Understanding a Hypothesis (Definition, Null, and Examples)

You come home exhausted and plop down on the couch. You don’t know why you are feeling so weary. You think about several possible reasons. Is it because you stayed up late last night? Is it because you skipped breakfast? Or is it because you had to take the stairs due to a power outage? Or is it because of all the above reasons?

What you are doing is hypothesizing about why you are feeling tired.

If you enjoy reading detective stories, you would have already come across a hypothesis. A good whodunit mystery confounds the reader with multiple hypotheses about who committed the crime.

What is a Hypothesis?

The term hypothesis is often used in a scientific context as a possible explanation for an occurrence.

The word originated from ancient Greek and means “putting under” indicating its early association with experimentation.

A hypothesis should	Rule
Explain what you expect to happen	Rule #1
Be clear and understandable	Rule #2
Be testible	Rule #3
Be measurable	Rule #4
Contain and independent and dependent variable	Rule #5

A hypothesis is:

An assumption that serves as a starting point for further research
A supposition made on the basis of insufficient evidence
A tentative and logical statement that can be tested for its authenticity
An idea that seeks to explain why a phenomenon takes place
A prediction about the outcome of a study according to known facts
A proposal about the possible relationship between two or more variables

A scientist testing a hypothesis is no different from a detective investigating a crime scene. Famous detectives such as Sherlock Holmes combine the evidence with their powers of prediction to identify the criminal from several potential suspects.

The scientist examines each hypothesis rigorously for any inconsistencies through experiments before it can receive the stamp of approval.

Scientists accept a hypothesis as a theory only after it has been validated several times in different conditions. This includes use of scientific methods and protocols involving observation and analysis of results.

A good hypothesis seeks to establish a causal relationship between two or more variables, primarily between the independent and the dependent variable.

Example:

Brushing your teeth at least twice in a day reduces the incidence of dental caries.

The independent variable or cause in the above example is the number of times you brush in a day. The dependent variable or effect is the incidence of dental caries or cavities.

A scientist or researcher tests a hypothesis by changing the independent variable and measuring its effect on the dependent variable.

A relationship between a single independent and dependent variable is known as a simple hypothesis.

The mathematical expression of this relationship is:

Y = f(x)
where x is the independent variable and Y is the dependent variable and
where x is the input and Y is the output or a function of x

So, brushing your teeth at least twice daily is an input and the reduction of dental caries is an output or a function of the action of brushing your teeth.

If there are multiple independent variables or in some cases more than a single dependent variable, the statement is a complex hypothesis.

Example:

Brushing your teeth at least twice a day and using dental floss reduces the incidence of cavities and periodontitis.

In the above example the two independent variables are brushing teeth and using dental floss. The dependent variables are reduction in cavities and periodontitis or gum infection. In this example the two independent variables are common for the two dependent variables.

The equation of a complex hypothesis can be written as:

Y = f(x₁+x₂+x₃…)

Y₁ = f(z₁+z₂+z₃…)

where z is a different set of independent variables for Y1 as the dependent variable

Developing a Hypothesis

A hypothesis is a frame of reference or a window through which you observe a phenomenon. The phenomenon is the dependent variable. Your job is to determine the independent variables that are causing the event.

Cultivate the habit of looking for patterns in anything that happens. Train your mind to think in terms of stimulus and reaction or cause and effect.

This will enable you to glean insights from the knowledge you gather. You will then be able to write a strong hypothesis that focuses on the variables that matter over the noise.

The six steps to developing a hypothesis are:

Ask a question
Preliminary research
Formulate the hypothesis
Refine the hypothesis
Phrase your hypothesis in three ways
Write a null hypothesis

Ask a Question

The first step is to write a research question.

To write an effective research question be as curious as possible. Start with asking yourself a ton of questions.

Begin with broad and open-ended questions before narrowing it down to more specific ones.

You can use the 5W1H method to get into the mode of writing a research question.

What took place?
When did it happen?
Where did it occur?
Why did it take place?
Who did it affect?
How did it happen?

The research question needs to be clear, objective, well-defined and measurable.

Example:

Do people who take health supplements log in fewer sick days at work in a year than those who don’t?

After you have framed the right question you can make an educated guess to answer it. This answer will be your preliminary hypothesis. Your hypothesis will attempt to answer the research question with observable facts through various experiments.

Preliminary Research

You don’t have to start from scratch. You can draw from preexisting knowledge and well-established theories to discount fallacious premises at the outset.

Resources that you can refer to include case studies, research papers and theses published in academic or scientific journals. A thorough background research will help you to look at the research question from several angles.

