Unfair Outcomes Research Articles

Obstacles to the Full Realization and Adoption of Artificial Intelligence (AI)

Artificial Intelligence (AI) holds transformative potential across various sectors, yet its journey to full realization and widespread adoption is impeded by several significant obstacles. These challenges span technical, ethical, societal, research and development, implementation, and regulatory domains, necessitating a comprehensive and multi-faceted approach to address them.Technical Challenges: AI systems require vast amounts of high-quality, labeled data, which is often unavailable or of poor quality. AI models also struggle to generalize beyond specific tasks, and their opaque decision-making processes hinder trust. Scaling AI to handle large datasets and interactions, while ensuring robustness and security against adversarial attacks, poses major technical hurdles.Ethical and Societal Challenges: AI can perpetuate and amplify biases, leading to unfair outcomes and privacy issues. AI-driven automation risks job displacement, necessitating measures for worker support and retraining. The rapid development of AI outstrips regulatory bodies' abilities to create appropriate frameworks, complicating responsible deployment.Research and Development Challenges: AI struggles to integrate information from different modalities and lacks human-like common sense and reasoning. Improving AI's learning efficiency to reduce its dependence on vast data is a significant research focus.Implementation Challenges: Integrating AI into legacy systems, addressing the talent shortage, and managing the high costs of AI development and deployment are substantial barriers.Ethical and Regulatory Uncertainty: The absence of universally accepted ethical guidelines and varying regulatory landscapes create uncertainty and potential misuse.Addressing these multifaceted obstacles requires a concerted effort from researchers, developers, policymakers, and society to create robust, fair, and transparent AI systems, unlocking AI's full potential and societal benefits.

Exploring bias risks in artificial intelligence and targeted medicines manufacturing

BackgroundThough artificial intelligence holds great value for healthcare, it may also amplify health inequalities through risks of bias. In this paper, we explore bias risks in targeted medicines manufacturing. Targeted medicines manufacturing refers to the act of making medicines targeted to individual patients or to subpopulations of patients within a general group, which can be achieved, for example, by means of cell and gene therapies. These manufacturing processes are increasingly reliant on digitalised systems which can be controlled by artificial intelligence algorithms. Whether and how bias might turn up in the process, however, is uncertain due to the novelty of the development.MethodsExamining stakeholder views across bioethics, precision medicine, and artificial intelligence, we document a range of opinions from eleven semi-structured interviews about the possibility of bias in AI-driven targeted therapies manufacturing.ResultFindings show that bias can emerge in upstream (research and development) and downstream (medicine production) processes when manufacturing targeted medicines. However, interviewees emphasized that downstream processes, particularly those not relying on patient or population data, may have lower bias risks. The study also identified a spectrum of bias meanings ranging from negative and ambivalent to positive and productive. Notably, some participants highlighted the potential for certain biases to have productive moral value in correcting health inequalities. This idea of “corrective bias” problematizes the conventional understanding of bias as primarily a negative concept defined by systematic error or unfair outcomes and suggests potential value in capitalizing on biases to help address health inequalities. Our analysis also indicates, however, that the concept of “corrective bias” requires further critical reflection before they can be used to this end.

Responses to the updated Nutri-Score algorithms in Norway: A qualitative study among food system actors in the NewTools-project.

Nutri-Score is a front-of-pack label grading foods and beverages from A to E indicating nutritional quality based on the foods' favorable and unfavorable components, and a contender in the ongoing debate on the possible implementation of a harmonized mandatory front-of-pack nutrition label in the European Union. NewTools is a research project on scoring systems for foods involving 28 partners representing actors involved in the Norwegian food system. This study aimed to explore views reported by Norwegian food system actors on the advantages and disadvantages with the updated Nutri-Score algorithms for food and beverages (2022-2023). This included Nutri-Score's performance in ranking foods according to the national food-based dietary guidelines and to the nutritional challenges in Norway. A total of 28 project partners and 15 other food system stakeholders following the NewTools-project were invited to provide responses on the Nutri-Score algorithms and their application on foods and beverages in the Norwegian food composition table. Thirteen written responses were received and analyzed with qualitative content analysis. The responses to the updated Nutri-Score varied in content, reflecting mainly concerns. Examples of perceived concerns included excessive penalty of salt content; insufficient differentiation based on fat content in meat, sausages, cheese, and milk; and several unreasonable comparisons across food categories. They also expressed a concern that Nutri-Score may stimulate to increased food processing, and some reported inconsistencies between Nutri-Score's classification of foods and national nutrition guidelines and policies. Several concerns with the updated Nutri-Score algorithms were raised, including the weighting of specific nutrients, unfair outcomes when comparing across food categories, and inconsistencies with established Norwegian nutrition guidelines and policies. The results should be interpreted with caution, as some perspectives from the Norwegian food system may be missing.

