Table of Contents
Types of Academic Research Paper
Not all research papers have the same structure or purpose. Understanding your paper type before you begin shapes everything from the literature review to the conclusion.
| Type | Purpose | Typical structure |
|---|---|---|
| Empirical research article | Reports original data collection and analysis | AIMRaD (Abstract, Introduction, Methods, Results, Discussion) |
| Theoretical/conceptual paper | Advances a new framework or critique | Problem → literature engagement → theoretical argument → implications |
| Review article | Synthesises existing literature on a question | Search strategy → inclusion/exclusion → synthesis → conclusions |
| Case study paper | In-depth analysis of one case to produce theoretical insight | Context → analysis → theoretical implications → generalisability |
Research Question vs Hypothesis
The choice between a research question and a formal hypothesis reflects your epistemic approach and discipline:
A research question is appropriate when you are exploring a phenomenon whose parameters are not yet established, when qualitative data is primary, or when the purpose is interpretive rather than hypothesis-testing. Social sciences and humanities typically use research questions.
A hypothesis is a falsifiable prediction derived from theory. It is the appropriate framing when your study is designed to test whether a predicted relationship holds in empirical data. Natural and health sciences, and quantitative social science, typically use hypotheses.
| Research question | Hypothesis | |
|---|---|---|
| Appropriate for | Exploratory, interpretive, qualitative | Confirmatory, experimental, quantitative |
| Example | "How do first-generation university students construct academic identity during their first year?" | "Students who participate in peer mentoring programmes will show significantly higher first-year retention rates than matched controls." |
| Evaluated by | Depth of interpretation, explanatory adequacy | Statistical significance and effect size |
For quantitative research: frame using PICO (Population, Intervention, Comparison, Outcome). For qualitative research: SPIDER (Sample, Phenomenon of Interest, Design, Evaluation, Research type) helps construct a well-scoped research question and guides your database search strategy.
AIMRaD Structure
Abstract
150–250 words. Background sentence, the gap, your approach, key findings, implications. Written last but read first — it determines whether a reader proceeds.
Introduction
Context → gap in knowledge → your research question/hypothesis → brief signpost of approach and structure. The introduction tells the reader why this paper exists.
Methods
Everything another researcher would need to replicate your study. Design, participants, instruments, procedure, analysis. Justify every significant methodological choice.
Results
What the data show — without interpretation. Structured around your research questions or hypotheses, not around your data collection instruments.
Discussion
Interpretation of results in relation to the literature and research question. Where do your findings confirm, contest, or extend existing knowledge? What are the limitations and implications?
Introduction: Establishing the Intellectual Problem
The funnel model is standard for research paper introductions: begin broadly at the level of the field or societal problem, then narrow progressively to the specific gap your paper fills, ending with a precise statement of your research question or hypothesis.
Level 2 (narrowing to the gap): However, the majority of this literature focuses on executive overreach and judicial capture, with comparatively little attention to the role of electoral system design in either accelerating or inhibiting backsliding dynamics.
Level 3 (research question): This paper asks: to what extent does proportional representation constrain executive-led democratic backsliding in post-2000 parliamentary democracies?
Integrating the Literature
The literature review in a research paper is not a separate section in most empirical papers — it is distributed across the introduction (to establish the gap) and the discussion (to situate findings). The goal is synthesis, not citation accumulation.
Synthesis vs citation
Citation is stating what individual scholars have said. Synthesis is showing how multiple scholars' positions relate to each other and to your own argument:
Methods Transparency
The methods section must be transparent enough to permit replication. This means reporting:
- Design — experimental, quasi-experimental, observational, survey, case study, ethnographic
- Participants / data sources — sampling strategy, inclusion/exclusion criteria, final N, demographic profile if relevant
- Instruments — questionnaire items, interview schedule, coding framework, variable definitions and operationalisation
- Procedure — data collection timeline, conditions, ethical approvals obtained
- Analysis — statistical tests used (and software version), qualitative coding approach, inter-rater reliability if applicable
"Semi-structured interviews were conducted with 12 participants." This describes but does not justify. Add: "Semi-structured interviews were selected because the research question requires exploratory, participant-driven accounts of X, which closed-format instruments cannot capture (Braun & Clarke, 2006). A sample of 12 was determined following the saturation principle widely applied in thematic analysis."
Reporting Results
Results are reported before interpretation. The boundary between Results and Discussion must be clear. Common violations:
- Interpreting findings in the Results section ("This shows that…" → belongs in Discussion)
- Reporting only significant results and omitting null findings (selective reporting bias)
- Presenting tables without narration — every figure and table must be described in prose
- Using results as a summary of your methodology (results must answer the research question, not describe the process)
Discussion: From Results to Knowledge
The Discussion is where the paper's intellectual contribution is forged. It must accomplish four tasks:
- Interpret — what do the results mean? How do they answer the research question?
- Contextualise — how do the findings compare to prior research? Where do they confirm, challenge, or extend existing knowledge?
- Explain unexpected findings — null results and surprises require explanation, not suppression
- Acknowledge limitations — every study has constraints; identifying them honestly signals scholarly maturity and does not undermine the paper's contribution
Conclusion and Contribution
The conclusion synthesises the paper's argument and states its contribution to the field. It should be concise (one substantive paragraph in a journal article; a full section in a dissertation chapter). It must:
- Answer the research question/hypothesis directly
- State what the paper adds to existing knowledge
- Specify implications for theory, practice, or policy
- Propose concrete directions for future research
Common Errors in Research Writing
| Error | Academic consequence | Fix |
|---|---|---|
| Confusing correlation and causation | Overclaims the significance of findings | Use causal language only when causal mechanisms are established; otherwise "is associated with," "predicts," "is correlated with" |
| Undersized sample undisclosed | Reviewers reject on validity grounds | Justify sample size using power analysis (quantitative) or saturation argument (qualitative) |
| Literature review in Discussion | Structural confusion; reviewers note lack of integration | Literature belongs in Introduction (to establish gap) and Discussion (to contextualise findings) |
| Methods described not justified | Reviewer cannot assess research quality | Justify every significant methodological choice with a citation to methodological literature |
| Abstract written as introduction | Fails to communicate key findings | Abstract must include findings and implications — not just background and questions |