How to start a writing?

With this we are ready to start, we go in parts:


There are people who advise to start writing by the Methods, since it is, by far, the easiest part. On the other hand, I recommend starting with the Introduction because that is how we force ourselves to create a mental outline of our thesis, it makes us start ordering and reading the bibliography and, in summary, it starts us up.

The Introduction should serve to introduce the reader to the topic and establish the bases of our research. To achieve this, a useful rule is to go from the general to the specific. If for example we talk about the study of a mutation in a disease, we will start with a section on the disease (epidemiology, clinical, pathophysiology …), then another on the affected gene (structure, function …) and, finally, we would describe what is what is known so far of the role of our mutation in this or other diseases.

From experience, for someone who reviews a thesis, nothing is as discouraging as opening the book and seeing pages and pages of text. It is necessary to synthesize as much information as possible in figures and tables. In the text we only have to highlight the most relevant studies that have been published, but without going into discussing their contents, that’s what the Discussion is for.

Tip: a methodology that works well for the Introduction is to look for two or three recent reviews in prestigious journals on our subject and, from there, go to the original sources that cite these reviews. We will then complement this with our own PubMed search.

Extension: between 30 and 50 pages.


We must describe very succinctly and point by point what we wanted to find out with our doctoral work. We can put a small introductory paragraph that includes a hypothesis, something like: “With the previous premises (referring to the introduction), which suggest that […], we propose the following working hypothesis: […]. To try to prove this hypothesis, the specific objectives of this thesis were: […].

Advice: brief if good twice good.

Extension: very short, always less than two pages.


As said before, it’s the easiest part. It is about (i) describing the study population, animal or human, if any; (ii) reflect in detail the material and techniques that we have used in the laboratory and (iii) cite the statistical methods applied. Always in this order.

Tip: when describing the experimental techniques we will do it in the sequence that we have established in the list of Objectives.

Extension: it depends a lot on the degree of laboratory work that the thesis has had. You could say that 20-25 pages is a reasonable maximum, although there are jobs with much shorter methods.


When you have been working on a very specific topic for a long period of time you tend to lose perspective and decrease the ability to present the problem in a coherent way to someone who is not intimately related to the subject. Therefore we must present the results in the most aseptic and structured way possible, always respecting the order in which we have followed in Objectives and Methods.

Tip: if the results are clearly divided into blocks, eg. the study of three different genes, it is not superfluous to make a small summary table for each part, with three or four main results. They are not conclusions, simply review the most relevant figures. This helps the reader not get lost in a tangle of numbers and that the main message of the thesis, something that we must never lose sight of, is slowly creeping up.

Extension: this section can not be guided on how many pages would be appropriate, it depends entirely on the work done.


The right time to write the conclusions is, in my opinion, just when the results are finished. It is not necessary to give exact figures again unless they are extremely important. That is, if we have obtained a result that quantifies the risk of lung cancer conferred by the X mutation with an odds ratio of 2.03 (1.32 – 6.25), we can conclude that “Individuals who carry the X mutation have approximately twice the risk of getting lung cancer. ” In this way the message is clearer than if we hide it behind numbers.

Tip: limit the number of conclusions, always less than ten, grouping if necessary those that are more related.
Extension: two pages maximum.