R2 — Never use agree/disagree statements in the widget

Rule: Do not use agree/disagree statements anywhere in the widget. Use item-specific response options or counts.

Why: Saris 2010 (Saris et al. 2010 — item-specific response options far outperform agree/disagree on data quality) — item-specific quality coefficients 0.74-0.89 vs agree/disagree 0.18-0.51; the gap is method bias (acquiescence), not noise. >100 studies show acquiescence affects ~10-20% of responses. Lelkes & Weiss 2015 is a partial counter on criterion validity — but rewriting alone is not magic; the real win is converting to OBSERVATIONS, which agree/disagree formats actively prevent.

How to apply: Forbidden phrasing: "I have a strong online presence (agree/disagree)", "My website performs well (agree/disagree)". Required phrasing: name the construct directly with item-specific options.