How to Write Research Questions and Hypotheses (With Examples)

Research questions and hypotheses state clearly what your study wants to find out. Methods, analysis and discussion are all shaped by them, so a vague question makes the whole thesis vague.

Qualities of a good research question

• Clear: the variables and population are obvious.
• Answerable: the data you will collect can answer it.
• Focused: one question does not ask several things.
• Significant: it contributes to the literature or practice.

Weak: How does social media affect students?

Strong: Is there a significant relationship between university students’ daily social media use and their self-esteem scores?

Types of questions

• Descriptive: "What is the level of…?"
• Comparative: "Do … scores differ by gender?"
• Relational: "Is there a relationship between … and …?"
• Predictive: "Does … significantly predict …?"

Qualitative questions are more open-ended: "How do teachers make sense of their remote-teaching experiences?"

What is a hypothesis?

A hypothesis is a testable tentative answer to a research question. In quantitative studies it is tested statistically.

• Null hypothesis (H₀): there is no difference or relationship.
• Alternative hypothesis (H₁): there is a difference or relationship.

Directional and non-directional

• Non-directional: "There is a significant difference between the two groups."
• Directional: "Post-test scores of the experimental group are significantly higher than those of the control group."

You need strong support from the literature to state a directional hypothesis.

Tips for writing hypotheses

• Each hypothesis should test one relationship or difference.
• Number them (H1, H2, H3…) and address them in the same order in the results.
• State in the methods which analysis tests each hypothesis.

Questions and hypotheses with Academiana

Academiana’s "Research questions" and "Generate hypotheses" actions make suggestions based on your project topic and sources, showing the basis of each hypothesis in the literature. The quality check warns you if a hypothesis is not addressed in your analysis or results.