经济与社会科学2016journals.plos.org
统计上控制混杂构念很困难（2016）
Statistically controlling for confounding constructs is hard (2016)
进阶 · 约 60 分钟读完前置知识：回归分析、测量误差、心理测量/构念效度
导读
论文讨论统计上“控制混淆构念”为何困难：研究者常在回归中控制某个测量变量，以为控制了理论构念，但测量有误差且只覆盖构念的一部分。作者指出这类测量层面的增量效度论证很容易失败，大量社会科学研究可能因此把本不区分的两个构念误判为不同。
HN 关注点：因为“控制住 X 之后”是实证研究中最常见也最容易被误用的说法，论文说明测量误差会让这种控制失效，并波及社会科学、遗传学等大量研究。
后来怎么样
该文使“控制测量变量≠控制理论构念”成为心理学方法学中的常见警告，并常被用来批评依赖增量效度论证的研究结论。
HN 评论精选
这几条评论里：支持 2 补充 2
支持论文核心问题是：回归中控制的只是带误差的测量变量，不等于控制理论构念；并推荐作者参与的解释。
这篇论文讨论了一个我认为有学术素养的人群应远更关注的重要问题；这种知识至少应在中学教授，尤其现在论文触手可及。一个更通俗的叙述版总结在 https://www.talyarkoni.org/blog/2016/06/11/the-great-minds-journal-club-discusses-westfall-yarkoni-2016/ 。其中说：问题在于统计控制——至少人们通常的做法——是测量层面的技术。也就是说，当你在把癌症回归到培根上时控制酒精使用率，你并不是真正控制了酒精使用；你实际控制的只是酒精使用的一个特定操作化，它可能覆盖不了整个构念，而且通常还有测量误差。我强烈建议也读该页上一位作者的评论。
Veedrac查看英文原文
This paper discusses a major issue that I think deserves way more attention from the academically literate crowd. This is the kind of knowledge that should be taught in secondary school, at least in the modern age with papers at everyone's fingertips. A more approachable summary is in narrative form at https://www.talyarkoni.org/blog/2016/06/11/the-great-minds-journal-club-discusses-westfall-yarkoni-2016/ > “But there’s a problem: statistical control–at least the way people typically do it–is a measurement-level technique. Meaning, when you control for the rate of alcohol use in a regression of cancer on bacon, you’re not really controlling for alcohol use. What you’re actually controlling for is just one particular operationalization of alcohol use–which probably doesn’t cover the entire construct, and is also usually measured with some error.” I strongly suggest reading the comment on that page from one of the authors as well.
补充用多基因风险评分的真实案例说明：复杂模型中简单控制群体结构仍会失败，并可能被滥用。
最近有两篇关于多基因风险评分（PRS）的精彩论文；PRS 现在在人类遗传学中很流行，用来替代 GWAS，因为 PRS 可以纳入 GWAS 忽略的微小遗传变异效应。两篇论文的编辑评论见 https://elifesciences.org/articles/45380 。简言之，做 GWAS 时，人们用前两三个主成分作为协变量来控制群体结构。而在 PRS 中，模型太复杂，用主成分控制并不能消除群体结构的全部影响——这就是为什么一些著名的大规模人类遗传学研究无法重复。（这有社会层面的重大影响：Twitter 上满是“种族现实主义者”即种族主义者，他们用虚假的 PRS 来支撑自己预设的结论。）
a_bonobo查看英文原文
There was an amazing couple of papers recently on polygenic risk scores (now trendy in human genetics in place of GWAS, since PRS can include tiny effects from genetic variants that GWAS ignores) Editorial on both papers here: https://elifesciences.org/articles/45380 In summary, when you run a GWAS, you control for population structure by using the first two or three principal components as covariates. In PRS, the model built is so complex that controlling using PCs doesn't fix all influence by population structure - which is now why some famous large-scale human genetic studies fail to replicate. (This has large implications on a societal level - Twitter is full of 'race realists' ,i.e., racists, who are using spurious PRS to prop up their foregone conclusions)
补充用冰淇淋/游泳例子解释：协变量测量不精确时，“控制后仍相关”不能证明独立构念；贝叶斯网络可能更合适。
这很有意思。用他们的例子概括论文思想：如果人们在冰淇淋销量高的日子游泳更多，可能得出冰淇淋导致游泳。但若有温度信息，可以拟合统计模型预测游泳；模型会学到温度能完美预测游泳，完全不用冰淇淋特征，于是可说控制温度后冰淇淋不再与游泳相关。但如果温度估计有噪声或不精确，比如只知道温度是否高于75度而不知道确切值，这个技巧就失效：温度仍能预测游泳很多但不完全，所以冰淇淋销量仍有用，控制温度后仍与游泳相关。我想解决之道是否是贝叶斯网络：可检验游泳、冰淇淋销量、观测温度都由某个未观测第三变量导致的模型；它应比“冰淇淋导致游泳和温度”等替代模型拟合更好，但简单线性模型无法表示这类依赖。
Houshalter查看英文原文
This is really interesting. Trying to summarize the idea of the paper using their example: If people swim more on days when ice cream sales are high, one might conclude that ice cream causes swimming. But if you have information about the temperature, you can fit a statistical model that predicts swimming. The model will learn that temperature predicts swimming perfectly, and the feature of ice cream isn't used at all. So you could say after controlling for temperature, ice cream no longer correlates with swimming. But what if your temperature estimate is noisy or imprecise? For example, if you only have information if the temperature was hotter than 75 degrees, but not exactly what it was. Then this trick no longer works. The temperature variable predicts swimming a lot, but not completely. So ice cream sales are still useful, and so still correlate with swimming after controlling for temperature. I wonder if the way to solve this is using bayes nets. You could test a model where the variables "swimming", "ice cream sales", and "observed temperature", are all caused by some unobserved third variable. This model should fit the data better than alternative models, like ones where "ice cream sales" cause swimming and temperature. But simple linear models can't represent dependencies like this.
