# What Is A Hypothesis? Explained.

### Content Outline

- What is a Hypothesis?
- Characteristics of Hypothesis
- Process of Generating a Hypothesis
- Types of hypotheses
- Errors in Hypothesis
- Summary

### What is a Hypothesis?

A hypothesis (plural form: hypotheses) is a statement that explains the relationship between two or more
variables. This statement is tested during a research study, and the predicted relationship can be proved to
be either valid or invalid within the framework or particular context of the study. It is important to note
that a hypothesis is validated (or proved to be invalid) within the specified time frame of the study. A
hypothesis can be regarded as the proposed (or tentative) answer to a research question. For example, if
the research question is “Do supervisory practices help counsellors involved in HIV/AIDS counselling in
Mumbai cope with burnout?”, the hypothesis could be “Supervisory practices assist in reducing burnout
amongst counsellors involved in HIV/AIDS counselling in Mumbai’. Reducing a problem to a hypothesis
format makes it possible to scientifically solve the problem.

In quantitative studies, a hypothesis is usually stated in advance of the study, and then is tested during the study. Stating a hypothesis in advance helps researchers to keep the study focused and also helps them interpret the results of a study against an established framework. A hypothesis is formulated based on existing knowledge, and by the process of being tested during a research study, serves as a tool to advance knowledge on a particular subject. Even if a given hypothesis is proved to be false during a study, the result is regarded as an advancement of knowledge because we know that within the context of the study, the stated variables do/do not share a relationship. This knowledge is useful to other researchers who may want to test that relationship in other contexts.

Hypotheses are usually derived from theory and may be based on already conducted research on the subject. Thus, the development of a hypothesis is closely linked to the literature review conducted for the study. For example, in a study on child labor, the research question could be: What is the link between dropout rates in primary school and children being involved in labour and the hypothesis could be: As dropout rates in primary schools increase, the rate of child labor among 6-10 year old children also increases.

Hypotheses usually include the independent variable, the dependent variable, the predicted outcome and the population of a study. In the above example, school dropout is the independent variable, child labor is the dependent variable, and 6-10 year old children are the population. The predicted outcome is that child labor will increase if school abandonment increases.

In quantitative studies, a hypothesis is usually stated in advance of the study, and then is tested during the study. Stating a hypothesis in advance helps researchers to keep the study focused and also helps them interpret the results of a study against an established framework. A hypothesis is formulated based on existing knowledge, and by the process of being tested during a research study, serves as a tool to advance knowledge on a particular subject. Even if a given hypothesis is proved to be false during a study, the result is regarded as an advancement of knowledge because we know that within the context of the study, the stated variables do/do not share a relationship. This knowledge is useful to other researchers who may want to test that relationship in other contexts.

Hypotheses are usually derived from theory and may be based on already conducted research on the subject. Thus, the development of a hypothesis is closely linked to the literature review conducted for the study. For example, in a study on child labor, the research question could be: What is the link between dropout rates in primary school and children being involved in labour and the hypothesis could be: As dropout rates in primary schools increase, the rate of child labor among 6-10 year old children also increases.

Hypotheses usually include the independent variable, the dependent variable, the predicted outcome and the population of a study. In the above example, school dropout is the independent variable, child labor is the dependent variable, and 6-10 year old children are the population. The predicted outcome is that child labor will increase if school abandonment increases.

Hypotheses usually indicate the direction of the relationship between the independent and dependent

variable. These relationships are usually presented as below:- Positive association (+): Two or more variables are said to be positively associated when they change in the same direction as each other. Thus, if one variable increases, the other also increases OR if one decreases, the other also decreases. For example, the higher the levels of stress faced by students, the greater the chances of their chances of scoring good marks.
- Negative association (-): Two or more variables are said to be negatively associated when they change in opposite directions from each other. Thus, as one variable increases, the other decreases. For example, the more sleep deprived a person is, the higher the chances of their making mistakes in tests.

In order to phrase the correct relationship between the Independent and the Dependent variable, it is often
helpful to think of the hypothesis in terms of an IF-THEN statement. Since the Independent variable is the
variable that is influencing the dependent variable, an IF-THEN statement can help express this
relationship correctly in a hypothesis. For example, if we consider our earlier example of the child labor hypothesis, we can say: “IF school abandonment increases, THEN the rate of child labor also increases.”
This gives us a clear indication that school abandonment influences child labor, and therefore, school
abandonment is the Independent variable.

