The Evidence Assembly and Analysis Model
A figure supplied as an image, with no caption, no attribution and no prose anywhere in the material that refers to it. What it does carry is a precise structure, and the structure can be read. This page reads it and stops there.
What the figure contains
The figure is supplied as an image with no accompanying text. It carries no caption in the supplied material, no attribution, and no surrounding paragraph explaining it. No week's prose refers to it. What follows is its transcription.
- Key Evaluation Questions
- Assembly of evidence
- Analysis and reporting of evidence
THE THREE BOXES AND WHAT EACH HOLDS
| Box | Heading | Inner labels | Internal arrows |
|---|---|---|---|
| 1 | Key Evaluation Questions | None - heading only | None |
| 2 | Assembly of evidence | Selecting sources of data · Gaining access to data · Obtaining data | Double-headed, between all three |
| 3 | Analysis and reporting of evidence | Data display · Data reduction · Conclusion drawing / verification | Double-headed, between all three |
The arrangement inside boxes 2 and 3 is triangular, not listed. The order in which the three labels are printed above carries no priority; the figure does not impose one.
Two kinds of arrow, and the whole of the argument is in the difference
The figure makes exactly two claims about sequence, and it makes them through its arrows rather than in words.
Between the boxes: one-directional
- Questions precede assembly.
- Assembly precedes analysis.
- No arrow returns upward, so the figure does not depict analysis feeding back into the questions or into assembly.
Inside each box: bidirectional
- Every pair of the three activities is joined by a double-headed arrow.
- The three are therefore interactive and repeated, not a sequence to be completed once.
- The figure sets no entry point among the three and no order in which they must be taken.
That asymmetry is worth holding onto, because it is the one piece of instruction the figure genuinely delivers. Its stages are ordered; its activities within a stage are not. A design that treats "select sources, then gain access, then obtain data" as three tasks in a schedule is reading the figure against its own arrows.
Box 1: Key Evaluation Questions
The first box is a heading with nothing under it. The figure gives it no sub-elements, no examples, no count and no criterion for what makes a question "key".
Structurally it does one job: it establishes that nothing enters the assembly stage until questions exist. That is a useful discipline and it is all the figure supports. The word "evaluation" is left to carry itself - the figure does not say whether an evaluation question differs from a research question, and it does not connect to anything else in the material.
Box 2: Assembly of evidence
Three labels, joined to each other in both directions. The figure names them and stops.
BOX 2, LABEL BY LABEL
| Label | What the figure supplies | What the arrows imply about it |
|---|---|---|
| Selecting sources of data | The label only. | It can be revised by what happens in the other two - a source chosen and then found to be inaccessible returns to this activity. |
| Gaining access to data | The label only. | It is not a gate passed once. The figure joins it to selection and to obtaining in both directions. |
| Obtaining data | The label only. | Obtaining can send you back to access, or to selection, rather than terminating the stage. |
The third column is this page's reading of the double-headed arrows. The figure attaches no text to any of the three labels.
The practical value here is the separation itself. Selecting a source, getting permission to use it, and actually obtaining the material are three distinct pieces of work with three distinct failure modes, and project research routinely collapses them into one line of a plan. The figure refuses that collapse.
None of the three is developed anywhere. The material's own coverage of the underlying activities sits elsewhere and is not linked to this figure: participant selection in Sampling and Selecting Participants, negotiated and covert access in Covert Research and Negotiating Access, and collection itself in Data Collection Methods.
Box 3: Analysis and reporting of evidence
The same triangular arrangement, with three terms that are the standard vocabulary of an interactive model of qualitative data analysis: data display, data reduction, and conclusion drawing or verification.
Read structurally, the box says that reduction, display and conclusion drawing proceed together and revise one another, and that verification is bound to conclusion drawing rather than following it as a separate stage. The figure carries them as a single label, separated by a slash.
The notes' surrounding material is worth having even though it does not explain the three activities. It defines memos, sets out thick description, and states that qualitative collection and analysis usually proceed simultaneously - all of which is covered in Thick Description, Memos and Field Data.
What the figure does not carry
- No attribution. The supplied image carries no citation, so the model cannot be traced to a named work from the material alone.