Detection of Bias in Machine Learning Models for Predicting Deaths Caused by COVID-19

The COVID-19 pandemic has significantly impacted global health, resulting in numerous fatalities and presenting substantial challenges to national healthcare systems due to a sharp increase in cases. Key to managing this crisis is the rapid and accurate identification of COVID-19 infections, a task that can be enhanced with Machine Learning (ML) techniques. However, ML applications can also generate biased and potentially unfair outcomes for certain demographic groups. This paper introduces a ML model designed for detecting both COVID-19 cases and biases associated with specific patient attributes. The model employs Decision Tree and XGBoost algorithms for case detection, while bias analysis is performed using the DALEX library, which focuses on protected attributes such as age, gender, race, and ethnicity. DALEX works by creating an "explainer" object that represents the model, enabling exploration of the model's functions without requiring in-depth knowledge of its workings. This approach helps pinpoint influential attributes and uncover potential biases within the model. Model performance is assessed through accuracy metrics, with the Decision Tree algorithm achieving the highest accuracy at 99% following Bayesian hyperparameter optimization. However, high accuracy does not ensure fairness, as biases related to protected attributes may still persist.

When Job Opportunities Go Up in Smoke

Abstract: Employment discrimination based on job-unrelated factors (e.g., gender, smoking status) can lead to unfair outcomes for applicants. In the present study, 400 Canadian and US hiring professionals evaluated a candidate’s resume and then cybervetted their social media page which disclosed their gender and smoking status (i.e., cigarette smoker, vaper, or nonsmoker). Revised evaluations post-cybervetting were lower for applicants discovered as smokers and vapers than for nonsmokers, but vapers were perceived as negatively as smokers. Negative evaluations for cigarette smokers depended on raters’ attitudes toward smoking. Applicant gender moderated the effect of vaping but not smoking on evaluations. Managers’ gender role beliefs did not play a moderating role. These findings indicate the need for improvements in cybervetting practices used in hiring.

EXPRESS: Fairness as a Robust Utilitarianism

This article focuses on the social choice problem in which decisions are based on the utility of multiple stakeholder types. The sum of these utilities – Utilitarian welfare - is one of the most important objectives in solving the social choice problem. While it is the most efficient solution, maximizing Utilitarian welfare may lead to unfair outcomes. However, Encouraging a Utilitarian decision-maker to adopt a fair decision is challenging due to the associated efficiency loss. This article takes a novel perspective by motivating a Utilitarian decision-maker to make fair decisions from an uncertainty-averse standpoint. We study the problem where the proportions of stakeholder types are uncertain and propose a distributionally robust optimization (DRO) model that maximizes the worst-case Utilitarian welfare over an ϕ-divergence-based uncertainty set. We provide three aspects of the relationship between fairness and the uncertain-averse Utilitarian welfare maximization. First, we establish that the worst-case Utilitarian welfare adheres to all five axioms of unfairness-averse cardinal welfare functions with two stakeholder types and satisfies four of these when this number exceeds two. Second, we demonstrate that with the maximal extent of uncertainty aversion, the DRO model identifies the Egalitarian welfare maximizer, which prioritizes fairness. Further, given serveral conventional assumptions, the proposed model selects the Nash welfare maximizer, an objective trade-off between efficiency and fairness, with moderate levels of uncertainty aversion. Lastly, we present numerical studies of two specific instances of the social choice problem – resource allocation and facility location – to show that, as uncertainty aversion increases, our model increasingly emphasizes fairness.

