Knowing the seven steps for building a conceptual framework does not tell you what other people’s finished frameworks actually contain. This is the missing half: eight completed research paradigms with every box filled in, four using the input-process-output model and four using the independent-dependent variable model, each followed by the narrative paragraph that has to sit beneath the figure.
Which model to use, in one table
| Use the IPO model when | Use the IV-DV model when |
|---|---|
| Your study produces something: a programme, a system, a manual | Your study measures a relationship between existing variables |
| Your last specific question asks what may be proposed | Your last specific question asks about a relationship or a difference |
| There is a procedure the study carries out | There is no intervention; you observe and measure |
| Common in education, IT capstones, hospitality operations | Common in psychology, nursing, business, criminology |
Choosing the wrong one is not fatal, but it shows. An IPO diagram for a purely correlational study leaves the process box holding nothing but “statistical treatment”, which is a procedure you would have run anyway and not a process the study performs on anything.
Four paradigms in the IPO model
Each example lists the contents of the three boxes, in the order they appear left to right, with the feedback arrow noted where one belongs.
1. Education: a reading support programme
| Input | Process | Output |
|---|---|---|
| Profile of teacher respondents; extent of use of reading intervention strategies; pupils’ oral reading performance levels | Administration of the survey questionnaire; retrieval and tabulation; statistical treatment using weighted mean and Pearson r; analysis of the relationship | A proposed reading support programme for Grade 3 teachers in the district |
Feedback arrow: from Output back to Input, because the proposed programme is intended to change the extent of strategy use that was measured.
2. IT capstone: an inventory and sales monitoring system
| Input | Process | Output |
|---|---|---|
| Problems in the existing manual process; user requirements; hardware and software requirements; the developer’s knowledge of the framework and database used | Requirements analysis; system design; development using the chosen methodology; testing; evaluation by users and IT practitioners | A deployed web-based inventory and sales monitoring system, evaluated against the stated criteria |
Feedback arrow: from Output back to Process, representing the revision cycle after user evaluation. Draw it only if you actually iterated; a feedback arrow you did not follow is a question at the defence.
3. Hospitality: a service improvement plan
| Input | Process | Output |
|---|---|---|
| Profile of guest respondents; perceived service quality across the five dimensions; level of guest satisfaction | Survey administration in the selected resorts; tabulation; weighted mean and correlation analysis; identification of the lowest-rated dimensions | A proposed service improvement plan for the participating resorts |
4. Nursing: an infection control orientation module
| Input | Process | Output |
|---|---|---|
| Profile of staff nurse respondents; level of knowledge of standard precautions; level of reported practice | Questionnaire administration; scoring of the knowledge items; computation of weighted means; correlation of knowledge with practice | A proposed orientation module addressing the practice areas with the lowest reported compliance |
Four paradigms in the independent-dependent model
5. Business: social media marketing and purchase intention
Independent variable box. Social media marketing practice, with the indicators listed inside: content frequency, responsiveness, promotional offers.
Dependent variable box. Customer purchase intention.
Moderating box, drawn below and joined to the arrow: Customer profile, with age, sex, educational attainment and length of patronage listed.
Arrows. One arrow from independent to dependent. One arrow from the moderator to the middle of that arrow, never to either box.
6. Psychology: academic stress and sleep quality
Independent variable box. Perceived academic stress, with its scale dimensions listed.
Dependent variable box. Sleep quality, with its scale components listed.
Moderating box. Respondent profile: year level and programme.
Arrows. One from stress to sleep quality; one from the moderator to the relationship. The direction of the main arrow states a hypothesis, so be sure it is the one you test.
7. Criminology: programme implementation and community participation
Left box. Level of implementation of council programmes: advocacy, rehabilitation referral, monitoring.
Right box. Level of community participation.
Arrows. A double-headed arrow is correct here and in most descriptive-correlational designs, because the study establishes association rather than direction. Using a single arrow claims causation you cannot support.
8. Accountancy: internal control and reporting reliability
Left box. Extent of adoption of internal control practices: control environment, control activities, monitoring.
Right box. Perceived reliability of financial reporting.
Moderating box. Enterprise profile: years in operation, number of employees, nature of business.
Arrows. One from left to right, one from the moderator to that arrow.
The narrative paragraph
A paradigm without a paragraph beneath it is half a section. The paragraph does three things: it names the model, it walks the reader through the boxes in order, and it states what the arrows claim.
Figure 1 presents the research paradigm of the study, which follows the independent-dependent variable model. The independent variable is social media marketing practice, measured through content frequency, responsiveness and promotional offers. The dependent variable is customer purchase intention. Customer profile is treated as a moderating variable, since purchase intention may vary with age, sex, educational attainment and length of patronage. The arrow from the independent to the dependent variable represents the association the study tested; no causal direction is claimed, as the design is descriptive-correlational.
