Intersectional Inequality: Race, Class, Test Scores, and Poverty
Intersectional Inequality: Race, Class, Test Scores, and Poverty
复制标题
交叉不平等：种族、阶级、考试成绩和贫困
DOI：
10.1177/0094306118779814hh
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
Sigal Alon
中科院分区：
文献类型：
--
作者：
Sigal Alon
交叉不平等：《种族、阶级、考试分数和贫困》一书探讨了社会不平等的核心本质，具体而言，就是社会中资产、商品和资源分配的重叠，这意味着一些人始终处于所有地位等级的顶端，而另一些人则处于多重劣势。多年来，这种现象在文献中被描述为交叉性（Hill柯林斯1990; Cotter等人1999）、累积缺点/优点（Merton 1968; DiPrete和Eirich 2006）、结晶（Grusky 2001）和重叠缺点/优点（Alon 2007）。许多人指出，关注这种非随机模式可以帮助我们理解分层系统的结构，以及从一代到下一代产生和维持分层系统的基本过程。然而，虽然经验智慧是“通过简单地检查我们通常控制的变量的非随机组合的性质，可以学到很多关于基本因果力的知识”（Lieberson 1985：211），传统的定量方法，即回归模型，旨在估计变量的净效应，同时“保持其他一切相等”。因此，我们通常使用的方法无法捕捉社会不平等的核心本质。在这本简短的书中，查尔斯·拉金和皮尔·菲斯利用了以前在《比较法》中开发的工具（Ragin 1987），模糊集社会科学（Ragin，2000）和重新设计社会调查：模糊集及超越（Ragin 2008）在理查德·赫恩斯坦和查尔斯·默里的《钟形曲线》（1994）中著名的辩论中发表意见。以及克劳德·菲舍尔及其同事的《设计造成的不平等》（1996）--关于考试成绩在多大程度上塑造了个人的生活机会，特别是在经历或避免贫困方面。在方法论上，根据Ragin和菲斯，钟形曲线和设计的不平等代表了社会科学研究的两种极端方法。一方面，钟形曲线的规范描述了一个“简单的画面”，只关注三个变量来预测贫困（考试成绩，社会经济地位和年龄）;而另一方面，设计的不平等模型是“除了厨房水槽之外的一切”，因为它包括29个独立变量。第二章对这两项研究进行了总结，第三章使用全国青年纵向调查数据集进行了重复分析。在评估这些模型的解释能力（有多少情况被正确分类）时，Ragin和菲斯的分析表明，与设计不平等的规范相比，钟形曲线的表现很差。然而，这一成就是以吝啬为代价的。Ragin和菲斯提出了一种中间路径方法，集中在六个理论上重要的解释变量（种族，性别，父母收入，父母教育，受访者的教育，考试成绩，和家庭组成），用于校准模糊集在第四章。这些变量被操纵，以便更好地捕捉社会世界的复杂性。贫穷二分变量（高于/低于官方贫穷线）被转换成两个虚拟变量，即贫穷和不贫穷，以便对贫穷的成员进行更细致的划分，而不把刚刚超过贫穷线的人与较富裕的人混为一谈。解释变量也得到同样的处理。在接下来的第5章到第7章中，拉金和菲斯展示了他们交叉研究方法的分析杠杆作用，因为他们权衡了种族、阶级、考试成绩和贫困之间复杂的、组合的和多维的联系。使用模糊集分析方法，他们评估集合重合度，即多重优势（不低考试分数，不低收入父母，受过教育，受过教育的父母）和多重劣势（不高考试分数，不高收入父母，不高教育，不高教育的父母）重合的程度。490评论
Intersectional Inequality: Race, Class, Test Scores, and Poverty deals with the core nature of social inequality, specifically, the overlap between the distribution of assets, goods, and resources in a society, which means that some individuals consistently appear at the top of all status hierarchies, while others suffer from multiple disadvantages. Over the years, this phenomenon was described in the literature as intersectionality (Hill Collins 1990; Cotter et al. 1999), cumulative dis/ advantages (Merton 1968; DiPrete and Eirich 2006), crystallization (Grusky 2001), and overlapping dis/advantages (Alon 2007). Many have pointed out that focusing on such nonrandom patterns can facilitate our understanding of the structuring of the stratification system and the underlying processes that produce and maintain it from one generation to the next. Yet, while the empirical wisdom is that ‘‘much can be learned about the basic causal force by simply examining the nature of the nonrandom combinations of variables for which we normally control’’ (Lieberson 1985:211), the conventional quantitative method—that is, the regression model—is designed to estimate the net effect of variables while ‘‘holding everything else equal.’’ Thus, the methods we typically use fail to capture the core nature of social inequality. In this short book, Charles Ragin and Peer Fiss harness the tools previously developed in The Comparative Method (Ragin 1987), Fuzzy-Set Social Science (Ragin 2000), and Redesigning Social Inquiry: Fuzzy Sets and Beyond (Ragin 2008) to weigh in on the well-known debate—featured in Richard Herrnstein and Charles Murray’s The Bell Curve (1994) and Claude Fischer and colleagues’ Inequality by Design (1996)— regarding the extent to which test scores shape the individual’s life chances, specifically in experiencing or avoiding poverty. Methodologically, according to Ragin and Fiss, The Bell Curve and Inequality by Design represent two extreme approaches for social science research. On the one side, The Bell Curve’s specification depicts a ‘‘simple picture’’ by focusing on only three variables to predict poverty (test scores, socioeconomic status, and age); while on the other side, Inequality by Design’s model is ‘‘everything but the kitchen sink,’’ as it includes 29 independent variables. Both studies are summarized in Chapter Two, and their analyses are replicated in Chapter Three using the National Longitudinal Survey of Youth dataset. Assessing the explanatory power of these models (how many cases are correctly classified), Ragin and Fiss’s analysis demonstrates that The Bell Curve’s performs poorly compared to Inequality by Design’s specification. Yet this achievement has come at the cost of parsimony. Ragin and Fiss suggest a middle-path approach by focusing on six theoretically important explanatory variables (race, gender, parental income, parental education, respondent’s education, test scores, and household composition) that are used to calibrate fuzzy sets in Chapter Four. These variables are manipulated so that they better capture the complexity of the social world. The dichotomous poverty variable (above/ below the official poverty threshold) is converted into two dummy variables, inpoverty and not-in-poverty, to allow a more nuanced membership in poverty that does not clump together those just above the poverty line with the more affluent. The explanatory variables receive the same treatment. In subsequent chapters (5 to 7) Ragin and Fiss demonstrate the analytic leverage of their intersectional approach as they weigh in on the complex, combinatorial, and multidimensional link between race, class, test scores, and poverty. Using fuzzy-set analytic methods, they assess the degree of set coincidence, that is, the degree to which multiple advantages (not-low-test-score, not-low-income-parents, educated, educated-parents) and multiple disadvantages (not-high-test-score, not-high-incomeparents, not-highly-educated, not-highlyeducated-parents) coincide. The findings 490 Reviews