4 ms·
Psychological, Educational, and Behavioral Treatment Meta-Analysis Confirmation
- gwern 13y agoI love these meta-meta-analyses, since they let you say things like 'Published studies yield mean effects 0.14 SDs larger than unpublished studies.' Isn't that useful to know?
- lutusp 13y agoFirst, a paper from 1993? This isn't news. If the paper were a person, it could vote. Second, a quote: "... there are good reasons to believe that its results are more credible than those of conventional reviews and to conclude that well-developed psychological, educational, and behavioral treatment is generally efficacious." Yes -- we can agree that people prefer talking to a sympathetic listener over staring at a wall. But without an effort to craft and then test a theory about the value of talking to other people, and whether talking to a therapist differs from talking to a bartender, it's not science, it's anecdote. This was true in 1993, and it's still true today.