Descriptive, Explanatory and Evaluative Research
Purpose is the first real decision in a research project and the one most often left implicit. The supplied teaching source names three purposes, defines each of them twice in slightly different words, and then leaves you to work out what follows — which is almost everything else in the design.
Three purposes, defined twice
The material gives the three purposes in two places. The first is a terminology table that sits alongside the qualitative, quantitative, inductive and deductive entries. The second is the topic-selection guidance, where the same three appear as the types your research can be. The wordings are close but not identical, and the differences are worth reading rather than smoothing over.
THE TWO WORDINGS SIDE BY SIDE
| Purpose | Terminology table | Topic-selection guidance | What the second adds |
|---|---|---|---|
| Descriptive | Seeks to describe what is; associated with exploratory | Finding out, describing what is | Foregrounds the finding-out work, which reads as the earlier, looser stage |
| Explanatory | Seeks to explain relationships between phenomena; associated with predictive | Explaining how or why things are as they are, and using this to predict | Makes prediction part of the purpose rather than an associated term |
| Evaluative | Seeks to test policy/management outcomes against benchmarks; no associated terms given | Evaluation of policies and programmes | Narrows the object to policies and programmes, and drops the benchmark |
Both wordings are the source's. The fourth column is our comparison.
One asymmetry stands out. Descriptive and explanatory each carry an associated term — exploratory and predictive respectively — and evaluative carries none. Whether that is a considered position or an omission, the material does not say.
The verbs that give the purpose away
You can usually identify the purpose of a study from the main verb in its question. The groupings below are built from the verbs inside the source's own definitions; the source does not set them out as question stems.
What each purpose sounds like
What is, how many, how often, who
Questions that map a situation. What tools are in use across the portfolio? How often is this process bypassed, and by whom? The output is an accurate picture of a state of affairs that was not previously documented.
Why, how, what is the relationship between
Questions that connect two or more things. Why do estimates in one delivery model drift more than in another? What is the relationship between procurement route and risk exposure? The output is an account of a relationship, and sometimes a basis for prediction.
Did it work, against what, and should it continue
Questions that judge something against a standard. Did the new assurance process deliver what it was introduced to deliver? The output is a verdict, and it is only as good as the benchmark you judged against.
Purpose is not methodology
This is the distinction most easily lost, because the same words are reused. Purpose is what the study is for. Methodology is the strategy you use to carry it out — in the source's own words, the design of the study. The material contains a separate table of named methodologies, and two of those names duplicate purpose words.
Descriptive appears in the methodology table with an entirely different definition: a methodology that relies on observations recorded for subsequent analysis, depends on human responses and observations, and can be distorted by selective observation and biased interviews. It is typed as qualitative. That is a method of working, not a reason for working, and it has its own page in this library; the page you are reading is about purpose only.
PURPOSE, AND THE METHODOLOGIES THE SOURCE NAMES THAT COULD SERVE IT
| Purpose | Methodologies from the source's table that fit naturally | Where they are covered |
|---|---|---|
| Descriptive | Descriptive · Historical · Comparative | Descriptive research methodology |
| Explanatory | Correlation · Experimental | Correlation and experimental methodologies |
| Evaluative | Evaluation · Action | Evaluation and action research methodologies |
The methodology names and their qualitative or quantitative typing are the source's. The alignment of methodology to purpose is our reading — the source never connects its two tables.
One topic, three purposes, three different projects
The source offers a list of broad project management themes as a starting point for topic selection, and stakeholder management is one of them — see research themes in project management. The worked comparison below takes that single theme and runs it through each purpose. The theme is the source's; the questions and their consequences are ours, built to show how much the purpose decides.
THE SAME TERRITORY, THREE WAYS
| Purpose | The question | Evidence it needs | The strongest claim it supports |
|---|---|---|---|
| Descriptive | How is stakeholder engagement actually being carried out across the current portfolio? | A census or sample of live projects; documented practice; observation of what is done rather than what is mandated | An accurate account of current practice, bounded to the population and the period you covered |
| Explanatory | Why does engagement effort vary so widely between projects of similar size? | Comparable measures across enough cases to test candidate factors, plus accounts from the people making the decisions | A defensible statement about relationships between factors — association more readily than cause |
| Evaluative | Has the engagement framework introduced last year produced the outcomes it was introduced to produce? | The stated intended outcomes, a benchmark to judge against, and outcome data from before and after | A verdict on the framework against a named benchmark, valid for the conditions in which it was applied |
An illustration of how purpose changes a design. The figures, populations and framework here are hypothetical; the source provides no worked example of this kind.
Notice what happens in the third row. The evaluative question cannot be answered at all until someone supplies the benchmark, and in most organisations the benchmark either does not exist or exists only as an unwritten expectation. Establishing it is often the first real task of an evaluative study, and it is work you must budget for.
