<< Self-Defense and Riots | Main | News Scan >>


Studies Show ...

| 1 Comment
Although not directly on topic, this WSJ article by Gautam Naik describes the problem of medical studies published in prestigious journals that are later found to be wrong, sometimes incompetent, and occasionally fraudulent.

If this problem is so widespread in the medical field, where studies are relatively objective and usually unpolitical, how much greater is the problem in social sciences, including studies about crime?  Social science is generally softer, and methods that would be unacceptably unreliable for medicine are routinely used in social science.  Quasi-experiments are accepted where true experiments are impossible.  Would the FDA ever approve a drug with only quasi-experimental data to support its safety and effectiveness?  The article notes that a study not being double-blind is "a situation many investigators consider tantamount to fraud" in the medical field, but it is routine in social science.

On top of that, you have the Political Correctness quotient.  Sources of fraud or at least fudging in the medical arena include the monetary interests of drug manufacturers and the reputational interests of researchers, but the people in charge of reviewing the study for publication typically have no interest in the outcome.  With studies on politically controversial topics, however, the massive PC bias of academia means that studies with Politically Correct bottom lines will get less scrutiny than those with Politically Incorrect bottom lines.

What "studies show" is not necessarily so.

1 Comment

Even more reason to be skeptical of "evidence-based practices" that the left is fond of quoting.

Leave a comment