Your report, thesis or article in a journal or a book must begin with an abstract; usually the abstract is in Finnish and English, possibly also in Swedish. The abstract precedes the actual text of your thesis or article (or other form of production), and the maximum length is one page or less. In journals and conference proceedings, the length is usually determined quite specifically; it is typically between 150 and 200 words.
The abstract should sell your thesis or article (or other proof of learnedness); it should convince someone browsing databases that your text is worth reading in its entirety. The abstract must be informative, crisp, down to the point and succinct. Within the given word limit, the abstract must summarize your work efficiently.
The abstract must read without the thesis report; it must function on its own. Note that the abstract is not the same thing as the introduction. The abstract is written (or finished) after the writing of the thesis report.
Examples of an abstract identity information (nimikeosio engl. kiel. abstraktiin) are provided below.
ABSTRACT
Kaisa Meri and Simo Virtanen
Evaluating the Effectiveness of Mobile Health Applications in Managing Diabetes
40 p., 2 appendices
Spring 2024
Diaconia University of Applied Sciences
Bachelor of Health Care
ABSTRACT
Elna Andersson, Lucas Ahlström and Eetu Martikainen
Supporting Families with Disabled Children: The Impact of Social Services on Family Wellbeing
46 p., 1 appendix
Autumn 2024
Diaconia University of Applied Sciences
Bachelor of Social Services
The content of the abstract varies according to whether the thesis is a research report or another form of carrying out the thesis work, for example, a product/project. The following aspects may be included:
To conclude, write your thesis abstract following this template:
The following instructions may be useful. This is not an exhaustive list or a synopsis of English grammar. Rather, the aim is to point out things that are known to be critical issues in abstract writing.
1. Use the passive voice. Do not write In this thesis I studied; instead, write In this thesis … was studied. You can also write This thesis studied/investigated…. Why should you avoid the passive voice? The passive voice makes the text more formal by effacing the writer’s persona. The text is important, not the writer.
2. Do not use s-genitives in a row. Do not write the informants’ answers’ main content; instead, write the main content of the informants’ answers.
3. The definite article is usually appropriate when referring to your informants, research methods, research, etc. Remember that your thesis is the thesis in your abstract, not thesis or a thesis: The aim of the thesis is to…
4. The past tense is used when describing what was studied, how the study was carried out or how the production was accomplished. As for the conclusions, the present tense can be used. In the conclusion, regarding further research avenues, the future tense is, of course, appropriate.
5. Do not use a comma after the subject, however long it may be. In speech, we are likely to use a pause (to inhale) should the subject be very long. In writing, a pause (i.e. a comma) is not needed. Thus, in the sentence The chairman of the meeting representing various countries with a say in the matter suggests that new proposal be made, no comma is needed after matter – even though the subject may seem exceedingly long.
6. Subject-predicate agreement is important: if the subject is in the singular, the predicate (the verb) must be in the singular, and if the subject is in the plural, the predicate must be in the plural. In the sentence above, the subject is in the singular (the chairman is but one person): the predicate (suggests) is also in the singular.
7. Each sentence in you abstract must be complete: each sentence must contain a subject and a predicate. An incomplete sentence would be, for example, the latter one in the following context: There are two point that should be emphasized. The social ramifications and the taxation issue.
8. You can (and you should) use subordinate clauses in the abstract. Remember that you cannot use a subordinate clause without the main clause. Which was an important result and Because that increased social unrest would be incorrect sentences on their own.
9. Do not use a comma before a subordinate clause beginning with the subordinator that. The sentence The literature pertaining to this specific topic indicates that the issues are still largely unresolved must not contain any commas; note also the subject-predicate agreement here.
10. Avoid run-on sentences. A run-on sentence is a case of two complete sentences being squashed together without a coordinating conjunction or punctuation. That is, chunks of thoughts are squashed together without demarcation. The sentence Some clients prefer online services however they are not accessible everywhere is unclear without proper punctuation: a semicolon must be placed after services and a comma after however: Some clients prefer online services; however, they are not accessible everywhere.
11. Avoid comma splices. A comma splice occurs when short and closely related clauses are linked together by a comma only, without a conjunction such as and or but. An example would be: The services are definitely needed, they are provided very professionally nowadays. The conjunction and would come to the rescue here: The services are definitely needed, and they are provided very professionally nowadays.
12. Remember: non-restrictive relative clauses (providing non-essential, superfluous information) need commas, whereas restrictive relative clauses (providing essential, indispensable information) do not. If you write Dr. Smith was the first researcher, who dealt with the social implications of this issue, the sentence suggests that Dr. Smith was the first researcher - ever. This is hardly true. As for commas with relative clauses, when in doubt, leave them out!
13. Use a semi-colon to separate items in a list or to separate two closely related independent clauses, not a dependent and an independent clause. A correct formulation would be Williams wrote several books; none of them, however, were as successful as the first, ‘The Triad’. An incorrect formulation would be Williams wrote several books; ‘The Triad’ being the first and the most successful.
14. Be careful with vague or confusing pronouns and antecedents. Is it clear to what or to whom pronouns refer? Is the referent suppressed? For example, as for the sentences The disaster was reported in the papers. They still didn't act, it is unclear what they refers to. It does not refer to papers, surely. One might write Government officials still didn't act, and the reader would understand. Be careful when you begin sentences or paragraphs with This; the reader should know (without scanning backwards) which noun This refers back to.
15. Avoid redundant and obvious expressions. Do not tell the reader what he or she does not need to know. For example, in the introduction In our modern world of today... the concepts today and modern overlap, and so do our and modern.
16. Use linking expressions, in small doses: e.g. firstly, secondly, thirdly, furthermore, in addition, consequently, on the one hand, on the other hand, finally, in conclusion. These expressions outline your text and make your argumentation clearer.
