Typical Mistakes You Should Avoid When Writing Your Thesis

By Joshua Meyer and Timo P. Gros

A bachelor’s/master’s thesis is often (one of) the first scientific work(s) that a student writes. Therefore, it is not unusual to encounter the same, common mistakes when proofreading or grading theses. In this non-exhaustive guide, we want to draw attention to some of the typical mistakes and how to best avoid them. To this end, we list a general rule to follow, and provide good and bad examples. Please follow the good examples rather than the bad ones. Also note that there are, as always, cases in which a rule can be broken, e.g., to provide emphasis.

Adhering to the rules stated on this page will save us time and nerves, and help you get better feedback that is more specific to your particular thesis. For more information, see also the organizational notes for writing your thesis with us.

Equations

Embedding

Punctuation

Assuming $\gamma$ and $T$ have already been introduced:

Citations

Usage as Nouns (or Words)

Non-breaking Spaces

Another example (assuming \rt is a defined command for the word ‘Racetrack’), where space vanishes. Through correct macro definition, e.g., via xspace, this case can be circumvented though.

Capitalization

Within a Sentence

Students tend to use capital letters for important words or constructs, e.g.,

instead of “Markov decision process”. In general, this is incorrect. If you want to emphasize the importance of an expression, rather use \emph{…} (or a similar notion) when the word occurs for the first time, and write it normally afterward.

After Colons

A colon is used to introduce something that is strongly related, such as a list of particulars, an amplification, or a conclusion. The style guides agree that, therefore, one should generally use the lowercase letter after the colon. Exceptions are names, etc., that would also be spelled with a capital letter mid-sentence.

Some style guides propose to use a capital letter after a colon when what follows is a complete sentence. However, this makes it more difficult to decide when to use lower- or uppercase. Thus, the simpler rule is to simply stick to lowercase:

In Headings

Depending on the publisher, different title and heading capitalizations are used. Generally,

• the first letter is capitalized, and
• all “proper” words are capitalized (for some definition of proper).

The style guides differ when it comes to, e.g., adverbs, conjunctions, prepositions, hyphenated words, … When writing your thesis, the most important thing is: decide for one and stay consistent.

For the thesis title, you have slightly more freedom, e.g., if you want to emphasize an acronym:

Figures

Reference

Figures that you do not use within your text are a waste of space. Without a reference, a reader will not know when the figure is relevant or what it is supposed to show. Exactly describe what you want the reader to see/to learn from the figure. Do not babble unnecessarily about the figure, but in every case refer to it. Sometimes it is even enough to just put it in parantheses at the end of a sentence.

Captions

Hyperlinks

PDFs

How to Use Latin Abbreviations

Some Latin abbreviations are commonly used in our field. However, they are often misused in students’ theses. Therefore, we here examine their correct usage.

A bad example would be

while it is correct to write

Conjunctive adverbs connect two sentences. Other examples are ‘however’, ’thus’, or ’though’. The correct usage therefore is

While it is common to use other conjunctive adverbs in the beginning of a sentence, e.g.,

it would not make any sense to do this with ’e.g.’ or ‘i.e.’:

Note that you cannot use ’etc.’ in combination with ’e.g.’ or with similar terms such as ‘such as’ or ‘for example’.

Usually, publisher guidelines define when all author names should be spelled out and when you should use ’et al.’ within the text (in the bibliography, usually all authors are listed). For your thesis, I suggest the following (which is also common for most publishers): spell out the name of one or two author(s), and use ’et al.’ for three or more authors, e.g.,

Introduce Abbreviations

Abbreviations make your text easier to read and understand. Let’s revisit a former example that gets even better when instead writing:

If you are using common constructs, e.g., Markov decision processes, make sure to use the common abbreviations (here MDP) and do not invent new ones. Only introduce new abbreviations if there are no commonly used ones in the literature. Still, do not use an abbreviation that you did not introduce, regardless of how common and well-known it is.

Define Your Commands

One of the worst things about a thesis is inconsistency. This is not just a question of style, e.g., whether to write racetrack or Racetrack, but can further lead to the reviewers misunderstanding your thesis, e.g., if an important aspect is referred to by two different names and the reader is not aware that those names mean the same. An easy mechanism to avoid such ambiguities is to define LaTeX commands for every important or regularly used expression.

Not only does this decrease the chances of inconsistencies, but it also makes changes that might be required mid-writing much easier.

Variable Names in Math Mode

In math mode, LaTeX assumes every letter to be its own variable. If we assume your variable to be named “TEST”, simply writing $TEST \cdot TEST$ results in this ugly version

while when using mathit, the letters are displayed closer together, i.e., $\mathit{TEST} \cdot \mathit{TEST}$ produces

Consistency

Minor decisions in your thesis, e.g., whether to write figure with a lowercase or Figure with an uppercase letter when referring to a figure, are not incredibly important. However, (as already mentioned,) one of the worst things about a thesis is inconsistency. When you make a minor decision for a “style”, stick to that decision throughout your thesis.

Use Spell and Grammar Checkers

Even though the quality of your language does not explicitly influence the grade of your thesis, it implicitly does. First, it can be quite hard to understand what you want to express if the thesis is not written in proper English. Second, it is hard to focus on the content of your thesis when there are many language errors.

Whether to Use Active or Passive Voice; Whether to Use the First Person Pronoun

This section will be shorter than it should be; I could write a whole book (or at least a paper) about it. I here summarize my point of view on this complicated matter. Note that whatever this summary states holds only in our area of science – if at all.

