Steps 7-8: Conducting the Research and Theorising the Data
Two consecutive steps, wildly unequal in the source. One gets two sentences and assumes you already know how to run it; the other contains the only account in the supplied material of how data becomes knowledge.
Step 7 in full: two sentences
The section headed "conduct research" is reproduced below in its entirety. Nothing has been left out.
Set that against the next step and the weighting becomes clear.
The imbalance is not an oversight so much as an assumption, and the assumption is about you. The material is written for practising project managers, and it takes execution as the thing you already know how to do. What it does not assume is that you know how to turn a dataset into a contribution to a field.
That assumption is defensible, but it leaves the action stage with no controls of its own. The rest of this section supplies them from elsewhere in the same source, which is safer than importing them from outside it.
What "effective project management is required" has to mean here
The source describes the project manager's role activities in some detail in earlier material. Mapped onto a research project in the field, they turn a one-line instruction into something you can actually run.
THE SOURCE'S ROLE ACTIVITIES, APPLIED TO THE ACTION STAGE
| Role activity, as the source defines it | What it controls while collection is running |
|---|---|
| Planning — deciding what needs to happen in accordance with the deliverables, planning tasks and schedules | The collection window, the order in which sources are approached, and what has to be finished before the writing half begins |
| Organising — allocating and procuring resources, making arrangements, allocating tasks | Booking interviews, obtaining system extracts, and securing the access that approval permits but does not arrange |
| Controlling and monitoring — checking progress against plans, reviewing, amending and reporting | Response rates, completed interviews and returned records tracked against the target, weekly rather than at the end |
| Leading and motivating — determining what needs to be done and generating willingness to do it | Participation. Most non-response in an internal study is not refusal, it is a diary problem nobody chased |
| Reporting — reviewing, amending and reporting progress | The supervisor conversation, and the early warning when a source is not going to deliver |
The role activity definitions in column one are the source's. Their application to the action stage in column two is synthesis; the source makes the connection only by the phrase "effective project management is required at this stage".
What the source does supply, scattered across other steps, is a serviceable set of controls. Every item below is drawn from a statement made somewhere in the supplied material; assembling them into a running checklist is the synthesis.
Controls to run during collection, drawn from elsewhere in the source
- The half-and-half time rule — collection stops in time to leave half the schedule for writing and rewriting
- Your answer to "how much data or information is enough", used as the stopping rule it was written to be
- The authority for each source confirmed before you approach it, not after
- The three ethics screening questions answered for each method actually used, not just each method planned
- Writing daily, even a paragraph or two, first thing where possible
- All drafts and references noted and dated as you go
- Back-up copies kept, and stored in a separate location from the originals
The back-up instruction is worth taking literally. The source illustrates it with a story of a researcher who lost a nearly finished doctoral thesis in a house fire — an anecdote rather than evidence, but the lesson costs nothing to apply. The design that all of this executes is covered in Step 5: Designing Your Research.
Step 8: the analysed data is theorised
The eighth step carries the intellectual centre of the whole process. It is worth reading twice, because it describes a movement rather than a task.
- Analyse the data
- Test it against the theory in the literature review
- Establish how it fits the theory in the field
- Knit the findings into the body of knowledge
- New theory, or an extension of existing theory
The first move in that chain is the one most often skipped. Analysis and theorising are different operations: analysis produces findings, theorising places them. A results chapter that reports what the data showed and stops has performed the first and not the second.
The second move explains why the literature review had to be built the way it was. You cannot establish where your data supports or contradicts the theory unless the theory was stated in a form specific enough to be contradicted. A review that summarises what people have written, without committing to what the field currently holds, gives the data nothing to be tested against; see The Conceptual and Theoretical Framework.
Note too the phrase "the theory in the field", which is wider than the theory in your review. Your review is a selection; the field is what the selection was drawn from. Findings that fit your review but sit oddly against the wider field are worth saying so about.
Supports, contradicts — and the case the source does not name
The source names two relationships between data and theory, joined by "and/or". Real datasets produce four. The fourth is the one that causes trouble, and the supplied material does not address it.
FOUR RELATIONSHIPS BETWEEN YOUR DATA AND THE THEORY
| What the data does | Status in the source | What it obliges you to do |
|---|---|---|
| Supports the theory in the literature review | Named | State where and how. Support that is not localised to a specific claim is not support, it is agreement in general. |
| Contradicts the theory | Named | State where and how, and consider the alternative explanations — your data, your instrument, your setting — before concluding the theory is wrong. |
| Supports in one part and contradicts in another | Implied by the source's "and/or" | Treat the boundary between the two as the finding. That boundary is usually where the extension of existing theory lives. |
| Neither supports nor contradicts | Not addressed anywhere in the supplied material | Say so. An inconclusive result is a result, and burying it is the failure. Report what would have been needed to settle it. |
Column two records the status of each case in the supplied source. Column three is synthesis.
On the inconclusive case, the exemplar volume offers a precedent worth knowing about. One of its seventeen studies examined a programme that was not successfully implemented within the research timeframe, and reported on why. The study was published as part of the volume; see Seventeen Worked Research Designs in Project Management.
A worked theorising pass
Suppose the literature review established a claim that stakeholder satisfaction on projects rises with the frequency of formal communication, and the study surveyed satisfaction alongside communication frequency, then interviewed a subset of respondents.
From finding to theoretical statement
State the finding
Satisfaction did not rise consistently with frequency. Two of the sites with the highest reporting frequency recorded the lowest satisfaction scores.