- No prose. No week refers to the figure, quotes it, or builds on it.
- No definitions, for any of the six inner labels or for the heading of box 1.
- No methods. Nothing states how any of the six activities is performed.
- No criteria and no stopping rule. The bidirectional arrows imply repetition and nothing says when the repeating ends.
- No roles, no timing, no sequence within a box, and no relationship to any other model in the material.
Using the figure without over-reading it
Treated as three gates and two rules rather than as a method, it is a serviceable check on a research plan. The following use is this page's synthesis; the source supplies the structure and nothing else.
Three gates, two rules
Gate 1 - questions exist and are written down
Nothing crosses into assembly until the questions are on paper. The figure's first arrow is one-directional, so a plan that begins with a data source and works backwards to a question is not this model.
Gate 2 - assembly is complete enough to analyse
Selection, access and obtaining have all been done, not merely planned. The figure gives no completeness test, so set one yourself and record it.
Gate 3 - analysis and reporting are one stage, not two
The figure places reporting with analysis, not after it. Whatever you do about that, decide it deliberately rather than by default.
Rule 1 - inside a stage, expect to loop
Plan for revisiting selection when access fails, and for revisiting reduction when a display shows something the reduction hid. Build the loop into the schedule as contingency, not as failure.
Rule 2 - between stages, do not loop silently
The figure shows no upward arrows. If your analysis does change your questions - which happens - record it as a change rather than quietly rewriting the questions to fit the findings.
What to carry forward
- Three stages in a fixed order, with three interactive activities inside each of the last two. That is the entire content of the figure.
- One-directional arrows between boxes, double-headed arrows within them. The asymmetry is the instruction.
- Selecting sources, gaining access and obtaining data are three separate pieces of work. Keeping them separate is the most useful thing the figure does.
- The six inner labels are not defined by the figure, and the material's only gloss on three of them is a few words each, quoted from elsewhere, with no procedure.
- The figure is unattributed and unreferenced by any prose in the material. Present it as a structure you found there, not as a model the material teaches.
Frequently asked questions
What does data reduction mean here?
The figure does not say. The record of the figure states that none of its six inner labels is explained anywhere in the supplied material; the qualitative week's notes separately carry a three-word parenthetical gloss, "extracting the essence", quoted from a 1994 text. Neither amounts to a procedure, so take a working definition from a qualitative analysis text and cite it.
Why are the arrows inside the boxes double-headed?
To show that the three activities in each box are interactive and repeated rather than a sequence to be completed once. It is the only substantive claim the figure makes about how the work proceeds, and it applies within a stage, not between stages - the arrows between boxes run one way only.
Is this the same as the research process models taught elsewhere in the material?
The source does not say. No week's prose refers to this figure at all, so it is not connected to the seven-stage process, the ten-step thesis model or the methodological pyramid. Treat any correspondence you notice as your own observation and say so.
What is a key evaluation question?
The figure gives the phrase as a box heading with no sub-elements, no examples and no criterion for what makes a question key. It does not distinguish an evaluation question from a research question. If you use the term, define it from a cited source.
Can I use this figure in my methodology chapter?
You can use it as a structure, provided you say the supplied version carries no attribution and you cite your own definitions for the six labels. What you cannot do is present the labels as defined terms from this material, because the material does not define them.
References and source attribution
- Miles, M & Huberman, M 1994, Qualitative data analysis - the work the qualitative research week's study notes quote for the three analysis activities that appear in box 3 of this figure.
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
- Quinlan, C. 2011, Business Research Methods, 1st ed., Cengage Publishing.
- The supplied teaching source: an evidence assembly and analysis flow diagram supplied as an image without caption, attribution or accompanying text, transcribed for this library and read here against the qualitative research week's study notes.
Suggested questions for Ask KEVOS
- Turn this three-stage figure into a checklist for my own research plan.
- What are the failure modes of the access stage in a study inside my own organisation?
- Give me cited definitions of data reduction, data display and conclusion drawing.
- How should I schedule an interactive stage where the activities revise each other?
- Compare this figure with the seven-stage research process taught in this library.