If the between-groups and within-groups variance estimates are close, F will be near one and the null hypothesis is:

1. The F ratio compares two estimates of the same population variance. 2. Variance between samples and variance within samples both estimate the population variance. 3. If they agree, their ratio F sits near one. 4. But if F is much larger than one, then the evidence is against the null hypothesis. 5. So an F near one leaves the null hypothesis uncontradicted. _Source: OpenStax Introductory Statistics 2e (CC BY 4.0), Ch 13 'The F Distribution and One-Way ANOVA', sections 13.1-13.4_