Our article “Understanding the mobility chances of children from working‐class backgrounds in Britain: How important are cognitive ability and locus of control?” examines the role of cognitive ability and peoples’ sense of control over their lives in mediating the effects of individuals’ social background on their educational attainment and on their labor market position (Betthäuser et al., 2020a). The article takes as its starting point the persistent view in both academic and policy circles that most of the differences in the educational attainment and labor market success between individuals from different socio‐economic backgrounds are due to differences in cognitive ability between them (see e.g., Marks, 2014; Murray, 2012). Using data from the 1970 British Birth Cohort Study,¹ we find that cognitive ability mediates a non‐negligible yet limited amount of the effect of individuals’ social background on their educational attainment (about 35%) and their labor market position (about 20%). This means that about 65% of the effect of individuals’ social background on their educational attainment, and about 80% of the effect on their labor market position, is channeled by factors other than cognitive ability. Contradicting the claims by Murray (2012), Marks (2014), and others, this finding highlights that the intergenerational reproduction of social (dis)advantage that prevails in even the most developed societies is deeply unmeritocratic and unfair. Consequently, we see an urgent need for researchers to identify and for policy makers to address the channels through which individuals’ parental class background shapes their life chances, above and beyond its effects on individuals’ cognitive ability.

In his commentary on our article, Gary Marks (2020, p. 3) concludes that the findings of our article “are technically correct but unimportant.” He argues that examining the role of cognitive ability in mediating the association between individuals’ social background and their educational and labor market outcomes is not a relevant exercise, since there are “only moderate associations of class origins with educational and occupational outcomes” (p. 2). Instead, he suggests that research should focus on the importance of individuals’ genetic predisposition and cognitive ability in affecting individuals’ educational attainment and labor market outcomes (p. 2). In short, Marks takes issue, not with the substance and the findings of our article, but with the research question we pose and with our motivation for addressing it.^2,3 Our reply, therefore, focuses on why, in our view, it is imperative for social scientists across different disciplines to critically examine the association between individuals social background and their educational and labor market outcomes, and to understand the role of different factors—including cognitive ability—in accounting for this association.

We strongly disagree with the claim by Marks that there are “only moderate associations of class origins with educational and occupational outcomes” (p. 2). Research on social stratification and mobility in sociology, economics, and psychology has demonstrated that individuals’ social background continues to yield a strong influence on both their educational and labor market chances (Bukodi and Goldthorpe, 2018; Chetty et al., 2014; Laurison & Friedman, 2016; Major & Machin, 2018; Von Stumm et al., 2009). To use the analogy of John Rawls (1971), the ticket that people draw in the “lottery of birth” continues to matter for their life chances. This is true even in the most democratically and economically advanced societies, such as Britain. To illustrate this, Figure 1 shows the extent to which individuals’ social background affects their educational attainment. More specifically, it depicts the chances of individuals with the same level of cognitive ability but from different social backgrounds to attain an upper secondary or a higher level of qualification in Britain. We show this separately for women (right panel) and for men (left panel) and for different birth cohorts, spanning the last five decades. Individuals are split into three groups, based on their parents’ social class, social status, and educational attainment. The most advantaged group (10% in the earliest cohort and 27% in the most recent cohort) are predominantly the children of parents in the managerial and professional salariat or at least in white‐collar occupations who have tertiary‐ or at least upper secondary‐level qualifications. The least advantaged group (50% in the earliest cohort and 30% in the most recent cohort) are predominantly the children of parents in wage‐earning, mainly blue‐collar occupations with no qualifications or at best only ones at a lower secondary level. What this figure clearly shows is that individuals’ social background significantly shapes their educational attainment and that this is so even when we compare people who have the same level of cognitive ability. By way of example, a woman with an intermediate level of cognitive ability born in 1990, who comes from the most advantaged social background is, on average, about forty percentage points more likely to attain a qualification at the upper‐secondary level or above, compared to a woman with the same level of cognitive ability who comes from the least advantaged social background. For men, this difference is even more pronounced. There is no doubt that this constitutes a very substantial effect on individuals’ social background on their educational chances. And notably, this gap in the educational attainment of people coming from different social backgrounds has persisted over time. To claim that there are only moderate associations between peoples’ social origins and their educational and occupational outcomes, as Marks does, is, therefore, a misrepresentation of the evidence. To suggest that this association does not warrant further research is unfounded and neglects the responsibility of researchers to focus on what in our view is one of the key challenges that societies continue to face: to equalize the highly unequal playing field faced by individuals from different social backgrounds.

