A Guide to Hypothesis vs. Prediction (With Examples)

By Indeed Editorial Team

Published 8 May 2022

The Indeed Editorial Team comprises a diverse and talented team of writers, researchers and subject matter experts equipped with Indeed's data and insights to deliver useful tips to help guide your career journey.

Hypothesis and prediction can be useful tools when conducting experiments. Scientists use them to choose what objectives to measure when conducting research. Understanding the differences between hypothesis and prediction can help you identify the essential variables in your work and what to analyse in your tests. In this article, we define hypothesis and prediction, list their different types, explain how they differ and provide examples that can help you to understand the two testing tools clearly.

Definitions of hypothesis vs. prediction

When reviewing hypothesis vs. prediction, you can first understand the meaning of those terms:

What's a hypothesis?

A hypothesis can be an explanation you make based on limited observation. It can provide the objective for starting an investigation. Since you may establish a theory on minimal evidence and facts, there's a chance that it might be incorrect. Thus, hypotheses require tests to verify their truth. For example, researchers can create hypotheses before conducting research in a particular area. They may base the assumption on observations that are currently unaccounted for by existing scientific theories.

Related: Research Skills: Definition, Examples and Importance

What's a prediction?

You can define a prediction as an estimate or a forecast of an event in the future. An individual might base predictions on observations made at a certain time. For example, if it has rained every evening in the last five days, you might predict that it may rain again the following day at the same time. You can use predictions to test the outcomes of a hypothesis.

Types of hypothesis

You can categorise a hypothesis in the following groups:

Simple hypothesis

A simple hypothesis finds the relationship between two variables. It involves one dependent and one independent variable. Some examples of simple hypotheses are:

  • Taking plenty of water leads to weight loss.

  • Doing exercise reduces health problems.

  • Drinking water helps in improving digestion.

Complex hypothesis

A complex hypothesis describes the connection between multiple variables. It involves two or more independent variables and at least two dependent variables. Some examples of complex hypotheses include:

  • Individuals who take a balanced diet and have eight hours of sleep are likely to be alert during the day, and may live longer.

  • Adults who live in large cities and own cars are likely to enjoy shopping and may frequently visit restaurants.

  • People who get out of bed early are more likely to perform better at work, and may volunteer for tasks compared to those who don't.

Null hypothesis

A null hypothesis proposes that there's little relation between two variables. It assumes that the difference you observe between the variables is due to chance. The null statement includes a no. You can use a statistical test to verify whether a null hypothesis is true. Some examples of null hypotheses are:

  • There's no change in a plant's growth rate if you build it a shade.

  • There's no increase in productivity if you take three cups of coffee.

  • There's no gravity in space.

Alternative hypothesis

An alternative hypothesis is a claim that contradicts the null hypothesis. Scientists may pair the null and alternative hypotheses to prove a little connection between the variables. If the experiment disproves the null theory, it accepts the alternative hypothesis. Some examples of alternative hypotheses are:

  • A plant's growth rate changes if you build it a shade.

  • Taking three cups of coffee increases productivity.

  • If you're aged 48 and below, then no one has gone to the moon in your lifetime.

Related: What Does a Research Scientist Do? (With Skills and Salary)

Types of predictions

You can categorise predictions into the following:

Inductive prediction

You can create an inductive prediction by projecting past events into the future. For example, if today was sunny and the last 10 days were sunny, tomorrow is likely to be sunny. This method disregards the cause of the event. You can predict what the outcome might be without knowing what influences the observed patterns. In the weather prediction example, you might ignore what causes the day to be sunny.

Deductive prediction

This prediction relies on the consequence that follows a hypothesis. For example, if you're looking for a friend at a party but can't find them, you can predict that they may be at the party if their car is present. This type of prediction tests the hypothesis, which can inform the outcome of a future event. For example, if you see your friend's car, the future result might be that you get to meet them later on during the party.

Related: Inductive vs. Deductive Reasoning: Differences and How to Improve

Abductive prediction

Abductive reasoning relies on understanding the subject before making a prediction. Instead of considering past observations only to estimate a future event, abductive prediction might involve learning about why the specific event occurs. For example, if a person drives fast and often exceeds the speed limit, you may predict they might cause an accident. Using inductive prediction, you also analyse why accidents happen before making the prediction.

Related: What Are Analytical Skills and Why Are They Important for Employment?

Differences between hypothesis vs. prediction

Some differences between a hypothesis and prediction include:

Expression

Researchers may write a hypothesis as a statement with specific variables. For example, the hypothesis can be drinking coffee before sleeping leads to loss of sleep. The variables are either independent or dependent. Dependent variables are the outcomes you can measure, such as loss of sleep. Independent variables are those the researcher controls, like asking the participants to take coffee before sleeping. A hypothesis can aim to show a clear relationship between the two variables. The researchers alter the independent variable by different amounts to help them understand this relationship.

Typically, you can express a prediction as an If-Then statement. For example, if you take coffee before sleeping, then you lose your sleep. The if section of the message states a true condition. If the experiment meets this condition, it can give the outcome the hypothesis investigates.

Formation

Hypothesis and predictions can differ in how researchers create them. You can generate a hypothesis by asking a question and having a wish to understand the relationship between the variables in your inquiry. For example, How does taking breakfast affect a person's productivity? is a question with two variables–taking breakfast and productivity. A researcher can create the hypothesis that taking breakfast increases a person's productivity. The scientist can test this hypothesis by controlling a group of participants to either accept or miss breakfast.

A prediction may arrive from observations you notice instead of a question. For example, you may note that you get less work completed when you miss breakfast. From this observation, you may predict that if you take breakfast, then your productivity increases.

Aim

Hypothesis and predictions differ in the objectives of a research experiment. A thesis provides a claim that your experiment can verify. The study you conduct can either approve or disapprove of the hypothesis, but identifies a clear outcome. Predictions use current observations and findings to conclude. They focus less on finding the relationship between variables than a hypothesis does. Predictions may only aim at estimating a future outcome.

Examples of hypotheses and predictions

Here are some examples to help you better understand hypotheses and predictions:

Diet and sleepiness

You may notice that your diet has resulted in you falling asleep when working. Experimenting can let you identify the connection between your diet and sleep. The independent variable can be the diet, while sleep is the dependent variable. You can change your diet and record how it affects your sleeping pattern. This process can help you test your hypothesis.

  • Hypothesis: Taking sugar makes one fall asleep during the day.

  • Prediction: If I avoid sugar, then I stop falling asleep during the day.

Ice cream sales and weather

An ice cream seller makes more money on a Saturday than on Sunday. Saturday was a sunny day, while it rained on Sunday. The seller may experiment to verify the effect of weather on people's willingness to buy ice cream. The weather is the independent variable, while the number of ice cream people purchase is the dependent variable. Since you can't control the weather, you can record the weather for each day and the number of successful sales as the vendor. The hypothesis and prediction can be:

  • Hypothesis: The number of ice cream sales is high during sunny weather.

  • Prediction: If tomorrow is sunny, then many people may buy ice cream.

Questions and customer complaints

A customer representative notices that if an answer to a customer question is in the Frequently Asked Questions (FAQ) section, fewer customers call to ask the question. The representative might investigate whether adding a question to the FAQs affects the number of calls about the question. Placing the question in the FAQs is the independent variable, while the number of calls related to a question is a dependent variable. This experiment may use the following prediction and hypothesis:

  • Hypothesis: A question in the FAQs decreases the number of calls customers have about the question.

  • Prediction: If the company adds a common question to the FAQs, then fewer customers call to ask the question.


Explore more articles