### Characteristics of Hypothesis

A hypothesis has the following characteristics:

- It is phrased as a statement not as a question.
- A hypothesis is not a moral or ethical question, but rather a testable statement.
- In quantitative research, a hypothesis is stated prior to conducting a study.
- It predicts the relationship between two or more variables.
- A hypothesis can be tested and verified (either validated meaning proved as true or invalidated meaning proved as false).
- It is not too general and not too specific.
- Must be stated in clear and unambiguous terms.
- Hypotheses are derived from theory.
- Hypotheses usually state the relationship between two or more variables

### Process of Generating a Hypothesis

As mentioned earlier, we generate hypotheses from existing theory. The process of generating a
hypothesis in quantitative research can be described as follows:

Theory --->Hypothesis ---> Observation ---> Confirmation

From the diagram, we see that Theory is the starting point of the process. A theory systematically defines how certain variables and/or concepts relate with each other. We can identify theories during the literature review process when we review existing research on a subject, and identify a theory on how certain principles or concepts relate to each other. We then use these theories to define a hypothesis which predicts the relationship between the variables being considered the study. The hypothesis is then tested through a process of observation during the study. Based on the observations, the hypothesis is then either confirmed or rejected.

Theory --->Hypothesis ---> Observation ---> Confirmation

From the diagram, we see that Theory is the starting point of the process. A theory systematically defines how certain variables and/or concepts relate with each other. We can identify theories during the literature review process when we review existing research on a subject, and identify a theory on how certain principles or concepts relate to each other. We then use these theories to define a hypothesis which predicts the relationship between the variables being considered the study. The hypothesis is then tested through a process of observation during the study. Based on the observations, the hypothesis is then either confirmed or rejected.

### Types of hypotheses

There are primarily two types of hypotheses – a) Null hypothesis and b) Alternative hypothesis. We will
explore each of these categories further.

**Null Hypothesis:**Also known as the Statistical Hypothesis, the Null Hypothesis is often noted as H0 (H zero) or HN (H null). This type of hypothesis expresses a variable relationship that has either been proved true or has not yet been tested but is being used as the basis for an argument. A Null Hypothesis usually states there is no relationship between two or more variables, as stated in the research hypothesis. The conclusion of a quantitative study is usually expressed in terms of the Null hypothesis. A study can either conclude that “the Null Hypothesis was rejected in favor of the Alternative Hypothesis” (see below) or “the Null Hypothesis was not rejected.” It is important to note that just because a Null Hypothesis was not rejected, this does not mean that the Null Hypothesis is valid. It only means that in the context of a particular study, there was not enough evidence to reject the Null Hypothesis or prove that it was invalid.**Alternative Hypothesis**: Also known as the Research or Scientific Hypothesis, the Alternative Hypothesis is noted as H1 (H one) or HA (H alternative).This type of hypothesis is a statement for what the test seeks to establish. This type of hypothesis is considered the opposite of the Null hypothesis and is usually arrived at when the Null hypothesis is rejected. Examples of the different types of hypotheses are as follows:

Let us assume that we need to conduct a study to explore the relationship between teacher training and student performance. Here are the possible hypotheses: Null Hypothesis: There is no relationship between teacher training and student performance. Alternative Hypothesis: IF there is an increase in training programs for teachers, THEN there is an increase in student performance.

### Errors in Hypothesis

Two types of errors are associated with hypothesis testing.

**Type 1 Error:**A type 1 error occurs when the Null Hypothesis is wrongly rejected. For example, in our earlier example of teacher training and student performance, we defined our Null Hypothesis as: There is no relationship between teacher training and student performance. Let us suppose our study results demonstrate that teacher training does NOT influence student performance. In this case, the Null Hypothesis that we defined cannot be rejected. If we were to still reject the Null Hypothesis, this would be a Type 1 error.**Type 2 Error:**A type 2 error occurs when the Null Hypothesis is not rejected even when it is proven as invalid. Again, using the teacher training example, let us suppose our study results demonstrate that the Alternative Hypothesis is valid; that is, increased teacher training leads to an increase in student performance. This means, in effect, that the Null Hypothesis has to be rejected in favor of the Alternative Hypothesis. However, if we were to state instead that the Null Hypothesis is not rejected, this would be a Type 2 error.

A hypothesis is thus an important element of a research study, because it responds to the research
question. By stating a hypothesis in advance, a researcher can connect his/her study to existing research,
and also helps in interpreting study results within the context of a predefined framework. Even if a study
proves a given hypothesis as false, this is still considered a valuable contribution to knowledge.

We have now completed our review of two key elements of quantitative research – research question and hypothesis. These two elements are considered foundational in quantitative research. The importance of these elements is visible in the fact that quantitative studies are focused on responding to the research question or testing the hypothesis.

We have now completed our review of two key elements of quantitative research – research question and hypothesis. These two elements are considered foundational in quantitative research. The importance of these elements is visible in the fact that quantitative studies are focused on responding to the research question or testing the hypothesis.

### Summary

- A hypothesis (plural form: hypotheses) is a statement that explains the relationship between two or more variables.
- A hypothesis is derived from theory.
- A hypothesis usually indicates the direction of the association of the independent and dependent variable. The association between the variables can either be positive or negative.
- There are two types of hypothesis – Null Hypothesis and Alternative Hypothesis.
- Two types of errors are associated with Hypothesis Testing. Type 1 errors happen when the Null Hypothesis is wrongly rejected. Type 2 errors happen when the Null Hypothesis is not rejected even when it is false.

## Comments