The substantial gap in life chances between people from different social backgrounds is problematic both from a normative perspective and from an efficiency point of view. Seen from a normative perspective, it is socially unjust if individuals’ life chances depend on ascriptive characteristics, which are out of their control, such as the social circumstances into which they are born (Rawls, 1971). From an efficiency perspective, it is undesirable that individuals’ educational attainment and the type of job they have depends on their family background, rather than their ability and skills (Betthäuser, 2017; Gray and Moshinsky, 1935). In contrast to what Marks claims, these are important grounds for why it should be a top priority for researchers across different disciplines to study how and why peoples’ social background continues to exert such a strong effect on their life chances. Examining, as we do in our article, the extent to which the gap in the educational attainment and labor market success between individuals from different socio‐economic backgrounds are due to differences in cognitive ability between them is one important step toward this larger goal.

In his commentary, Marks claims that we “misrepresent the research [we] take issue with” (p. 2). Here we would like to ask the reader to examine the arguments made by Murray, Marks, and others, which we set out to test empirically in our article. As we note in the article, Murray argues that “the reason that upper‐middle‐class children dominate the population of elite schools is that the parents of the upper‐middle class now produce a disproportionate number of the smartest children” (2012, p. 60). Murray suggests that this further explains why individuals in advantaged labor market positions largely come from higher social class backgrounds (2012, pp. 46–68). He also posits that the transmission of intelligence is largely genetic and is reinforced by increasing homogamy, that is, the growing tendency of people to form partnerships with individuals of similar social standing (2012, pp. 46–68). These views are echoed by a number of sociologists and social psychologists (see, e.g., Gottfredson, 2003; Marks, 2014; Plomin, 2018; Saunders, 1997, 2012). With regards to the effect of individuals’ social background on their educational attainment, for instance, Marks (2014, p. 88) writes that “the inclusion of [cognitive] ability in the analysis reduces the impact of socioeconomic background considerably and in some cases to statistical insignificance.” With respect to the effects of individuals’ social background on occupational and economic outcomes, he further contends that “the direct impact of socioeconomic background is even smaller, and smaller again after taking into account educational attainment and, to a lesser extent, cognitive ability” (Marks, 2014, p. 234). These arguments clearly advance the claim that the differences in the educational attainment and labor market success between individuals from different socio‐economic backgrounds are due to differences in cognitive ability between them.

We believe that it is important to empirically test the arguments made by Marks and others, particularly because of their political potency. A large mediating role of cognitive ability can be (mis‐) interpreted to imply that the pronounced inequality in educational and labour market attainment between individuals from different social backgrounds is somehow efficient or legitimate and does not require political intervention. For example, as we note in our article, Dominic Cummings, who was the special advisor to the British Secretary of State for Education, Michael Gove, and Chief Special Advisor to the Prime Minister, Boris Johnson, contends that “differences in educational achievement are not mainly because of ‘richer parents buying greater opportunity’” and suggests that they are instead due to richer parents having more capable children than poorer parents (Cummings, 2013, p. 74). Cummings further argues that any policy that aims at equalizing educational opportunities and increasing the quality of education would increase the effect of children's social background on their educational achievement, thereby questioning the importance of lowering the effect of children's social background on their educational attainment as a policy priority (ibid.). In our article and our related work, we show that these arguments are unfounded and that equalising educational opportunities can substantially reduce the effect of children's social background on their educational achievement and labour market chances (Betthäuser, 2017; Betthäuser et al., 2020a).

In sum, the evidence clearly shows that individuals’ social background continues to exert a strong influence on their life chances. Understanding how and why this occurs is of utmost importance from a scientific, from a normative, from efficiency, and from a policy point of view. As a step toward this larger goal, our article focuses on examining the extent to which differences in cognitive ability account for the gap in educational and labor market attainment between individuals from different social backgrounds. We find that cognitive ability plays a relatively modest role—a far more limited one than the Murray (2012), Marks (2014), and others suggest—in accounting for the substantial gap in life chances between individuals from different social backgrounds. Clearly, this raises the question through which other ways peoples’ social background continues to shape their life chances. We urge researchers from all disciplines to improve our understanding of this question through careful empirical analysis. Moreover, we believe that it is the responsibility of policy makers from across the political spectrum to use this knowledge to level the highly inequitable playing field that people from different social backgrounds continue to face.

