Types of Hypotheses in scientific research (and examples)

We review this concept so basic to science and academia.

There are different types of hypotheses in scientific research. From the null, general or theoretical hypothesis, to the complementary, alternative or working hypothesis.

What is a hypothesis?

But, what exactly is a hypothesis and what is it for? Hypotheses specify the possible characteristics and results that may exist between certain variables to be studied.

Using the scientific method, a researcher must try to verify the validity of his initial (or main) hypothesis. This is what is usually called the working hypothesis. On other occasions, the researcher has several complementary or alternative hypotheses in mind.

If we examine these working and alternative hypotheses, we find three subtypes: attributive, causal and associative hypotheses. The general or theoretical hypotheses serve to establish a relationship (negative or positive) between the variables, while the working and alternative hypotheses are those that effectively quantify this relationship.

On the other hand, the null hypothesis is the one that reflects that there is no appreciable link between the variables studied. If the working and alternative hypotheses cannot be proven to be valid, the null hypothesis is accepted as correct.

Although the aforementioned are considered the most common types of hypotheses, there are also relative and conditional hypotheses. In this article we will discover all the types of hypotheses, and how they are used in scientific research.

What are hypotheses used for?

Any scientific study must begin with one or more hypotheses to be confirmed or disproved. to be confirmed or refuted.

A hypothesis is nothing more than a conjecture that can be confirmed, or not, by means of a scientific study. In other words, hypotheses are the scientists' way of posing the problem, establishing possible relationships between variables.

Types of hypotheses used in a scientific study

There are several criteria that can be followed when classifying the types of hypotheses used in science. These are described below.

1. Null hypothesis

The null hypothesis refers to the fact that there is no relationship between the variables that have been the object of investigation.. It is also called "hypothesis of no relationship", but it should not be confused with a negative or inverse relationship. Simply put, the variables studied do not seem to follow any particular pattern.

The null hypothesis is accepted if the scientific study results in the working and alternative hypotheses not being observed.

Example

"There is no relationship between people's sexual orientation and their purchasing power".

2. General or theoretical hypotheses

General or theoretical hypotheses are those that scientists establish prior to the study and conceptually, without quantifying the variables.without quantifying the variables. Generally, the theoretical hypothesis arises from generalization processes through certain preliminary observations on the phenomenon they wish to study.

Example

"The higher the level of studies, the higher the salary". There are several subtypes of theoretical hypotheses. Difference hypotheses, for example, state that there is a difference between two variables, but do not measure its intensity or magnitude. Example: "There are more female students than male students in the faculty of psychology".

3. Working hypotheses

The working hypothesis is the one that serves to try to demonstrate a specific relationship between variables through a scientific study. through a scientific study. These hypotheses are verified or refuted by means of the scientific method, which is why they are sometimes also known as "operational hypotheses". Generally, working hypotheses arise from deduction: from certain general principles, the researcher assumes certain characteristics of a particular case. Working hypotheses have several subtypes: associative, attributive and causal.

3.1. Associative

The associative hypothesis specifies a relationship between two variables. In this case, if we know the value of the first variable, we can predict the value of the second.

Example

"There are twice as many students enrolled in the first year of high school as in the second year of high school".

Attributive

The attributive hypothesis is the one used to describe the facts that occur between variables. It is used to explain and describe real and measurable phenomena. This type of hypothesis contains only one variable.

Example

"The majority of homeless people are between 50 and 64 years of age".

Causal

The causal hypothesis establishes a relationship between two variables. When one of the two variables increases or decreases, the other increases or decreases. Therefore, the causal hypothesis establishes a cause-effect relationship between the variables studied. To identify a causal hypothesis, a cause-effect link, or statistical (or probabilistic) relationship, must be established. It is also possible to verify this relationship through the refutation of alternative explanations. These hypotheses follow the premise: "If X, then Y".

Example

"If a player trains 1 additional hour each day, his percentage of successful throws increases by 10%."

4. Alternative hypotheses

The alternative hypotheses attempt to provide an answer to the same question as the working hypotheses. However, as can be deduced from its name, the alternative hypothesis explores different relationships and explanations. Thus, it is possible to investigate different hypotheses during the course of the same scientific study. This type of hypothesis can also be subdivided into attributive, associative and causal.

More types of hypotheses used in science

There are other types of hypotheses that are not so common, but are also used in different types of research. They are the following.

5. Relative hypotheses

Relative hypotheses give evidence of the influence of two or more variables on another variable. on another variable.

Example

"The effect of the decline in GDP per capita on the number of people with private pension plans is smaller than the effect of the fall in public spending on the rate of child malnutrition."

• Variable 1: decline in GDP
• Variable 2: fall in public spending
• Dependent variable: number of people with private pension plans.

6. Conditional hypotheses

Conditional hypotheses are used to indicate that one variable depends on the value of two other variables. This type of hypothesis is very similar to causal hypotheses, but in this case there are two "cause" variables and only one "effect" variable.

Example

"If the player receives a yellow card and is also warned by the fourth official, he must be excluded from the game for 5 minutes".

• Cause 1: receiving a yellow card
• Cause 2: being cautioned
• Effect: to be excluded from the game for 5 minutes. As we can see, for the variable "effect" to occur, it is not only necessary that one of the two "cause" variables be fulfilled, but both.

Other types of hypotheses

The types of hypotheses we have explained are the most commonly used in scientific and academic research. However, they can also be classified on the basis of other parameters.

7. Probabilistic hypotheses

This type of hypothesis indicates that there is a probable relationship between two variables. That is, the relationship is fulfilled in most of the cases studied.

Example

"If the student does not spend 10 hours a day reading, he will (probably) fail the course."

8. Deterministic hypotheses

Deterministic hypotheses indicate relationships between variables that are always true, without exception.without exception.

Example

"If a player does not wear cleats, he will not be able to play the game".

Bibliographical references:

• Hernández, R., Fernández, C., and Baptista, M.P. (2010) Research Methodology (5th Ed.). Mexico: McGraw Hill Educación
• Salkind, N.J. (1999). Research Methods. Mexico: Prentice Hall.
• Santisteban, C. and Alvarado, J.M. (2001). Psychometric Models. Madrid: UNED