That final clause is worth copying. It answers, before it is asked, the question every panel asks about a diagram with an arrow in it.
The APA rules that apply to the paradigm
The paradigm is a figure, and the figure rules apply to it whether or not anyone has told you so. It carries a figure number, an italic title above the image, the image itself, and a note below where the reader needs one. It is listed in the list of figures, and it is referred to in the text as “Figure 1”, never as “the figure below”.
The rest of the manuscript-level formatting, including where figures sit relative to headings and what the list of figures has to contain, is set out in the guide to APA 7 format for a Philippine thesis.
Five faults a panel finds in a paradigm
- A box that names a variable and nothing else. If the box says only “Service Quality”, the panel cannot see what you measured. List the dimensions inside.
- Boxes that do not match the statement of the problem. Every variable in the diagram appears in the statement of the problem, and no variable appears in the diagram that is absent from it.
- A single arrow in a correlational study. It claims direction the design cannot establish. Use a double-headed arrow or say explicitly that no causation is claimed.
- A moderator joined to a box rather than to the arrow. A moderator acts on the relationship, which is what the arrow represents. Joining it to a box makes it a second independent variable, and then it needs its own hypothesis.
- An output box that never becomes an output. If the paradigm promises a proposed programme, Chapter 5 has to contain one. This is the single commonest broken promise in an IPO diagram.
Fault 2 is best caught with the alignment grid rather than by eye. Faults 3 and 5 are caught by reading the paradigm out loud as a sentence: “this study says that X is associated with Y, and produces Z”. If any part of that sentence is not true of your study, the diagram is wrong.
What the paradigm feeds
| Element of the paradigm | Where it reappears |
|---|---|
| Each box’s listed dimensions | The parts of the questionnaire, and the definition of terms |
| The main arrow | The hypothesis, and the correlation or difference test |
| The moderating box | The grouping variable in the difference test |
| The output box, in an IPO diagram | The proposal in Chapter 5 |
Each of those dimensions also needs a measurement rule, which is what the operational half of the definition of terms supplies. The theory that justifies the arrow in the first place is chosen before any of this; which theory fits which course is set out in the piece on theoretical framework examples by course. The seven-step procedure for building the diagram from scratch, if you have not drawn one yet, is in the guide to making a conceptual framework step by step.
Get the paradigm right before the instrument
The paradigm is a one-page figure that determines what your questionnaire has to contain, which tests you will run and what Chapter 5 has to deliver. Drawn late, it is a drawing of decisions already made badly. Tesify builds the boxes from your variables, writes the narrative paragraph, and keeps the dimensions consistent with the instrument and the statistical treatment you describe later. The framework and the reasoning stay yours.
Build your conceptual framework free on Tesify
Frequently asked questions
Should I use the IPO model or the independent-dependent model?
Use IPO when your study produces something, such as a proposed programme or a developed system. Use the independent-dependent model when your study measures a relationship between existing variables and produces no artefact.
Can the arrows be double-headed?
Yes, and in a descriptive-correlational study they usually should be, because the design establishes association rather than direction. A single arrow claims causation.
What goes in the input box?
The data the study starts with: the respondent profile and the variables to be measured. Not the questionnaire and not the statistical tools, which belong in the process box.
How many variables should the diagram show?
The ones in your statement of the problem and no others. Two to four boxes is typical for an undergraduate thesis, plus a moderator where one applies.
Is the research paradigm the same as the conceptual framework?
In Philippine usage, the paradigm is the diagram and the conceptual framework is the section containing it, including the narrative paragraph. Some manuals use the words interchangeably.
Where does the conceptual framework go?
Most Philippine manuals place it at the end of Chapter 2, after the theoretical framework. A minority place it in Chapter 1. Your manual decides.
What software should I use to draw it?
Anything that produces clean rectangles and arrows: a word processor’s shape tools, a spreadsheet, or a diagramming tool. What matters is that the final image is legible in print and that the text inside the boxes is the same size throughout.
Do qualitative studies need a paradigm?
Many Philippine manuals ask for one. A qualitative paradigm usually shows the phenomenon, the data sources and the themes expected to emerge, without arrows claiming relationships between measured variables.
Does the framework change after data collection?
In a quantitative study it should not, because the instrument was built from it. If your findings contradict the arrow, that is a finding to discuss in Chapter 5, not a reason to redraw the diagram.
What is the most common mistake?
An output box promising something the manuscript never delivers. If the paradigm says a proposed programme, Chapter 5 must contain a programme, not a list of recommendations.