Foundations of Research

Every research study, whether a clinical trial, an ethnographic fieldwork project, or a corporate survey, is built on a set of philosophical assumptions about the nature of reality, the nature of knowledge, and the relationship between the researcher and the research subject. These assumptions are not optional: they exist whether the researcher acknowledges them or not. Naming them is what separates rigorous scholarship from unreflective data collection.

The Three Core Philosophical Questions

Before choosing a research methodology, every researcher must answer three foundational questions, consciously or otherwise:

The Four Major Research Paradigms

A paradigm is a shared set of beliefs and assumptions within a research community about what is worth studying, what constitutes valid knowledge, and how inquiry should be conducted. The paradigms below account for the vast majority of academic research produced today.

How Paradigms Connect to Methodology

The table below illustrates the logical chain from philosophical worldview to data collection method. This chain, sometimes called the research onion (Saunders, Lewis and Thornhill, 2019) or the research framework (Creswell, 2018), must be internally consistent throughout your thesis proposal. Examiners trace this chain from Chapter 1 to Chapter 3.

Research ethics is not a bureaucratic hurdle imposed by institutions. It is the moral infrastructure of academic inquiry. Every methodological decision carries an ethical dimension: who participates, what risks they face, how their data is used, and who benefits from the findings. This lesson covers the frameworks, principles, and practical requirements every researcher must understand before submitting an ethics application or beginning fieldwork.

The Six Core Principles of Research Ethics

Ethics for Vulnerable Populations

Additional ethical protections apply when research involves populations whose capacity to give fully informed, voluntary consent may be limited or compromised. The accordions below outline the key requirements for four common categories.

Research Misconduct: Definitions and Examples

Research misconduct encompasses a spectrum of practices that violate the norms of honest, transparent scholarship. Understanding the precise definitions is essential, not only to avoid misconduct personally, but to recognise and critically evaluate it when encountered in published literature.

• Fabrication: inventing data or results and reporting them as though they were genuine
• Falsification: manipulating research materials, equipment, or processes, or altering data so that the research record does not accurately reflect what occurred
• Plagiarism: presenting another person's work or ideas as one's own, without proper attribution
• Selective reporting (p-hacking): running multiple analyses and reporting only those that achieve statistical significance, creating a misleading impression of the strength of evidence
• HARKing: Hypothesising After Results are Known; presenting post-hoc hypotheses as though they were pre-registered before data collection began
• Duplicate publication: publishing substantially the same study in multiple journals without disclosure to editors
• Ghost or gift authorship: omitting genuine contributors from the author list, or including individuals who made no intellectual contribution

The literature review is, without exception, the most misunderstood section of a thesis. The majority of students produce an annotated bibliography: a series of summaries of individual papers, each treated in isolation. This is not a literature review. A literature review is a critical, synthesised argument about the state of knowledge on a research topic, designed to establish the intellectual gap that the proposed study will fill.

Types of Literature Reviews

Different research contexts require different types of literature reviews. Selecting the appropriate type is itself a methodological decision that must be justified in your thesis proposal.

How to Critically Synthesise Literature

The word critical in critical synthesis does not mean finding fault with every source. It means evaluating the quality, relevance, and contribution of each source, and then constructing an argument about what the sources, taken together, reveal about the state of knowledge on the topic.

Step 1: Map the Landscape

Before writing, build a synthesis matrix. Rows represent your sources; columns represent your themes. Populate each cell with the source's position on that theme. This reveals patterns, areas of consensus, and zones of scholarly disagreement at a glance and prevents the common error of reporting sources sequentially.

Step 2: Identify the Scholarly Conversations

Academic literature is a multi-decade conversation. Identify the key debates: where do scholars agree? Where do they disagree? Which studies built on, contradicted, or extended prior work? Tracing these intellectual lineages is what transforms a list of citations into a coherent intellectual argument.

