榴莲视频

为实现性别薪酬平等,高校必须让社会科学家看到数据

<榴莲视频 class="standfirst">苏尼尔·米特拉·库马尔(Sunil Mitra Kumar)称,隐私问题和管理主义阻碍了高校充分利用它们的研究专长
四月 28, 2022
Door marked "private"
Source: Getty

点击阅读英文原文


2021年,英国高等教育的平均性别薪酬差距为。这一数字比开始实施强制报告制度的2017年高出约3个百分点,也高于英国所有行业和服务业14.9%的平均水平。我们必须进行改善。

从历史上看,高校一直站在调查性别薪酬差距和倡导变革的最前沿。在上世纪70年代的开创性工作中,加州大学伯克利分校(University of California, Berkeley)统计学教授伊丽莎白·斯科特(Elizabeth Scott)说服雇主公开了详细薪资数据。在煞费苦心地整理了这些数据后,斯科特利用回归模型建立了薪酬、性别和许多其他变量之间的关系,以了解性别薪酬差距的性质和原因。

如今,薪酬和就业方面的性别差异在经济学、社会学、性别研究、人类学等领域引起了广泛关注。这些学科提供了定量工具来确定上述差距的驱动因素(如学历水平对直接歧视的影响),也提供了定性工具来理解更微妙的根本原因(比如如何解释女性不愿意要求加薪)。

学者们还开创性地研究了工作和就业以外的不平等问题:例如,医疗保健方面的种族不平等,以及获得信贷方面的性别不平等。你可能会认为,促进学术薪酬平等这样的工作符合高校的既得利益,尤其是在这个人们日益意识到薪酬差距和不平等的时代。然而,与半个世纪前相比,伊丽莎白·斯科特的工作在今天似乎更具开创性。

简单地说,高校不愿意公开详细薪资数据,以让自己的教职工研究薪酬差距。这种沉默在一定程度上是可以理解的,这是由于工资信息普遍的敏感性。但更重要的因素是现代高校的管理方式。除了一些小的例外情况(比如在教育系中),教师的研究专长都坚定地指向外部,而具有敏感性的内部问题则由顾问和管理人员处理。处理——而非研究。

比如,在我们学校,一个非学术性的“人员数据”团队负责处理对薪资数据的查询。该团队的存在是整个学校扩大数据分析推动力的一部分,它使几种类型的数据更容易获得,如按种族和性别划分的薪酬和学生成绩的汇总统计。但是,对这背后的详细数据的访问是被禁止的。这就阻碍了实际的研究,使得薪酬差距更多地需要从法律合规、市场营销和高校的其他外向方面来衡量和报告。

详细数据对于回答两种互补类型的问题是必要的。第一个是定量的,与年龄、性别、种族和学科相结合,这些因素与薪酬和晋升结果有关。这些只能用多元回归技术来建模。例如,识别导致在申请晋升时犹豫不决的个人经历的重复模式,或识别最有可能通过面试、得到工作并获得更高薪水的个人形象。机器学习技术甚至可以走得更远,在分析中纳入几十个变量,比如职业休假的时间和长度。

第二种类型的问题是定性的,与这些组合为什么决定这些结果,以及如何决定有关。了解这些因素也有助于关键的下一步:确定哪些行动可以改善结果,并针对变化建立现实的预期。例如,资深导师是否需要有目的地与初级教师进行配对,并每年都与他们进行晋升相关的谈话?回答这些问题需要采访,但要获得成果,采访必须建立在一个全面的定量理解上。

对薪资信息的隐私的担忧是真实的,尤其是在高校这一领域的法律义务方面,但我们有办法处理这一问题。我们可以要求希望处理和分析数据的研究人员签署保密协议,并要求那些希望就薪酬和就业进行采访的研究人员获得道德许可。我们还可以聘请外部研究人员,以减轻采访自己同事时的复杂性。

其次,一个折衷的解决方案是建立这样的统计软件,让研究人员在没有访问基础数据的情况下指定和估计模型。第叁,数据模拟技术可以用于创建支持匿名的伪数据,但保留原始数据的相关性结构,从而支持统计分析。

