Paired candidate impressions differ within respondents

if (v.kind === "t") {
  const rows = v.groups.map((g) => ({
    label: g.label,
    b: g.mean,
    lo: g.ci[0],
    hi: g.ci[1],
    key: g.key,
  }));
  a.push(
    ...forest(rows, {
      x: 0,
      y: 8,
      w: 1000,
      title: "A  Group means · pointwise 95% CI",
      unit: v.unit,
      rowH: 155,
      labelW: 260,
      sign: false,
      range:
        v.id === "paired"
          ? [1, 7]
          : [
              Math.min(...rows.map((r) => r.lo)) - 1,
              Math.max(...rows.map((r) => r.hi)) + 1,
            ],
    }),
    T(1100, 47, "B  Difference", 32, C.ink, 700),
    G(
      [
        T(1100, 145, signed(v.difference), 66, C.blue, 750),
        T(1100, 190, v.unit, 28, C.muted),
        T(1100, 258, "95% CI", 26, C.muted),
        T(1100, 305, "[" + v.ci.map((x) => fmt(x, 3)).join(", ") + "]", 36, C.ink, 650),
        T(1100, 377, "t(" + fmt(v.df, 1) + ") = " + fmt(v.stat, 2), 27),
        T(1100, 423, pv(v.p), 27),
        T(1100, 491, v.effectName + " = " + fmt(v.effect, 3), 30, C.teal, 700),
        T(
          1100,
          533,
          "effect CI [" + v.effectCI.map((x) => fmt(x, 3)).join(", ") + "]",
          24,
          C.muted,
        ),
      ],
      v.key,
    ),
    T(
      15,
      580,
      "Group n: " +
        v.groups.map((g) => g.n).join(" / ") +
        " · " +
        (v.id === "paired" ? "same 944 respondents" : "independent groups"),
      27,
      C.muted,
    ),
    ...(v.id === "paired"
      ? [T(15, 625, "Scale: 1 very liberal → 7 very conservative", 26, C.muted)]
      : []),
  );
  note =
    v.id === "paired"
      ? "DoleLR−ClinLR on the same 1–7 numeric-coded scale. Not a causal candidate effect."
      : "Mean age differs by expected vote; Welch df reflects unequal variances. Nonrandom groups.";
  source =
    v.id === "paired"
      ? 'flowchart LR\n R["Same respondent"] --> A["Clinton impression"]\n R --> B["Dole impression"]\n A --> D["Within-person difference"]\n B --> D'
      : 'flowchart LR\n A["Clinton expected"] --> D["Mean-age difference"]\n B["Dole expected"] --> D\n D --> W["Welch CI"]';
  map = { R: v.key, A: v.groups[0].key, B: v.groups[1].key, D: v.key, W: v.key };
}