Theses have much in common with books and reports. They should follow the principles of logical structure, within the limits imposed by any required formats.

A generic structure for the discussion of a topic or project is the ‘bowtie’ model:

Begin with the big picture to give the work context, then focus on the specific problem being tackled. The body says what was actually done, then the conclusions place that in the broad context established in the introduction.

In sciences, a common way of summing this up is IMRaD – introduction/method/results/and/ discussion and conclusions.

In other fields, a common way of thinking about structure is CEC – claims/evidence/commentary:

• Claims – what are the contentions, ideas or interpretations being put forward?
• Evidence – what is out there to support the claims?
• Commentary – how does the evidence support the claims, and what can we learn from this?

When a thesis encompasses several topics – as many do – structure may require considerable thought. Should it contain separate methods/results/conclusions for each topic, or a single results section, but with 1 part for each topic – or aspects of both? Always think about what the reader will find easiest to navigate.

It might help to sketch possible structures :

Choice of structure depends on how much, and what, the topics have in common – do they share methods, for example? If so, the third structure might make sense.

Once the structure is worked out, writing becomes a matter of filling in each box.