Procedural Fairness as Stepping Stone for Successful Implementation of Algorithmic Decision-Making in Public Administration: Review and Outlook

Algorithmic decision-making (ADM) is becoming more and more prevalent in everyday life. Due to their promise of producing faster, better, and less biased decisions, automated and data-driven processes also receive increasing attention in many different administrative settings. However, as a result of human mistakes ADM also poses the threat of producing unfair outcomes. Looming algorithmic discrimination can undermine the legitimacy of administrative decision-making. While lawyers and lawmakers face the age-old question of regulation, many decision-makers tasked with designing ADM for and implementing ADM in public administration wrestle with harnessing its advantages and limiting its disadvantages. “Algorithmic fairness” has evolved as key concept in developing algorithmic systems to counter detrimental outcomes. We provide a review of the vast literature on algorithmic fairness and show how key dimensions alter people’s perception of whether an algorithm is fair. In doing so, we provide entry point into this literature for anybody who is required to think about algorithmic fairness, particularly in an public administration context. We also pinpoint critical concerns about algorithmic fairness that public officials and researchers should note.

Implementing equitable and intersectionality‐aware ML in education: A practical guide

AbstractThe greater the proliferation of AI in educational contexts, the more important it becomes to ensure that AI adheres to the equity and inclusion values of an educational system or institution. Given that modern AI is based on historic datasets, mitigating historic biases with respect to protected classes (ie, fairness) is an important component of this value alignment. Although extensive research has been done on AI fairness in education, there has been a lack of guidance for practitioners, which could enhance the practical uptake of these methods. In this work, we present a practitioner‐oriented, step‐by‐step framework, based on findings from the field, to implement AI fairness techniques. We also present an empirical case study that applies this framework in the context of a grade prediction task using data from a large public university. Our novel findings from the case study and extended analyses underscore the importance of incorporating intersectionality (such as race and gender) as central equity and inclusion institution values. Moreover, our research demonstrates the effectiveness of bias mitigation techniques, like adversarial learning, in enhancing fairness, particularly for intersectional categories like race–gender and race–income. Practitioner notesWhat is already known about this topic AI‐powered Educational Decision Support Systems (EDSS) are increasingly used in various educational contexts, such as course selection, admissions, scholarship allocation and identifying at‐risk students. There are known challenges with AI in education, particularly around the reinforcement of existing biases, leading to unfair outcomes. The machine learning community has developed metrics and methods to measure and mitigate biases, which have been effectively applied to education as seen in the AI in education literature. What this paper adds Introduces a comprehensive technical framework for equity and inclusion, specifically for machine learning practitioners in AI education systems. Presents a novel modification to the ABROCA fairness metric to better represent disparities among multiple subgroups within a protected class. Empirical analysis of the effectiveness of bias‐mitigating techniques, like adversarial learning, in reducing biases in intersectional classes (eg, race–gender, race–income). Model reporting in the form of model cards that can foster transparent communication among developers, users and stakeholders. Implications for practice and/or policy The fairness framework can act as a systematic guide for practitioners to design equitable and inclusive AI‐EDSS. The fairness framework can act as a systematic guide for practitioners to make compliance with emerging AI regulations more manageable. Stakeholders may become more involved in tailoring the fairness and equity model tuning process to align with their values.

Examining Ethical Aspects of AI: Addressing Bias and Equity in the Discipline

he rapid progress in implementing Artificial Intelligence (AI) across various domains such as healthcare decision-making, medical diagnosis, and others has raised significant concerns regarding the fairness and bias embedded within AI systems. This is particularly crucial in sectors like healthcare, employment, criminal justice, credit scoring, and the emerging field of generative AI models (GenAI) producing synthetic media. Such systems can lead to unfair outcomes and perpetuate existing inequalities, including biases ingrained in the synthetic data representation of individuals.This survey paper provides a concise yet comprehensive examination of fairness and bias in AI, encompassing their origins, ramifications, and potential mitigation strategies. We scrutinize sources of bias, including data, algorithmic, and human decision biases, shedding light on the emergent issue of generative AI bias where models may replicate and amplify societal stereotypes. Assessing the societal impact of biased AI systems, we spotlight the perpetuation of inequalities and the reinforcement of harmful stereotypes, especially as generative AI gains traction in shaping public perception through generated content.Various proposed mitigation strategies are explored, with an emphasis on the ethical considerations surrounding their implementation. We stress the necessity of interdisciplinary collaboration to ensure the effectiveness of these strategies. Through a systematic literature review spanning multiple academic disciplines, we define AI bias and its various types, delving into the nuances of generative AI bias. We discuss the adverse effects of AI bias on individuals and society, providing an overview of current approaches to mitigate bias, including data preprocessing, model selection, and post-processing. Unique challenges posed by generative AI models are highlighted, underscoring the importance of tailored strategies to address them effectively.Addressing bias in AI necessitates a holistic approach, involving diverse and representative datasets, enhanced transparency, and accountability in AI systems, and exploration of alternative AI paradigms prioritizing fairness and ethical considerations. This survey contributes to the ongoing discourse on developing fair and unbiased AI systems by outlining the sources, impacts, and mitigation strategies related to AI bias, with a particular focus on the burgeoning field of generative AI.

