P.

R
Types of Quantitative Research Design
1. Descriptive Research. When little is known about the research problem, then it is appropriate to use
descriptive research design. It is a design that is exploratory in nature. The purpose of descriptive
research is basically to answer questions such as who, what, where, when, and how much. So this
design is best used when the main objective of the study is just to observe and report a certain
phenomenon as it is happening.

2. Correlational Research. The main goal of this design is to determine if variable increases or
decreases as another variable increases or decreases. This design seeks to establish an association
between variables. It does not seek cause and effect relationship like descriptive research; it measures
variables as it occurs. It has two major purposes: (a) to clarify the relationship between variables and
(b) predict the magnitude of the association. However, the extent of the purpose of correlational
research depends on the scope and delimitation of the study.

3. Ex Post Facto. If the objective of the study is to measure a cause from a pre-existing effects, then Ex
Post Facto research design is more appropriate to use. In this design, the researcher has no control over
the variables in the research study. Thus, one cannot conclude that the changes measured happen
during the actual conduct of the study.
The last two types of quantitative research designs are identifiable for the existence of treatment or
intervention applied to the current research study. Intervention or treatment pertains to controlling or
manipulating the independent variable in an experiment. It is assumed that the changes in dependent
variables were caused by the independent variable.
There are also two groups of subjects, participants, or respondents in quasi-experimental and
experimental research. The treatment group refers to the group subjected to treatment or intervention.
The group not subject to treatment or intervention is called the control group.

4. Quasi-Experimental. The term means partly, partially, or almost – pronounced as kwahz-eye. This
research design aims to measure the causal relationship between variables. The effect measured is
considered to have occurred during the conduct of the current study. The partiality of quasi-
experimental design comes from assigning subjects, participants, or respondents into their groups. The
groups are known to be already established before the study, such as age educational background and
nationality. Since the assignment of subjects, participants, or respondents are not randomly assigned
into an experimental or control groups, the conclusion of results is limited.

5. Experimental Research. This research design is based on the scientific method called experiment
with a procedure of gathering data under a controlled or manipulated environment. It is also known as
true experimental design since it applies treatment and manipulation more extensively compared to
quasi-experimental design. Random assignment of subjects or participants into treatment and control
group is done increasing the validity of the study. Experimental research, therefore, attempts to affect a
certain variable by directly manipulating the independent variable.

Power Analysis. This approach is founded on the principle of power analysis. There are two principles
you need to consider if you are going to use this approach: these are statistical power and effect size.

The probability of rejecting the null hypothesis is called statistical power.

Simple Random Sampling. It is a way of choosing individuals in which all members of the accessible
population are given an equal chance to be selected. These are fish bowl technique, roulette wheel, or
use of the table of random numbers. This technique is also readily available online.

Stratified Random Sampling. The same with simple random sampling, stratified random sampling
also gives an equal chance to all members of the population to be chosen

Cluster Sampling. This procedure is usually applied in large-scale studies, geographical spread out of
the population is a challenge, and gathering information will be very time-consuming. Similar to
stratified random sampling, cluster sampling also involves grouping of the population according to
subgroups or clusters.

Systematic Sampling. This procedure is as simple as selecting samples every nth (example every 2nd,
5th) of the chosen population until arriving at a desired total number of sample size.
PROBABILITY SAMPLING Probability sampling means that every member of the target population
has a known chance of being included in the sample. Probability sampling methods include Simple
Random, Stratified Random, Cluster Sampling, Systematic Sampling, Multi-stage Sampling, etc.

NON-PROBABILITY SAMPLING is defined as a sampling technique in which the researcher

selects samples based on the subjective judgment of the researcher rather than random selection. It is a
less stringent method. This sampling method depends heavily on the expertise of the researchers. It
includes Qouta Sampling, Snowball Sampling, Judgment Sampling, Convenience Sampling, etc.

Research Instruments are basic tools researchers used to gather data for specific research problems.
Common instruments are performance tests, questionnaires, interviews, and observation checklist.

Characteristics of a Good Research Instrument

Concise. Have you tried answering a very long test, and because of its length, you just pick the answer
without even reading it? A good research instrument is concise in length yet can elicit the needed data.

Sequential. Questions or items must be arranged well. It is recommended to arrange it from simplest to
the most complex. In this way, the instrument will be more favorable to the respondents to answer.

Valid and reliable. The instrument should pass the tests of validity and reliability to get more
appropriate and accurate information.
Easily tabulated. Since you will be constructing an instrument for quantitative research, this factor
should be considered. Hence, before crafting the instruments, the researcher makes sure that the
variable and research questions are established. These will be an important basis for making items in
the research instruments

Common Scales Used in Quantitative Research

Likert Scale. This is the most common scale used in quantitative research. Respondents were asked
to rate or rank statements according to the scale provided.

Types of Validity of Instrument

Face Validity. It is also known as “logical validity.” It calls for an initiative judgment of the
instruments as it “appear.” Just by looking at the instrument, the researcher decides if it is valid.

Content Validity. An instrument that is judged with content validity meets the objectives of the study.
It is done by checking the statements or questions if this elicits the needed information.

Construct Validity. It refers to the validity of instruments as it corresponds to the theoretical construct
of the study. It is concerning if a specific measure relates to other measures.

Concurrent Validity. When the instrument can predict results similar to those similar tests already
validated, it has concurrent validity.

Predictive Validity. When the instrument is able to produce results similar to those similar tests that
will be employed in the future, it has predictive validity. This is particularly useful for the aptitude test.

Reliability of Instrument
Test-retest Reliability. It is achieved by giving the same test to the same group of respondents twice.
The consistency of the two scores will be checked.

Equivalent Forms Reliability. It is established by administering two identical tests except for
wordings to the same group of respondents

Page 2
Internal Consistency Reliability. It determines how well the items measure the same construct. It is
reasonable that when a respondent gets a high score in one item, he will also get one in similar items.

Data analysis in research is a process in which gathered information are summarized in such a manner
that it will yield answers to the research questions.

Descriptive Statistical Technique provides a summary of the ordered or sequenced data from your
research sample.

Test of Relationship between Two Variables

• Pearson’s r (parametric)
•Phi coefficient (non-parametric for nominal and dichotomous variables)
•Spearman’s rho (non-parametric for ordinal variable)

Test of Difference between Two Data Sets from One Group

•T-test for dependent samples (parametric)
•McNemar change test (non-parametric for nominal and dichotomous variables)
•Wilcoxon signed-rank test (non-parametric for ordinal variable

Test of Difference between Two Data Sets from Two Different Groups
• T-test for independent samples (parametric)
•Two-way chi-square (non-parametric for nominal variable)
•Mann-Whitney U test (non-parametric for ordinal variable)

Test More than Two Population Means

•Analysis of Variance or ANOVA (parametric)

Test the Strength of Relation or Effect or Impact

•Regression (parametric)

Page 3