Sections of a typical Chapter 3
| Section | What it covers |
|---|---|
| Introduction | Restates the purpose and research questions |
| Research design and rationale | Quantitative, qualitative, or mixed methods, the specific design, and why it fits the questions |
| Role of the researcher | Mainly qualitative: your relationship to the setting and how you managed bias |
| Population and sample | Who could take part, how many, how they were selected, and why that number |
| Instrumentation | Surveys, scales, interview protocols, or data sources, with evidence of validity and reliability |
| Data collection procedures | Step by step: recruitment, consent, collection, and storage |
| Data analysis plan | Statistical tests or qualitative coding approach, tied to each research question |
| Validity, reliability, or trustworthiness | How you protected the quality of the results |
| Ethical procedures | IRB approval, informed consent, confidentiality, and data security |
| Summary | Recaps the chapter and previews Chapter 4 |
Match the design to the research questions
| If your question asks… | A fitting design |
|---|---|
| Is there a relationship between variables? | Quantitative correlational |
| Does a program or condition cause a difference? | Experimental or quasi-experimental |
| How do people describe an experience? | Qualitative phenomenology or descriptive design |
| How does a process unfold in one setting? | Qualitative case study |
| What and why, together? | Mixed methods (explanatory or exploratory sequential) |
Build an alignment table
Many committees ask for a table that links each research question to its data and analysis. It is also the fastest way to spot gaps in your plan.
| Research question | Data source | Analysis |
|---|---|---|
| RQ1: Relationship between shift length and burnout | Survey: shift hours; Maslach Burnout Inventory | Pearson correlation |
| RQ2: Difference in burnout by unit type | Survey: unit type; burnout score | One-way ANOVA |
| RQ3: How nurses describe support | Semi-structured interviews | Thematic analysis |
Population, sample, and sample size
- Define the population and the inclusion and exclusion criteria.
- Name the sampling method: random, convenience, purposive, snowball, or stratified.
- Justify the size. Quantitative studies usually report a power analysis (for example, with G*Power); qualitative studies explain saturation or follow accepted ranges for the design.
Quality: validity, reliability, and trustworthiness
| Quantitative | Qualitative |
|---|---|
| Instrument validity and reliability (for example, Cronbach's alpha) | Credibility: member checking, triangulation |
| Internal validity: controlling confounds | Transferability: rich description of the setting |
| External validity: generalizing to the population | Dependability: an audit trail of decisions |
| Statistical conclusion validity: meeting test assumptions | Confirmability: reflexive notes on researcher bias |
Tips that save revisions
- Write the proposal version in future tense; change to past tense after data collection.
- Describe procedures in the order they happened, with enough detail to repeat them.
- Put full instruments, consent forms, and protocols in the appendices and refer to them.
- Never collect data before IRB approval; describe the approval in the ethics section.
Formats differ by university, program, and field. Your graduate school's thesis and dissertation manual and your committee always take priority over any general guide, including this one.
Frequently asked questions
What is the difference between methods and methodology?
Methods are the specific steps and tools you used. Methodology is the reasoning behind the design. Chapter 3 includes both, which is why it is called either name.
How long is Chapter 3?
Often 15 to 30 pages in a dissertation and shorter in a master's thesis.
Do I need a pilot study?
Not always. A pilot is common when you create a new instrument or interview protocol; your committee will tell you if one is required.