Historical and Comparative Research Methodologies
Where, who, when and what kind of activity. Four questions carry the whole of the source's historical methodology, and the comparative methodology is defined largely by its habit of travelling with them.
What the supplied source defines
Two things are worth noticing about the pairing. The historical definition is one of the more complete in the source's methodology table — it supplies both a purpose and a structure. The comparative definition supplies neither; it describes a relationship to another methodology and an axis of comparison, and stops. Both sit among the seven methodologies mapped in Research Methodologies: An Overview.
The historical definition also echoes the source's own definition of research, taken from a dictionary: the systematic investigation into and study of materials and sources in order to establish facts and reach new conclusions. Historical methodology is, on this reading, the general research definition applied to the past. The wider definition sits in What Is Research?.
The four framing questions
The four questions are the spine of the methodology. They tell you what to collect, they tell you how to file it, and they tell you when a claim about the past is not yet supported.
The four questions the source names
Where the event happened
Location in the widest sense — the site, the organisation, the jurisdiction, the market. In project research this is usually where the boundary of your study gets drawn, and where the limits of your conclusions get set.
Who was involved
The actors, and by extension the interests. Establishing who was involved is also how you establish whose account you are reading, which is the first move in weighing a record's reliability.
When the event occurred
Sequence before date. What preceded what is usually the load-bearing question, because a claim that one thing caused another cannot survive a corrected chronology.
What kind of activity was involved
The category of the thing that happened — a decision, a delivery, a failure, a negotiation, a change of policy. This is the question that lets you compare one episode with another at all.
THE FOUR QUESTIONS AS AN EVIDENCE GRID
| Question | What you are establishing | Kinds of record that answer it | The claim it lets you defend |
|---|---|---|---|
| Where | Setting, boundary and scope of the episode | Documents and records, archives, site and organisational registers | That your findings apply to this setting, and a statement of where they stop |
| Who | Actors, roles, interests and authorship of the record | Correspondence, minutes, life history, narrative accounts, diaries | That an account represents a particular position, not a neutral view |
| When | Sequence, duration and coincidence | Dated records, schedules, registers, versioned documents | That one thing preceded another, which is the minimum for any causal claim |
| What kind | Category and comparability of the activity | Classification of the record itself, plus contemporaneous descriptions | That two episodes are like enough to be compared |
The record types listed are drawn from the data collection labels in the source's pyramid figure. The source defines none of them; they are used here as vocabulary, not as definitions.
Synthesis, not narration
The word doing the most work in the source's definition is synthesis. Historical methodology is not the retelling of what happened. It is the assembly of separate pieces of evidence into a conclusion that none of them states on its own.
That distinction is what separates a research project from a good project report. A report narrates: this happened, then this happened. A study synthesises: these four records, made independently and for different purposes, are only jointly explicable if the following was the case.
- Locate records
- Establish who made each and why
- Fix the sequence
- Cross-check across independent sources
- Synthesise into a claim
- State what the evidence cannot support
The last chip is the one that gets skipped. Because historical evidence is always partial, an explicit statement of what your records cannot establish is part of the finding, not an apology for it. The same discipline appears at the analysis stage of the process generally, where the source describes theorising data as establishing where and how your data supports and contradicts the theory in the literature.
The comparative extension
The source gives comparative methodology one sentence, and that sentence contains two quite different designs. Both compare societies; one compares across time, the other across cases at the same time.
Comparison in the past
- Two or more settings, both studied historically
- The comparison is between how each developed
- Chronology has to be established separately for each case before they can be set side by side
- Typical project question: why did the same reform take hold in one delivery organisation and not another
Comparison in parallel in the present
- Two or more settings observed as they currently are
- The comparison is between present states rather than developmental paths
- Needs the settings to be genuinely comparable on the what kind question
- Typical project question: how do two divisions running the same framework actually apply it
Take those gaps to the published methods literature rather than filling them from intuition. The catalogue of undefined methodology labels, including the archival and documentary entries, is set out in The Methodology Catalogue.
Evidence about the past is mostly someone else's evidence
This is where historical methodology collides with the source's own definitions of data, and the collision is worth understanding before you write a data chapter.
Apply those definitions to a study of the past and something awkward appears. Almost everything a historical study works from is, by this definition, secondary: it was gathered by other people, for other purposes, and has been interpreted by others. Yet the source characterises primary data as timely, which no evidence about a past event can be.
The source does not resolve this, and neither should you in your write-up. What you can do is be explicit about it: state that your evidence is documentary, state who produced each record and for what purpose, and treat that provenance as part of your analysis rather than a limitation buried in a final paragraph. The two categories are compared in Primary and Secondary Data Sources.
Reading a record you did not create
A worked outline
A delivery organisation has run a standard assurance framework across two business units for several years. In one unit it is used as intended; in the other it has quietly reduced to a document exercise. The research question is how the divergence came about.
