RESEARCH REPORT

3rd Quarter
LESSON 1: CHAPTER 3 -This design seeks to establish an
QUANTITATIVE RESEARCH association between
- is more systematic and controlled than variables.
qualitative. However, both research - It does not seek cause and effect
methods have a statement of the relationship like descriptive
problem to investigate. research; it measures variables
- Each type of research has a unique as it occurs.
design to follow that will also lead the - It has two major purposes: (a)
direction of sampling procedure, data to clarify the relationship
gathering, and data analysis. Each between variables and (b) predict
research type also aims to answer the magnitude of the association.
specific research questions; how it will - However, the extent of the
be answered is determined by its design purpose of correlational research
- important considerations also are the depends on the scope and
variables of the study. The type of delimitation of the study.
variables provides paradigm (your 3) EX POST FACTO
conceptual framework), which helps the - If the objective of the study is to
researcher decide on what will be the measure a cause from a pre-
design of the study. existing effect, then Ex Post Facto
research design is more
RESEARCH DESIGN appropriate to use
- logical and coherent overall strategy that - the researcher has no control
the researcher uses to integrate all the over the variables in the research
components of the research study study. Thus, one cannot conclude
- the research designs in quantitative that the changes measured
research are mostly pre-established. happen during the actual conduct
Hence having an appropriate research of the study.
design in quantitative research, the - The last two types of quantitative
researcher will have a clearer research designs are identifiable
comprehension of what he is trying to for the existence of treatment or
analyze and interpret. intervention applied to the
current research study.
Intervention or treatment
TYPES OF QUANTITATIVE RESEARCH
pertains to controlling or
DESIGN
manipulating the independent
1) DESCRIPTIVE RESARCH
variable in an experiment. It is
- When little is known about the
assumed that the changes in
research problem, then it is
dependent variables were caused
appropriate to use descriptive
by the independent variable.
research design.
- There are also two groups of
- It is a design that is exploratory in
subjects, participants, or
nature.
respondents in quasi-
- The purpose of descriptive
experimental and experimental
research is basically to answer
research. The treatment group
questions such as who, what,
refers to the group subjected to
where, when, and how much.
treatment or intervention. The
- when the main objective of the
group not subject to treatment or
study is just to observe and
intervention is called the control
report a certain phenomenon as it
group.
is happening.
4) QUASI- EXPERIMENTAL
2) CORRELATIONAL RESEARCH
- means partly, partially, or
- The main goal of this design is to
almost – pronounced as kwahz-
determine if variable increases or
eye.
decreases as another variable
- aims to measure the causal
increases or decreases.
relationship between variables.
RESEARCH REPORT
3rd Quarter
The effect measured is understood that this population
considered to have occurred possesses common characteristics about
during the conduct of the current which the research aims to explore.
study. - There are two types of population:
- The partiality of quasi- target population and accessible
experimental design comes from population.
assigning subjects, participants, - The actual population is the TARGET
or respondents into their groups. POPULATION, for example, all Senior
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
High School Students enrolled in
not randomly assigned into an
Science, Technology, Engineering, and
experimental or control groups,
Mathematics (STEM) in the Division of
the conclusion of results is
Cagayan de Oro City. While the
limited.
ACCESSIBLE POPULATION is the
5) EXPERIMENTAL RESEARCH
portion of the population in which the
- based on the scientific method
researcher has reasonable access, for
called experiment with a
example all Senior High School enrolled,
procedure of gathering data
STEM strand at Marayon Science High
under a controlled or manipulated
School – X.
environment.
- also known as true
experimental design since it SAMPLE REPRESENTATIVE
applies treatment and - Identified when the whole population is
manipulation more extensively too costly or time-consuming or
compared to quasi-experimental impractical to consider
design. SAMPLING
- Random assignment of subjects - representative subset of the population
or participants into treatment and - A good sample should have
control group is done increasing characteristics of the represented
the validity of the study. population – characteristics that are
- attempts to affect a certain within the scope of the study with fair
variable by directly manipulating accuracy.
the independent variable. - the larger the sample, the more
reliable the sample be, but still, it will
SAMPLING PROCEDURE AND THE depend on the scope and delimitation
SAMPLING and research design of the study.
- stated the subjects, participants, or
respondents of your study. APPROACHES IN IDENTIFYING THE SAMPLE
- described the characteristics which SIZE
qualify to be the source of your research 1) HEURISTICS
data. - rule of the thumb for sample size
- how many of the subjects, participants, - The early established approach
or respondents should be selected as a by Gay (1976) stated by Cristobal
source of data?
POPULATION AND SAMPLING
- The first step in determining the sample
size is identifying the population of the
topic of interest.
- The POPULATION is the totality of all
the objects, elements, persons, and
characteristics under consideration. It is
RESEARCH REPORT
3rd Quarter
and Dela Cruz-Cristobal (2017, p dependent variables, the effect size
172), sample sizes for different suggests the extent of the relationship
research designs are the between these two variables.
following: Henceforth, the higher the effect size,
- Lunenberg and Irby (2008), as means the greater the level
cited by Barrot (2017, p 107), difference between the
also suggested different sample experimental and control groups.
sizes for each quantitative EXAMPLE: your research study reveals
research design. that there is a difference in the pretest
LITERATURE REVIEW and posttest scores of the students in
- Another approach is by reading similar or the given anxiety test after
related literature and studies to your implementing a psychosocial
current research study. intervention. With the effect size, you
- Using this approach increases the will have an idea of how small or large
validity of your sampling procedure. the difference is
FORMULAS
- being established for the PROBABILITY SAMPLING IN QUANTITATIVE
computation of an acceptable RESEARCH
1) SIMPLE RANDOM SAMPLING
- a way of choosing individuals in
which all members of the
accessible population are given
an equal chance to be selected.
- Various ways to obtain samples:
sample size. The common
fish bowl technique, roulette
formula is Slovin’s Formula
wheel, or use of the table of
SLOVIN’S FORMULA
random numbers.
n = N/ 1 + Ne2
2) STRATIFIED RANDOM SAMPLING
where:
- gives an equal chance to all
n = sample size
members of the population to be
N = total population (actual population)
e = margin of error

