关于「预算约束与陈述偏好:必需品、奢侈品与「双轨」假说的检验」,《择偶多目标与帕累托边界》给出的定量结论是什么?

结论:程序模拟 80 项小样本研究,每项样本 20 到 60 人,每项有 4 种可报告的测量,只报告其中效应最大的那一种。当真实转移量为零时,全部研究的汇总效应是 0.162,只看「发表」的研究是 0.189;真实转移为 0.05 时,两者分别是 0.216 与 0.247;为 0.10 时是 0.272 与 0.325。反过来,一项预先登记、样本 584 人的研究,对 0.20 的转移有 0.998 的检验功效,对 0.10 有 0.676,对 0.05 只有 0.227。

数据口径:全书以「高固定投资方」「被索取方」等角色指称,两性差异只写效应量与重叠;书中人群、特质与参数均为合成或示意,全部计算由书中程序生成;本书不是婚恋咨询、心理咨询或临床判断,也不作任何命理推导。 第五章用预算约束区分必需品与奢侈品,说明陈述偏好预测力有限,并指出排卵周期偏好转移的两篇元分析结论相反、大样本研究未见转移,因此「双轨」只作角色通用、可以证伪的情境假说。

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: 0.05, 0.10, 0.162, 0.189, 0.20, 0.216

true delta  pooled(all, flexible)  pooled(published)  k pub
      0.00                  0.162              0.189     31
      0.05                  0.216              0.247     34
      0.10                  0.272              0.325     42
power of one preregistered study, n = 584, alpha 0.05 two-sided:
  delta 0.05: power 0.227
  delta 0.10: power 0.676
  delta 0.20: power 0.998
flexible small studies can pool to a clear positive shift when the
true shift is zero; a large preregistered design can detect 0.20
查看完整验算程序
# Chapter 5. p53: how small studies and selective reporting can split
# a literature on cycle shifts in preference (illustrative mechanism).
# 80 within-person studies of n = 20 to 60 each test a shift of true
# size delta (standardised). Each study reports one of several
# outcome measures; with flexible reporting the largest of 4 noisy
# estimates is reported. A "published-only" pool keeps significant
# positive results plus 30% of the rest. Compare pooled estimates with
# one preregistered study of n = 584 (the size in Jones et al. 2018).
import math
import random
from statistics import NormalDist

ND = NormalDist()


def run(delta, seed_rng):
    pooled_all, pooled_pub, w_all, w_pub = [], [], [], []
    for _ in range(80):
        n = seed_rng.randint(20, 60)
        se = 1.0 / math.sqrt(n)
        ests = [delta + seed_rng.gauss(0.0, se) for _ in range(4)]
        est = max(ests)
        z = est / se
        w = 1.0 / (se * se)
        pooled_all.append(w * est)
        w_all.append(w)
        if z > 1.96 or seed_rng.random() < 0.30:
            pooled_pub.append(w * est)
            w_pub.append(w)
    a = math.fsum(pooled_all) / math.fsum(w_all)
    b = math.fsum(pooled_pub) / math.fsum(w_pub)
    return a, b, len(w_pub)


print("true delta  pooled(all, flexible)  pooled(published)  k pub")
rng = random.Random(41531)
for delta in (0.0, 0.05, 0.10):
    a, b, k = run(delta, rng)
    print("{:10.2f}  {:21.3f}  {:17.3f}  {:5d}".format(delta, a, b, k))
print("power of one preregistered study, n = 584, alpha 0.05 two-sided:")
for delta in (0.05, 0.10, 0.20):
    z = delta * math.sqrt(584)
    power = 1 - ND.cdf(1.96 - z) + ND.cdf(-1.96 - z)
    print("  delta {:.2f}: power {:.3f}".format(delta, power))
print("flexible small studies can pool to a clear positive shift when the")
print("true shift is zero; a large preregistered design can detect 0.20")

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《双轨择偶多目标冲突、亲代投资不对称与伴侣筛选帕累托边界母典》第 5 章「预算约束与陈述偏好:必需品、奢侈品与「双轨」假说的检验」(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-421 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026421,
  author    = {Fan, Yafu},
  title     = {Dual-Mating Multi-Objective Conflicts, Parental-Investment Asymmetry and the Pareto Frontier of Partner Selection},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论