我们的文章“了解英国工人阶级背景的儿童的流动机会：认知能力和控制点有多重要？” 研究认知能力和人们对生活的控制感在调节个人社会背景对其教育程度和劳动力市场地位的影响方面的作用（Betthäuser 等，  2020a）。这篇文章以学术界和政策界的一贯观点为出发点，即来自不同社会经济背景的个人之间在教育程度和劳动力市场成功方面的差异，大部分是由于他们之间的认知能力差异造成的（参见例如，马克斯，  2014 年；默里，  2012 年）。使用 1970 年英国出生队列研究^{1 的}数据，我们发现认知能力在个人社会背景对其教育程度（约 35%）和劳动力市场地位（约 20% ）。这意味着个人的社会背景对其教育程度的影响，约 65%，对劳动力市场地位的影响，约 80%，是由认知能力以外的因素引起的。与 Murray ( 2012 ), Marks ( 2014 )的主张相矛盾) 和其他人，这一发现突出表明，即使在最发达的社会中，社会（不利）优势的代际再生产也是非常不择优和不公平的。因此，我们认为研究人员迫切需要确定和政策制定者解决个人父母阶级背景塑造他们生活机会的渠道，超越其对个人认知能力的影响。

在他对我们文章的评论中，Gary Marks ( 2020，第 3) 得出结论，我们文章的发现“在技术上是正确的，但并不重要”。他认为，研究认知能力在调节个人社会背景与其教育和劳动力市场结果之间的关联方面的作用并不是一项相关的工作，因为“阶级起源与教育和职业结果之间只有中等程度的关联”（第 19 页）。 2）。相反，他建议研究应该关注个人的遗传倾向和认知能力在影响个人教育程度和劳动力市场结果方面的重要性（第 2 页）。简而言之，马克斯提出的问题不是我们文章的实质和结果，而是我们提出的研究问题以及我们解决它的动机。^{2 , 3}因此，我们的答复集中在为什么我们认为不同学科的社会科学家必须批判性地研究个人社会背景与其教育和劳动力市场结果之间的关联，并了解不同因素的作用——包括认知能力——说明这种关联。

我们强烈不同意 Marks 的说法，即“阶级起源与教育和职业成果之间只有中等程度的关联”（第 2 页）。社会学、经济学和心理学对社会分层和流动性的研究表明，个人的社会背景继续对其教育和劳动力市场机会产生强大的影响（Bukodi 和 Goldthorpe，2018 年；Chetty 等人，  2014 年；Laurison & Friedman，2016 年；Major & Machin，  2018 年；Von Stumm 等人，  2009 年）。使用约翰·罗尔斯（John Rawls，1971 年）的类比)，人们在“出生彩票”中抽到的票对他们的生活机会仍然很重要。即使在最民主和经济最先进的社会，如英国，也是如此。为了说明这一点，图 1 显示了个人的社会背景对其教育程度的影响程度。更具体地说，它描述了具有相同认知能力水平但来自不同社会背景的个体的机会在英国获得高中或更高水平的资格。我们分别针对女性（右图）和男性（左图）以及跨越过去五年的不同出生队列展示了这一点。根据父母的社会阶层、社会地位和教育程度，个人被分为三组。最有优势的群体（最早的队列中有 10%，最近的队列中有 27%）主要是父母在管理和专业薪水或至少在白领职业中拥有高等教育或至少高中的子女级别资格。最不利的群体（最早的队列为 50%，最近的队列为 30%）主要是挣工资的父母的子女，主要是没有资格或最多只有初中水平的蓝领职业。这个数字清楚地表明，个人的社会背景显着影响了他们的教育程度，即使我们比较那些拥有相同水平的认知能力. 例如，1990 年出生、具有中等认知能力的女性来自最有利的社会背景，获得高中或以上学历的可能性平均高出约 40 个百分点，与来自最不利社会背景的具有相同认知能力水平的女性相比。对于男性来说，这种差异更加明显。毫无疑问，这对个人的社会背景对其受教育机会产生了非常重大的影响。值得注意的是，随着时间的推移，来自不同社会背景的人在教育程度方面的这种差距一直存在。因此，声称人们的社会出身与他们的教育和职业成果之间只有适度的关联，正如马克斯所做的那样，对证据的歪曲。认为这种关联不值得进一步研究是没有根据的，并且忽视了研究人员的责任，即关注我们认为社会继续面临的主要挑战之一：平衡来自不同社会的个体所面临的高度不平等的竞争环境。背景。