POWER ANALYSIS
- This approach is founded on the principle
of power analysis.
- There are two principles you need to chosen.
consider if you are going to use this - the population is first divided into
approach: these are statistical power strata or groups before selecting
and effect size. the samples. The samples are
- STATISTICAL POWER is the probability chosen from these subgroups and
of rejecting the null hypothesis. It not directly from the entire
suggests that indeed there is a population.
relationship between the independent - best used when the variables of
and dependent variables of the research the study are also grouped into
study. The ideal statistical power of a classes such as gender and grade
research study is 80%. It will be used level.
to identify the sufficient sample size for
measuring the effect size of a certain 3) CLUSTER SAMPLING
treatment. - usually applied in large-scale
- EFFECT SIZE is the level of difference studies, geographical spread out
between the experimental group and the of the population is a challenge,
control group and gathering information will be
- If the statistical power tells that very time-consuming.
relationship between independent and
RESEARCH REPORT
3rd Quarter
- involves grouping of the  Multi stage sampling
population according to 2) NON-PROBABILITY SAMPLING
subgroups or clusters. - researcher selects samples based
- where multiple clusters of people on the subjective judgment of the
from the chosen population will researcher rather than random
be created by the researcher in selection. It is a less stringent
order to have homogenous method.
characteristics  Quota sampling
EXAMPLE: a researcher would  Snowball sampling
like to interview of all public  Judgment sampling
senior high school students  Convenience sampling
across Mindanao. As a researcher
cluster will be selected to satisfy LESSON 3: RESEARCH INSTRUMENT,
the plan size. In the given VALIDITY AND RELIABILITY
example, the first cluster can be RESEARCH INSTRUMENTS
by region, the second cluster can - are basic tools researchers used to
be by division, and the third gather data for specific research
cluster can be by district problems.
- Common instruments are performance
4) SYSTEMATIC SAMPLING
tests, questionnaires, interviews, and
- as simple as selecting samples
observation checklist. The first two
every nth (example every 2nd,
instruments are usually used in
5th) of the chosen population
quantitative research, while the last two
until arriving at a desired total
instruments are often in qualitative
number of sample size
research. However, interviews and
- selection is based on a
observation checklists can still be used
predetermined interval.
in quantitative research once the
information gathered is translated into
numerical data.
- In constructing a quantitative research
instrument, it is very important to
remember that the tools created should
require responses or data that will be
numerically analyzed.

