The guide on how to write a hypothesis outlines the different types of hypothesis and how to formulate a hypothesis for research or a thesis.
Find more career guides, tips and advice
The core of a thesis or research essay is the research question. It sets out the content which will eventually be discussed in the work. For example, it could read: “Do children learn to read early if they were regularly read to in infancy?” In order to examine this question in the bachelor’s thesis, it is advisable to develop hypotheses which can subsequently be empirically examined. Hence, it is crucial for the success of your paper to know how to write a hypothesis.
What is a hypothesis?
The term hypothesis comes from Greek and means “insinuation” there. Hypotheses are therefore assumptions and assumptions, which therefore refer to the research question. They can be proven or refuted and thus make an important contribution to the answer to the research question. For example, the following hypothesis could be used for the example research question:
Children who have been read regularly in infancy read independently earlier than those who were not regularly read.
So, a hypothesis is a well-founded assumption that you make at the beginning of your empirical investigation. And which is a possible answer to the research question of your thesis or research essay. You test this assumption in your qualitative or quantitative analysis. Then you confirm or refute them.
When you make your hypothesis, you either assume a correlation or no connection between your two variables. Your independent variable is the cause, the dependent on the possibly occurring effect. The hypothesis helps you to proceed systematically when collecting and processing data.
Types of hypothesis
There are undirected and directed hypotheses.
Example of undirected hypothesis
For an undirected hypothesis, you only make a connection between two variables:
Reading to a child affects how well they lean to read.
Example-directed hypothesis
In the directed hypothesis, on the other hand, you evaluate the presumed connection:
The more you read to a child, the better they learn to read.
If you need to make a claim for your bachelor’s thesis or master’s thesis, it is best to use a directed hypothesis. Because this is more meaningful for your empirical investigation.To make a hypothesis, you must consider the following criteria:
- Both variables must be measurable.
- If you form several hypotheses, they must not contradict each other.
- Scientific assumptions must be refuted.
- Hypotheses must be objective and concise.
Whether you form the assumed relationship between two facts as if-then- or je-desto relationships is up to you. The only important thing is that you define a directed context. Only directed hypotheses make it possible to compare several scientific studies with each other.
Preparing well
Although hypotheses are based on assumptions to be examined, they should nevertheless have a theoretical foundation and should not be ad hoc assumptions (see Corsten et al. 2017, p. 6). Therefore, an intensive examination of the research field is very important in advance. During the detailed literature research, one should therefore inform the reader as precisely as possible about the current state of research. Often one discovers results on questions that one would otherwise have asked oneself and also receives suggestions as to what could still be investigated.
From the research question to the hypotheses
First of all to write a hypothesis, one should look at already existing findings and then familiarize yourself with the current state of research. Are there any research results that can be included accordingly? Or do certain assumptions have to be verified themselves? It is possible that the existing material already gives rise to a direction in which one’s own research can eventually go. And sometimes, in the literature research, additional aspects are also included, which can be taken into account in one’s own work.
Different types of hypotheses are: working hypotheses, correlation hypotheses, differential hypotheses, distribution alimpes (see Weber 2005, 2.2.1.3).
How to write a hypothesis: Formulation of hypothesis
Hypotheses must always describe a relationship between at least two factors. In our example, it is the presumed correlation between regular reading and later independent reading.
It is also important to note that the hypothesis should have a sufficiently large amount of information and be logically structured. It should be empirically verifiable and falsifiable, as well as precise and unambiguous and theoretically sound (cf. Kornmeier 2008, p. 131).
The assumptions to be tested should therefore be formulated as precisely as possible. In this way, the reader of the work clearly becomes clear about what is meant and what has been studied.
Particularly helpful and meaningful are hypotheses that contain not only a connection, but also a condition or direction. Here are two examples from educational research:
- “The more students in a class, the worse the performance of individuals.”
- “If teaching material isn’t relevant to exams, then students forget it more quickly.”
The first assumption, with its direction statement, refers to a direction of the context of class size and student performance. The second indicates a condition through the if-then statement.
However, hypotheses with directed and conditional connection are usually not only more meaningful and valuable in terms of content. They are also more concise in their structure (If-Then, for example). On the one hand, this is very conducive to the criteria of objectivity and comparability of research. On the other hand, a good formulation of assumptions is generally a central aspect. Even the best approach can be lost in a hypothesis that is too long, nested or contradictoryly formulated. It is therefore important to keep them as concise and comprehensible as possible.
Nevertheless, it must be complete sentences, not key points. In addition, the researcher must always remain objective. Own opinion, thoughts and speculations have nothing to do with an assumption. The individual hypotheses must not contradict each other either. However, they may build on each other or complement each other (see Kornmeier, 2016, 125ff.). Good assumptions must always bring added value, they are not simply intended to reflect well-known facts. At best, they are somewhat sharpened, and thus stimulate discussion.
