technique is called
multi stage cluster sampling.
