Designing Research – Methods & Instrumentation (Comprehensive Study Notes)

House Rules & Session Overview

• Take down notes, listen actively, ask questions.

• Two major blocks of the talk (35 min each):

  • Part 1 – Research Design (3 : 31 – 4 : 06 PM)

  • Part 2 – Instrumentation (4 : 07 – 4 : 42 PM)

• Theme of Forum 2025: “Starting the Research Journey: Building the Foundations for Academic and Social Impact.”

• Immediate course output required: A Simple Research Plan containing topic/title, chosen design, rationale, participants, data-gathering technique, and 2 sample questions.

Begin With the End in Mind

• Stephen Covey & OBE principle: Clarify intended outcomes first, then design backwards.

• Expected Learning Outcomes for this session:

  • Knowledge – explain methods & instruments in your own words.

  • Skills – build a correct, coherent research design.

  • Values – commit to honesty & responsibility in doing/reporting research.

  • Assessments: quick reflections, small-group plan, commitment statements.

Fundamentals: Design vs. Instrumentation

• Research Design = overarching strategy/blueprint/road-map; answers “WHAT & HOW will we study?”

• Instrumentation = concrete tools & procedures that collect the data; answers “WITH WHAT will we gather answers?”

• Methodology chapter normally includes: design, participants/sampling, data-gathering techniques, instruments, and analytic plan.

Three Master Research Designs

1 Quantitative Research Design

• Data are numeric (ages, scores, Likert values).

• Statistical analysis (descriptive & inferential) with hypothesis testing expected.

• Large, randomly selected samples → generalizable to population.

• Main types:

  • Descriptive (exploratory, comparative, correlational, evaluative, developmental).

  • Experimental – manipulation of IVs; requires hypothesis tests.

  • True Experimental: full randomization.

  • Quasi-Experimental: lacks full randomization.

• Metaphor: tasting a spoonful of soup to know the whole pot.

2 Qualitative Research Design

• Data are words/phrases (interviews, documents).

• Analytic focus: coding, theming, sense-making.

• Small, purposively selected samples → not for statistical generalization; goal = depth & uniqueness.

• Major genres: Case Study, Historical, Phenomenology, Ethnography, Grounded Theory, Action Research.

• Metaphor: reading Maria’s diary tells a lot about Maria, not entire class.

3 Mixed-Method Research Design

• Combines numeric & narrative to obtain both “what” (quant) and “why” (qual).

• Generalization only from quantitative strand.

• Core designs:

  • Convergent Parallel (collect both strands simultaneously, then merge).

  • Explanatory Sequential (quant → qual to explain).

  • Exploratory Sequential (qual → quant to test).

  • Q-Methodology (collect qual perceptions; analyze with quantitative factor analysis).

Quick-Check Questions & Answers

1. Three designs just discussed? Quantitative, Qualitative, Mixed.

2. Data types required? Numbers / Words / Both.

3. Most powerful? Mixed (because it triangulates).

4. Which will you use? Any is acceptable if aligned with objectives.

Deep-Dive: Specific Quantitative Sub-Types & Examples

• Descriptive-Exploratory – Factors Affecting Writing Skills of BCP Students.

• Descriptive-Comparative – Math Performance of Students Taught Online vs F2F.

• Descriptive-Correlational – Social Media Usage & Academic Performance.

• Descriptive-Evaluative – Effectiveness of AI-Assisted Teaching Strategy.

• Descriptive-Developmental – Development & Validation of Accounting 101 Modules.

• True Experimental – Effect of Peer Tutoring on Math Achievement (random groups).

• Quasi-Experimental – Impact of Salary Increase on Motivation (pre/post within same employees).

Deep-Dive: Specific Qualitative Sub-Types & Examples

1. Case Study – Leadership Styles & Employee Satisfaction in a Small Business.

2. Historical Study – Evolution of Women’s Rights (1900-2000).

3. Phenomenology – Lived Experiences of First-Time Urban Mothers.

4. Ethnography – Cultural Practices of a Rural Indigenous Community.

5. Grounded Theory – Building Theory of Student Engagement in Online Learning.

6. Action Research – Improving Classroom Participation via Collaborative Learning.

Deep-Dive: Mixed-Method Title Samples

• Convergent Parallel – Impact of Online Learning on Engagement (survey + interviews).

• Explanatory Sequential – Tech-Company Employee Turnover (quant survey → in-depth interviews).

• Exploratory Sequential – Teachers’ Perceptions of AI Integration (qual first).

• Q-Methodology – Public Attitudes Toward Impeachment (PQMethod software).

• Real publication examples: Digital Learning using Q-Methodology (Blay 2024); Service-Learning ripple effects (accepted for Buenos Aires 2025).

