Sign In
50 Hypothesis Examples Across Every Research Field: Psychology, Biology, Education, Medicine, and More
A hypothesis is a testable prediction about the relationship between two or more variables. Most quantitative research papers state one. This guide gives 50 hypothesis examples organized by research field. Each example names the independent variable, the dependent variable, and what makes the hypothesis testable. Jump to your field, find a model close to your topic, and use it to structure your own.
Quick Answer: What Makes a Strong Hypothesis
A strong hypothesis is testable (a study can support or refute it), falsifiable (some possible result could prove it wrong), specific (variables are named and the predicted direction is clear), and grounded in prior research (it flows from existing theory or findings).
Weak: "Exercise affects health."
Strong: "Adults with hypertension who engage in 30 minutes of moderate aerobic exercise five days per week show a greater reduction in systolic blood pressure after 12 weeks than adults who maintain a sedentary lifestyle."
Jump to a Section
Hypothesis Examples in Psychology
1. Sleep and Memory
Hypothesis: Adults who sleep fewer than six hours per night perform worse on short-term memory tasks than adults who sleep seven to nine hours, as measured by the Digit Span Backward test.
2. Social Media and Self-Esteem
Hypothesis: Adolescents who spend more than three hours per day on social media report lower self-esteem scores on the Rosenberg Self-Esteem Scale than adolescents who spend fewer than one hour per day.
3. Cognitive Behavioral Therapy and Anxiety
Hypothesis: Participants who complete a 12-week cognitive behavioral therapy program report significantly lower generalized anxiety scores at follow-up than participants in a waitlist control group, as measured by the GAD-7.
4. Stress and Decision-Making
Hypothesis: Individuals under high occupational stress make more impulsive financial decisions than individuals under low occupational stress, as measured by a standardized risk tolerance instrument.
5. Childhood Adversity and Adult Resilience
Hypothesis: Adults who experienced high levels of childhood adversity but had at least one stable, supportive caregiver report higher resilience scores than adults who experienced similar adversity without a supportive caregiver.
Hypothesis Examples in Education
6. Class Size and Academic Performance
Hypothesis: Students in classes of fewer than 20 students score higher on end-of-year standardized assessments than students in classes of 30 or more students, controlling for socioeconomic status.
7. Parental Involvement and Reading Levels
Hypothesis: Children whose parents read to them at least five times per week score at a higher reading level by age seven than children whose parents read to them fewer than twice per week.
8. Teacher Feedback and Writing Quality
Hypothesis: Students who receive specific, written feedback on their essays improve their writing scores more over a semester than students who receive only grades without written feedback.
9. Technology in the Classroom
Hypothesis: Middle school students who use tablet-based learning tools for mathematics instruction score higher on math assessments than students using traditional textbooks, after controlling for prior math ability.
10. Bilingual Education and Cognitive Flexibility
Hypothesis: Students enrolled in dual-language immersion programs perform better on cognitive flexibility tasks than students in monolingual programs at the same grade level.
Hypothesis Examples in Medicine and Public Health
11. Exercise and Blood Pressure
Hypothesis: Adults with hypertension who engage in 30 minutes of moderate aerobic exercise five days per week show a significantly greater reduction in systolic blood pressure after 12 weeks than adults who maintain a sedentary lifestyle.
12. Diet and Inflammation
Hypothesis: Patients who follow a Mediterranean diet for six months show lower levels of C-reactive protein, a biomarker of inflammation, than patients who follow a standard Western diet.
13. Vaccination Rates and Herd Immunity
Hypothesis: Communities where more than 85% of the population is vaccinated against measles experience significantly lower measles outbreak rates than communities where vaccination rates fall below 75%.
14. Nurse-to-Patient Ratios and Recovery Times
Hypothesis: Patients in hospital wards with a nurse-to-patient ratio of 1:4 or lower have shorter average recovery times than patients in wards with a ratio of 1:8 or higher, after controlling for diagnosis severity.
15. Sleep Deprivation and Immune Function
Hypothesis: Adults who sleep fewer than six hours per night for four consecutive weeks show lower antibody responses to influenza vaccination than adults who sleep seven to nine hours per night.
Submitting a research paper or dissertation?
Choose an editor with experience in your field. Every edit is done by a native English editor from the United States, the United Kingdom, or Canada, with no AI at any stage. Your first 300 words are edited at no cost.
