关于「薄切片的信度与效度:三秒判断有多稳,又有多准」,《三秒薄切片信任》给出的定量结论是什么?

结论:第一章区分信度与效度,示意参数下三秒判断只达到效度上限的 0.612,与长期评价的一致率为 0.579。

数据口径:书中参数为示意;全部计算由书中程序生成;本书只写如何让真实、可兑现的信号在前三秒清晰一致,不写任何伪造资质或假权威的手法,不宣称任何气味的健康功效,是非个人化经营建议。 第一章区分信度与效度,示意参数下三秒判断只达到效度上限的 0.612,与长期评价的一致率为 0.579。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 0.579, 0.612

A. validity r(t) with ceiling r_inf0.40
t(s) t02.0 t0= 5.0 t0=15.0
10.231 0.163 0.100
30.310 0.245 0.163
100.365 0.327 0.253
300.387 0.370 0.327
600.393 0.384 0.358
3000.399 0.397 0.390
B. share of ceiling reached at 3 s
t02.0: 0.775
t05.0: 0.612
t015.0: 0.408
C. median-split agreement with the long-run criterion
r0.0: agree 0.500, variance shared 0.000
r0.2: agree 0.564, variance shared 0.040
r0.3: agree 0.597, variance shared 0.090
r0.4: agree 0.631, variance shared 0.160
r0.6: agree 0.705, variance shared 0.360
D. 3 s, t05: agree 0.579; wrong on 0.421 of cases
查看完整验算程序
# Chapter 1. p12: how validity of a thin-slice judgment grows with exposure.
# Model (illustrative): a judgment made after t seconds carries signal plus
# noise whose variance falls as t0 / t, so its correlation with a long-run
# criterion is r(t) = r_inf * sqrt(t / (t + t0)). r_inf is the ceiling set
# by how far behaviour reflects the criterion at all; t0 is the exposure at
# which half of the reliable variance is in. Values of r_inf and t0 are
# assumptions for illustration, not estimates from any one study.
# For a median split, agreement with the criterion is 1/2 + asin(r)/pi
# (bivariate normal orthant probability).
import math

R_INF = 0.40
T0S = (2.0, 5.0, 15.0)
TIMES = (1, 3, 10, 30, 60, 300)


def r_of_t(t, t0, r_inf=R_INF):
    return r_inf * math.sqrt(t / (t + t0))


def split_agreement(r):
    return 0.5 + math.asin(r) / math.pi


if __name__ == "__main__":
    print("A. validity r(t) with ceiling r_inf = {:.2f}".format(R_INF))
    print("   t(s)  " + "  ".join("t0={:>4.1f}".format(t0) for t0 in T0S))
    for t in TIMES:
        row = "   ".join("  {:.3f}".format(r_of_t(t, t0)) for t0 in T0S)
        print("  {:>4}   {}".format(t, row))
    print("B. share of ceiling reached at 3 s")
    for t0 in T0S:
        print("   t0 = {:>4.1f}: {:.3f}".format(t0, r_of_t(3, t0) / R_INF))
    print("C. median-split agreement with the long-run criterion")
    for r in (0.0, 0.2, 0.3, 0.4, 0.6):
        print("   r = {:.1f}: agree {:.3f}, variance shared {:.3f}".format(
            r, split_agreement(r), r * r))
    a3 = split_agreement(r_of_t(3, 5.0))
    print("D. 3 s, t0 = 5: agree {:.3f}; wrong on {:.3f} of cases".format(
        a3, 1.0 - a3))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《三秒薄切片信任捕获:视听嗅三通道同频一致性与瞬间权威建立母典》第 1 章「薄切片的信度与效度:三秒判断有多稳,又有多准」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-419 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026419,
  author    = {Fan, Yafu},
  title     = {Three-Second Thin-Slicing Trust: Visual-Auditory-Olfactory Congruence and Instant Authority Signalling},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论