The conceptual model in your thesis

It may prove useful to create a conceptual model for your thesis before conducting any research. In this model, you show how factors from your research are expected to relate to each other. This way, you visualise your hypotheses before you start testing them in your research. Use our example to discover what a conceptual model for your thesis might look like. 

Increasingly, it becomes compulsory to include a conceptual model in your thesis, at least if you are conducting evaluative research. With this, you make it clear to your reader at a glance which connections between variables you want to investigate.

The meaning of a conceptual model is as follows: it is a visual representation of the different variables in your research and the relationship between them. You indicate which relationships you expect. With such a model you give direction to your research. The model helps you, for instance, to shape your research plan, determine your research design and to answer your research questions later on. 

Not all research requires a conceptual model. You usually use such a model only in testing research, for example when you conduct an experiment. In literature research or descriptive research, for example, such a model is often unnecessary. 

In a conceptual model, you portray each variable from your research. A variable is a factor you can measure and/or manipulate for your research.

A conceptual model is meant to reveal possible relationships. So, you indicate which variable has possible effects on which other variable and how exactly it may affect it. This may be causation (one variable affects the other) or correlation (when one variable changes, the other changes with it). 

These three types of variables are particularly important:

• Independent variables: variables that potentially affect another variable. 

• Dependent variables: variables that are potentially affected by (and therefore dependent on) another variable. 

• Moderating variables: variables that may co-determine whether this relationship exists and/or how strong the relationship is (think IQ, age, gender, level of education, etc.).

• Mediating variable: variables that depend on the independent variable and influence how strong the relationship is between the independent and dependent variables.

In addition, there are sometimes control variables. More on that later!

You can think of the conceptual model as a translation of your hypotheses. In your hypotheses, you predict relationships between variables based on existing literature. You visualise these in the conceptual model. You then test these hypotheses (and thus the model) by conducting research. 

Let's clear up what aconceptual model should look like by using a (highly simplified) example.

Suppose you are curious to know whether students do their exams faster if they drink a cup of coffee just before the exam. 

So, you are interested in two variables: the time it takes students to complete their exams and students' coffee consumption. 

Coffee consumption is the independent variable; exam time is the dependent variable. Namely, you want to know whether exam time depends on (= is determined by) students' coffee consumption.

In the conceptual model, the first step is to place the independent and dependent variables. You do this by using blocks and arrows or lines. In each block, you put the name of the variable. With the arrows and lines, you indicate the direction and type of relationships. 

As we explained earlier, there are two types of relationships:

1. Cause-effect: one variable directly affects the other variable.

2. Correlation: when one variable changes, the other automatically changes with it (even without cause and effect).

Cause-effect relationships are indicated with an arrow; for correlations, use lines.

For example, your simplified conceptual model could look something like this:

Independent variable:                                                  Dependent variable:

coffee consumption                          →                         exam time

This process is usually not as simple as the conceptual model example above suggests. Models often include moderating, mediating and control variables. 

For example, a moderating variable could be study time. Suppose students who spend more time preparing always take their exams faster. Then in this example, study time might be moderating the relationship between the other variables.

One possible mediating variable is focus. The more coffee someone drinks, the higher their focus is, and thus the shorter their exam time will be. This variable possibly explains the cause-effect relationship between coffee consumption and exam time. However, always test with a statistical analysis whether there is indeed this mediating relationship.

Control variables are variables that may also affect the results, independently of the effect you are investigating. They may explain the effect found, so you should consider them before drawing conclusions about a cause-effect relationship or correlation. For example, there are more variables influencing someone's exam time than just coffee consumption. Think about IQ, reading ability or study background. These are all control variables to include in the conceptual model as well as in your research.

Usually, you give the conceptual model a place at the end of your introduction or theoretical framework. You then insert it after your research question and subquestions or hypotheses. After all, these elements all come together in the conceptual model.

Most study programmes have guidelines on where to place the conceptual model in your thesis. So make sure to ask your supervisor which structure you should follow. Moreover, check our article on thesis structure. There you will find an overview of all components that should be included in your thesis.

Have you created your conceptual model and is your theoretical framework ready? Good job! You might be wondering if you've done it correctly and if you've covered all the parts that must be included. Not to worry! Our editors can check your thesis for language, as well as the structure and a common thread. 

We point out ambiguities in your argumentation or parts that are missing. This way, you can hand in your thesis later with peace of mind.