Choose Your EditorHypothesis Examples in Biology and Life Sciences
16. Plant Growth and Light Exposure
Hypothesis: Tomato plants exposed to 16 hours of light per day produce more fruit per plant over a 90-day growing period than tomato plants exposed to 8 hours of light per day under otherwise identical conditions.
17. Antibiotic Resistance
Hypothesis: Bacterial cultures exposed to sub-therapeutic doses of ampicillin for 14 days develop resistance to ampicillin at higher rates than cultures not exposed to the antibiotic.
18. Temperature and Enzyme Activity
Hypothesis: The enzyme amylase shows peak activity at 37 degrees Celsius and significantly reduced activity at temperatures above 50 degrees Celsius or below 10 degrees Celsius.
19. Habitat Fragmentation and Species Diversity
Hypothesis: Forest fragments smaller than 10 hectares support fewer native bird species than forest fragments of 100 hectares or more in the same geographic region.
20. Genetic Variation and Disease Susceptibility
Hypothesis: Individuals carrying the APOE4 allele have a significantly higher lifetime risk of developing Alzheimer's disease than individuals with the APOE3 allele, independent of age and sex.
Hypothesis Examples in Economics and Finance
21. Income Inequality and Social Mobility
Hypothesis: Countries with higher Gini coefficients show lower rates of intergenerational income mobility than countries with lower Gini coefficients, after controlling for GDP per capita.
22. Minimum Wage and Employment
Hypothesis: A 10% increase in the minimum wage is associated with a statistically significant change in employment among workers aged 16 to 24 in small retail businesses.
23. Gender and Financial Risk Tolerance
Hypothesis: Gender differences in financial risk tolerance are explained by differences in income uncertainty and net worth between men and women, rather than by gender itself.
24. Interest Rates and Consumer Spending
Hypothesis: A one percentage point increase in the federal funds rate is associated with a measurable decrease in consumer spending on durable goods within two quarters.
25. Financial Literacy and Retirement Savings
Hypothesis: Adults who score in the top quartile on a standardized financial literacy assessment accumulate significantly higher retirement savings by age 65 than adults who score in the bottom quartile, controlling for income.
Hypothesis Examples in Sociology
26. Social Networks and Mental Health
Hypothesis: Adults with three or more close social relationships report lower rates of depression and anxiety than adults who report having fewer than two close social relationships, as measured by the PHQ-9 and GAD-7.
27. Neighborhood Environment and Educational Attainment
Hypothesis: Students raised in neighborhoods with high concentrations of poverty complete fewer years of formal education than students raised in mixed-income neighborhoods, after controlling for individual family income.
28. Gender and Unpaid Labor
Hypothesis: Women in dual-income heterosexual households perform significantly more unpaid domestic labor per week than their male partners, even when both partners work equivalent paid hours.
29. Religious Participation and Civic Engagement
Hypothesis: Adults who attend religious services at least once per week report higher rates of volunteer activity and charitable giving than adults who attend rarely or never.
30. Urban Density and Social Trust
Hypothesis: Residents of high-density urban neighborhoods report different levels of generalized social trust than residents of low-density suburban neighborhoods in the same metropolitan area.
Hypothesis Examples in Environmental Science
31. Air Pollution and Respiratory Health
Hypothesis: Children living within one mile of a major highway show higher rates of asthma diagnosis than children living more than three miles from a major highway in the same city.
32. Ocean Acidification and Coral Bleaching
Hypothesis: Coral reefs exposed to water with a pH below 7.9 show higher rates of bleaching over a 12-month period than coral reefs in water with a pH above 8.1 under otherwise comparable conditions.
33. Green Space and Urban Temperature
Hypothesis: Urban neighborhoods with more than 20% green space coverage experience lower average summer temperatures than neighborhoods with less than 5% green space coverage, controlling for building density.
34. Plastic Pollution and Marine Biodiversity
Hypothesis: Marine sampling sites with high concentrations of microplastic particles support fewer species of zooplankton than sites with low concentrations in the same ocean region.
35. Deforestation and Rainfall Patterns
Hypothesis: Regions that experienced more than 30% deforestation between 1990 and 2010 show measurably reduced annual rainfall compared to regions with less than 5% deforestation in the same climate zone.
Hypothesis Examples in Political Science
36. Voter Turnout and Electoral Competitiveness
Hypothesis: Voter turnout in state legislative elections is significantly higher in districts where the margin of victory in the previous election was less than 5% than in districts where the margin exceeded 20%.
