As we demonstrated in the above, equating mean differences with group differences is mostly unjustified because even “highly significant” differences between means usually involve large similarities between the groups in question.
← all excerpts
A new way to look at the data: Similarities between groups of people are large and important.
1
—
—