Crisis in Psychology?

[tackattack]

My thoughts are that the results of conclusions in psychology are poorly defined. What is the conclusion of diagnosing and "fixing" schizophrenia.... no schizophrenic episodes? No, the results of all of psychology is a better quality of life for an individual/society. That's so nebulous as it's undefinable without a lens that eliminates the individual. When we can map thoughts to 1 and 0s then maybe a discussion of scientific and mathematical reproducibility within psychology would be prudent. I don't think it's valid to determine whether the construct is discriminantly and convergently valid without first defining a really comprehensive nomological network.

[John 6IX Breezy]

Here is a short comment published in Science Magazine, presenting some arguments against the original Open Science studies that sparked the replication debate. They discuss issues with statistical errors, power, and bias in the replication studies, and conclude that reproducibility is quite high when corrected. (Some of these problems came to mind when I was thinking about why replication is problematic—such as taking samples from a population that differs from the original in any way.)

[John 6IX Breezy]

I want to add an argument given the above paper: That what is described as a replication crisis is more accurately described as a generalizability crisis.

Science has always had a problem with induction: The very act of experimentation isolates nature such that it requires generalizations to say anything beyond a lab. Newton's laws of motion, for example, are generalizations made about the universe drawn from a handful of localized observations. So I would compare the "unsuccessful" replication studies in psychology that were sampled from a population that differed from the original, to a chemist that fails to replicate the boiling point of water, yet makes his observations at a different altitude than the original. The issue isn't replication here, it's the generalizability of scientific observations.