Step 3: Evaluate Quality and Recency

Not all sources carry equal evidential weight. Peer-reviewed journal articles published in high-impact journals carry more weight than textbooks, opinion pieces, or grey literature. For rapidly evolving fields, prioritise sources from the last five to seven years. For foundational theory, seminal works may be decades old and remain essential. Apply a quality appraisal framework such as CASP or JBI for any systematic component of your review.

Step 4: Identify and Name the Gap

The most important sentence in your literature review is the one that names the specific, unfilled space in the literature that your study will occupy. This is the research gap. It must satisfy three conditions: it must be genuinely present in the existing literature, it must be important enough to warrant a formal study, and it must be addressable by your proposed research design and methodology.

Source Selection and Database Strategy

A defensible literature review requires a transparent and documented search strategy. For a thesis proposal, search at minimum the following databases, chosen to match your discipline:

• Multidisciplinary databases: Web of Science, Scopus, Google Scholar
• Discipline-specific databases: PsycINFO for psychology, CINAHL for nursing, ERIC for education, PubMed for medicine, JSTOR for humanities, EconLit for economics
• Grey literature: WHO, World Bank, government repositories, and institutional databases as appropriate to your topic

Document your search strategy in full: the databases searched, the search terms used, the Boolean operators applied, the date range filters set, and the inclusion and exclusion criteria. This documentation belongs in Chapter 2 for narrative reviews and in a dedicated methodology appendix for systematic reviews. It demonstrates rigour and allows future researchers to replicate or update your search.

A research problem is not a topic of interest or a broad area of inquiry. It is a specific, identifiable discrepancy between what is known and what needs to be known, or between what exists and what should exist, that justifies a formal investigation. Identifying and articulating this discrepancy with precision is one of the most intellectually demanding tasks in the entire research process.

Sources of Research Problems

Research problems typically emerge from one of five sources. Understanding these sources helps researchers identify and justify their own problems with greater specificity and credibility.

Anatomy of a Research Problem Statement

A well-constructed problem statement accomplishes three objectives in a logical sequence. Every sentence should serve one of these three functions and no sentence should be wasted on restating general knowledge or expressing the researcher's personal enthusiasm for the topic.

Function 1: Establish the Context

Situate the problem within the broader field or real-world context using current, empirical evidence. Quantify the problem where possible. Avoid vague or self-evident statements such as "education is important for society" or "leadership affects performance." These statements waste space and signal that the researcher has not yet identified a specific, defensible problem.

Function 2: Identify the Specific Gap or Tension

Name precisely what is missing, unresolved, or contradicted in the existing literature or practice. This is the pivot point of the entire thesis proposal. The gap must be specific enough to be addressable in a single study and narrow enough that your proposed methodology can genuinely address it.

Function 3: State the Consequence of the Gap

Explain why the gap matters: what is at stake practically, theoretically, or socially if it remains unaddressed. This section justifies the study's significance and gives the reader a reason to care about the research before the formal research questions have even been stated.

From Problem to Research Questions

Research questions operationalise your research problem. They translate the abstract gap in the literature into concrete, answerable inquiries. The quality of your research questions determines the quality of your entire study: they drive methodology, instrument design, data analysis, and the framing of conclusions.

Characteristics of Well-Formed Research Questions

• Specific: narrow enough to be answered by a single study of manageable scope
• Researchable: answerable through the collection and analysis of empirical data, not through philosophical argument or value judgement alone
• Significant: the answer would genuinely advance knowledge, inform practice, or guide policy
• Ethical: the study can be conducted without causing undue harm to participants or communities
• Aligned: each question maps directly to a stated objective, a data collection instrument, and a method of analysis

Research Questions Versus Hypotheses

Quantitative studies typically state both research questions and hypotheses. A hypothesis is a specific, testable prediction about the relationship between two or more variables, stated before data collection begins. Hypotheses are tested statistically and are either supported or not supported by the data. They are never proven in the absolute sense. Qualitative studies use research questions only and do not formulate hypotheses, because the inductive nature of qualitative inquiry means the researcher does not begin with a predicted outcome.