37. Media Exposure and Political Polarization
Hypothesis: Adults who consume news primarily from partisan sources report more extreme political attitudes than adults who consume news from a variety of ideologically diverse sources, as measured by a standardized ideological consistency scale.
38. Campaign Spending and Electoral Outcomes
Hypothesis: Incumbent candidates who outspend their challengers by a ratio of 3:1 or more win re-election at significantly higher rates than incumbents who are outspent or spend comparably.
Hypothesis Examples in Business and Management
39. Employee Autonomy and Job Satisfaction
Hypothesis: Employees who report high levels of autonomy in their daily work report higher job satisfaction scores than employees who report low autonomy, controlling for compensation and job level.
40. Remote Work and Job Satisfaction
Hypothesis: Knowledge workers who work remotely full-time report different levels of job satisfaction than comparable workers in full-time office environments, as measured by a standardized job satisfaction scale.
41. Diversity and Team Performance
Hypothesis: Project teams with gender and ethnic diversity scores in the top quartile of their organization deliver projects on time and within budget at higher rates than teams in the bottom quartile.
Hypothesis Examples in Nursing and Healthcare
42. Patient Education and Medication Adherence
Hypothesis: Patients who receive structured medication education from a nurse at discharge show higher medication adherence rates at 30 days than patients who receive standard written discharge instructions only.
43. Hand Hygiene Compliance and Infection Rates
Hypothesis: Hospital wards that implement structured hand hygiene monitoring programs show lower rates of healthcare-associated infection than wards without such programs over a 12-month period.
44. Mindfulness Training and Nurse Burnout
Hypothesis: Nurses who complete an eight-week mindfulness-based stress reduction program report lower burnout scores at follow-up than nurses in a waitlist control group, as measured by the Maslach Burnout Inventory.
Hypothesis Examples for High School and Science Fair Projects
These four examples use materials a school lab or kitchen can supply. Each one also works as a science fair hypothesis. The if-then versions below show how to rewrite them in the format many teachers ask for.
45. Sugar and Energy Levels
Hypothesis: Students who consume a high-sugar breakfast report lower self-rated energy levels two hours after eating than students who consume a high-protein breakfast.
46. Music and Study Performance
Hypothesis: Students who study with lyrical music in the background score lower on a subsequent recall test than students who study in silence or with instrumental music.
47. Exercise and Mood
Hypothesis: Participants who engage in 20 minutes of aerobic exercise report higher self-rated mood scores immediately after exercise than participants who rest for 20 minutes.
48. Plant Growth and Fertilizer Type
Hypothesis: Bean plants treated with organic fertilizer grow taller over a 30-day period than bean plants treated with synthetic fertilizer under identical light and water conditions.
Hypothesis Examples in Linguistics and Communication
49. Language of Instruction and Comprehension
Hypothesis: Students who receive mathematics instruction in their first language score higher on mathematics assessments than students who receive equivalent instruction in a second language, at equivalent second-language proficiency levels.
50. Email Length and Response Rate
Hypothesis: Professional emails of fewer than 100 words receive responses within 24 hours at higher rates than professional emails of more than 300 words sent to the same population of recipients.
How to Write Your Own Hypothesis
Follow these five steps to write a clear, testable hypothesis for your research paper.
Step 1: Start with a research question
Every hypothesis begins as a question. Start with something specific.
- Too broad: Does exercise affect health?
- Better: Does 30 minutes of aerobic exercise five days per week reduce systolic blood pressure in adults with mild hypertension?
Step 2: Review the existing literature
Your hypothesis should be grounded in what's already known. Read the relevant research. Identify what has been established and what remains unclear. Your hypothesis addresses the gap.
Step 3: Identify your variables
Name your independent variable (what you manipulate or observe) and your dependent variable (what you measure).
- Independent variable: Hours of sleep per night.
- Dependent variable: Score on a standardized memory test.
Step 4: Write a directional prediction
State the expected direction of the relationship with words such as "more than," "less than," "higher," "lower," or "greater." Use a non-directional prediction ("differs from," "is associated with a change in") only when the published literature is genuinely mixed.
- Weak: Sleep affects memory test scores.
- Strong: Adults who sleep seven to nine hours per night score significantly higher on standardized memory tests than adults who sleep fewer than six hours.
