The hypothesis is a proposition that has not yet been corroborated and from which an investigation can be developed.
That is, a hypothesis is a statement that may or may not be true. However, it is formulated based on an indication or a series of facts, to which certain assumptions can be added.
The hypothesis can be based on scientific evidence or a set of arguments that have support. Otherwise, it would not make sense to start an analysis work on it.
It should be noted that the hypothesis is an essential element in the scientific method, since it starts from a hypothesis in order, through experimentation, to verify or refute it.
It is worth clarifying, however, that the use of hypotheses is not only framed in the academic context, but also in the everyday environment. For example, if a group of friends meets to meet in a cafeteria and one is taking a long time to arrive, those who are already gathered could hypothesize what could have happened (an accident, unexpected vehicular traffic, the person decided at the last minute who was not going to attend, etc).
Characteristics of the hypotheses
The hypotheses must meet the following general characteristics:
- The variables with which you want to test a hypothesis must be real and be related to that hypothesis. For example, going back to the example of the meeting with friends, if a friend is late or does not appear, the reason cannot be related to the fact that there has been a fire in the town next to the city where he lives.
- For a hypothesis to be tested, it must be based on circumstances that can be observed and are not the result of imagination.
- The variables on which a hypothesis is based must be able to be quantified and studied.
- The more variables a hypothesis includes, the greater the difficulty in testing it.
- The hypotheses must be able to be both approved and rejected, if they only contemplate one of these options, it cannot be considered hypotheses.
Types of hypotheses
Depending on the way it was formulated, a hypothesis can be:
- Inductive: It results from an analysis process that goes from the particular (a concrete case) to the general. For example, if on the arrival of a tourist in a city a person notices that bars and restaurants close at 12 am, he can suggest as a hypothesis that there is a law that orders the closure of said premises from the midnight.
- Deductive: They are those that are studied through the deductive method, that is, when it goes from the general (such as laws or principles) to the particular (the reality of a specific case). For example, we know that there are no direct flights between Madrid and Iquitos in Peru. So, if Mariana tells us that she traveled from Madrid to Iquitos last month, the most reasonable hypothesis is that on the way (by air) she has made stops.
- Analog: They start from comparisons. For example, Marco traveled eight hours by bus from Madrid to Barcelona, stopping twice during the journey. Then, months later, if a friend tells him that they will do the same route, Marco will think that during that trip they will also make a couple of stops.
On the other hand, the types of hypotheses regarding its scope can be:
- General: They are applicable for all the cases involved. They are divided into:
- Universal: They are approaches for the entire universe studied. For example, the hypothesis that a disease could have been transferred from animals to people.
- Probabilistics: They are expressed as a possibility or as a percentage. For example, if, based on historical data, the vast majority of high school students are expected to pass the geography course.
- Individuals: They apply to a singular case. For example, when I hypothesize that Carlos got the flu because his father, who lives with him, had also had the flu a few days before.
Hypothesis in statistics
In statistics, we have two types of hypotheses:
- Null hypothesis: It is the statement that the researcher intends to reject.
- Alternative hypothesis: It is the conclusion that the researcher wants to reach.
These concepts will become clearer with the following example.
How do you formulate a hypothesis?
Before formulating any hypothesis, the researcher or researchers must carry out a documentation process. Throughout this process, they must collect information in order to draw the conclusions that will be taken as hypotheses.
Once the necessary information has been collected, it will be studied to establish what are the reasons that have caused this situation to occur. For example, if we want to know why the reservoirs in a region are at 50%, we will have to obtain data about rainfall throughout the year, the evolution of irrigation, household consumption, etc.
Finally, based on these data, the corresponding hypotheses will be raised to try to test them.
Examples of hypotheses
Suppose a researcher wants to show that the level of education is relevant to the economic growth of a country.
Thus, the null hypothesis of your econometric analysis may be that the average years of education of the population have no correlation with the country's per capita gross domestic product. On the contrary, the alternative hypothesis is that both variables do depend on each other, the second on the first.
Another example, let's imagine that we are workers in a nursing home and after observing the people who stay there, we deduce that the life expectancy of women is higher than that of men.
To do this, we will establish as a null hypothesis that men have a life expectancy equal to or greater than that of women. Regarding the alternative hypothesis, we will indicate that women have a higher life expectancy than men.