诚然,对于薪资数据如何处理的争论,与学术界对于管理主义的更广泛的讨论有关。但与管理主义提出的一些问题不同,理解和解决薪酬差距是我们力所能及的。暴露在聚光灯下可能会让人不舒服,但允许研究这些问题的学者研究自己的高校,将产生实质性的见解,并促进全面的平等。

苏尼尔·米特拉·库马尔是伦敦大学国王学院(King’s College London)印度研究所和国际发展部经济学高级讲师。

本文由陆子惠为泰晤士高等教育翻译。

<榴莲视频 class="pane-title"> 后记

Print headline:?For gender pay equality, social scientists must have access to the data

请先注册再继续

为何要注册?

  • 注册是免费的,而且十分便捷
  • 注册成功后,您每月可免费阅读3篇文章
  • 订阅我们的邮件
注册
Please 登录 or 注册 to read this article.
<榴莲视频 class="pane-title"> 相关文章

即便具有男性同等绩效评级的女性也会遭受巨大薪酬不公。安·布鲁尔(Ann Brower)和亚历克斯·詹姆斯(Alex James)称,这个问题必须得到解决

2月 27日
<榴莲视频 class="pane-title"> Reader's comments (7)
Great article - and I fully agree. This is the problem with many institutions’ “outward-facing” reviews, which seem to find that the pay gap is all down to seniority and that men and women doing equal work are not being paid differently. Without the actual detailed data, there is no way for us to determine whether this is correct. The focus on compliance rather than real, granular research obscures the actual scale, dynamics and causes of the pay gap - and is a waste of the fantastic resource of researchers within HE institutions who could make a real difference to understanding the issue. “Handled - but not researched” describes perfectly what is happening at the moment.
I think opening up the data is essential in understanding and addressing the pay inequality gaps in Higher Education and this is precisely why it will never happen. What would be the outcome when Researchers uncover differences in pay within gender, race, age groups etc. as well as between them- Universities might then have to consider the premise of equality for all rather than simply looking at the agenda of the week or the best marketing opportunity ? Could salary data also be used for cost benefit analysis ( Do we really need another level of management , does the VC warrant a ?500k salary, what have the bonuses been paid for etc. ?) It is a can of worms that will never be opened.
Sitting on job grading panels as a Trades Union rep it's clear most new jobs we deal with are now pretty much aligned with the Universities equality targets requirements, regrading however is a different matter, with some applications still being obviously loaded for favoured people by their management (we have torn a number of those to shreds, even when H.R. has tried to persuade us not to). So we would welcome a thorough examination of the data, at ALL level's, especially for those jobs at the highest levels where the remuneration is agreed by 'personal negotiation' or set by council, not through any grading scheme. Talking of which, we are unfortunate to be locked into using the HAY scheme (thanks to the Reverend), with it's built in biases benefiting admin roles, rather than the sector developed HERA scheme, I wonder if that is also a factor?
The comment above.."obviously loaded for favoured people by their management (we have torn a number of those to shreds, even when H.R. has tried to persuade us not to)." hits the nail on the head. In many places HR managers do not have the spine to overrule anything that the deans or other pesudo managers want them to do, even when it is obvious that it may not be entirely consistent with policy or precedence.
Indeed, and it becomes even more difficult and opaque when the managers can use the Staff Achievement Award scheme to bypass the Grading system and Trades Union oversight to have extra increments, including those in whats called the HRZ (higher responsibility zone, parallel pay points to the next grade up without regrading) and/or 'personal' awards (?1000 max and NOT recorded as pay in the normal way).
"Granular data are necessary to answer two complementary types of questions. The first is quantitative and relates to the combinations of age, gender, ethnicity and discipline that correlate with pay and promotion outcomes." I wonder if Sunil is aware we already do exactly this at Kings'? This is published in the Gender and Ethnicity report and is shared on a granular level with the managers with responsibility for pay decisions. Reward specialists in many organisations use multivariate regression analysis, this technique is not restricted to academics.
Nepotism and mindless boxticking still rule in many places and in too many instances than necessary. Worse still are those who are overlooked for promotions etc. because they dare to speak up against mismanagement. The governance systems in universities must be totally overhauled.