Step 5: Check that it's testable and falsifiable
Ask yourself: could a study disprove this? If no possible result could show the hypothesis to be wrong, rewrite it. A hypothesis that can't be falsified isn't a scientific hypothesis. For where this step sits in the larger research process, see our guide to what the scientific method is and how it works.
If-Then Hypothesis Examples
An if-then hypothesis names the change you'll make (the "if" clause) and the result you expect (the "then" clause). Science fair guides and introductory lab courses often ask for this format. It works best for experiments, where you control the independent variable directly. Here are four examples from this page rewritten in if-then form.
- Plant growth (#16): If tomato plants get 16 hours of light per day instead of 8, then they will produce more fruit over 90 days.
- Fertilizer (#48): If bean plants get organic fertilizer instead of synthetic fertilizer, then they will grow taller over 30 days.
- Study music (#46): If students study in silence instead of with lyrical music, then they will score higher on a recall test.
- Enzymes (#18): If amylase is heated above 50 degrees Celsius, then its activity will drop well below its rate at 37 degrees.
Some teachers ask for an if-then-because version, which adds the reason behind the prediction. For example: "If tomato plants get 16 hours of light, then they will produce more fruit, because more light allows more photosynthesis." For observational studies, where you can't assign the condition, the standard statement form used in the 50 examples reads more naturally.
Types of Hypotheses
Knowing the main types of hypotheses helps you choose the right structure for your study.
| Type | What it predicts | Example |
|---|---|---|
| Simple | A relationship between one IV and one DV | "Students who sleep 8+ hours score higher on tests than students who sleep under 6 hours." |
| Complex | A relationship involving multiple IVs or DVs | "Students who sleep 8+ hours and eat breakfast daily score higher and report lower stress than students who do neither." |
| Null (H₀) | No relationship between the variables | "There is no significant difference in test scores between sleep groups." |
| Alternative (H₁) | A relationship exists | "Students who sleep 8+ hours score significantly higher than students who sleep under 6 hours." |
| Directional | The direction of the relationship (higher, lower) | "Increased screen time is associated with higher reported anxiety in adults aged 18 to 35." |
| Non-directional | A relationship exists, direction unspecified | "There is a relationship between screen time and anxiety in adults aged 18 to 35." |
Null and Alternative Hypothesis Examples
Any research hypothesis on this page can be restated as a null and alternative pair for statistical testing. The null hypothesis (H₀) says there's no effect or no difference. The alternative hypothesis (H₁) states the effect you predicted. Your statistical test shows whether the data give enough evidence to reject H₀.
| Example | Null hypothesis (H₀) | Alternative hypothesis (H₁) |
|---|---|---|
| Sleep and memory (#1) | Digit Span Backward scores don't differ between adults who sleep under six hours and adults who sleep seven to nine hours. | Adults who sleep under six hours score lower on the Digit Span Backward test than adults who sleep seven to nine hours. |
| Exercise and blood pressure (#11) | The drop in systolic blood pressure after 12 weeks is the same for exercising and sedentary adults with hypertension. | Exercising adults with hypertension show a larger drop in systolic blood pressure after 12 weeks than sedentary adults. |
| Class size (#6) | Assessment scores don't differ between small and large classes, controlling for socioeconomic status. | Students in small classes score higher on assessments than students in large classes, controlling for socioeconomic status. |
| Antibiotic resistance (#17) | Resistance rates are the same in exposed and unexposed bacterial cultures after 14 days. | Cultures exposed to sub-therapeutic ampicillin develop resistance at higher rates than unexposed cultures. |
| Fertilizer (#48) | Plant height at 30 days doesn't differ between organic and synthetic fertilizer groups. | Bean plants given organic fertilizer are taller at 30 days than bean plants given synthetic fertilizer. |
| Minimum wage (#22) | A 10% minimum wage increase has no effect on employment among workers aged 16 to 24 in small retail. | A 10% minimum wage increase changes employment among workers aged 16 to 24 in small retail (in either direction). |
A directional alternative hypothesis, like the first five rows, calls for a one-tailed test. A non-directional one, like the minimum wage row, calls for a two-tailed test. The same logic applies to categorical data, as our guide to chi-square goodness of fit and independence tests shows.