从规范的角度和效率的角度来看，来自不同社会背景的人之间生活机会的巨大差距都是有问题的。从规范的角度来看，如果个人的生活机会取决于他们无法控制的归因特征，例如他们出生的社会环境，那么在社会上是不公正的（Rawls，  1971）。从效率的角度来看，个人的受教育程度和他们所从事的工作类型取决于他们的家庭背景而不是他们的能力和技能是不可取的（Betthäuser，  2017；Gray and Moshinsky，  1935）。与马克斯声称的相反，这些是为什么不同学科的研究人员应该优先研究人们的社会背景如何以及为什么继续对他们的生活机会产生如此强烈的影响的重要理由。正如我们在文章中所做的那样，检查来自不同社会经济背景的个人之间的教育程度和劳动力市场成功的差距在多大程度上是由于他们之间认知能力的差异导致的，这是朝着这个更大目标迈出的重要一步。

在他的评论中，马克斯声称我们“歪曲了[我们]提出异议的研究”（第 2 页）。在这里，我们想请读者检查 Murray、Marks 和其他人提出的论点，我们开始在文章中进行实证检验。正如我们在文章中所指出的，默里认为“中上阶层儿童在精英学校人口中占主导地位的原因是中上阶层的父母现在培养出不成比例的最聪明的孩子”（2012，p . 60）。Murray 认为，这进一步解释了为什么处于有利劳动力市场地位的个人主要来自较高的社会阶层背景（2012 年），第 46-68 页）。他还假设，智力的传递主要是遗传的，并且通过日益增加的同性婚姻而得到加强，即人们与具有相似社会地位的个人形成伙伴关系的日益增长的趋势（2012 年，第 46-68 页）。这些观点得到了许多社会学家和社会心理学家的赞同（参见，例如 Gottfredson，  2003 年；Marks，  2014 年；Plomin，  2018 年；Saunders，  1997 年，2012 年）。关于个人的社会背景对其受教育程度的影响，例如，Marks ( 2014，第 88) 写道，“在分析中纳入 [认知] 能力可以显着降低社会经济背景的影响，在某些情况下甚至会降低统计意义。” 关于个人的社会背景对职业和经济结果的影响，他进一步认为“社会经济背景的直接影响甚至更小，而且在考虑到教育程度和认知能力后又小了一点” （马克，  2014 年，第 234 页）。这些论点清楚地提出了这样一种说法，即来自不同社会经济背景的个人之间的教育程度和劳动力市场成功的差异是由于他们之间认知能力的差异。

我们认为，对 Marks 和其他人提出的论点进行实证检验很重要，特别是因为他们的政治效力。认知能力的巨大中介作用可以被（错误地）解释为意味着来自不同社会背景的个人之间在教育和劳动力市场上的明显不平等在某种程度上是有效的或合法的，并且不需要政治干预。例如，正如我们在文章中指出的，英国教育大臣迈克尔戈夫的特别顾问多米尼克卡明斯和首相鲍里斯约翰逊的首席特别顾问， 2013 年，第。74)。卡明斯进一步认为，任何旨在平等教育机会和提高教育质量的政策都会增加儿童社会背景对其教育成就的影响，从而质疑将降低儿童社会背景对其教育成就的影响作为一项政策的重要性。优先权（同上）。在我们的文章和我们的相关工作中，我们表明这些论点是没有根据的，教育机会均等可以大大降低儿童的社会背景对其教育成就和劳动力市场机会的影响（Betthäuser，2017 年；Betthäuser 等人，2020a）。

总之，证据清楚地表明，个人的社会背景继续对其生活机会产生强大的影响。从科学、规范、效率和政策的角度来看，了解这种情况如何以及为什么会发生至关重要。作为朝着这个更大目标迈出的一步，我们的文章重点研究了认知能力的差异在多大程度上解释了不同社会背景的个人之间的教育和劳动力市场成就差距。我们发现认知能力的作用相对较小——远比 Murray ( 2012 )、Marks ( 2014)) 和其他人建议——解释来自不同社会背景的个人之间生活机会的巨大差距。显然，这提出了一个问题，人们的社会背景通过哪些其他方式继续塑造他们的生活机会。我们敦促所有学科的研究人员通过仔细的实证分析来提高我们对这个问题的理解。此外，我们认为，来自不同政治领域的政策制定者有责任利用这些知识来平衡来自不同社会背景的人们继续面临的高度不公平的竞争环境。