Positivism and Post-Positivism
Two positions in one lineage: the demand for objectively collected factual data, and the retreat from proof to probability. What each obliges you to fix before you collect anything, and what it lets you say once you have.
The two definitions the source actually gives
The supplied teaching source treats epistemological positions as a terminology set rather than a philosophy syllabus. Each position gets one line of definition and a list of associated terms. Positivism is the first entry in that table; post-positivism is the second.
In the prose beside that table, the source names positivism as one of the three positions generally considered, alongside constructionism and interpretivism. Positivism also appears among the eight fundamental philosophies at the base tier of the methodological pyramid. Post-positivism does not appear in the pyramid at all.
THE TWO POSITIONS AS THE SOURCE PRESENTS THEM
| Term | Definition given in the source | Associated terms listed | In the pyramid? |
|---|---|---|---|
| Positivist | Hypotheses (suppositions, propositions) tested using objectively collected factual data; if successful leads to scientific laws | Scientific method, logical empiricist, objectivist | Yes — base tier, as "Positivism" |
| Post-positivist | Hypotheses found to be consistent with data establishing probable facts or laws | None listed | No |
The associated terms are labels. The supplied source defines none of them, so this page does not define them either.
The claim ladder: from hypothesis to probable law
Read the positivist definition as a sequence, because that is how it behaves in a research design. Each rung depends on the one below it holding, and a weakness low down cannot be repaired higher up.
- Prior theory
- Stated hypothesis
- Objectively collected data
- Test
- Supported / not supported
- Probable fact or law
This is deductive reasoning in the source's own terms: data collected to test a prior hypothesis or proposition, with hypothetical-deductive and confirmatory listed as its associated terms. If your questions are still forming and you expect explanations to emerge from the data, you are working inductively and the positivist frame is the wrong one to declare. The distinction is set out at inductive and deductive reasoning.
What the position fixes before you collect anything
The practical difference between a positivist study and every other kind is when the decisions are made. Almost everything is settled before the first response arrives, because a hypothesis tested against data you shaped after seeing it is not a test.
Settled before data collection opens
- The hypothesis or proposition, written as a statement the data could contradict
- Each construct in that statement, defined so two people would code the same case the same way
- The measure — the same instrument, the same wording, the same scale for every case
- The population you are speaking about, and how cases will be drawn from it
- The analysis you will run, chosen before you know what it will show
- A procedure record complete enough that another team could repeat the study without asking you a question
The source is careful in one respect worth copying: it never says a positivist study must be quantitative. What it does say is that quantitative data collection methods rely on random sampling and structured instruments that fit diverse experiences into predetermined categories, and that these methods produce results which can be summarised, compared and generalised. That is a close fit with the positivist demand, but the fit is an argument you make, not a rule you inherit. See qualitative and quantitative research compared.
Read against the five characteristics of scientific research
The same teaching source lists five characteristics of scientific research. Used as a design audit, they turn the positivist position from a label into a set of things a reviewer can check.
THE FIVE CHARACTERISTICS AS A DESIGN AUDIT
| Characteristic (source) | What it asks of your design | Where project studies come unstuck |
|---|---|---|
| Conducted within rules of science | A declared protocol, followed even when it is inconvenient | Changing the question mid-collection because early responses are dull |
| Based on logic | A stated line from theory to hypothesis to measure | A measure that does not actually operationalise the construct in the hypothesis |
| Systematic analysis of evidence | Every case handled the same way, including inconvenient ones | Quietly dropping the responses that do not fit |
| Can be replicated | Instrument, sampling frame and analysis documented in full | An instrument that exists only in the researcher's head, or was revised without a record |
| Knowledge is cumulative | Results positioned against the literature that preceded them | A finding reported as novel because the literature review stopped early |
The characteristics are from the supplied source. The middle and right columns are this library's reading of them for a project setting.
Where post-positivism changes the wording
The two definitions differ in one clause, and the clause is the point. Positivism tests a hypothesis and, on success, heads towards a law. Post-positivism finds a hypothesis to be consistent with the data and establishes something probable. The design work is nearly identical; the sentence you can write at the end is not.
Positivist phrasing
- Hypotheses are tested
- Data is objectively collected factual data
- Success points towards scientific laws
- Associated with the scientific method, logical empiricism, objectivism
- Reads as: the test settles it
Post-positivist phrasing
- Hypotheses are found consistent with data
- The source does not restate the data requirement — read it forward from the positivist line
- Establishes probable facts or laws
- No associated terms are offered
- Reads as: the test raises the probability
Which methodology the position points you towards
Worked example: comparing delivery methods with a structured instrument
One design in the exemplar volume of master's project management research shows the shape clearly. It compared project delivery methods in large-scale residential architecture, and gathered primary data by structured web-based questionnaire from 52 industry professionals.