Ethical Considerations in the Development and Deployment of AI Systems

Purpose: Complex artificial intelligence algorithms may make it hard to understand how they reach certain conclusions or decisions. Lack of transparency raises concerns about bias, discrimination, and opacity, all of which may detract from trust in AI systems. Businesses and developers should prioritize creating AI systems that are easy to understand and explain so that users can understand the reasoning behind their results. Second, fairness and nondiscrimination are fundamental principles. It is possible for AI systems to unintentionally provide biased or unfair outcomes by reinforcing or amplifying biases already seen in training data. Make sure that AI systems are trained on diverse and representative datasets and that they are tested for bias often; this is of the highest significance.
 Materials and Methods: By analyzing the co-occurrence of keywords, we can see that there are recurring themes when it comes to AI ethics. These topics include big data, social value, algorithms, and ethical aspects.
 Findings: Critical works that have had a major impact on the field may be found using citation analysis. The results shed light on how AI ethics is always changing as a result of several factors coming together, such as the social effect of technology and the management of stakeholders.
 Implications to Theory, Practice and Policy: Researchers, legislators, and practitioners may all benefit from the study's findings, which will help direct the development of AI in a way that is ethical and consistent with human values.

Changing Dimensions of Victims’ Rights in The Era of Technology and Innovation within the Criminal Justice System

The current study investigates how victims' rights are changing, as the technology advances. The incorporation of innovation and technology into legal systems is critical for ensuring international peace and order in accordance with Sustainable Development Goal 16, which emphasises universal access to justice. Governments must regulate social behaviour and ensure the safety and property of their constituents. In India, incidents arise in which accused persons gain from a lack of evidence or evidence that has been tampered with, leaving victims with unfair outcomes, and potentially disrupting social order. This study seeks to investigate the underlying causes of maladministration of justice and violations of victims' rights. It will also investigate potential cures, particularly in terms of technological improvements. This study uses a doctrinal legal research technique to investigate relevant national and international laws and policies. This study contributes to the discussion on improving legal mechanisms by shining a light on existing justice delivery systems and pushing for technological integration. The study acknowledges and expands on prior scholarly work in this topic to provide comprehensive insights.

Fairness and Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, and Mitigation Strategies

The significant advancements in applying artificial intelligence (AI) to healthcare decision-making, medical diagnosis, and other domains have simultaneously raised concerns about the fairness and bias of AI systems. This is particularly critical in areas like healthcare, employment, criminal justice, credit scoring, and increasingly, in generative AI models (GenAI) that produce synthetic media. Such systems can lead to unfair outcomes and perpetuate existing inequalities, including generative biases that affect the representation of individuals in synthetic data. This survey study offers a succinct, comprehensive overview of fairness and bias in AI, addressing their sources, impacts, and mitigation strategies. We review sources of bias, such as data, algorithm, and human decision biases—highlighting the emergent issue of generative AI bias, where models may reproduce and amplify societal stereotypes. We assess the societal impact of biased AI systems, focusing on perpetuating inequalities and reinforcing harmful stereotypes, especially as generative AI becomes more prevalent in creating content that influences public perception. We explore various proposed mitigation strategies, discuss the ethical considerations of their implementation, and emphasize the need for interdisciplinary collaboration to ensure effectiveness. Through a systematic literature review spanning multiple academic disciplines, we present definitions of AI bias and its different types, including a detailed look at generative AI bias. We discuss the negative impacts of AI bias on individuals and society and provide an overview of current approaches to mitigate AI bias, including data pre-processing, model selection, and post-processing. We emphasize the unique challenges presented by generative AI models and the importance of strategies specifically tailored to address these. Addressing bias in AI requires a holistic approach involving diverse and representative datasets, enhanced transparency and accountability in AI systems, and the exploration of alternative AI paradigms that prioritize fairness and ethical considerations. This survey contributes to the ongoing discussion on developing fair and unbiased AI systems by providing an overview of the sources, impacts, and mitigation strategies related to AI bias, with a particular focus on the emerging field of generative AI.