Active Versus Passive Voice

Ancient science tells us to favor the passive voice over the active one. However, this is no longer true – and it has not been for quite some time. Sadly, this knowledge has not reached (all) schools yet, where the passive voice is still taught to be favorable.

Most of the style guides of journals, experts, … agree that the active voice is not only shorter but makes a sentence more readable. However, while you should use the passive voice carefully, it is not strictly forbidden to use it. Try to keep its usage to a minimum. Compare the active sentence

with the passive one

The active voice is not only 20% (2 words out of 10) shorter, but also easier to read. There are many examples (you can easily find references on the web) where the saving is even significantly higher.

Whether to Use the First Person Pronoun

Once again, you might be surprised: the vast majority of journals, experts, … agree that there are no restrictions on using the first person pronoun. This includes both the singular (‘I’) and the plural (‘we’) forms, depending on whether you are a single author or a group of authors, respectively. (As you are writing your thesis, of course, you are a single author.) However, in our field, it is uncommon to use ‘I’. Therefore we conclude:

There are several examples to justify this rule. We here only want to look at a common phrase

which informs the reader not only about the experiment but that we did it. In comparison, the passive voice

clearly does not answer who conducted the experiment. Science should be as precise as possible. There is no reason to conceal who conducted the experiment.

Whether to Use the First Person Pronoun and the Active Voice Together

When looking at the last example, the answer is clearly yes: we were the ones to conduct the experiment, we want to share that information and, therefore, ‘we’ needs to be the subject of the sentence. Still, let’s revisit another example:

Is the information that we created the summary necessary to the sentence? The list of the author(s) is at the top of the page anyway, and for the concrete sentence, the information is not important at all. It is better to make the ‘summary’ the subject:

In conclusion:

Editorial We

The ’editorial we’ means the inclusion of a further group of people in addition to the author(s), e.g., the author(s) and the reader together, a larger group of scientists (including the author), even all scientists in general, etc. – and exactly this ambiguity constitutes a problem. Remember, science should be as precise as possible; the ambiguity introduced by the ’editorial we’ represents the opposite. However, consider this example:

Clearly, ‘we’ here is editorial: it means the author(s) and the reader. This sentence is hard to write otherwise – at least when sticking to the active voice. Therefore, our rule comes as a compromise:

The difference between ‘we’ representing only the author(s), and ‘we’ representing author(s) and reader together can be distinguished by the context of the sentence.

Tenses

There are several different guides that, in a very complicated and very detailed manner, explain which tense to use depending on the chapter and even on the type of content. Mostly, they do not agree with each other. This section describes the solution I have found for myself. I claim neither that this solution is perfect nor that it is the only way to do it – I merely describe a rough guideline which might be helpful.

Present Tense

The present tense is the main tense of your thesis. Use it when writing about your work, e.g., describing figures, presenting results, analyzing your findings, discussing your results, etc. A few examples are:

Also, the present tense is used to describe things that are generally accepted in your field, or to refer to findings from other papers (both with a reference), e.g.,

Past Tense

Use the past tense when writing about experiments you conducted before writing your thesis, e.g.,

Note that these things you describe in the past tense must be finished before you can write your thesis – in contrast to the analysis, results, etc., you describe in the present.

Also use the past tense when writing about methodology published in former publications, e.g.,

or when writing about things that were disproved, e.g.,

Lastly, use the past tense when summarizing things you already wrote earlier (which usually is the case in the conclusion), e.g.,

Future Tense

Use the future tense only when writing about things not included in the paper but left open (which usually is the case in future work), e.g.,

Putting It All Together

It is hard to summarize all of this into a simple rule, but I think it comes down to this:

There is no need for overly complex tense rules.

Being Concise

Again, this section will be shorter than it should be. There are courses and papers just about how to be concise in science. One of the oldest and best known books about English writing describes writing concisely as: “A sentence should contain no unnecessary words, a paragraph no unnecessary sentences, for the same reason that a drawing should have no unnecessary lines and a machine no unnecessary parts. This requires not that the writer make all sentences short or avoid all detail and treat subjects only in outline, but that every word tell.” [3] At the same time, the same book advises: “Do not take shortcuts at the cost of clarity.” [3] Our ruling joins the two pieces of advice of Strunk and White:

Note that this guide is an explanation and not a publication. In many parts, it is not concise, e.g., repeating some things, or using more words than needed to ensure clarity in a way you would not use in a publication.

Closing note: Please note that having this list of typical mistakes to avoid does not mean that I did not make (some) of these mistakes in the past or that I will never make them again in the future. However, both you and I should try to stick to these rules.

References

1. Gros, T.P., Hermanns, H., Hoffmann, J., Klauck, M., Steinmetz, M.: Deep Statistical Model Checking. In: Proceedings of the 40th International Conference on Formal Techniques for Distributed Objects, Components, and Systems (FORTE'20) (2020), available at https://doi.org/10.1007/978-3-030-50086-3_6
2. Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A.A., Veness, J., Bellemare, M.G., Graves, A., Riedmiller, M., Fidjeland, A.K., Ostrovski, G., et al.: Human-level control through deep reinforcement learning. Nature 518(7540), 529–533 (2015)
3. Strunk Jr, W., White, E.B.: The Elements of Style Illustrated. Penguin (2007)
4. Sutton, R.S., Barto, A.G.: Reinforcement learning: An introduction. MIT press (2018)