Null and alternative hypotheses
The null hypothesis H0 is the default claim of no effect or no difference. The alternative H1 is what you suspect instead. A test asks whether the data is too surprising to fit H0.
Last updated
Analogy
A court starts with 'innocent' (the null). The evidence must be strong enough to overturn it.
Example
Claim: mean battery life is 10 hoursH0: mean = 10H1: mean is not 10 (two-sided)Or H1: mean < 10 (one-sided)
In practice
H0: a drug does nothing. H1: it changes recovery time. Trials try to show H0 is hard to believe.
Practice
Flashcards
- What does the null hypothesis usually claim?No effect and no difference.
Questions people ask
Why do we test the null instead of proving the alternative?
A single unusual result cannot prove a general claim, but data can be shown to clash with a precise default. The null is exact, such as no difference, so we can calculate how likely our data would be if it were true. Strong clashes count as evidence.
When should I use a one-sided test?
Use it only when you decided, before seeing the data, that you care about one direction, such as a new method being better. A two-sided test checks for any difference and is the safer default. Choosing the side after peeking inflates false alarms.