Structuring the study on the four questions
Where — set the two boundaries
Define each unit as a case, including which projects, which sites and which parts of the governance structure fall inside it. Record what sits outside, because your conclusions will stop at that line.
When — build one chronology per case before comparing
Assemble dated records for each unit separately: adoption decisions, template versions, restructures, changes of leadership, audit findings. Comparing before the sequences are fixed produces confident nonsense.
Who — attribute every record
For each document, establish who produced it and for what purpose. A gate review written for an external auditor and one written for an internal steering group are different kinds of evidence about the same event.
What kind — classify the activity, not the artefact
Sort episodes by what actually happened — a decision taken, a requirement waived, an escalation made — rather than by the type of document that recorded it. This is what makes the two units comparable at all.
Synthesise across the two chronologies
Set the two sequences side by side and look for the points at which they diverge. The comparative design earns its place here: the divergence points are the candidate explanations.
State the limits with the finding
Name what the record cannot show — decisions taken verbally, staff who have left, periods with thin documentation — as part of the conclusion rather than as a disclaimer at the end.
The supplied source offers one adjacent worked design of its own, for a question about changing work health and safety laws, which pairs the discovery of policy documents with a before-and-after comparison of volunteering numbers. It is set out in Designing Your Research and is a useful demonstration that document work and comparison over time routinely travel together.
Failure modes and a pre-flight check
Before you commit to a historical or comparative design
- You have confirmed the records exist and that you have authority to access them, not merely that they should exist.
- Each of the four questions has a named source of evidence behind it, and you can say which records answer which.
- Your chronology is built from dated records rather than from recollection, and you have noted where dating is uncertain.
- For a comparative design, you can state in one sentence why the cases are comparable on the what kind question.
- You have a plan for contradictory records that does not consist of preferring the more convenient one.
- The claim you expect to make is one that synthesis could support, not one that a single document would have to prove.
What to carry forward
- Historical methodology is the synthesis of evidence about the past, structured by four questions: where, who, when and what kind of activity.
- Comparative methodology is defined in a single clause and usually travels with historical work, comparing societies in the past or in parallel in the present.
- Synthesis is the operative word. If your output narrates rather than assembles a claim, it is a report and not a study.
- Almost all evidence about the past is secondary by the source's own definition, which makes provenance part of your analysis rather than a limitation.
- The source supplies no procedure for source criticism, case selection or comparability. Take those to the published methods literature.
- State what the record cannot show alongside your finding, not after it.
Frequently asked questions
What are the four questions in the source's definition of historical research?
Where the event happened, who was involved, when the event occurred, and what kind of activity was involved. The source presents them as the considerations through which evidence is synthesised to establish facts and draw conclusions about the past, and it labels the methodology qualitative.
Can comparative research be used on its own?
The supplied source does not say. It defines comparative research as often used with historical research, comparing different societies whether in the past or in parallel in the present, and never addresses whether it stands alone. If you intend to use it independently, justify that from the published methods literature rather than from the teaching material.
Is historical research primary or secondary data?
By the source's own definitions almost all of it is secondary: existing data gathered by other people or organisations for other purposes, and interpreted by others. That sits awkwardly with the source's description of primary data as timely, since no evidence about a past event can be. Rather than resolving the tension, state clearly that your evidence is documentary and treat the provenance of each record as part of your analysis.
How is this different from a lessons-learned review?
A lessons-learned review narrates what happened and extracts recommendations. A historical study synthesises independent pieces of evidence into a conclusion that none of them states on its own, establishes the sequence from dated records, weighs each record by who made it and why, and states explicitly what the evidence cannot support.
How many cases do I need for a comparative study?
The source gives no number and no criteria for case selection. What it does require is that the cases be comparable, which follows from the what kind question in the historical definition. Justify your case count and your selection from the published methods literature, and be able to say in one sentence why the cases are alike enough to compare.
References and source attribution
- Clarke, R. J. 2005, Research methods and methodologies, pp. 35–45 (the work the supplied source adapts its methodologies table from).
- Quinlan, C. 2011, Business Research Methods, 1st ed., Cengage Publishing. (Chapter 1, p. 50 — the methodological pyramid.)
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
- Naoum, S. G. 2013, Dissertation Research & Writing for Construction Students, 3rd ed., Routledge.
- The Australian Concise Oxford Dictionary, p. 972 (the definitions of research quoted in the supplied source).
- The supplied teaching source: Week 2 study notes and slide deck on introduction to research methods, including the research methodologies table.
Suggested questions for Ask KEVOS
- How do I structure a study of a past project failure using the four framing questions?
- What makes two organisations comparable for a comparative research design?
- Is documentary evidence primary or secondary data in this source?
- How do I handle two records of the same event that contradict each other?
- What is the difference between a historical research study and a lessons-learned report?
- How do I state the limits of archival evidence without undermining my findings?