Too Good To Be True

A prescient (2015) too-good-to-be-true critique of social psychology. Contrasts three studies claiming huge outcomes from tiny interventions (a plan-making course -> +0.8 GPA; a growth-mindset pen-pal letter -> higher GPA; an 8-week time-management program -> better at nearly everything) with the previous post's $58k, 10-year intervention that produced only modest effects. The dilemma: 'either psychological research sucks or everything else sucks' -- and Scott bets on the former, finding flaws in each (tiny subgroups; effects that vanish on basic correlation / even reverse; dropped exam-period context), invoking the xkcd 'are they?' chart.

Quality 70 · Strong. Strong, upper (70). Prescient and well-argued: it red-flags the exact implausibly-large social-psych effects (growth mindset etc.) that the replication crisis later deflated, with a crisp dilemma framing and concrete flaw-finding. Capped by brevity and a partly-skim ('I don't want to sound like I'm definitively crushing them') basis.

Claude’s paradigm shift 53 · Moderate. Notable-edge (53). The 'implausibly large effects from tiny interventions are a red flag' heuristic, applied before the replication crisis was mainstream, is a fresh and non-obvious skeptical frame for 2015.

Real-world impact 2 · Minor. 2 -- minor/within-blog. A prescient contribution to the broader replication-skepticism discourse, but a single post's footprint is within-blog; the field-wide impact belongs to the wider movement.