What the purpose fixes downstream
Purpose is not a label you attach at the end. It is the first constraint in a chain, and each link narrows the next.
- Purpose
- Question
- Methodology
- Methods
- Analysis
- The claim you may make
- Question form. The purpose fixes the verb, and the verb fixes what an answer has to look like. See deciding your research question.
- Methodology. Different kinds of research questions require different kinds of approaches — the source is explicit about that. A descriptive purpose served by an experimental design is a coherence failure before any data is collected.
- Evidence. A descriptive purpose needs coverage; an explanatory purpose needs comparison; an evaluative purpose needs a before, an after and a standard.
- Reasoning. The source associates descriptive work with exploratory and inductive reasoning, and explanatory work with prediction, which leans deductive. See inductive and deductive reasoning.
- The claim. Description supports this is what we found. Explanation supports this appears to be related to that. Evaluation supports measured against this benchmark, it did or did not deliver.
The order also runs in reverse, which is a useful diagnostic. If the claim you want to make at the end is evaluative, and your design has no benchmark in it, the problem is in the design and not in the writing. Fixing that is the substance of designing your research.
Choosing your purpose
Which purpose fits what you actually have
Before you lock the purpose in
- You can name your purpose in one word and your question contains the verb that matches it
- You have not smuggled a second purpose into a subordinate clause
- If evaluative: the benchmark is named, sourced and defensible, and you can obtain outcome data that predates the thing you are evaluating
- If explanatory: you have enough comparable cases for the comparison to mean anything
- If descriptive: the population you claim to describe is the one you can actually reach
- The methodology you are proposing serves the purpose you have named, and you have not chosen it by duplication of the word
What to carry forward
- Descriptive, explanatory and evaluative are purposes — reasons for doing the study. They are a different vocabulary from the methodologies that carry the work out.
- The source defines each purpose twice, and the wordings differ. The topic-selection version adds prediction to explanatory and narrows evaluative to policies and programmes.
- The word descriptive does double duty as both a purpose and a named methodology in the same material. Say which you mean whenever you use it.
- An evaluative study is only as strong as its benchmark, and the source never tells you where one comes from. Budget time to establish it.
- Fix the purpose before the methodology. Almost every design incoherence in a practitioner thesis can be traced back to a purpose that was never explicitly chosen.
Frequently asked questions
What is the difference between descriptive research as a purpose and descriptive research as a methodology?
The purpose is a reason for doing the study — it seeks to describe what is. The methodology of the same name is a way of working: relying on observations recorded for later analysis, dependent on human responses, and vulnerable to selective observation and biased interviews. The supplied source defines both and never connects them, so use the word with a qualifier.
Can a single project be both descriptive and explanatory?
In practice most explanatory studies contain a descriptive component, because you have to establish what is happening before you can explain it. What matters is which purpose carries your contribution. Declare that one as the purpose and treat the other as groundwork.
Where do the benchmarks for an evaluative study come from?
The supplied source does not say. It defines the evaluative purpose as testing policy or management outcomes against benchmarks and stops there. In practice you either adopt a published model, use the outcomes the initiative itself declared at approval, or construct a baseline from historical data — and whichever you choose has to be justified before you use it.
Is exploratory a fourth purpose?
Not in this material. Exploratory appears as one of the three forms the research process takes, and separately as an associated term for descriptive research and for inductive reasoning. It is not listed as a purpose in its own right, which is one of several places where the source's vocabularies overlap.
Does the purpose determine whether my study is qualitative or quantitative?
It influences it but does not decide it. The source keeps purpose and medium as separate vocabularies: a descriptive study can count things or record accounts, and an evaluative study can do both. The choice of medium follows from what your question needs as evidence, not from the purpose label alone.
References and source attribution
- Naoum, S. G. 2013, Dissertation Research & Writing for Construction Students, 3rd ed., Routledge (set reading on topic selection, pp. 9–11).
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London (listed among the source's references for topic development).
- Clarke, R. J. 2005, Research methods and methodologies, pp. 35–45 — the work the source adapts for its table of research methodologies, including the descriptive, correlation, experimental, evaluation and action entries referred to here.
- An edited volume of master's degree research projects in project management, 2018, academic publisher, 378 pp. — the exemplar volume; the illustrative evaluative design is paraphrased from a published abstract.
- The supplied teaching source: master's-level teaching material introducing research methods for project management, supplied as study notes and accompanying slides.
Suggested questions for Ask KEVOS
- What is the difference between an explanatory and an evaluative research purpose?
- Rewrite my research question so it has a single clear purpose.
- Where can I get a defensible benchmark for an evaluative study?
- Show me how one project management theme could be studied three different ways.
- Why does the source use the word descriptive for both a purpose and a methodology?