JentooPsychology

AP® Psychology · Science Practices: Research Methods and Data

2.B Hypotheses and Operational Definitions

A good hypothesis is a testable prediction that could turn out to be wrong, and an operational definition spells out exactly how an abstract idea like happiness or aggression will be measured.

Watch it drawn3 questionsPractice

A hypothesis is a specific, testable prediction about the relationship between variables. For it to do any scientific work, there has to be some possible result that would count against it. This is falsifiability: if a claim can stretch to explain any outcome at all, improvement or decline, more of something or less, it is not making a real prediction. It is just a label stuck onto whatever happens. A hypothesis that risks being wrong is the one that can teach you something when it survives a real test.

Once a hypothesis names a concept like happiness, stress, or aggression, that concept has to become something countable. An operational definition is the exact procedure or measurement a researcher uses to turn an abstract idea into data: a score on a 10-item life-satisfaction survey, cortisol level in saliva, number of hits thrown in a play session. Two researchers can both claim to study 'happiness' and still measure completely different things, which is why every study needs its operational definitions spelled out before anyone can judge, replicate, or compare it.

This sits right before sampling and experimental design in the research methods practices, because a hypothesis and its operational definitions are what everything downstream depends on. If the definitions are vague or shift midstudy, the eventual data cannot be trusted no matter how careful the sampling or statistics are. Good operational definitions are also what let a second lab repeat the study and check whether the finding holds up.

Key terms

Hypothesis
A specific, testable prediction about how variables relate to each other.
Falsifiability
The property that some possible observation could prove a hypothesis wrong.
Operational definition
The exact procedure or measurement used to turn an abstract concept into something observable and quantifiable.
Construct
An abstract concept, like happiness or aggression, that cannot be observed directly and must be measured indirectly.

Watch it explained

  1. How it worksHow does an operational definition turn happiness into something a second lab can measure?
  2. Worked exampleShow me three operational definitions of aggression and how each one changes the result.
  3. Common mistakeWhy isn't a hypothesis that explains everything a strong hypothesis?

Check yourself

Your turn1 of 2
A researcher proposes the hypothesis 'this therapy helps everyone in every situation, and any outcome, whether the client improves or gets worse, can be explained by the theory.' A colleague argues this is actually a WEAK hypothesis, despite sounding broadly applicable. What is the colleague's reasoning?

End of topic.

Next: 2.B Independent, Dependent, and Confounding Variables →