To reduce the risk of type 1 errors (as there were over 100 potential inferential tests that could be conducted), the p value for a significant result was set at 0.01 and a staged approach was often applied that began with testing larger combined variables and only testing the subgroups of borderline significant results.
← all excerpts
Integrative Medicine in General Practice in Australia: A Mixed-Methods Study Exploring Education Pathways and Training Needs.
1
—
—