支持如果协变量只是带测量误差的估计，“控制”并不能可靠区分构念；作者措辞谨慎但结论影响很大。
我一直想知道人们说“控制……”到底是什么意思，仿佛能区分原因 A 和原因 B 对结果 C 的作用。原来，如果你能无噪声地测量 A 和 B，“控制……”这套做法是成立的；但如果你只有实际 A、B 的带噪估计 A'、B'，许多研究者使用的“控制……”技术就根本不好用。有趣的引文：“当前结果表明，对于一类非常常见的增量效度论证，这种策略失败风险很高。问题范围相当大：历史上横跨众多科学领域、数十万项研究依赖测量层面的增量效度论证，来支持关于理论构念之间关系的强结论。”“我们得出结论，社会科学中许多先前发现很可能高风险地错误判断两个构念不同，而它们实际上可能并非不同。”注意作者多礼貌。他们没直说许多（大多数？）多元回归结果是胡扯，只说“高风险出错”。
ivan_ah查看英文原文
I always wondered what exactly people mean when they say “controlling for ...” as if somehow they can tell apart cause A and cause B for the effect C. It turns out if you can measure A and B without noise, the whole "controlling for..." business is legit, but if you only have noisy estimates A' and B' of the actual A and B, then the "controlling for..." technique used by many researchers does not work very well at all. interesting quotes: > The present results demonstrate that, for a very common class of incremental validity arguments, such a strategy runs a high risk of failure. The scope of the problem is considerable: literally hundreds of thousands of studies spanning numerous fields of science have historically relied on measurement-level incremental validity arguments to support strong conclusions about the relationships between theoretical constructs. > We conclude that many previous findings in the social sciences are at high risk of having concluded that two constructs are distinct when they may not in fact be so. Note how polite the authors are. They don't say it straight up that many (most?) multiple regression results are bullshit, they just say "at high risk of being wrong."
相关资源
从这篇论文所有 HN 讨论帖的全部评论里整理出来的链接。
- 评论区提到的论文
历次讨论
- 2016Hacker News 讨论帖44 分 · 19 评
- 2019Hacker News 讨论帖143 分 · 37 评
同领域还值得读
一切的回报率，1870–2015
The Rate of Return on Everything, 1870–2015
利用1870–2015年多国的长期数据，估算股票、住房等各类资本的回报率，并将其与经济增长率进行比较。
经济与社会科学2019 · 487 分 · 260 评进阶 · 约 100 分钟编辑精选一切资产的回报率，1870–2015
The Rate of Return on Everything, 1870–2015
构建了 1870–2015 年 16 个国家的长周期资产回报数据集，覆盖股票、住房、政府债券和短期票据，系统比较各类资产的真实回报与波动性。
经济与社会科学2018 · 215 分 · 151 评进阶 · 约 120 分钟编辑精选一切资产的回报率，1870–2015（2019）
The Rate of Return on Everything, 1870–2015 (2019)
这篇发表于《经济学季刊》的论文构建了 1870–2015 年间 16 个主要发达国家的股票、债券、房地产等资产收益率的长周期数据库。
经济与社会科学2024 · 190 分 · 148 评进阶 · 约 80 分钟编辑精选想法是否越来越难找？
Are ideas getting harder to find?
探讨“想法是否越来越难找到”：通过衡量多个领域的“研究生产率”，发现维持相同的增长需要投入指数级增长的研究资源，即研究效率在持续下降。
经济与社会科学2016 · 182 分 · 106 评入门 · 约 60 分钟编辑精选比特币与区块链的经济极限
The Economic Limits of Bitcoin and the Blockchain
芝加哥大学 Eric Budish 的论文从经济学角度分析比特币与区块链的根本限制：核心论点是只要攻击网络（双花或砸盘做空）的收益超过挖矿成本，51% 算力攻击就有利可图，而挖矿…
经济与社会科学2018 · 176 分 · 37 评进阶 · 约 90 分钟编辑精选