Abstract

A recent Public Opinion Quarterly article argued that men whose first children were girls rather than boys—daughters rather than sons—were more likely to support gender-equality policies. This note reexamines the coding and model specification of that analysis, focusing on the definition of “first daughters” and on the inclusion, perilously for causal inference, of control variables that are more probably consequences than causes of the independent variable of interest. In the tested alternative specifications, the predicted effect of first-child sex diminishes, usually falling short of standard benchmarks for statistical significance.

中文翻译：

---

重温“初女效应”

摘要

最近《民意季刊》的一篇文章认为，第一个孩子是女孩而不是男孩的男孩（女儿而不是儿子）更有可能支持性别平等政策。本说明重新检查了该分析的编码和模型规范，重点关注“第一个子项”的定义，以及出于因果推论危险地包含控制变量的可能性，这些控制变量可能比引起关注的自变量的原因多。在经过测试的替代规范中，对初生孩子的性行为的预期影响会减弱，通常在统计学意义上低于标准基准。