Hypothesis Definition
According to the definition, a hypothesis is an assumption one makes based on existing knowledge. To elaborate, it is a statement that translates the initial research question into a logical prediction shaped on the basis of available facts and evidence. To solve a specific problem, one first needs to identify the research problem (research question), conduct initial research, and set out to answer the given question by performing experiments and observing their outcomes. However, before one can move to the experimental part of the research, they should first identify what they expect to see for results. At this stage, a scientist makes an educated guess and writes a hypothesis that he or she is going to prove or refute in the course of their study.
A hypothesis can also be seen as a form of development of knowledge. It is a well-grounded assumption put forward to clarify the properties and causes of the phenomena being studied.
As a rule, a hypothesis is formed based on a number of observations and examples that confirm it. This way, it looks plausible as it is backed up with some known information. The hypothesis is subsequently proved by turning it into an established fact or refuted (for example, by pointing out a counterexample), which allows it to attribute it to the category of false statements.
As a student, you may be asked to create a hypothesis statement as a part of your academic papers. Hypothesis-based approaches are commonly used among scientific academic works, including but not limited to research papers, theses, and dissertations.
Note that in some disciplines, a hypothesis statement is called a thesis statement. However, its essence and purpose remain unchanged – this statement aims to make an assumption regarding the outcomes of the investigation that will either be proved or refuted.
Characteristics and Sources of a Hypothesis
Now, as you know what a hypothesis is in a nutshell, let’s look at the key characteristics that define it:
- It has to be clear and accurate in order to look reliable.
- It has to be specific.
- There should be scope for further investigation and experiments.
- A hypothesis should be explained in simple language—while retaining its significance.
- If you are making a relational hypothesis, two essential elements you have to include are variables and the relationship between them.
The main sources of a hypothesis are:
- Scientific theories.
- Observations from previous studies and current experiences.
- The resemblance among different phenomena.
- General patterns that affect people’s thinking process.
Types of Hypothesis
Basically, there are two major types of scientific hypothesis: alternative and null.
- Alternative Hypothesis
This type of hypothesis is generally denoted as H1. This statement is used to identify the expected outcome of your research. According to the alternative hypothesis definition, this type of hypothesis can be further divided into two subcategories:
- Directional — a statement that explains the direction of the expected outcomes. Sometimes this type of hypothesis is used to study the relationship between variables rather than comparing between the groups.
- Non-directional — unlike the directional alternative hypothesis, a non-directional one does not imply a specific direction of the expected outcomes.
Now, let’s see an alternative hypothesis example for each type:
Directional: Attending more lectures will result in improved test scores among students.
Non-directional: Lecture attendance will influence test scores among students.
Notice how in the directional hypothesis we specified that the attendance of more lectures will boost student’s performance on tests, whereas in the non-directional hypothesis we only stated that there is a relationship between the two variables (i.e. lecture attendance and students’ test scores) but did not specify whether the performance will improve or decrease.
- Null Hypothesis
This type of hypothesis is generally denoted as H0. This statement is the complete opposite of what you expect or predict will happen throughout the course of your study—meaning it is the opposite of your alternative hypothesis. Simply put, a null hypothesis claims that there is no exact or actual correlation between the variables defined in the hypothesis.
To give you a better idea of how to write a null hypothesis, here is a clear example:
Lecture attendance has no effect on student’s test scores.
Both of these types of hypotheses provide specific clarifications and restatements of the research problem. The main difference between these hypotheses and a research problem is that the latter is just a question that can’t be tested, whereas hypotheses can.
Based on the alternative and null hypothesis examples provided earlier, we can conclude that the importance and main purpose of these hypotheses are that they deliver a rough description of the subject matter. The main purpose of these statements is to give an investigator a specific guess that can be directly tested in a study. Simply put, a hypothesis outlines the framework, scope, and direction for the study. Although null and alternative hypotheses are the major types, there are also a few more to keep in mind:
Research Hypothesis — a statement that is used to test the correlation between two or more variables.
For example: Eating vitamin-rich foods affects human health.
Simple Hypothesis — a statement used to indicate the correlation between one independent and one dependent variable.
For example: Eating more vegetables leads to better immunity.
Complex Hypothesis — a statement used to indicate the correlation between two or more independent variables and two or more dependent variables.
For example: Eating more fruits and vegetables leads to better immunity, weight loss, and lower risk of diseases.
Associative and Causal Hypothesis — an associative hypothesis is a statement used to indicate the correlation between variables under the scenario when a change in one variable inevitably changes the other variable. A causal hypothesis is a statement that highlights the cause and effect relationship between variables.