The number 52 is what that study did. It is not a threshold, a minimum or a standard, and the supplied source sets none. Any page that tells you otherwise has invented it.
That design can conclude that, among the professionals surveyed, one delivery method was rated better on the compared dimensions. It cannot conclude that the method performs better on all projects, because the sample was not drawn to represent all projects, and it cannot explain why practitioners hold that view — the structured instrument closed that door in exchange for comparability. Other designs in the same volume are reviewed at seventeen worked research designs.
Where the frame meets the social world
Project research is social science research, and the same source lists five characteristics that pull against a positivist design. They are worth reading as constraints on the claim, not as reasons to abandon the position.
- It deals with people, either in groups or individually
- It uses the methods and traditions of social science
- People are less predictable than non-human phenomena
- The social world changes
- People are aware of research being conducted on them
What you can claim, and what you cannot
CLAIM BOUNDARIES UNDER EACH POSITION
| Claim | Positivist | Post-positivist |
|---|---|---|
| The hypothesis was tested against data collected to a fixed procedure | Yes | Yes |
| The data are consistent with the hypothesis | Yes | Yes — this is the position's own wording |
| The finding is probable rather than settled | Understated by the position's phrasing | Yes — stated explicitly |
| The relationship is proven | Not supported by the source's wording | Not supported by the source's wording |
| The result holds for the wider population | Only if the sampling supports it | Only if the sampling supports it, and then as probable |
| Why participants behave as the numbers suggest | Outside this design | Outside this design |
What to carry forward
- Positivism, in this source, is a one-line commitment: prior hypothesis, objectively collected factual data, a test that could have gone the other way.
- Post-positivism keeps the machinery and softens the claim to probable facts or laws — usually the more honest phrasing for a project-scale study.
- The position is declared in the proposal but paid for in the design: fixed measure, defined population, pre-chosen analysis, documented procedure.
- Your findings are bounded by your sampling. Write the conclusion sentence early and check the design can carry it.
- The source supplies labels and definitions, not thresholds. Any number you use must come from a methods text you cite.
Frequently asked questions
Is positivism the same thing as quantitative research?
No, and the supplied source never equates them. Positivism is an epistemological position about how knowledge is created; quantitative research is a research design working in numbers. They pair well because quantitative collection relies on structured instruments and predetermined categories, which suits hypothesis testing, but the pairing is an argument you make in your methodology chapter rather than a rule the source hands you.
Can I run a positivist study without a hypothesis?
Not on the source's definition, which is built on hypotheses being tested. If you have no prior proposition and expect explanations to emerge from your data, you are reasoning inductively and should look at an exploratory framing or another epistemological position instead of declaring one you are not using.
What is the practical difference between positivism and post-positivism in a write-up?
It shows up in the verbs of your conclusion. A positivist framing invites "the hypothesis was tested and supported"; a post-positivist framing invites "the data are consistent with the hypothesis, establishing a probable relationship". For most project-scale studies with non-random samples, the second is the defensible sentence.
How many respondents do I need for a positivist project study?
The supplied teaching source gives no number, and neither does this page. Sample size follows from the population, the analysis you intend and the precision you need, and you should cite a methods text for the rule you apply. Figures quoted from published exemplars describe what those studies did, not what yours must do.
My organisation wants a clear yes or no answer. Does that make positivism the right position?
It makes it the tempting one. The test is whether the thing being decided can be measured the same way for every case — if it can, a positivist design will give a defensible answer within stated limits. If the decision turns on what people mean by a term or how they experience a process, a design that forces those into predetermined categories will return a confident answer to the wrong question.
References and source attribution
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
- Quinlan, C. 2011, Business Research Methods, 1st ed., Cengage Publishing (Chapter 1, p. 50 — the methodological pyramid reproduced in the teaching material).
- Veal, A. J. 2005, Business Research Methods: A Managerial Approach, Longman (source of the scientific and social science research characteristics used in the teaching material).
- An edited volume of master's degree research projects in project management, 2018, academic publisher, 378 pp. (the exemplar volume; design summaries paraphrased from published abstracts).
- The supplied teaching source: Week 2 study notes and slide deck, 'Introduction to Research Methods', for a master's-level research subject in project management.
Suggested questions for Ask KEVOS
- Rewrite my research aim as a testable hypothesis and tell me what result would contradict it.
- My sample is a convenience sample of colleagues. What conclusion wording can I defend?
- Show me the difference between a positivist and a post-positivist version of the same conclusion paragraph.
- Which parts of my design have to be fixed before I open data collection?
- Is my measure actually operationalising the construct in my hypothesis, or something adjacent to it?