- Dividing the population size by

the sample size, the interval will
be obtained.
EXAMPLE: from a total
population of 75, you have 25
CHARACTERISTICS OF A GOOD RESEARCH
samples; using systematic
INSTRUMENT
sampling, you will decide to
1) CONCISE
select every 3rd person on the list
- Have you tried answering a very
of individuals.
long test, and because of its
length, you just pick the answer
SAMPLING TECHNIQUES
without even reading it? A good
1) PROBABILITY SAMPLING
research instrument is concise in
- every member of the target
length yet can elicit the needed
population has a known chance of
data.
being included in the sample.
2) SEQUENTIAL
 Simple random
- Questions or items must be
 Stratified random
arranged well. It is recommended
 Cluster sampling
to arrange it from simplest to the
 Systematic sampling
most complex. In this way, the
RESEARCH REPORT
3rd Quarter
instrument will be more favorable respondents. This scale seems to
to the respondents to answer. be more advantageous since it is
3) VALID AND RELIABLE
- The instrument should pass the
tests of validity and reliability to
get more appropriate and
accurate information. more flexible and easier to
4) EASILY TABULATED construct.
- Since you will be constructing an NOTE: Another important consideration in
instrument for quantitative constructing a research instrument is how to
research, this factor should be establish its validity and reliability.
considered. Hence, before
crafting the instruments, the VALIDITY
researcher makes sure that the - A research instrument is considered valid
variable and research questions if it measures what it supposed to
are established. These will be an measure
important basis for making items - When measuring oral communication
in the research instruments. proficiency level of students, speech
WAYS IN DEVELOPING RESEARCH performance using rubric or rating scale
INSTRUMENT is more valid than students are given
- There are three ways you can consider multiple choice tests
in developing the research instrument - It has several types: face content,
for your study. First is adopting an construct, concurrent, and predictive
instrument from the already utilized validity
instruments from previous related
studies. The second way is modifying TYPES OF VALIDITY OF INSTRUMENT
an existing instrument when the 1) FACE VALIDITY
available instruments do not yield - also known as “logical validity”
the exact data that will answer the - It calls for an initiative judgment
research problem. And the third way is of the instruments as it “appear.”
when the researcher made his own Just by looking at the instrument,
instrument that corresponds to the the researcher decides if it is
variable and scope of his current valid.
study. 2) CONTENT VALIDITY
- An instrument that is judged with
COMMON SCALES USED IN QUANTITATIVE content validity meets the
RESEARCH objectives of the study.
1) LIKERT SCALE - It is done by checking the
- This is the most common scale statements or questions if this
used in quantitative research. elicits the needed information.
Respondents were asked to rate Experts in the field of interest can
or rank statements according to also provide specific elements
the scale provided. 3) CONSTRUCT VALIDITY
Example: A Likert scale that - validity of instruments as it
corresponds to the theoretical
construct of the study.
- It is concerning if a specific
measure relates to other
measures
measures the attitude of students 4) CONCURRENT VALIDITY
towards distance learning. - When the instrument can predict
results similar to those similar
2) SEMANTIC DIFFERENTIAL tests already validated, it has
- In this scale, a series of bipolar concurrent validity.
adjectives will be rated by the 5) PREDICTIVE VALIDITY
RESEARCH REPORT
3rd Quarter
- When the instrument is able to - These numerical data are usually subject
produce results similar to those to statistical treatment depending on the
similar tests that will be nature of data and the type of research
employed in the future, it has problem presented. The statistical
predictive validity. treatment makes explicit the different
- This is particularly useful for the statistical methods and formulas needed
aptitude test. to analyze the research data.
RELIABILITY PLANNING YOUR DATA ANALYSIS
- The consistency of the measures or - Before choosing what statistical test is
results of the instruments appropriate for your research study it is
important to determine what statistical
RELIABILITY OF INSTRUMENT formation is applicable to your current
1) TEST-TEST RELIABILITY study. In immersing yourself into
- achieved by giving the same test planning your data analysis, you have to
to the same group of respondents decide what basic descriptive statistical
twice. The consistency of the two technique you are going to use.
scores will be checked. DESCRIPTIVE STATISTICAL TECHNIQUE
2) EQUIVALENT FORMS RELIABILITY - provides a summary of the ordered or
- established by administering two sequenced data from your research
identical tests except for sample.
wordings to the same group of - Frequency distribution, measure of
respondents central tendencies (mean, median,
3) INTERNAL CONSISTENCY mode), and standard deviation are the
RELIABILITY sets of data from descriptive statistics.
- It determines how well the items INFERENTIAL STATISTICS
measure the same construct. It is - used when the research study focuses
reasonable that when a on finding predictions; testing
respondent gets a high score in hypothesis; and finding interpretations,
one item, he will also get one in generalizations, and conclusions.
similar items. - Since this statistical method is more
- There are three ways to complex and has more advanced
measure the internal
consistency; through the split-
half coefficient, Cronbach’s alpha,
and Kuder-Richardson formula