Survey on Machine Learning Biases and Mitigation Techniques

Machine learning (ML) has become increasingly prevalent in various domains. However, ML algorithms sometimes give unfair outcomes and discrimination against certain groups. Thereby, bias occurs when our results produce a decision that is systematically incorrect. At various phases of the ML pipeline, such as data collection, pre-processing, model selection, and evaluation, these biases appear. Bias reduction methods for ML have been suggested using a variety of techniques. By changing the data or the model itself, adding more fairness constraints, or both, these methods try to lessen bias. The best technique relies on the particular context and application because each technique has advantages and disadvantages. Therefore, in this paper, we present a comprehensive survey of bias mitigation techniques in machine learning (ML) with a focus on in-depth exploration of methods, including adversarial training. We examine the diverse types of bias that can afflict ML systems, elucidate current research trends, and address future challenges. Our discussion encompasses a detailed analysis of pre-processing, in-processing, and post-processing methods, including their respective pros and cons. Moreover, we go beyond qualitative assessments by quantifying the strategies for bias reduction and providing empirical evidence and performance metrics. This paper serves as an invaluable resource for researchers, practitioners, and policymakers seeking to navigate the intricate landscape of bias in ML, offering both a profound understanding of the issue and actionable insights for responsible and effective bias mitigation.

Economic Mathematical Models: Examining Their Impact on Individuals

Economic mathematical models are essential tools for comprehending intricate economic phenomena, shaping policies, and aiding decision-making. This paper provides a comprehensive analysis of the positive and negative effects of these models on individuals in the context of economic analysis and decision-making. The benefits of economic mathematical models lie in their ability to enhance understanding through data-driven approaches and advanced algorithms, enabling evidence-based policy formulation and solutions to economic challenges. Moreover, economic models play a crucial role in policy formulation, enabling proactive analysis to minimize unintended consequences and maximize policy effectiveness, benefiting both individuals and society. These models also aid decision-making by providing quantitative frameworks to evaluate different options, reducing uncertainty and risk. However, it is essential to acknowledge the limitations of these models, as they are simplifications of reality and may lead to biases and inaccuracies in predictions and recommendations. Ethical implications arise when decision-makers prioritize economic efficiency over social welfare, potentially leading to unfair outcomes for certain individuals or groups. To address these concerns, this paper presents case studies illustrating the successes and shortcomings of economic models in real-world applications. It also explores alternative methodologies, such as qualitative research and mixed-method approaches, to complement economic analysis. By fostering informed discussions on the appropriate role of economic mathematical models, this research contributes to the development of more equitable and effective economic policies that benefit society as a whole.

The role of luck in the success of social media influencers

MotivationSocial media platforms centered around content creators (CCs) faced rapid growth in the past decade. Currently, millions of CCs make livable incomes through platforms such as YouTube, TikTok, and Instagram. As such, similarly to the job market, it is important to ensure the success and income (usually related to the follower counts) of CCs reflect the quality of their work. Since quality cannot be observed directly, two other factors govern the network-formation process: (a) the visibility of CCs (resulted from, e.g., recommender systems and moderation processes) and (b) the decision-making process of seekers (i.e., of users focused on finding CCs). Prior virtual experiments and empirical work seem contradictory regarding fairness: While the first suggests the expected number of followers of CCs reflects their quality, the second says that quality does not perfectly predict success.ResultsOur paper extends prior models in order to bridge this gap between theoretical and empirical work. We (a) define a parameterized recommendation process which allocates visibility based on popularity biases, (b) define two metrics of individual fairness (ex-ante and ex-post), and (c) define a metric for seeker satisfaction. Through an analytical approach we show our process is an absorbing Markov Chain where exploring only the most popular CCs leads to lower expected times to absorption but higher chances of unfairness for CCs. While increasing the exploration helps, doing so only guarantees fair outcomes for the highest (and lowest) quality CC. Simulations revealed that CCs and seekers prefer different algorithmic designs: CCs generally have higher chances of fairness with anti-popularity biased recommendation processes, while seekers are more satisfied with popularity-biased recommendations. Altogether, our results suggest that while the exploration of low-popularity CCs is needed to improve fairness, platforms might not have the incentive to do so and such interventions do not entirely prevent unfair outcomes.

