Null and Alternate Hypotheses

Null hypothesis

Null hypothesis is the default state, or presumed to be true unless evidence shows otherwise (thus confirming the alternate hypothesis).

"A null hypothesis is a type of hypothesis professionals use in statistical theory, which states that there's no significant relationship between two sets of observed data. In statistics, the null hypothesis presumes to be correct until there's enough evidence to suggest otherwise." lifted from Indeed.com

Alternate hypothesis

If statistically significant evidence is shown that there is sufficient information to reject the null hypothesis, then the alternate hypothesis must be true.

Example

Null hypothesis - Filipinos are generally perceived to be happy and friendly people.

Alternate hypothesis - Filipinos are not perceived to be happy or friendly people.

Data collection - Sampling data of perceptions on Filipinos are collected.

α = 0.05 (5% confidence level).

Result: If the P value is less than or equal to your significance level (alpha) then you should reject your null hypothesis.

e.g. P value is 0.0005, which is less than 0.05. Therefore, we reject the null hypothesis: Filipinos are not perceived as happy or friendly people.

References

• https://www.indeed.com/career-advice/career-development/reject-null-hypothesis

• T-test with paired means