The race × week FE design gives β = +6.95 pp on 60 cells

🟢 The strictest timing-controlled spec (race × week FE with comparator restricted to independent-media-sponsored polls) gives β = +6.95 pp (p = 0.008) on 60 race-week cells.

Identifying variation comes exclusively from polls fielded in the same race within the same week, comparing self-sponsored vs independent polls of the same candidate. Magnitude matches the within-candidate FE (+7.75) and the placebo (+6.70) despite sharing no identifying variation with either.

Sources.

Own analysis: build/table/regressions.csv (Spec 3c); AN-002
Cross-refs: H1 headline-sponsor-bias; headline-sponsor-bias finding

Cited analyses

AN-002 green robustness

Restricting comparators to media-or-pollster-self polls in the same race within the same week (Spec 3c) gives β = +6.95 pp (p=0.008) on 60 race-week cells. The tightest design and the most robust to the timing-of-commission alternative.