关于「书名是程式化写法:双函数混合与两性分布的重叠」,《亲密决策阈值不对称》给出的定量结论是什么?

结论:第一章由效应量算重叠率:元分析报告的性驱力效应 0.69 下,两条分布共有面积 0.730,随机配对中 0.313 方向相反;用性别标签猜一个人只对 0.635。

数据口径:书中参数均为示意或合成;全部计算由书中程序生成;本书不是心理咨询、婚恋咨询或临床判断,也不是个人化经营建议。 第三章在合成拆解里,感知安全、体验与污名承担原始差距的 0.842,并说明零接受率会把优势比放大到失真。

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: 0.313, 0.69, 0.730

A. overlap table (equal variances)
d OVL CL U3 R2
0.100.960 0.528 0.540 0.002
0.200.920 0.556 0.579 0.010
0.350.861 0.598 0.637 0.030
0.500.803 0.638 0.691 0.059
0.540.787 0.649 0.705 0.068
0.690.730 0.687 0.755 0.106
0.800.689 0.714 0.788 0.138
1.200.549 0.802 0.885 0.265
B. reading d0.69
shared area 0.730; reversed pairs 0.313
variance left inside groups 0.894
C. d needed for a given common-language probability
CL 0.75 needs d0.95
CL 0.90 needs d1.81
CL 0.95 needs d2.33
CL 0.99 needs d3.29
查看完整验算程序
# Chapter 1. p11: Cohen d to overlap, common-language and variance shares
# Two normal distributions with equal variance and standardized mean gap d.
# OVL = 2 * Phi(-|d| / 2)      overlapping coefficient (shared area)
# CL  = Phi(d / sqrt(2))       P(random member of group 1 > group 2)
# U3  = Phi(d)                 share of group 1 above the group 2 mean
# R2  = d^2 / (d^2 + 4)        variance share explained by group label
# d = 0.54 and 0.69 are the adjusted and raw sex-drive estimates of the
# Frankenbach et al. 2022 meta-analysis (raw 95 % CI 0.58 to 0.81);
# the other rows are reference
# values (Cohen's small, medium, large; Hyde's close-to-zero band).
import math
from statistics import NormalDist


def phi(x):
    return 0.5 * (1.0 + math.erf(x / math.sqrt(2.0)))


def row(d):
    ovl = 2.0 * phi(-abs(d) / 2.0)
    cl = phi(d / math.sqrt(2.0))
    u3 = phi(d)
    r2 = d * d / (d * d + 4.0)
    return ovl, cl, u3, r2


print("A. overlap table (equal variances)")
print("   d     OVL    CL     U3     R2")
for d in (0.10, 0.20, 0.35, 0.50, 0.54, 0.69, 0.80, 1.20):
    ovl, cl, u3, r2 = row(d)
    print("  {:4.2f}  {:.3f}  {:.3f}  {:.3f}  {:.3f}".format(
        d, ovl, cl, u3, r2))

print("B. reading d = 0.69")
ovl, cl, u3, r2 = row(0.69)
print("  shared area {:.3f}; reversed pairs {:.3f}".format(ovl, 1.0 - cl))
print("  variance left inside groups {:.3f}".format(1.0 - r2))

print("C. d needed for a given common-language probability")
z = NormalDist()
for p in (0.75, 0.90, 0.95, 0.99):
    print("  CL {:.2f} needs d = {:.2f}".format(p, math.sqrt(2.0) *
                                            z.inv_cdf(p)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《亲密决策阈值不对称:男性边际成本核算与女性情境张力唤醒的演化经济学母典》第 1 章「书名是程式化写法:双函数混合与两性分布的重叠」(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-424 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026424,
  author    = {Fan, Yafu},
  title     = {Asymmetric Intimacy Thresholds: Male Marginal-Cost Accounting versus Female Contextual-Tension Arousal in Mating Economics},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论