Mshengu v Estate Late Mshengu 9223/2016P - Considering the ownership of house property in customary law

The Recognition of Customary Marriages Amendment Act 1 of 2021 was enacted to address the proprietary consequences of customary marriages. This note examines the implications of the Amendment Act in light of the Mshengu v Estate Mshengu 9223/2016P judgment, which was decided shortly after the Amendment Act came into effect. Three key issues are analysed: first the potential conflict between the Amendment Act and the Reform of Customary Law of Succession and Regulation of Related Matters Act 11 of 2009 in relation to the ownership of house property; second the challenges in classifying property as house or family property; and third the impact of the devolution of property on the rights of other family members. The analysis emphasises the importance of soliciting input from communities who live according to customary law and highlights the need for legislation that is flexibly drafted to accommodate nuanced customary law practices and provide avenues for redress in cases where statutory provisions yield unfair outcomes.

Algorithm justice – moving towards equitable artificial intelligence

Research methodology The case is presented as descriptive in nature and primarily involves exploratory research. Case overview/synopsis Ashraf, a young graduate from Bangalore, India, started a chain of lifestyle shops, his family business in Khartoum, Sudan. To modernize the shops, Ashraf approached a small finance bank for financial assistance. However, after submitting the required documents and with a good credit score, he was denied a loan. The bank officials had mentioned that the loan automation software did not approve the application. Hence, the bank personnel said that they could not do anything further. Disappointed, Ashraf sought the help of his professor, John, to understand why the software rejected his application. Professor John explained to Ashraf the advantages and disadvantages of automation. In the process, Ashraf understood the significance and compelling need to address “Algorithm Bias,” a situation in which specific attributes of an algorithm cause unfair outcomes. The case place students in Ashraf’s position to help them understand the advantages and issues of applying automation through artificial intelligence. Complexity academic level The case suits graduate-level courses like business analytics, financial analytics and business intelligence. Learning objectives Through the case, the students will be able to: Understand the role of algorithms in business and society. Understand the causes, effects and methods of reducing algorithm bias. Demonstrate the ability to detect algorithm bias. Define policies to mitigate algorithm bias.

A Study on Social Exclusion in Human-Robot Interaction

Recent research in human-robot interaction (HRI) points to possible unfair outcomes caused by artificial systems based on machine learning. The aim of this study was to investigate if people are susceptible to social exclusion shown by a robot and, if they are, how they signal the feeling of being rejected from the group. We review the research on social exclusion in the context of human–human interaction and explore its relevance for HRI. Then we present the results of our experiment to simulate social exclusion in the context of HRI: the participants (for whom it was their first encounter with a robot) and the Nao robot were asked to cooperate in solving the bomb defusal task, during which the robot favored one participant with whom it had a longer interaction before the task. The robot was controlled using the Wizard-of-Oz methodology throughout the experiment. Our results show that the discriminated participants reported a higher feeling of exclusion. Though some other hypotheses were not confirmed, we present several qualitative observations from our experiment. For example, it was noticed that the participants who behaved more openly and were more extraverted acted more comfortably when interacting with the robot.

Disrupting the discourse: Applying critical race theory as a conceptual framework for reflecting on learning and teaching in higher education

The Higher Education Sector in Britain, United Kingdom (UK) is currently being taken to task regarding issues of structural inequality and unfair outcomes for student learners from non-white backgrounds, also referred to as the degree awarding gap. How do we disrupt the narrative concerning race, and more specifically as part of our learning, teaching and assessment practices? Using Critical Race Theory, a project, ‘Disrupt the Discourse,’ an initiative launched in 2021 is a small scale case study piloting the application of the aforementioned theory as part of learning and teaching in a Higher Education Institution based in London. The project explored issues of curriculum design, and assessment practices in response to the issue of the degree awarding gap. The initiative explicitly explores uncomfortable conversations about race as part of learning and teaching practice and by working with a team of anti-racist scholars, a curriculum framework and digital toolkit to explore the lived experiences of student and staff was created. Feedback from the pilot was encouraging, academics from different curriculum disciplines and cultural backgrounds saw the initiative as instrumental in re considering assessment practices, curriculum content and pedagogy as part of learning and teaching practice.

