Working With Google Trends
Read the original on Slate Star Codex →
Summary
[Epistemic status: low] A methodological exploration of using Google Trends for research, cataloguing the pitfalls: individual keywords move in opposite directions ('psychiatry' down, 'psychiatrist' up); movie-release spikes ('Suicide Squad'); the school-term sawtooth pattern that flags student searches vs 'organic' ones; the depression-vs-depressed two-factor distinction (students studying vs people seeking help). Tries averaging + factor analysis (two factors: intellectual-interest vs mental-health-need) and tests for (and finds no) general 'intellectual decline of the internet.' Ends with practical tips.
Why this score
Quality 60 · Strong. 60 — Solid/Strong boundary. A useful, honest methods exploration with genuinely helpful heuristics (the sawtooth=students insight, the depression-vs-depressed distinction), but exploratory, hedged ('epistemic status: low'), in-the-weeds, and inconclusive. Firm at the boundary.
Claude’s paradigm shift 42 · Moderate. 42 — Moderate. A methodological exploration in a 'not original' spirit; the specific heuristics are mildly useful.
Real-world impact 1 · Negligible. 1 — negligible. A methods note; no material reach.