Research Projects from Start to Finish: Introductory Guide for Myanmar

1. Defining Research and Its Fundamental Types

• Definitions of Core Concepts:

  • Data: Information that you collect for your research project.

  • Theory: A collection of ideas that try to explain something.

  • Evidence: The available information or facts that show that a belief or theory is true (valid).

• Informal Research:

  • Describes the everyday act of finding information (e.g., browsing product reviews online or researching a company before a job interview).

  • It involves analyzing information to make a decision but is typically unplanned.

• Formal Research:

  • A systematic, organized process of finding new knowledge to answer a specific question.

  • Follows a structured plan to ensure information is true and correct.

  • Aims to make a small, specific, and original contribution to global knowledge.

• Primary vs. Secondary Research:

  • Desk Research (Secondary Research): Finding, compiling, summarizing, and analyzing data that already exists (e.g., Ardeth Thawnghmung’s $2016$ paper on the Myanmar elections analyzed news and previous papers).

  • Primary Research: The collection of new, original data directly by the researcher.

• Rationale for Research:

  • Aims to increase understanding, make good decisions, and solve problems.

  • Research prevents unwanted results from uninformed actions (e.g., starting a business without cost research or a government banning imports without studying local production capacity).

• Basic vs. Applied Research:

  • Basic Research: Performed to expand knowledge or test existing theories; usually academic.

  • Applied Research: Performed to use new knowledge in a practical way to solve specific problems (e.g., studying the impact of Covid-$19$ on youth livelihoods in Shan State to help that population).

  • Applied research is critical for policy- and decision-making.

2. Phases of the Research Process

While projects vary, they generally follow four major phases:

• Phase 1: Exploration:

  • Investigating areas of interest, reading literature (books, articles, news), and speaking with community members or experts.

  • Developing the research problem and specific research question.

• Phase 2: Research Design:

  • Choosing methodologies and methods.

  • Planning tools, identifying data sources, defining the participant sample, and assessing risks.

• Phase 3: Research Execution:

  • Conducting a pilot study to test methods.

  • Data collection, analysis (summarizing), and interpretation (explaining meaning).

• Phase 4: Follow-up:

  • Reporting findings (oral or written) and dissemination (sharing with relevant audiences).

  • Application: Using the information to influence decision-makers, communities, or future research.

3. From Research Topic to Research Question

• The narrowing process follows this hierarchy:

  • Area of Study: A broad field of knowledge (e.g., Education).

  • Topic: A specific issue within that field (e.g., Corporal punishment in education settings).

  • Research Problem: The explanation of why the study is needed (e.g., corporal punishment is outlawed but stories suggest it still happens in rural areas; a lack of data exists).

  • Research Question: A specific question starting with What, Why, or How. It must be narrowed by timeframe, geographic location, or demographic (e.g., ‐Do teachers in rural primary schools in Rakhine State continue to use corporal punishment? If so, why?").

• Gender Considerations: Researchers must consider how topics affect different genders differently (e.g., how water shortages impact men vs. women).

4. Understanding the Literature Review

• Definition: A comprehensive summary of ideas, issues, and research findings on a particular topic.

• Purpose:

  • To understand the background and determine if the question has already been answered.

  • If many sources answer the question, the question must be changed.

  • If no direct sources exist, researchers should look for similar topics (e.g., the effect of flooding in other regions vs. a specific township).

• Sources of Literature:

  • Physical: Libraries and organization collections.

  • Online: Google Scholar, Wiley Online Library, Academia, Research Gate, Core, CrossRef.

  • Myanmar-specific: The Online Burma/Myanmar Library, Tea Circle, Andrew Selth’s bibliography.

  • Access tools: Sci-hub and Z-lib.

• Checking Reliability:

  • Is the info current? Are there references? Is the author listed? Is the purpose to inform, persuade, or entertain?

• Referencing and Plagiarism:

  • Plagiarism: Using ideas/words without credit; can lead to failure or expulsion.

  • Styles: The guide uses APA Style ($7^{th}$ edition).

  • Verbatim quotes require quotation marks and page numbers: (ActionAid Myanmar, $2020$, p. $17$).

  • Paraphrased ideas require a citation: ActionAid Myanmar ($2020$).

  • Common knowledge (e.g., the dry zone suffers from drought) does not require a reference.

5. Choosing the Right Methodology

• Qualitative Methodology:

  • Goal: Understand the "quality"/meaning of situations.

  • Data: Words.

  • Focus: Depth, subjectivity, and rich detail.

  • Sample: Smaller groups.

  • Example study: Land confiscation in Kayah State (interviews with farmers).

• Quantitative Methodology:

  • Goal: Measure "quantity"/how many.

  • Data: Numbers.

  • Focus: Width, objectivity, and generalizations.

  • Sample: Large groups.

  • Example study: Public trust in the Myanmar Police Force (MPF) using surveys of $401$ participants.

• Mixed Methodology:

  • Combines both approaches for a more complete answer (e.g., Lim et al. $2013$ study on health worker trauma in Karen State using surveys and interviews).