Hypothesis vs Prediction
When speaking of hypotheses, another term that comes to mind is prediction. These two terms are often used interchangeably, which can be rather confusing. Although both a hypothesis and prediction can generally be defined as “guesses” and can be easy to confuse, these terms are different. The main difference between a hypothesis and a prediction is that the first is predominantly used in science, while the latter is most often used outside of science.
Simply put, a hypothesis is an intelligent assumption. It is a guess made regarding the nature of the unknown (or less known) phenomena based on existing knowledge, studies, and/or series of experiments, and is otherwise grounded by valid facts. The main purpose of a hypothesis is to use available facts to create a logical relationship between variables in order to provide a more precise scientific explanation. Additionally, hypotheses are statements that can be tested with further experiments. It is an assumption you make regarding the flow and outcome(s) of your research study.
A prediction, on the contrary, is a guess that often lacks grounding. Although, in theory, a prediction can be scientific, in most cases it is rather fictional—i.e. a pure guess that is not based on current knowledge and/or facts. As a rule, predictions are linked to foretelling events that may or may not occur in the future. Often, a person who makes predictions has little or no actual knowledge of the subject matter he or she makes the assumption about.
Another big difference between these terms is in the methodology used to prove each of them. A prediction can only be proven once. You can determine whether it is right or wrong only upon the occurrence or non-occurrence of the predicted event. A hypothesis, on the other hand, offers scope for further testing and experiments. Additionally, a hypothesis can be proven in multiple stages. This basically means that a single hypothesis can be proven or refuted numerous times by different scientists who use different scientific tools and methods.
To give you a better idea of how a hypothesis is different from a prediction, let’s look at the following examples:
Hypothesis: If I eat more vegetables and fruits, then I will lose weight faster.
This is a hypothesis because it is based on generally available knowledge (i.e. fruits and vegetables include fewer calories compared to other foods) and past experiences (i.e. people who give preference to healthier foods like fruits and vegetables are losing weight easier). It is still a guess, but it is based on facts and can be tested with an experiment.
Prediction: The end of the world will occur in 2023.
This is a prediction because it foretells future events. However, this assumption is fictional as it doesn’t have any actual grounded evidence supported by facts.
Based on everything that was said earlier and our examples, we can highlight the following key takeaways:
- A hypothesis, unlike a prediction, is a more intelligent assumption based on facts.
- Hypotheses define existing variables and analyze the relationship(s) between them.
- Predictions are most often fictional and lack grounding.
- A prediction is most often used to foretell events in the future.
- A prediction can only be proven once – when the predicted event occurs or doesn’t occur.
- A hypothesis can remain a hypothesis even if one scientist has already proven or disproven it. Other scientists in the future can obtain a different result using other methods and tools.
How to Write a Hypothesis
Now, as you know what a hypothesis is, what types of it exist, and how it differs from a prediction, you are probably wondering how to state a hypothesis. In this section, we will guide you through the main stages of writing a good hypothesis and provide handy tips and examples to help you overcome this challenge:
1. Define Your Research Question
Here is one thing to keep in mind – regardless of the paper or project you are working on, the process should always start with asking the right research question. A perfect research question should be specific, clear, focused (meaning not too broad), and manageable.
Example: How does eating fruits and vegetables affect human health?
2. Conduct Your Basic Initial Research
As you already know, a hypothesis is an educated guess of the expected results and outcomes of an investigation. Thus, it is vital to collect some information before you can make this assumption.
At this stage, you should find an answer to your research question based on what has already been discovered. Search for facts, past studies, theories, etc. Based on the collected information, you should be able to make a logical and intelligent guess.
3. Formulate a Hypothesis
Based on the initial research, you should have a certain idea of what you may find throughout the course of your research. Use this knowledge to shape a clear and concise hypothesis.
Based on the type of project you are working on, and the type of hypothesis you are planning to use, you can restate your hypothesis in several different ways:
Non-directional: Eating fruits and vegetables will affect one’s human physical health.
Directional: Eating fruits and vegetables will positively affect one’s human physical health.
Null: Eating fruits and vegetables will have no effect on one’s human physical health.
4. Refine Your Hypothesis
Finally, the last stage of creating a good hypothesis is refining what you’ve got. During this step, you need to define whether your hypothesis:
- Has clear and relevant variables;
- Identifies the relationship between its variables;
- Is specific and testable;
- Suggests a predicted result of the investigation or experiment.
Hypothesis Examples
Following a step-by-step guide and tips from this article, you should be able to create good hypotheses with ease. To give you a starting point, we have also compiled a list of different research questions with one hypothesis and one null hypothesis example for each:
