The notion of hypothesis has gained great relevance , since in the scientific method it is quite important at the beginning of any inquiry. It is worth saying that every hypothesis must have important characteristics that make it valid as a hypothesis. Otherwise, it may be invalid or give wrong research results.

It is worth saying that, in the field of psychology , hypotheses are also used quite regularly. It must be said that they have even marked a way of thinking: every problem is faced by having the hypothesis beforehand, that is, considering a kind of position with respect to the problem addressed.


The hypothesis is a supposed explanation that is based on some facts that it serves as a basis. In addition, based on the information available to a researcher, it is the series of data that defines a problem, where an explanation of the solution to said problem is proposed.

The Dictionary of the Royal Spanish Academy ( describes that the term “hypothesis” comes from the Greek ὑπόθεσις  hypóthesis : “assumption of something possible or impossible to derive a consequence from it”.

In the scientific world, the concept of a working hypothesis describes only a methodical structure that lacks any objective content. Even the strictest scientists consider hypotheses to be fantasies of only practical value. Seen this way, the hypothesis is an artificiality of the human mind, necessary to systematize the acquired knowledge.

In this sense, the working hypotheses are the first explanations of the phenomenon . They are totally provisional structures, which can be accepted and discarded according to the needs of the investigation of the object.

Once these hypotheses have been raised, the researcher proceeds to verify them by looking for the phenomena that corroborate them. This is how it acquires the status of a real or scientific hypothesis.

Characteristics of the hypotheses

The hypothesis possess a number of characteristics . They are not just guesses or random assumptions. Actually, there are certain criteria for understanding hypothesis features. To this is added that not all hypotheses are the same. In this case, let’s see below some of the characteristics of this concept:

Generability and specificity

The hypothesis must have an optimal level of generalization, insofar as it goes beyond the presumed explanation of the particular. But, to be specific, it must facilitate the division of operations and predictions.

Empirical reference, verifiability and refutability

A hypothesis without reference in the real world constitutes only a value judgment , therefore it is essential that the hypothesis can be testable or verifiable. The validity then of a hypothesis depends on whether it can be subjected to empirical verification from a scientific perspective.

Now, in the realm of logic , hypotheses have value because they come out gracefully from any refutation, without contradictions.

Reference to a theory

Although it is feasible to build an investigation and formulate hypotheses without being explicitly linked to a theoretical framework, this does not lead to the formation of new science.


The hypotheses must be operational ; it is the only way to test them. For this to be possible, it is essential that they are correctly formulated, without inaccuracies, that allow deductions to be made, such as: a clear link between the variables and what they imply; in addition to the objective description of the indices to be used.

For the purposes of the so-called “operability” there are two levels in it, namely: a ) the “occurrence”, which arises from prior knowledge without due justification and represents an initial study of theoretical research; b) “testability ” is the methodology that allows the hypothesis to be corroborated. The true or false character is determined when the hypotheses are based on some proven theory or law, in which case a formal contrast would be made.

Types of hypotheses

The typology is diverse. They are grouped synthetically , although to be honest there is no single categorization of the hypotheses. In this case, we present below a somewhat general way of cataloging the:

According to its origin

In this case, the mode of thought in which the hypotheses are obtained is taken as a premise. Also, the type of initial data from which these hypotheses are formulated. Let’s see the categorization below:

  • Inductive hypotheses . These hypotheses arise from data and observation, which serve to produce theories. Its methodology is bottom-up, in the sense of its analysis logic.
  • Deductive hypotheses . Contrary to inductive hypotheses, deductive hypotheses follow a top-down logic. Starting from theory, they are used to check them in practice. Due to their formulation and meaning, they seek a more complete knowledge system than the inductive ones.
  • Statistical hypothesis . Part of an assumption that is established on one or more parameters. It can be stated in two ways: a) “null” and b) “alternate”.

The null statistic is the assertion of one or more point values ​​for population measures. Meanwhile, “alternate” statistics institute the link between variables or the discrepancy between experimental treatments. The “alternate” constitutes the statement that the investigating subject aspires to validate, although its truth cannot be demonstrated. That is to say, the alternate statistic is, worth the redundancy, the alternative to the null statistic. Together, they would make up the series of logical probabilities for the relationships that would be under analysis.

It is just in one way to understand the concept of hypotheses. There are other ways to do it, as explained below.

According to the number of variables and the descriptions made of them.

In this case, the origin of the hypothesis does not matter so much. In reality, what is relevant is nothing more than the way to use the hypothesis . Therefore, we can assert that we are facing a very practical type of cataloging. This way of ordering hypotheses and recognizing them is quite effective for those who are already using them in an investigation:

  • Descriptive hypotheses (with a single variable). They represent the presence or absence of some phenomena in the population. Although they must be verified, they do not explain the phenomena, objects of analysis. They work to verify the existence or not of a population trait or quality and reveal new hypotheses that expose the presence or not of a phenomenon.
  • Descriptive hypotheses (they link two or more variables in an associated or co-varied way). The independent variable change determines a proportional change in the dependent variable. Formulated in the form “A major or minor X … major or minor Y”. This correlation does not mean causality, but it is the starting point for determining causal hypotheses. An example would be: -the higher the cost of living, the less chance of gaining the confidence of the people in their rulers-.
  • Hypotheses that link two or more variables in terms of dependency. These hypotheses allow exposing and forecasting social processes. In causal links, the following must be met: a) the existence of concomitant variation (co-variation) between the variables; b) the co-variation must not be the result of other factors or variables; and c) the independent variable must happen before the dependent one.
  • Difference or intervention hypothesis. The hypothesis attempts to determine the consequences of some treatment (use of vaccines, for example) with the experimental group versus no treatment or no intervention with the control group.
  • Difference hypothesis in post-facto research . The analyst will try to observe the difference between groups based on one or more variables that the subject already has.
  • Equality hypothesis. It is the hypothesis that is based on the following law: if the notion A is equal to another B, and B is equal to the notion C, then the first notion A is equal to the third notion C.
  • Symmetry hypothesis. They are two hypotheses or premises, understood as specific universal judgments, symmetrical due to their mutual implication. Consequently, they are equivalent in both senses: if it is A then it is B; if it is B then it is A; if it is not A then it is not B and vice versa.
  • Hypothesis by homology . They are the hypotheses in which some link already established in similar hypotheses is transported to the conclusion. They are hypotheses that seek to verify a parallel process, but not the same.
  • Hypothesis by analogy. It is the type most used to formulate hypotheses. It is common to resort to analogies to establish virtual common systems of operation and to elaborate explanatory metaphors. An analogical hypothesis seeks to prove that what is true in one set of facts and processes can be true in another, because both have certain properties in common.

Originally, this hypothesis is based on mathematical reasoning, which seeks to verify the fourth term of a proportion when all three terms are known. Working with analogies strengthens the rational imagination and allows the advancement of knowledge.

  • Linear projective enumeration hypothesis . It is the type of hypothesis that establishes the direction that will govern the process for each of the elements of a defined class.
  • Complementary reconstruction hypothesis . They are the hypotheses that are built based on documents, testimonies, figures, among others. They are valid when you want to reconstruct social facts.
  • Hypothesis as a cyclic sequence . They are hypotheses based on biological processes. That is, the social process is understood as a life cycle: birth, development, rise, decline and death. In a logic scheme A goes to B, then to C, then to D.

As can be seen, in this case the types of hypotheses are very varied. Also, there are some that overlap each other. In this case, it is important to keep in mind that the hypothesis is always a tool . It should not be a straitjacket for the investigator.

The hypothesis, step by step

The construction of a hypothesis must be done properly . Be careful with making misconfigured hypotheses. A hypothesis that is not well stated leads to a wrong result in any investigation.

Likewise, a meaningless hypothesis causes the researcher to divert his effort along lines of inquiry that are meaningless. Not in vain it is necessary to bear in mind that the hypothesis is always found at the beginning of any investigation. Therefore, it is something like the first step, the first decision regarding the direction of the investigative process that is going to be carried out.

How is a hypothesis made correctly? The truth is that a correctly formulated hypothesis necessarily goes through the following steps:

  1. Initial thought. This stage is focused on the maturation of the problem. At first everything seems fuzzy and disjointed; the hypothesis and objectives have yet to be clarified.
  2. Plausibility. The first audit and revision of the sources serves to assess the problem. That will depend on whether it is plausible to investigate it.
  3. Acceptability. It is the stage of acceptance of the idea once its viability has been verified. Once the idea is accepted as valid and viable, it must be converted into a hypothesis to be subjected to empirical verification.
  4. Operationality of the hypothesis . In their origin, the hypotheses have a conceptual character. In order to be able to work with them, they must be made operable by transforming them into observable variables, the latter being always necessary.

In this way, it will be possible to determine the indicators to be measured and the links that can be created in said indicators . The indicators must be the manifestations that best represent the variables of the analysis.

This system is useful for gathering or designing the most convenient information gathering instruments . For the process to be effective, from the beginning a clear link must be established between the questions of the inquiry, the hypotheses or objectives, the indicators of the independent and dependent variables.

It should be noted that in any research process starting hypotheses are needed. In the same way that a detective has “suspicions”, an investigator poses “hypotheses” that allow him to rule out, demonstrate or say interesting questions about what he is investigating.

Website | + posts

Alexa Clark specializes in Cognitive Behavioral Therapy. She has experience in listening and welcoming in Individual Therapy and Couples Therapy. It meets demands such as generalized anxiety, professional, love and family conflicts, stress, depression, sexual dysfunction, grief, and adolescents from 15 years of age. Over the years, She felt the need to conduct the psychotherapy sessions with subtlety since She understands that the psychologist acts as a facilitator of self-understanding and self-acceptance, valuing each person's respect, uniqueness, and acceptance.

Related Posts

Leave a Reply

Your email address will not be published. Required fields are marked *