17. Do not begin a sentence with And or But. This would be too informal.
18. Do not use abbreviations; preferably do not use acronyms. If you must include them, explain them. Instead of As regards the KIKY agreement and the SOTE reform…, write As regards the Competitiveness Agreement (KIKY) and the Social and Health Services (SOTE) Reform…
19. Contractions are a matter of correctness rather than style. The plural of man is men, and the possessive of men is men's, not mens'. Do not confuse it is, contracted as it's, with its, the possessive adjective. An example of the confusion would be The cup lost it's handle. Do not confuse the possessive with the plural, either. An example would be The boy's came home late. The possessive of their is theirs, not their's.
20. Do not hyphenate. Word will do it for you if needed.
21. Avoid colloquial expressions: the abstract should contain the word children instead of kids, for instance. Also, avoid very common words with a vague meaning: e.g. think, thing, people, say, sure, money.
22. If you use non-English words in the text (as data examples, for example), use italics.
23. Do not translate Finnish terms unless they have obvious or official translation equivalents in English. If you mention a Finnish organization in the abstract, make sure that you use the appropriate English term for it. For example, Valvira must be translated as “National Supervisory Authority for Welfare and Health”.
24. Use the language check tool in Word. Do not rely on Google Translate or other translation software; they do not produce a publication-ready translation. Digital tools can be useful for language revision purposes (see below).
25. Have someone read you text and give you constructive criticism.
26. Read abstract in theses and published work in your field.
A final point here concerns the overall ethos of abstract writing. Firstly, the abstract is not the place to show creativity in language use. The language of the abstract, within a discipline or a field of study, is constrained and almost ritualistic – the same terms and phrases occur again and again. They do as the discipline deals with only so many issues and questions. New aspects and results emerge but many of the research questions remain the same. Secondly, one often seen adjectives such as innovative and important in an abstract. These words may be justified but usually the writer is advised to let the reader or the supervisor be the judge as to how innovative the thesis is. Indeed, the writer is advised to use such hedging expressions as apparently, evidently, probably, somewhat, to some extent, presumably and so on to convey the idea that the thesis reports only what seems plausible in the light of the evidence. Indeed, one of researchers’ favorite expressions is the somewhat inconclusive verb to suggest: a typical expression is that the results suggest that…
Useful thesis-related vocabulary and phraseology is presented below. Note that these are not sub-headings in the abstract; they just describe things that may be included the abstract.
AIM
- the aim of this thesis is to
- the taking-off point of this thesis is to…
- this thesis aims to….
- the thesis work arose from the social observation that…
- the question of children’s position in this situation was behind the development work idea to…
- a development work proposal in the social work with immigrants was the basis of…
- the idea for the thesis centered around the development needs in early childhood education…
MATERIAL/DATA/BACKGROUND
ANALYSIS/METHODS/PROCES
RESULTS/EVALUATION
CONCLUSION AND FURTHER RESEARCH
KEYWORDS
YSO - General Finnish ontology
One may wonder if Google Translate, for example, can be used when producing the English abstract. The answer is, as so often, that it depends. Google Translate can be used to produce a preliminary version of the English text but one does well to remember that it is just that, a preliminary and approximate version. The same goes for the language checker application available in Word.
Automatic language checkers and translation tools today are, for the most part, statistics-based algorithms, and often they are very powerful. The language model needed for the target language (English) is trained from a very large training corpus. Typically, the training corpus contain hundreds of millions or billions of words in English (newspaper texts, for instance). Sequences which occur often in the corpus can be considered correct in other texts, and uncommon sequences might be errors. The standard method is the N-gram, which means that windows of three units (words) are slid over the text to be analysed (your abstract), and the stored N-gram sequences in the memory architecture of the algorithm are compared with those in your abstract.
Language modeling is the art of determining the probability of a sequence of words. What this means is that Word and Google Translate operate in a probabilistic fashion, trying to figure out the most likely word or phrase which the writer is fumbling for. Typically, a neural language model learns the probability distribution of the next word given a fixed number of preceding words which act as the context. The probabilities are based on the vast training corpora of the algorithms but when it comes to more specific text genres or more creative use of language, the non-infallibility of the algorithms begins to show. They cannot handle, in a fully trustworthy way, specialized texts dealing with highly specific topics. Thus, when using these applications with and for producing English text on a specific area, one needs to be critical and double-check the terms. This can be achieved by using up-to-date online vocabulary resources.
One should also remember that, when using Google Translate and/or Word for writing, data privacy in a critical issue. The algorithm producers need huge amounts of multilingual data – data is their capital. They are thus likely to store the data their online systems receive for the purpose of constantly updating the training corpora. Your text, which in some cases might be confidential, could end up as part of the training corpus.
Apart from Word and Google Translate, there are other digital writing tools to consider, some of them being very promising. Pro Writing Aid, Grammarly and White Smoke are some of the most recently introduced digital writing tools. These algorithms, especially the Pro Writing Aid, appear to be very effective in correcting and revising the text online in real time. In addition, the writer will receive, if he or she so chooses, tips and suggestions as to how the grammar and style can be further improved.
All in all, the writer of an English abstract has a number of digital writing tools at his/her disposal. All of them can be useful but none of them is the definite solution – for now. In the best case, when experimenting with theses algorithms, the writer will come to develop his/her own language instinct. However, optimistically or defeatistically, one may conclude that, with neural networks and deep learning being integrated into these algorithms, a day may come when the machine can be the ultimate language teacher and evaluator.
Original article: Juhani Toivanen 2020
Updated by Liisa Keränen and Juhani Toivanen 2024