6. Data Collection Methods

• Interviews:

  • Structured: Fixed questions in a set order; easy to compare but less flexible.

  • Unstructured: Open conversation with a goal; flexible but harder to analyze.

  • Semi-structured: A mix with set topics and flexible follow-ups (e.g., used in women rice farming studies in Ayeyarwady/Bago).

  • Disadvantage: Interviewer bias/identity (age, sex, ethnicity) can influence responses.

• Focus Group Discussions (FGDs):

  • $5$ to $8$ participants discussing a topic with a facilitator.

  • Strengths: Faster than individual interviews; allows observation of social dynamics.

  • Weaknesses: Potential for groupthink; dominated by high-status individuals.

• Questionnaires:

  • Best for large samples and uncomplicated data (e.g., poverty and flood relationship in Bago city).

  • Weaknesses: No chance for follow-up; requires participants to be literate or requires time-intensive manual data collection.

• Oral History:

  • In-depth, first-hand accounts of historical events (e.g., Women’s oral histories from the former Soviet Union).

7. Creating Research Instruments

• Types of Questions:

  • Open-ended: Participants answer in their own words; produces rich qualitative data.

  • Close-ended: Limits answers (e.g., Multiple choice, Yes/No, Likert scale usually $1$ to $5$).

• Drafting Tips:

  • Use simple language; avoid leading questions (e.g., use "effects" instead of "problems").

  • Avoid double-barreled questions (asking two things in one).

  • Collect data on gender and age to ensure inclusive results.

• Instructions:

  • Must include study purpose, researcher ID, contact details, and participant rights.

8. Pilot Studies

• Definition: A small practice version of the data collection.

• Purpose: To test if instructions are clear, check timing, and practice interviewing skills.

• Sample: Usually $2$ to $3$ people similar to the target population.

• Outcomes: May lead to deleting repetitive questions, changing wording due to sensitivity, or adjusting the overall research plan.

9. Selection of the Sample

• Research Population: Defined by criteria (e.g., avocado farmers in Taunggyi with over $5$ years experience).

• Sampling Strategies:

  • Probability (Random): Every member has an equal chance. Best for generalization. Includes Simple Random, Systematic (intervals), and Stratified (by group like income or ethnicity).

  • Non-probability: Not random. Includes Convenience (easy access), Purposive (specifically choosing experts), and Snowball (recruiting via other participants).

• Sample Size: Crucial for quantitative generalization; qualitative focus is on the "richness" of the data.

10. Digital Security and Ethics

• Ethical Pillars:

  • Respect for rights, honesty in reporting, and ensuring benefits outweigh risks.

• Informed Consent:

  • Participants must agree voluntarily with full understanding.

  • Components: Purpose, timeframe, risks, anonymity/confidentiality details, and the right to stop at any time.

  • Oral consent is acceptable for sensitive topics where signed names pose a risk.

• Risk Assessment:

  • Categorize risks into social, legal, and health/safety for both researchers and participants.

• Digital Security Tips:

  • Passwords (not biometric), end-to-end encryption (e.g., Signal),

  • Encrypted email (Proton Mail, Virtru), and secure storage (sync.com, mega.io).

11. Analyzing Qualitative Data

• Preparation: Transcribing (writing down recordings word-for-word).

• Data Reduction:

  • Phase 1: Reading for first impressions.

  • Phase 2: Coding: Using keywords to label sections (e.g., "Loneliness").

  • Phase 3: Categories: Grouping codes together (e.g., "Mental Health").

  • Phase 4: Themes: Finding broad patterns across categories (e.g., "Impacts of Lockdown").

• Display: Use tables or diagrams to show patterns and illustrative quotes.

12. Analyzing Quantitative Data

• Variable Types:

  • Interval (Scale): Ranked with equal distances (e.g., Age, Income).

  • Ordinal: Ranked but distances vary (e.g., Education level).

  • Nominal (Categorical): Labels/Names only (e.g., Ethnicity, Religion).

• Univariate Analysis (One variable):

  • Frequency and Percentages (e.g., $89\%$).

  • Mean (Average): Calculated as $\frac{\sum x}{n}$. Example age mean: $33.4$.

  • Range: Minimum and maximum values (e.g., $22-50$).

• Bivariate Analysis (Two variables):

  • Contingency Tables: Measure frequency across two nominal variables (e.g., Gender vs. Trust).

  • Pearson’s r: Measures correlation strength ($0$ to $1$; positive or negative). Example: height and weight correlation of $r = 0.95$.

  • Important Note: Correlation is not causation; coincidences happen.

13. Presenting and Applying Findings

• Recommendations: Must be "SMART" (Specific, Measurable, Achievable, Relevant, Time-bound).

• Dissemination Strategies:

  • Social media, panel discussions, press releases, newsletters, and sharing results with the study participants.

• Oral Presentation Best Practices:

  • Dos: strong intro/outro, be concise, practice, maintain eye contact.

  • Don’ts: read from slides, use small fonts, arrive late, speak in monotone.

• Application: Using research for community school satisfaction improvements, further research cycles, or evidence-based decision-making.