Hypothesis vs Theory vs Law
These three terms are often confused. Here's how they differ.
| Term | What it is | Evidence required | Example |
|---|---|---|---|
| Hypothesis | A testable prediction based on prior knowledge | Grounded in theory or earlier findings; not yet tested | "Sleep deprivation reduces memory performance." |
| Theory | A well-tested explanation of why something happens | Supported by many studies, replicated across contexts | The theory of evolution by natural selection |
| Law | A description of what consistently happens, often as an equation | Reliably observed under the same conditions | Newton's law of universal gravitation |
A common myth is that a hypothesis becomes a theory, and a theory becomes a law, once enough evidence piles up. That isn't how science uses the terms. A law describes what happens, and a theory explains why it happens. A hypothesis is a specific prediction you test along the way.
Hypothesis Non-Examples and Common Mistakes
A non-example looks like a hypothesis but fails one of the four tests: testable, falsifiable, specific, and grounded in prior knowledge. The table below shows six common non-examples and a testable rewrite of each.
| Non-example | Why it fails | Testable version |
|---|---|---|
| "Exercise is good for you." | Too vague: no named variables or measure | Adults who walk 30 minutes a day for 12 weeks show a larger drop in resting heart rate than adults who don't. |
| "Does social media affect teens?" | A research question, not a prediction | Teens who use social media more than three hours a day score lower on the Rosenberg Self-Esteem Scale than teens who use it under one hour. |
| "Plants need water to survive." | An established fact, so there's nothing new to test | Bean plants watered daily grow taller over 30 days than bean plants watered every third day. |
| "Chocolate is the best ice cream flavor." | An opinion that no data can falsify | When offered five flavors at the same price, more than half of customers at a campus ice cream shop choose chocolate. |
| "Students who study more might do better, or they might not." | Every outcome fits, so it can't be wrong | Students who study ten or more hours a week earn higher final exam scores than students who study under three hours. |
| "Sleep, diet, exercise, and screen time all affect grades." | Too many variables for one prediction | High school students who sleep eight or more hours on school nights earn higher grade point averages than students who sleep under six hours. |
Beyond these patterns, watch for the following mistakes when you draft or revise.
- Confusing the null and alternative hypothesis. The null hypothesis always states no relationship. The alternative hypothesis states a relationship.
- Claiming a study "proved" the hypothesis. Data support or fail to support a hypothesis. Write "the results support the hypothesis," not "the results prove it."
- Predicting a finding the literature doesn't support. If published evidence is mixed, use a non-directional hypothesis. If the evidence consistently shows the opposite of your prediction, reconsider it before you design the study.
- Changing the hypothesis after seeing the data. State your hypothesis before you collect or analyze data. Rewriting it to fit the results is known as HARKing (hypothesizing after the results are known).
How to Use "Hypothesis" in a Sentence
Hypothesis is a noun that means a testable prediction or a proposed explanation. Its plural is hypotheses (pronounced hy-POTH-uh-seez). The verb is hypothesize, and the adjective is hypothetical. Here are example sentences that show each form in context.
- Research paper:The data supported our hypothesis, but the effect was smaller than expected.
- Methods section:We tested the hypothesis that longer recess periods improve afternoon attention.
- Statistics:The team rejected the null hypothesis at the 0.05 significance level.
- Plural:The study compared two competing hypotheses about why the reef declined.
- Verb:Researchers hypothesize that sleep loss weakens the immune response to vaccines.
- Adjective:In a hypothetical scenario, the new drug would cut recovery time in half.
- Everyday use:My hypothesis is that the dog, not the toddler, knocked over the plant.
- Science class:Until someone tests it, the idea remains only a hypothesis.
In American English, write "a hypothesis," not "an hypothesis," because the h is pronounced. Pair the word with verbs such as test, support, reject, and revise. Avoid "prove" in academic writing, since a single study can support a hypothesis but can't prove it.
100%
Human editing, no AI
2 Hours
Fastest turnaround
5.0 / 5
Google Reviews rating
BBB A+
Accredited since 2010
65+
Countries served
24/7
Available year-round
Woman-Founded. Purpose-Driven. People First.
Editor World was founded in 2010 by Patti Fisher, a graduate of The Ohio State University, after seeing firsthand the need for high-quality, personalized editing support for writers at every level. Every client who submits a document at Editor World connects directly with a real editor, receives a personal response, and is treated as an individual rather than a transaction. That is the mission Editor World has maintained for 16 years, and it is reflected in every review we receive.
Writing up your hypothesis for a paper or dissertation?