To illustrate, here are the above examples of condition and direction assumptions in a deliberately less formulated variant:
- “In many cases, the number of pupils in individual school classes differs significantly. A connection between the class size and the individual performance of the respective pupils in the classroom is to be assumed.”
- “Whether a particular learning material is queried in a class work or in a final examination is often told by their teachers to students in their schools. Students don’t even learn this material and probably forget about it right after class.”
It becomes clear that word choice and sentence construction have a decisive influence on the quality of hypotheses. They should not be too academic, but they should not be too colloquial. After all, it depends on more than a good writing style — and incorrect formulations cannot be saved in a bachelor’s thesis or master’s thesis.
Where to put a hypothesis in a paper
But how do you integrate the hypotheses into a thesis? It makes sense to list the assumption directly in the introduction. First, one should explain the overarching research question, and then go into the individual hypotheses. After all, these are closely linked to the research question and serve to ultimately answer it.
One can and should give oneanother room to examine in detail the theoretical or practical derivation of the hypotheses. It must be demonstrated that the hypotheses are based on sound scientific knowledge. They must not be based on personal opinions. In the course of the research process, the hypotheses are then either confirmed or refuted. This can already be addressed in the middle part of the research work.
However, an in-depth reflection of the work with the hypotheses should only be carried out at the end. Here we summarize again which hypotheses have been confirmed, which have not been confirmed and where one sees reasons for them. It is also important, of course, to show how the hypotheses have helped to work on and answer the research question.
Basically, hypotheses are scientific tools. They help the researcher to structure his data and ensure that a common thread is not lost. But well-thought-out and formulated assumptions make a significant contribution to the quality of research work.
In empirical work, the use of hypotheses is practically mandatory. But even in theoretical work, their application can be useful and profitable. It is possible to confirm or refute hypotheses, for example, by means of literature. It is therefore generally advisable to spend sufficient time in formulating hypotheses. In this sense: “The better the hypothesis, the easier it is to answer the research question.
How to write a hypothesis: Checking hypotheses
Once the hypotheses have been formulated and selected, the empirical examination is required. To do this, the appropriate research method must first be selected, because a hypothesis “must say something about reality that can be verified by empirical data.” (Ernst et al, 2002, p. 87).
In our first example above, the survey would be suitable for parents of children who are already reading independently, based on a qualitative questionnaire. This would be developed on the basis of the underlying hypotheses and would try to examine and verify them as much as possible. For example, with regard to hypothesis 1, the questions might be: “How many times have you read to your toddler during the week” and “When did your child start reading on her own?”
If the evaluation of the results reveals a significant correlation between the reading frequency in infancy (often) and the start of independent reading (earlier than in children who have not been read often), hypothetised 1 can be confirmed. If no connection can be discerned, this hypothesis can be rejected. Thus, all the hypotheses formed are gradually reviewed and thus contribute a piece to the answer to the research question.
It is not important whether the hypotheses put forward turn out to be true or false in the course of the research. Both are a result that can bring one closer to answering the research question. It cannot be inferred from this that the quality of the allegations is the result. What is much more important is the technical content of the adoption. Is it directly, understandable, related to the research question? Was it derived from systematic observations or considerations? Is there a scientific theory behind it? Is the statement verifiable at all?
How to write a hypothesis: Gathering results
In the result part of the bachelor’s thesis, the evaluations of the research are then compiled and explained in detail. Which hypothesis could be clearly substantiated or rejected? The final part of how to write a hypothesis properly is the conclusion section. Conclusions can then be drawn in this section, and, if necessary, other approaches that result from the results and which can be followed up.
How to write a hypothesis: in summary
Hypotheses are important pillars of scientific work, and knowing how to write a hypothesis is crucial for your academic results and career. Hypotheses derive from the research question and assume a causal connection. This can then be tested using empirical methods. It is important in the formation of hypotheses in advance a thorough research of the current state of research, that the hypothesis has a sufficiently large information content, that it is logically structured, empirically verifiable and falsifiable, and that it is precisely and theoretically sound. The results of the tested hypotheses are summarized in the results section and discussed in the conclusion.
Here again the most important points, how to formulate hypotheses and what to look out for:
Hypotheses…
- must be related to the subject of research
- should be short, concise sentences
- must not contain speculation or false assumptions
- must be relegable in principle
- may not logically exclude or contradict themselves
- are more meaningful if they contain a condition or direction (if-then, Je-Desto)
- are formulated inductively or deductively depending on the research objective
- must not contain a personal opinion or anticipate research results
- should be explained in the introduction and reflected and confirmed or refuted in the final part
Hypotheses thus help the researcher to focus on the essential questions. They give the whole process a structure. They also facilitate the choice of an appropriate empirical method. It is therefore necessary to invest sufficient time in making hypotheses. The more detailed this preparatory work is, the easier it is to carry out the examination itself afterwards.