Misconceiving Merit: Paradoxes of Excellence and Devotion in Academic Science and Engineering by Mary Blair-Loy and Erin Cech

Reviewed by: Misconceiving Merit: Paradoxes of Excellence and Devotion in Academic Science and Engineering by Mary Blair-Loy and Erin Cech Julie R. Posselt and Gloria C. Anglón Mary Blair-Loy and Erin Cech. Misconceiving Merit: Paradoxes of Excellence and Devotion in Academic Science and Engineering. Chicago: University of Chicago Press, 2022. 232 pp. $27.50. ISBN 9780226820156. Yale professor Nicholas Christakis recently sparked a flare up on social media by challenging the rights of graduate students to unionize, tweeting, "Graduate students are primarily students and trainees, not ordinary workers. Academia is a calling." Although the view of scholarly life as a calling may be contested viz a viz unionization, sociologists Mary Blair-Loy and Erin Cech document in Misconceiving Merit: Paradoxes of Excellence and Devotion that there is indeed a widespread expectation that scholars should single-mindedly dedicate themselves to their work as if it is a vocation or spiritual calling. Even more insidious than the belief that academia is a calling is the belief that scientific excellence following from unbounded devotion to one's [End Page 410] work is pure, unbiased, and unpoliticized. These beliefs work in tandem; many assume academia is a meritocratic profession in which the most hardworking scientists are rewarded for their excellent contributions. Yet, decades of research reveal otherwise. Opportunities, rewards, and penalties vary by social identities; people are neither equally rewarded for similar efforts nor similarly penalized for deviations from devotion and excellence. The result? STEM (science, technology, engineering and math) professionals do not mirror the race, ethnicity, and gender demographics represented in the United States. U.S. white and Asian men represent 63% of the tenured or tenure-track science and engineering (S&E) positions at 4-year colleges; 18% of S&E doctoral students; 18% of S&E master's students; and 28% of full-time S&E undergraduates at 4-year colleges. In this volume, the authors pose the question, "How can a set of professions that so highly values objectivity and fairness in assessments of merit reproduce these unfair outcomes?" (p. 22). Their answer is that STEM faculty's misplaced adherence to meritocracy is filtered through two schemas (i.e., widely held, hegemonic beliefs): The work devotion schema is "a cultural mandate that defines work as a calling deserving of undivided allegiance" (p 3), while the scientific excellence schema is "a cultural yardstick that STEM academics use to measure the competence and worthiness of their colleagues" (p. 4). These schemas are "anchors of the cultural definition of merit in STEM" (p. 19), shaping the professional culture of science. STEM faculty dutifully uphold–even embrace– these beliefs. Scientists live their own lives according to unrealistic expectations around devotion to work and norms for excellence. And, importantly for the opportunity structure and inequalities within it, they judge themselves and their colleagues through these schemas, despite evidence that they uphold gender, racial/ethnic, and LGBTQ inequalities in science. Specifically, white and Asian cis-gender men are rewarded more than marginalized groups, including white women, and black and Latino men and women, for similar work. Blair-Loy and Cech arrived at these conclusions with a case-oriented methodology. They collected data in 2012 from STEM faculty at one U.S. research university, including over 500 university personnel records, surveys from approximately 250 of the people from whom they had records, and semi-structured interviews with 86 faculty who completed the survey. The authors also constructed a Scholarly Productivity Indices database of all participants by tracking the number of publications from a SCOPUS bibliographic database, as well as the number of grants awarded, dollars, and PI (principal investigator) or Co-PI designations based on university records. The authors used the survey responses to develop scales of how faculty view themselves and an idealized successful person (self-conceptions) on assertiveness, relational, and diversity promotion. The heart of the book documents the existence of these schemas and offers evidence that professionals use them to "explain away underrepresentation and divert blame for its resolution from STEM faculty and institutions" (p. 123). Critical to this argument is evidence that the schemas, while widely held, intersect with patterns of discrimination in the real world to reproduce inequalities. For...