What Is a Hypothesis?

What is a hypothesis?  A hypothesis is a clear statement of what is intended to be investigated. It is a tentative prediction about the nature of the relationship between two or more variables.

A hypothesis can be tested.  They are verifiable or falsifiable, and a prediction of certain outcomes.  A hypothesis is considered valuable even if it has been proven false.

A hypothesis is NEVER a moral or ethical question.  Those questions are a lot of fun, but belong in the philosophy department.  

To better understand how to develop a proper hypothesis, consider this situation:

You are a nutritionist working in a zoo, and one of your responsibilities is to develop a menu plan for the group of monkeys. In order to get the nutrition they need, the monkeys have to be given fresh leaves as part of their diet. 

You must choose leaves from the following species: (a) A (b) B (c) C (d) D and (e) E. 

Your research has shown that in the wild the monkeys eat mainly B leaves, but you suspect that this could be because they feel safest feeding in B trees.  Your assumption is that eating leaves from the other four trees would make them vulnerable to predators.

You design an experiment to find out which type of leaf the monkeys like best: You offer the monkeys all five types of leaves in equal quantities, and observe what they eat.  

From studying the popularity of these different leaves, there are many correct experimental hypotheses you could create.

For example:

When offered all five types of leaves, the monkeys will eat more of the leaves from tree B.

This statement satisfies both criteria for experimental hypotheses because it is a prediction of what will be learned after the experiment and it is testable, that is, once you have collected and evaluated your observations of what the monkeys eat when all five types of leaves are offered, you know whether or not they ate more B leaves than the other types. 

Let’s look at some incorrect hypothesis to better understand how to form a correct hypothesis.

Incorrect hypotheses would include:

When offered all five types of leaves, the monkeys will preferentially eat the type they like best.

This statement certainly sounds predictive, but it does not satisfy the second criterion: there is no way you can test whether it is true once you have the results of your study. Your data will show you whether the monkeys preferred one type of leaf, but not WHY they liked it best. In fact, the above statement is an assumption that is inherent in the design of this experiment rather than a hypothesis.

When offered all five types of leaves, the monkeys will preferentially eat B leaves because they can eat these safely in their natural habitat.

This statement is problematic because the second part of the statement “because they can eat these safely in their natural habitat” also fails to satisfy the criterion of testability. You can tell whether the monkeys preferentially eat certain leaves, but the results of this experiment cannot tell you why. 

In their natural habitat, howler monkeys that feed in B trees are less vulnerable to predation than monkeys that feed on A, C, D, or E.

This is a perfectly good experimental hypothesis, but not for the experiment described in

the question. You could use this hypothesis if you did a study in the wild looking at how

many howler monkeys get are attacked by predators whilst feeding on the leaves of certain trees. However, for the experimental feeding study in the zoo, it is neither a prediction nor testable. 

When offered all five types of leaves, which type will the monkeys eat preferentially?

This is a question, and questions fail to satisfy criterion number one.  Questions are not predictive statements, and therefore, a question is not a hypothesis.

Creating a proper hypothesis takes some practice, but remember: a proper hypothesis is formed based on knowledge of other relationships as well as observation so get out there and explore your universe!

Credit: Arizona State University