LESSON 4: PLANNING AND DATA

ANALYSIS
DATA ANALYSIS
mathematical computations, you can use
- process in which gathered information
computer software to aid your analysis.
are summarized in such a manner that it
UNIVARIATE ANALYSIS
will yield answers to the research
- analysis of one variable
questions. During quantitative data
BIVARIATE ANALYSIS
analysis gathered information were
- Analysis of two variables such as
break down and ordered into categories
independent and dependent variables
in order to draw trends or patterns in a
MULTIVARIATE ANALYSIS
certain condition.
- analysis of the multiple relations
- In quantitative research, the numerical
between multiple variables.
data collected is not taken as a whole. In
NOTE: in choosing statistical techniques in
order to understand it better, it is
quantitative research, the purpose or objective
analyze into components based on the
of the research study should be considered
chosen research variables and research
TEST OF RELATIONSHIP BETWEEN TWO
questions you are going to answer.
VARIABLES
 Pearson’s r (parametric)
RESEARCH REPORT
3rd Quarter
 Phi coefficient (non-parametric for statistical measure used to express a
nominal and dichotomous variables) proportion of a whole.
 Spearman’s rho (non-parametric for 4) MEAN- The average of a set of numbers
ordinal variable) calculated by adding all and dividing by
TEST OF DIFFERENCE BETWEEN TWO DATA the number of values.
SETS FROM ONE GROUP 5) STANDARD DEVIATION- A measure of
 T-test for dependent samples how spread out the data is in a set, is
(parametric) calculated by finding the average
 McNemar change test (non-parametric squared deviation from the mean.
for nominal and dichotomous variables) 6) TABLE- A visual organization of data in
 Wilcoxon signed-rank test (non- rows and columns.
parametric for ordinal variable) 7) FIGURE- A visual representation of data,
TEST OF DIFFERENCE BETWEEN TWO DATA such as a graph, chart, or diagram.
SETS FROM TWO DIFFERENT GROUPS 8) PARAMETRIC TEST- A statistical test
 T-test for independent samples assuming a specific underlying
(parametric) distribution for the data (normal
 Two-way chi-square (non-parametric for distribution for a t-test).
nominal variable) 9) NON-PARAMETRIC TEST- A statistical
 Mann-Whitney U test (non-parametric for test that doesn't require assumptions
ordinal variable) about the underlying data distribution.
TEST MORE THAN TWO POPULATION 10) CORRELATION- A statistical measure
MEANS indicating the strength and direction of
 Analysis of Variance or ANOVA the relationship between two variables
(parametric) (positive, negative, or zero).
TEST THE STRENGTH OF RELATION OR 11) REGRESSION- A statistical method
EFFECT OR IMPACT modeling the relationship between two
 Regression (parametric) variables, allowing the prediction of one
variable based on the other.
LESSON 5: PRESENTING RESEARCH
METHODOLOGY
RESEARCH METHODOLOGY
- describes the techniques and procedures
used to identify and analyze information
regarding a specific research topic. It is a
process by which researchers design
their study so that they can achieve their
objectives using the selected research
instruments.
PARTS OF RESEARCH METHODOLOGY
 Research Design Research
 Population and Sample
 Sampling Procedure
 Research Instruments
 Validity and Reliability of Instruments
Research Intervention (if applicable)
 Data Collection Procedure Data Analysis

TERMINOLOGIES
1) DATA- A collection of information about
a specific topic.
2) FREQUENCY- The number of times
something occurs in a data set.
3) PERCENT- "Percent" means "out of
every 100". Percentage refers to a