Do keep an open mind or a blank slate to avoid falling in the trap of preconceived notions and prejudices. Your initial research should help you focus on the areas where you are most likely to find the answers.

You can come up with a blueprint or outline highlighting the variables that you think are most relevant to your research question.

Think how changing the attributes of a single variable potentially affects others. You may need to operationalize or define how you are going to measure the variables and their effects.

Formulate the Hypothesis

It’s time now to put together your hypothesis into words.

A sound hypothesis states:

Who or what is being studied?
The relationship between the variables
A measurable and reproducible outcome
The possibility to prove it as true or false

Example:

Teenagers in the 14-16 age group who eat a high-protein diet are taller by two inches than the average height for that age group.

Refine it

The next step is to ensure your statement ticks all the boxes for a strong hypothesis.

Is the hypothesis:

Precise and quantifiable without any ambiguity
Lucid and focused on the results described in the research question

Does the hypothesis include:

An independent and dependent variable
Variables that can be changed or controlled
Terms that even a layman can understand
A well-defined outcome

Phrase your Hypothesis in Three Ways

A hypothesis is often written in an If-then format. This format describes the cause and effect relationship between an independent variable and a dependent variable.

Phrase your hypothesis as “If {you make changes to an independent variable} then {you will observe this change in the dependent variable}.”

Example:

If employees are given more autonomy to take work-related decisions then their overall performance improves.

Another way to write a hypothesis is by directly stating the outcome between the two variables.

Example:

More autonomy in terms of taking work-related decisions helps to improve an employee’s overall performance.

You can also state a hypothesis as a comparison between two groups.

Example:

Employees who are offered more autonomy to take work-related decisions show better overall performance than those who work in a micro-managed environment.

Write a Null Hypothesis

The next step is to frame a null hypothesis, especially if your study requires you to analyze the data statistically. A null hypothesis by default takes a converse position to the researcher’s hypothesis.

Your statement is known as the alternative hypothesis while its opposite outcome is referred to as the null hypothesis.

If you expect a change according to a relationship between the variables the null hypothesis denies the possibility of any change or association between the variables. If you expect the conditions to remain constant the null hypothesis states that change will take place.

The null hypothesis is referred to as H_0. Your hypothesis which is the alternative is written as H₁ or H_a.

Example:

H₁: A player who is more than two meters tall has a better chance of winning the National Basketball Association Most Valuable Player Award.

H₀: The height of a player does not affect his prospects of winning the National Basketball Association Most Valuable Player Award.

Hypothesis Examples

Examples of Research Questions

Which loop diuretic drug is more effective for treating heart failure?
Does attending online learning sessions help students to improve their exam scores?
Does talking on the phone while driving cause more accidents?
Does increasing the pressure affect the rate of reaction between gases?
Is a person more likely to be obese if she or he eats unhealthy foods at least four times in a week?

Examples of a Hypothesis

The clinical trial of the new drug Furosemide proved that it is better at treating heart failure than other loop diuretic drugs such as Bumetanide.
The students who attended online learning sessions had better exam scores than those who skipped the sessions.
Drivers who talk on the phone are likely to have an accident than those who don’t.
Increasing the pressure affects the concentration of gases and it acts as a catalyst in speeding up the rate of reaction.
People who eat processed foods frequently are more likely to be obese than people who limit their intake of such foods.

Examples of a Null Hypothesis

The clinical trial proved that there is no difference between the effectiveness of Furosemide and other loop diuretic drugs, such as Bumetanide, for treating heart failure.
There is no difference in the exam scores of students who attended online learning sessions and those who did not attend.
There is no difference in the rate of accidents experienced by drivers who talk on the phone compared with those who don’t talk on the phone while driving.
The elevation of pressure has no effect on the rate of reaction between gases.
The food consumed and its frequency of consumption do not affect the probability of a person becoming obese.

What are Null Hypotheses?

The null hypothesis states the opposite outcome to the researcher’s hypothesis.

In most cases, the null hypothesis’s default position is a prediction that no relationship exists between any two or more variables. The null hypothesis denies the possibility of a causal relationship existing between an assumed independent and dependent variable.

The symbol of the null hypothesis is H₀.

The notion of a null hypothesis fulfills the requirement of the falsifiability of a hypothesis before it can be accepted as valid.

A null hypothesis is often written as a negative statement that posits that the original hypothesis is false. It either claims that the results obtained are due to chance or there is no evidence to prove any change.

Example:

Original Hypothesis: Use of nitrogen fertilizers helps plants grow faster as compared to use of phosphorus or potassium fertilizers.

Null Hypothesis (H₀): The fertilizer used has no bearing on the rate of plant growth

What are Alternative Hypotheses?

