Hypothesis comes from Greek hypothesis = basis, assumption
- It can be general: presumption, assumption; opposite of knowledge
- or a verifiable statement about a possible fact that goes beyond mere observation
Hypotheses can be formulated about all, some or individual facts of a certain area.
A hypothesis is:
- a statement / assumption to be tested, between two or more variables
- a preliminary prediction concerning the relationship between two or more variables
- a description of the problem or the question predicting the expected results
- a claim without a question
There are different types of hypotheses:
- simple (bivariate) / multiple (multivariate)
- non directional (direction of the relationship)
- statistical / null hypothesis
- inductive (based on concrete experience) vs. deductive (derived from theory)
Hypotheses are the link between theory and research:
In PREMIER, transparent scientific work should be the focus of every organization. Therefore the first working hypothesis has to be included in the pre-registration. In this way, so-called HARKing can also be prevented.
"HARKing" means "formulating hypotheses after the results are known": a hypothesis based on the interpretation of the data is presented as if it had already existed before the data were obtained. HARKing may also occur when a researcher tests an a priori hypothesis but then omits that hypothesis from their research report after they find out the results of their test.
The pre-registration is preferably stored in an electronic laboratory journal, so that all steps can be traced at any time.
In order not to get caught up in a hypothesis and to try to prove it, even though it may be incorrect, it is necessary to make a counter-hypothesis or a null-hypothesis. The null hypothesis states that there is no effect or difference or that a certain connection does not exist. The testing of the counter-hypothesis helps to avoid false conclusions.
How can the hypothesis be tested?
The hypothesis is examined through a series of tests (study), collecting and analyzing the obtained data. Before starting the experiment, it is fundamental to define the number of test subjects (number of samples) needed to draw reliable, valid conclusions and the at least two groups to be compared. These groups can be independent of each other, or they can be matched pairs.