This module highlights the importance of groundwork in defining the research problem, establishing

background, formulating hypotheses, and reviewing literature. It emphasizes differences in design

between quantitative and qualitative research, guided by conceptual frameworks and variable selection.

Meaning of Quantitative Research Design

Just as a blueprint is essential in building a house, research designs act as blueprints in research,
ensuring valid, objective, reliable, and accurate answers to research questions.

Research Design

Research design is a systematic strategy guiding researchers, integrating essential components. In

quantitative research, pre-established designs enhance comprehension, providing clarity in analysis and
interpretation, as emphasized by Fraenkel and Wallen.

Types of Quantitative Research Design

Quantitative Research Designs have five general classifications: descriptive, Correlational, ex post facto,
quasi-experimental, and experimental.

Descriptive Research design is ideal when there’s limited knowledge about the problem, providing an
exploratory approach to answer fundamental questions and observe and report phenomena as they
occur.

Correlational Research aims to identify associations between variables without seeking cause-and-effect
relationships. It clarifies relationships and predicts the magnitude of association, with the purpose
influenced by the study’s scope and delimitation.

Ex Post Facto Research design is suitable when the goal is to measure a cause from a pre-existing effect.
The researcher lacks control over variables, making it inappropriate to conclude that changes measured
occur during the study’s actual conduct.

The last two types of quantitative research designs involve the presence of treatment or intervention in
the current study. Treatment involves controlling or manipulating the independent variable, assuming it
causes changes in dependent variables. Both quasi-experimental and experimental research include two
groups: the treatment group, subjected to intervention, and the control group, not undergoing
treatment.

Quasi-experimental design, partly derived from “quasi,” measures causal relationships between
variables, considering effects during the study. Partiality results from pre-established group assignments
based on factors, limiting conclusions due to non-random subject assignment.

Experimental research design, rooted in the scientific method, employs controlled environments for
data gathering through extensive treatment and manipulation. Random subject assignment to treatment
and control groups enhances study validity, aiming to manipulate the independent variable for influence.

Population and Sample

1. Identifying Population
 - The population encompasses all objects, elements, persons, and characteristics related to the
research topic.

- Two types of populations: target (entire group of interest) and accessible (portion with researcher’s
reasonable access).

2. Sampling Process

- When considering the entire population is impractical, a representative sample is identified.

- Sampling is a systematic process to select a group for analysis, aiming for representation of the target
population.

3. Population vs. Sample

- The whole population might be impractical; hence, a sample is chosen for analysis.

- The sample, a subset of the population, increases generalizability and applicability of findings.

4. Sample Characteristics

- A good sample should accurately represent the population within the study’s scope.

- Larger samples generally increase reliability, but the size depends on study scope, delimitation, and
research design.

Approaches in Identifying the Sample Size

Heuristics. This approach refers to the general rule or rule of the thumb for sample size. This is the
earliest established approach for sample sizes for different research designs.
Literature Review Approach to Sample Size

- Reading similar or related literature and studies aids in determining sample size.

- Recall insights gained during the literature review process.

- Application of findings from existing studies enhances the validity of the sampling procedure in the
current research study.

Formulas

Formulas play a crucial role in determining an acceptable sample size, offering a systematic approach
based on available information about the population. Various formulas cater to different levels of
knowledge about the population, and online tools simplify their application. The Cochran formula is
particularly useful when dealing with a large sample size, providing a structured method for calculating
the appropriate size for a study. Utilizing these formulas ensures a methodical and data-driven approach
to sample size determination in research studies.

Cochran’s Formula:

Where: e is the desired level of precision (margin of error)

P is the (estimated) proportion of the population which has the attribute in question

Q is 1 – p

Z is the z-value found in the Z table

For example, your study is for your city, and you want to determine how many

Households whose breadwinner work onsite during COVID pandemic. And you do not have

Enough information about the population. In that case, you are going to assume that a half of
The breadwinners in the city are working onsite. So, this assumption will give you a maximum

Variability, so, p = 0.5. And you want 95% confidence, and at least 5 percent – plus or minus

– Precision; and that gives you Z values 1.96. As shown below, your sample size will be 384.

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.

Statistical power, the likelihood of rejecting the null hypothesis, indicates a relationship between
independent and dependent variables. Typically set at 80%, it aids in determining a sufficient sample size
for measuring the effect size, representing the magnitude of the difference between experimental and
control groups. Effect size is crucial for assessing the practical significance of observed outcomes in a
research study.

Statistical power indicates a relationship between independent and dependent variables, while effect
size quantifies the magnitude of this relationship. A higher effect size signifies a greater difference
between experimental and control groups, providing valuable insight into the practical significance of
observed differences in a research study, such as anxiety test scores before and after an intervention.

Probability Sampling in Quantitative Research

Simple random sampling ensures every accessible population member has an equal chance of selection.
Methods like the fishbowl technique, roulette wheel, or random number table can be used. The key
principle is unbiased, equal opportunity for each individual within the accessible population to be
chosen for the sample.

Stratified random sampling ensures an equal chance for all population members. Before selection, the
population is divided into strata. Samples are chosen from these subgroups rather than the entire
population. This method is effective when study variables can be categorized, allowing for a targeted and
representative sample selection.

You can simply follow the steps from this given example:

A population of 600 Junior High School students includes 180 Grade 7, 160

Grade 8, 150 Grade 9, and 110 Grade 10. If the computed sample size is 240,

The following proportionate sampling will be as follows.

The number of members per subgroup is divided by the total accessible sample size.

The percentage result of members per subgroup will be multiplied from the computed total

Sample size. After obtaining the sample size per strata, then simple random sampling will be

Done for the selection of samples from each group.

Cluster sampling is used in large-scale studies with dispersed populations. Similar to stratified random
sampling, it involves dividing the population into clusters. The researcher creates homogenous clusters,
simplifying data collection. This approach is efficient when direct sampling from the entire population is
impractical, making it a suitable choice for managing geographical challenges and time constraints in
extensive studies.

For example, a researcher would like to Interview of all public senior high school students across
Mindanao. A cluster of research will be selected to satisfy the plan size. In the given example, The first
cluster can be by region, the second cluster can be by division, and the third cluster can be by district
another way of doing cluster sampling is Illustrated on the figure on the right side.

Systematic sampling samples at regular intervals, such as every 2nd or 5th, until reaching the desired size.
The predetermined interval is based on dividing the population size by the sample size. For example, in a
population of 75 with a sample size of 25, every 3rd person is chosen. This method provides an efficient
way to achieve a representative sample.
Quantitative Research Instrument

Similar to building a house, meticulous preparation is essential in constructing quantitative research

instruments. This ensures careful tool preparation, emphasizing the need for responses or data that can
be numerically analyzed. The analogy underscores the importance of precision and thoughtful planning
to yield meaningful and analyzable research results.

Research instruments, essential for data collection, encompass performance tests, questionnaires,
interviews, and observation checklists. Quantitative research predominantly employs performance tests
and questionnaires, while qualitative research often uses interviews and observation checklists.
Importantly, interviews and observation checklists remain viable in quantitative research when the
gathered information is translated into numerical data. This highlights the versatility of research
instruments across quantitative and qualitative methodologies.

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 arrangement of questions or items in a research instrument is crucial. It is advisable to

organize them from the simplest to the most complex. This arrangement enhances respondent
engagement and makes the instrument more favorable for effective and comprehensive responses.

Valid and reliable. The instrument should pass the tests of validity and reliability to get more
appropriate and accurate information.

Easily Tabulated. For effective construction of a quantitative research instrument, ensure variables and
research questions are established beforehand. The instrument should be easily tabulated, emphasizing
clarity and simplicity for efficient data analysis.

Ways in Developing Research Instrument

Developing a research instrument offers three approaches: adopting an instrument from prior related
studies, modifying an existing one if needed, and creating a new instrument tailored to the variables and
scope of the current study.

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.

Example: A Likert scale that measures the attitude of students towards distance learning.
Semantic Differential. In this scale, a series of bipolar adjectives will be rated by the Respondents. This
scale seems to be more advantageous since it is more flexible and easier to construct.

Another important consideration in constructing a research instrument is how to establish its validity
and reliability.

Types of Validity of Instrument

Face validity, or logical validity, involves the researcher’s judgment of an instrument’s validity based on
appearance.

Content validity. Content validity ensures an instrument aligns with study objectives, checking if
statements elicit needed information. Construct validity corresponds to theoretical constructs, and
concurrent validity predicts results similar to validated tests.

Predictive validity occurs when an instrument’s results align with future tests, especially beneficial for
aptitude tests.

Validity. Valid research instruments measure what they are supposed to. For oral communication
proficiency, performance-based assessments like rubrics are more valid than multiple-choice tests.
Validity types include face, content, construct, concurrent, and predictive.

Reliability of Instrument

Test-retest reliability is ensured by administering the same test to the same group twice, checking score
consistency.

Equivalent forms reliability is confirmed by administering two identical tests with different wordings to
the same group.

Internal consistency reliability assesses how well items measure the same construct. If a respondent
scores high on one item, a similar score is expected on related items. Measurement methods include
split-half coefficient, Cronbach’s alpha, and Kuder-Richardson formula.

Nature of Research Intervention

In experimental research, the researcher determines the nature of intervention or treatment, deciding
what will happen to study subjects, who will receive it, and the extent of its application.

Steps in Describing the Research Intervention Process

In experimental research papers, a dedicated section details the research intervention procedure,
providing clear explanations.

Write the Background Information. The background information introduces and justifies the
intervention’s relevance, including context and treatment duration in studies.

Describe differences and similarities between experimental and control groups, outlining what will and
will not occur in each, clarifying research group parameters.

Describe the Procedures of the Intervention. Detail the intervention procedures, explaining how the
experimental group experiences the condition, outlining the process for achieving desired study results.

Explain the rationale for intervention procedures with clear and concrete reasons, establishing necessity.
Include the theoretical and conceptual basis to validate the chosen procedures.

Quantitative Data

Data, information, or facts become valuable in research when they address the problem. Quantitative
data is numerical, but qualitative data, when categorized and numbered, can be useful in analysis.

Techniques in Collecting Quantitative Data

Observation gathers information about a condition using senses. Recorded by direct or indirect means,
such as gadgets, an observation checklist aids in systematic data collection.

Survey involves data collection through interviews or questionnaires. Questionnaires consist of written
responses, while interviews involve oral responses. In quantitative research, responses are expected to
have numerical value, either nominal or ordinal.

Experiment. In experimental design, treatment or intervention is applied, and the study assesses its
effects after chosen subjects undergo the intervention.

Three Phases in Data Collection

Data collection is crucial in research. Precision in the collection procedure is vital to avoid inaccuracies
and ensure gathered data addresses research questions, maintaining validity. The research paper details
data gathering procedures, including steps before, during, and after data collection, offering suggested
but not limited steps.
Data Analysis

Data analysis in research involves summarizing information to answer questions. In quantitative analysis,
data is broken down into categories, revealing trends. Statistical treatment is applied to analyze
numerical data systematically.

Planning your Data Analysis

Before selecting a statistical test, it’s crucial to identify the appropriate statistical formation for your
study. In data analysis planning, decide on a basic descriptive statistical technique for coding and
tabulating data.

Descriptive statistical techniques, including frequency distribution and measures of central tendencies,
summarize ordered data. Inferential statistics, for predictions and hypothesis testing, often requires
computer software for complex computations. Selecting a statistical test involves identifying whether to
use a parametric or non-parametric test. Considerations are based on Statistics and Probability concepts.

In addition, in choosing statistical techniques in quantitative research, the purpose or objective of the
research study should be considered.

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

-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)