Design, Sampling, and

Instrument

Practical Research:
Lesson 5
Quantitative Research
Lesson Objectives
• Describe different quantitative research
designs
• Identify different probability and non-
probability sampling techniques
• Determine the different types of rating scale
• Differentiate validity from reliability
• Identify the steps in developing an instrument
• Develop an instrument for quantitative
research
 Research Design

Practical Research:
Lesson 5
Quantitative Research
Research Design
• A procedural plan adopted by researchers to
answer questions objectively, accurately,
economically and with validity (Kumar, 1999).
• A blueprint or a plan for action, specifying the
methods and procedures for collecting and
analyzing the needed information, fulfilling the
research objectives, and finding the solutions
(Rani, 2004).
Designs in Quantitative Research

•Descriptive
Non-
experimental •Correlational

•Quasi-experimental
Experimental •True experimental
Non-Experimental Design
• A set of techniques used to conduct a
quantitative research without manipulation of
any variable (Creswell, 2018).
• Measurement of variables is done naturally
without any treatment or condition set.
Non-Experimental Designs
• Descriptive research- describes and interprets
the current status of people, places, conditions,
or events.
• Correlational research- discovers and assesses
the relationships between two or more
variables; the relationship examined is linear,
not causal.
Experimental Design
• A set of techniques used in quantitative
research when manipulation of variables is
performed (Creswell, 2018).
• Treatments or conditions are tested through
which a researcher measures their effects on
the participants.
Experimental Design
• Quasi-experimental research- determines cause-
effect relationship; somewhat similar to true
experimental research, but is different in a way
that it does not manipulate variables nor randomly
assigns groups.
• True experimental research- ascertains cause-
effect relationships among variables; manipulates
the independent variable and randomly assigns
groups.
 Sampling

Practical Research:
Lesson 5
Quantitative Research
Sampling in Quantitative Research
• Sampling refers to the statistical process of
selecting a subset of a population of interest
for purposes of making observations and
statistical inferences about that population
(Bhattacherjee, 2010).
Population vs. Sample

Source: https://www.omniconvert.com/what-is/sample-size/
Population vs. Sample
• Population is the entire group that you
want to draw conclusions about.
• Sample is the specific group of
individuals that you will collect data
from.

McCombes (2019)
Sampling frame and sample size
• Sampling frame is the actual list of individuals
that the sample will be drawn from. Ideally, it
should include the entire target population
(and nobody who is not part of that
population).

• Sample size refers to the number of

participants included in a study.
McCombes (2019)
Sampling can be probabilistic or non-
probabilistic.

https://keydifferences.com/difference-between-probability-and-non-probability-sampling.html
Probability Sampling
• Probability sampling, also known as random
sampling is a method in quantitative research
for selecting respondents so that “each
individual has an equal probability of being
selected from the population, ensuring that the
sample will be representative of the
population” (Creswell, 2018).
Probability Sampling Techniques
Simple random

Systematic

Stratified

Cluster
Simple Random
• Least biased sampling technique
• Can yield most generalizable results
• Each unit of the population has an equal
chance of being selected in the sample.
• Units are selected randomly from the
target population.
Systematic
• Units are selected at a regular interval.
• The interval could be order, space or
time.
Stratified
• Subgroups are identified based on
certain characteristics, e.g. age,
nationality, course.
• The subgroups are called as strata
(stratum for singular).
Cluster
• Very much applicable in cases where
units of population are scattered over a
wide geographical location.
• Subgroups are identified through
clustering, e.g. city, district, institution.
• The subgroups are called as clusters.
Probability Sampling Techniques

Source: https://www.scribbr.com/methodology/sampling-methods/
Non-probability Sampling
• Also referred to as judgement sampling.
• No random selection is made; selection is
based on the researcher’s subjective
judgment.
• Best used when the target population is
not precisely defined.

Alvi (2016)
Non-probability Sampling Techniques
• Volunteer/voluntary response – The participants
volunteer themselves to be part of the study.
• Convenience – Also called as accidental or opportunity
sampling; participants include those who are easy to
approach.
• Purposive – Participants are selected because they
meet specific criteria set for the purpose of the study.
• Snowball – Also called as chain sampling; participants
are asked to refer to other elements in the population.

Alvi (2016)
Sampling Procedure
• Defining the target population
• Choosing a sampling frame
• Choosing a sampling frame using a well-
defined sampling technique
Slovin’s Formula in Calculating Sample Size

n = N / (1+Ne²)

whereas:
n = no. of samples
N = total population
e = error margin / margin of error

Note: Margin of error may be set at 5% (0.05).