An alternative hypothesis states the researcher’s supposition of a causal relationship between any two or more variables. Alternative hypotheses are based upon an observable effect and seek to predict how changing an independent variable will affect the dependent variable.

An alternative hypothesis is symbolized as H₁ or H_a. It’s often written together with a null hypothesis with the two statements existing as a dual pair of opposite assumptions. Only a single statement among two can be true.

Alternative hypotheses try to determine that the results are obtained due to significant changes related to the variables and not due to chance.

Example:

Research Question: Does washing hands thoroughly with soap before eating a meal reduce the rate of recurrence of respiratory ailments?

Alternative Hypothesis (H₁): Washing hands with soap before eating reduces the rate of recurrence of respiratory ailments by 30% compared with those who neglect hand hygiene.

Null Hypothesis (H₀): Washing hands with soap before eating has no effect on the rate of recurrence of respiratory ailments.

What is Hypothesis Testing?

After you have formulated a hypothesis, you need to choose a research and testing method.

Use a descriptive approach when experiments are difficult to conduct. A descriptive method incorporates case studies and surveys to collect data.

You can employ statistical tools such as a correlational study to measure the relationship between variables.

A correlational study calculates the probability of whether a linkage between two variables can be determined or do the changes occur purely due to chance. Do note that correlation is not equivalent to causality.

This method lets you arrive at a conclusion by generalizing the data obtained without performing any actual experiments. A hypothesis proved using this approach is known as a statistical hypothesis.

The other approach is the experimental method in which causal relationships are established between different variables through demonstrations. A working or empirical hypothesis often makes use of the experimental method to determine the relationships between the variables.

The steps for testing a hypothesis experimentally are:

Design of experiments
Collating data
Analysis of observable facts
Summarizing the conclusions
Validating the hypothesis as a theory

How to Write a Good Hypothesis

To find ideas for a hypothesis, you can look through discussion sections in academic and scientific journals or browse online publications. You will come across questions that can be investigated further.

Simple Steps

The steps to write a strong hypothesis are:

Choose your frame of reference or direction for determining the cause
Such an approach is known as a directional hypothesis
If you are unable to determine a starting point or the current theories are ambiguous and contradictory, you can choose a non-directional approach
This method involves stating the facts and observations randomly and then seeking to find a pattern
Identify the key variables
A variable is any attribute that can have measurable values such as temperature, time, or length
Tentatively label some variables as independent and some as dependent
State the relationship between the variables using clear and objective language
Operationalize or define how you will measure the variables for testability
Write the statement in the If-then format. You can also write it as a declarative sentence
Avoid jargon and use simple words that can be understood by a layman
Write a null hypothesis to satisfy the condition of falsifiability

Example:

If you watch television for more than three hours a day, then your ability to concentrate diminishes.

How to Write a Scientific Hypothesis

A good scientific hypothesis is:

Consistent: Use preexisting knowledge as a springboard for further research
Testable: Include words that are quantifiable or measurable
Concise: Cut down on verbose phrases and use precise words
Scalable: Formulate the statement in a universal context based on the variables
Promising: State unexplained occurrences as loose ends that can be investigated further

Simple steps

Record your observations and facts about the topic
Evaluate your statements for possible links to determine the cause and effect
Document all potential explanations to analyze further
Write the null hypothesis along with your own hypothesis
This satisfies the requisite condition for a valid hypothesis. It can either be confirmed or disproved

Example:

If you plant cotton in black soil, then the production is boosted by 20% as compared to the output from red soil.

How to write a Psychology Hypothesis

A psychology hypothesis often begins with how the environment or certain parameters within it influence or cause a specific behavior.

Simple steps

To write a sound psychology hypothesis:

Choose a topic that you are genuinely interested in
Do not ramble. Keep it short and simple
Use previous research and your own study to direct your vision
Ascertain and define the variables
You can write the hypothesis either as an If-then statement
Other alternatives are to write the hypothesis as a direct sentence or a comparative supposition

Use the following questions to guide your understanding of the topic.

Is your hypothesis based on a preexisting theory or your own research?
Can your hypothesis be tested for falsifiability?
What are the independent and dependent variables?

Example:

People who exercise regularly are less at risk from depression than people who lead a sedentary life.

Sources

Tags:

Hypothesis

Fact checked:
Content is rigorously reviewed by a team of qualified and experienced fact checkers. Fact checkers review articles for factual accuracy, relevance, and timeliness. Learn more.

About the author

Dalia Y.: Dalia is an English Major and linguistics expert with an additional degree in Psychology. Dalia has featured articles on Forbes, Inc, Fast Company, Grammarly, and many more. She covers English, ESL, and all things grammar on GrammarBrain.