Locate it against the claim
This contradicts the claim as stated. The contradiction is specific: it concerns frequency, not communication as such, and it appears at the high-frequency end.
Bring the second dataset to bear
Interviews indicated that respondents valued knowing when information would arrive more than how often it arrived. The counting strand found the anomaly; the accounts strand explains it.
Consider the alternatives
Before the theory is amended, rule out the instrument, the sample and the setting. A satisfaction scale that measures something else, or two sites that differ in some other respect, would explain the same pattern.
Write the theoretical statement
Frequency of communication is a poor proxy for the underlying variable, which appears to be predictability. That is an extension of existing theory rather than a replacement of it — the original claim holds where frequency and predictability move together.
Attach it to the field
Name the body of work the extension attaches to, and say what would be needed to test it properly. This is the knitting the source describes.
Notice that the theorising is what produced the contribution, not the analysis. The analysis produced an anomaly. Placing the anomaly against a specific claim, explaining it with the second dataset and proposing a revised variable is the work step 8 is describing.
Knitting findings into the body of knowledge
The closing sentences of step 8 connect back to the definition of research the source opens with: all research is about the generation of knowledge, and as every research project concludes and makes its contribution, the body of knowledge grows. Step 8 is where that contribution is actually made; see What Is Research? Definition, Purpose and the Body of Knowledge.
It is also where a promise made much earlier falls due. Among the seven things the theoretical framework does is to outline the theory to which your research will ultimately contribute. If that outline was written honestly at framework stage, step 8 has a target. If it was not, this is where the absence shows.
One practical consequence: do not let "conclusions" shrink into a summary of results. The source's own test is whether new theory, or an extension of existing theory, has come out the other end.
Before you move to writing
The transition to step 9 has a precondition attached to it in the source: you cannot write a report before you have collated your information, interpreted it and come to some rational conclusions.
The step 8 exit check
- The analysis technique is named, and the text you took it from is cited
- Each finding is stated separately from its interpretation
- Each finding has been located against a specific claim in the literature review, not against the literature in general
- Support and contradiction are both reported, including the contradictions you did not expect
- Alternative explanations — instrument, sample, setting — have been considered before any theory is amended
- Anything the data could not settle is reported as unsettled
- The theory your findings attach to is named, and you can say whether the contribution is new theory or an extension
With the theorising done, the remaining steps are the report and the disposition of the data itself; see Steps 9-10: Writing the Report and Storing the Data.
What to carry forward
- Step 7 is two sentences. The source assumes you can execute and spends its words on interpretation instead.
- Supply the missing controls from elsewhere in the same source: the source's suggested time split, the stopping rule, the authority checks, daily writing, dated drafts and off-site back-ups.
- Analysis produces findings; theorising places them against the theory in your literature review. They are different operations.
- Locate support and contradiction against specific claims. Agreement in general is not a finding.
- The source has no account of an inconclusive result and supplies no analysis technique at all. Report the first honestly and cite the second.
- The output the source expects is new theory or an extension of existing theory — not a summary of what the data showed.
Frequently asked questions
How much guidance does the source give on actually running the research?
Two sentences: effective project management is required at this stage, and this is the action stage where you apply your research methods. There is nothing on data quality, non-response, record-keeping during collection, or what to do if the design has to change once collection has started. The controls have to be assembled from statements the source makes at other steps.
What is the difference between analysing data and theorising it?
Analysis produces findings — what the data shows. Theorising establishes where and how those findings support or contradict the theory in your literature review, and therefore how they fit with the theory in the field. The source treats theorising as the step that produces the new knowledge; a results chapter that stops at findings has done only the first half.
What do I do if my data contradicts the theory in my literature review?
Report it. The source explicitly includes contradiction: you establish where and how your data supports and/or contradicts the theory. Localise the contradiction to a specific claim, then rule out the alternative explanations — your instrument, your sample, your setting — before concluding that the theory needs revising. A contradiction that survives that scrutiny is usually the most valuable finding in the study.
What if my data neither supports nor contradicts the theory?
The supplied source does not address this case; its framing is binary. Say so plainly, describe what the data could and could not settle, and state what would have been needed to settle it. An inconclusive finding reported honestly is a contribution; one that is quietly dropped is a defect.
Does the source explain how to analyse the data?
No. It states that techniques for analysing non-numerical data are discussed in detail in a textbook, and that textbook is not among the supplied material. No analysis technique of any kind appears in the supplied files. Name the technique you use and cite the text you took it from.
How do I know whether I have produced new theory or an extension of existing theory?
The source names both outcomes without distinguishing them. A workable test is whether the claim you are making could be added to the existing account without dismantling it. If the original claim still holds under specified conditions and yours refines those conditions, that is an extension; if the original claim cannot survive your finding, you are proposing something new and the burden of evidence is correspondingly higher.
References and source attribution
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
- 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.
- An edited volume of master's degree research projects in project management, 2018, academic publisher, 378 pp. — the exemplar volume, one of whose studies reports on a programme not successfully implemented within the research timeframe.
- The supplied teaching source: consolidated weekly teaching notes and slide material on developing a research topic, used here as the basis for all statements attributed to "the source".
Suggested questions for Ask KEVOS
- Turn my findings into theoretical statements against the specific claims in my literature review.
- My data contradicts the literature — help me rule out the instrument, the sample and the setting first.
- Build a collection-stage control checklist for my study from the source's own instructions.
- How do I report a finding that neither supports nor contradicts the theory?
- Is my contribution new theory or an extension of existing theory, and how should I word it?