 Research Instrument

Practical Research:
Lesson 5
Quantitative Research
Measurement in quantitative research

• Measurement is essential in quantitative research

because of the following reasons:
(1) allows us to delineate fine differences between
people in terms of the characteristic in question;
(2) gives us a consistent device for making
distinctions; and
(3) provides the basis for more precise estimates of
the degree of relationships between concepts.
Processes involved in measurement in
quantitative research
• Conceptualization – the process by which
concepts are specified in research based on
what the researcher means.
• Operationalization – the process of connecting
concepts to observations.
Levels of measurement
1. Nominal – identifies variables whose values
have no mathematical interpretation; they vary in
kind or quality, but not in amount (e.g. color,
gender, blood type).

2. Ordinal – the numbers assigned to cases specify

only the order of the cases, permitting greater
than and less than distinctions (e.g. academic
rank, job positions, grade level).
Levels of measurement
3. Interval – the values of variable represent fixed
measurement units but have no absolute, or fixed,
zero point (e.g. temperature, IQ test, personality
test).

4. Ratio – measurement represents fixed

measuring units and an absolute zero point (zero
means no amount of whatever the variable
indicates) (e.g. age, income, length of residency).
Levels of measurement

Source: https://conjointly.com/kb/levels-of-measurement/
Research instrument
• A research instrument is a tool used to obtain data
on a topic of interest from the target
respondents/participants.
• A research instrument can be a test, survey, scale,
questionnaire, or even a checklist.
• In quantitative research, a research instrument is
structured in a way that responses can be quantified
and subjected for statistical analysis.
Rating scale
• A rating scale is used to capture a respondent’s
reactions or responses to a given item in the
scale.
• A nominal scale captures a yes/no response
while an interval-scale item captures a value
between “strongly agree” to “strongly disagree”.
Types of rating scale
Binary

Likert

Semantic Differential

Guttman
Binary scale
• A binary scale is a nominal scale consisting of
binary items that assume one of two possible
values, for example, yes or no, true or false, and
so on.
Example of a binary scale

Source: Bhattacherjee, A. (2012). Social science research: Principles, methods, and

practices.
Likert scale
• Simply-worded statements to which respondents
can indicate their extent of agreement or
disagreement on a five- or seven-point scale
ranging from “strongly agree” to “strongly
disagree”.
Example of a Likert scale

Source: Bhattacherjee, A. (2012). Social science research: Principles, methods, and

practices.
Semantic differential scale
• A composite or multi-item scale in which
respondents are typically asked to indicate their
opinions or feelings toward a single statement
using different pairs of adjectives framed as
polar opposites or two extremes.
Example of a semantic differential scale

Source: Bhattacherjee, A. (2012). Social science research: Principles,

methods, and practices.
Guttman scale
• Also a composite or multi-item scale, but is
different in a way that it uses a series of items
arranged in an increasing order of intensity of
the concept of interest, from least intense to
most intense.
Example of a Guttman scale

Source: Bhattacherjee, A. (2012). Social science research:

Principles, methods, and practices.
Validity and reliability
• Validity refers to the issue of whether an indicator (or
a set of indicators) that is used to measure a concept
really measures that concept.
• Reliability refers to the consistency of a measure of a
concept.
• A research instrument may be valid but not reliable,
or reliable but not valid.
• A good research instrument should be both valid and
reliable.
Types of validity

Face Content Construct

• Instrument • Instrument • Instrument

has a measures all measures
“physical aspects of the what it
look” that it concept. intends to
will work. measure
(theoretically-
aligned).
Types of reliability

Internal
Interrater Test-retest
consistency
• The • The • The
consistency consistency consistency
of the of an of an
instrument instrument instrument
itself across raters across time
or observers
Validity and reliability

Source: https://gradehub.com/blog/validity-and-reliability/
Some decisions to consider before
developing an instrument
• Adopt – using an existing instrument as is without any
changes at all; permission from the author needed;
reliability check needed.
• Adapt – using an existing instrument but with some
changes to suit the needs of the present research;
permission from the author needed; validation and
reliability check needed.
• Self-develop – developing a new instrument to measure
the research problem at hand; validation and reliability
check needed.
Steps in developing an instrument
• Review your variables, research problem, and
research questions; make sure they are aligned.
• Decide on the type of questions, the type of
instrument, and the number of items to develop.
• Write the items that would answer your research
questions; don’t forget to also state the instructions
about how to answer the instrument.
Steps in developing an instrument
• Have the instrument validated by an expert; make
necessary changes based on the suggestions by the
expert.
• Prepare the final draft of your instrument, and pilot
test it among a small group of target respondents.
• Perform statistical analysis to determine reliability of
the instrument.
• Revise if necessary, and then finalize the instrument.