An editor who knows your field can check that your hypothesis, methods, and results tell one consistent story. Editor World offers research paper editing, journal article editing, dissertation editing, and academic editing. Researchers writing in a second language can choose ESL editing. A certificate of editing is available as an optional add-on.
Start With a Free Sample EditFrequently Asked Questions
What is a hypothesis in simple terms?
A hypothesis is an educated prediction about what you expect to find in your research. It states the relationship you expect between two or more variables before you collect data. A strong hypothesis is testable, falsifiable, specific, and grounded in prior research or theory. It names the independent variable, the dependent variable, and the predicted direction of the relationship in one clear statement.
What are the four qualities of a strong hypothesis?
A strong hypothesis has four qualities. First, it's testable, meaning a study can be designed that either supports or refutes it. Second, it's falsifiable, meaning some possible result could prove it wrong. Third, it's specific, naming the variables and predicting the direction or nature of the relationship. Fourth, it's grounded in prior knowledge, flowing from existing theory or previous research findings.
What is the plural of hypothesis?
The plural of hypothesis is hypotheses, pronounced hy-POTH-uh-seez. The word comes from Greek, so it follows the same pattern as thesis and theses, analysis and analyses, and crisis and crises. Write "two hypotheses," never "two hypothesises." The related verb is hypothesize, and the adjective is hypothetical.
How do you use hypothesis in a sentence?
Use hypothesis as a noun for a testable prediction or a proposed explanation. For example: "The team tested the hypothesis that longer recess improves afternoon attention." Hypotheses are tested, supported, or rejected, so pair the word with those verbs. Avoid saying a study proved a hypothesis. Research can support a hypothesis, but a single study can't prove it.
What is an if-then hypothesis?
An if-then hypothesis names the change you'll make and the result you expect. For example: "If bean plants get organic fertilizer instead of synthetic fertilizer, then they will grow taller over 30 days." The if clause holds the independent variable, and the then clause holds the dependent variable. The format suits experiments and science fair projects best.
What is the difference between a hypothesis and a thesis statement?
A hypothesis is used in scientific and empirical research. It predicts a testable relationship between variables. A thesis statement is used in argumentative writing. It states the central argument of an essay. Both are specific claims, but a hypothesis must be testable and falsifiable. A thesis statement doesn't have to be.
Does every research paper need a hypothesis?
Not every paper. Qualitative research, theoretical papers, and literature reviews often don't state a formal hypothesis. Quantitative and experimental research almost always does. Before you decide, check the conventions in your discipline and the requirements of the journal or program you're writing for.
Can a hypothesis be a question?
No. A research question asks what you want to find out. A hypothesis states what you expect to find. Turn your research question into a predictive statement before you write your hypothesis. For example, "Does sleep affect memory?" is a research question. "Adults who sleep more perform better on memory tests" is a hypothesis.
How long should a hypothesis be?
One to two sentences in most cases. A hypothesis should be specific and direct. If it takes more than two sentences to state, it probably contains too many variables or too much qualification. Most published hypotheses fit in a single sentence that names the independent variable, the dependent variable, and the predicted direction of the relationship.
What is the difference between a directional and non-directional hypothesis?
A directional hypothesis predicts the direction of the relationship, using words such as "higher," "lower," "more," or "less." A non-directional hypothesis predicts that a relationship exists without saying which way it goes. Use a directional hypothesis when prior research consistently points one way. Use a non-directional one when the evidence is mixed or the study is exploratory.
What is the difference between the null hypothesis and the alternative hypothesis?
The null hypothesis (H₀) states that there's no relationship between the variables being studied. It's the statement a statistical test evaluates. The alternative hypothesis (H₁) states that a relationship exists, and it's what the researcher expects to find. If the data give strong enough evidence to reject the null hypothesis, that result supports the alternative.
What is an example of a null hypothesis?
A null hypothesis states that there's no effect or no difference. For example: "Plant height after 30 days doesn't differ between bean plants given organic fertilizer and bean plants given synthetic fertilizer." The matching alternative hypothesis predicts that one group grows taller. A statistical test then shows whether the data justify rejecting the null hypothesis.
Related guides: Our guide to the scientific method explains where a hypothesis fits among observation, experiment, and conclusion. For testing hypotheses about categorical data, see our walkthrough of chi-square goodness of fit and independence tests.
Reviewed by the Editor World editorial team under the direction of Patti Fisher, PhD, founder. Editor World has provided 100% human editing by native English editors since 2010. About Editor World