Writing the Methodology “Story” (Sample Excerpt)

• Begin with aim/objective.

• Justify chosen design & why others don’t fit.

• Cite prior researchers who used same design.

• Example: Descriptive-Evaluative design for formative evaluation of Mathematics in the Modern World GE course at DLS-CSB.

PART 2 – Instrumentation & Sampling

Population → Sample Logic

1. Identify target population $N$.

2. Decide whether census feasible; if not, compute sample size $n$.

3. Select participants using an appropriate sampling technique.

4. Collect data with valid & reliable instruments.

Sample Size Determination

• Key statistical levers:

  • Confidence Level (95 % or 99 %).

  • Margin of Error (±1 % – ±5 %).

• Cochran’s equations:

  • Initial infinite-population size $n_0 = \frac{Z^2 p q}{e^2}$.

  • Finite-population correction $n = \frac{n<em>0}{1 + \frac{n</em>0 - 1}{N}}$.

• Online calculators / tables (Cochran, calculator.net, G*Power).

• Rule-of-thumb: always aim for full population; otherwise justify sample.

Illustrative Calculations

• Publicus survey: $N \approx 200{,}000$ voters, $e = 2.5\%$, obtained $n = 1{,}502$.

• BCP College Students:

  • $N = 33{,}742$, $CL = 95\%$, $e = 3\%$ → $n \approx 1{,}035$.

  • Stratified by year level:

  • 1st yr 16 % → $166$; 2nd yr 28 % → $290$; 3rd yr 31 % → $321$; 4th yr 25 % → $259$.

Sampling Techniques

1. Probability (Unbiased)

   • Simple Random (lottery, random-number table).

   • Systematic (every $k$-th name after random start).

   • Stratified Random (split into strata, random within each).

2. Non-Probability (e.g., Purposive – choose cases with needed attributes; used widely in qualitative work).

Data-Gathering Methods (Choose to match Questions)

1. Surveys – most popular in quantitative studies.

2. Interviews – semi-structured, in-depth.

3. Focus Group Discussions (FGDs) – moderated group dialogue.

4. Observation/Experiment – direct behavioral or lab data.

5. Usage/Data Mining – extraction from existing documents or digital traces.

Crafting a Survey Instrument

Structural Parts

1. Cover Letter / Informed-Consent section.

2. Demographic profile block (optional personal identifiers).

3. Clear instructions & layout.

4. Thematic blocks of items (MCQ, open-ended, Likert, etc.).

5. Thank-you statement.

Item-Writing Tips

• Mix item types; keep language concise & non-leading.

• Provide neutral/“don’t know” options.

• Use consistent point scale; odd number (3/5/7) gives midpoint.

• Limit total questions (≤ 5 RQs, questionnaire not overlong).

• Ensure every RQ can be answered by the questionnaire.

Likert Scale Mechanics

• Respondents rate a statement on agreement/intensity continuum.

• Example 5-point coding: 1 = Strongly Disagree … 5 = Strongly Agree.

• Possible interpretive bands (Sample 1):

  • $4.20–5.00$ – Very Great Extent

  • $3.40–4.19$ – Great Extent

  • $2.60–3.39$ – Moderate Extent, etc.

Sample Item Set on “Enjoyment in Online Teaching”

1. “I find online teaching to be an enjoyable experience.”

2. “I look forward to interacting with my students virtually.”

3. “Adapting to new online methods gives me enthusiasm,” … etc.

Instrument Validation & Reliability

• Researcher-Made Instruments

  • 3-step validation: self-review → expert panel → semantic (pilot with similar group).

  • For Likert scales: compute internal consistency with Cronbach’s $\alpha$.

  • $\alpha \ge 0.90$ – Excellent; 0.80!\le!\alpha<0.90 – Good; $0.70$ Acceptable; <0.70 needs revision.

  • Rule of thumb: at least “10 respondents × items” (e.g., 5 items → 50 pilot cases).

• Ready-Made Instruments

  • Secure author permission; may or may not need re-validation.

Data-Collection Roll-Out

1. Finalize validated instrument.

2. Randomly pick required sample list.

3. Distribute survey / schedule interviews / arrange FGDs.

4. Monitor responses; send follow-ups.

5. Begin data coding & analysis upon completion.

Ethical & Practical Reminders

• Honesty & responsibility in data handling and reporting.

• Obtain informed consent; ensure confidentiality.

• Avoid wasting resources (too large samples) or risking under-powered study (too small samples).

• Consult research adviser and statistician as needed.

Inspirational Close

“Design with intention, and your research will always speak with clarity.” – Dr. Basilia E. Blay