Category: Resources
Keynote abstracts
Advancing MAIHDA (Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy: Towards Contextual Intersectional Multilevel Modelling?
Professor Anne Laure Humbert
University of Gothenburg
Quantitative research has often struggled to engage meaningfully with intersectionality. Traditional methods, such as cross-tabulations or models with interaction terms, can rarely capture the structural, relational, and contextual dimensions of intersectional inequalities. The MAIHDA approach (Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy) has offered a promising alternative, particularly by simplifying complex model specifications and highlighting variation between intersectional groups. However, its application to date has largely remained within the paradigm of ‘descriptive intersectionality’.
This talk presents how MAIHDA can be extended to support ‘analytical intersectionality’, an approach concerned not only with documenting intersectional inequalities, but also with analysing the processes and structures that produce and sustain them. Drawing on data from the Joint European and World Values Surveys, I demonstrate how intersectional multilevel models can be used to explore variation across identity groups and national contexts. I argue for a shift in how the MAIHDA approach can be used: from modelling individual differences to interrogating structural inequalities. This extended methodological framework, what might be called (contextual) intersectional multilevel modelling or (C)I-MLM, opens up new possibilities for critical, theoretically informed, and methodologically rigorous research on intersectional inequalities in the economy and society.
Biography
Professor Anne Laure Humbert, PhD, is a researcher at the University of Gothenburg, Sweden. Anne is very experienced in gender equality research at national, EU and international level, policy analysis and assessment as well as gender statistics. She specialises in applying quantitative methods to comparative social and economic analysis. Anne is a regular public speaker on gender equality and intersectionality, and she enjoys the opportunity to make connections between theory, policy and practice. Her recent methodological work has focused on the applications of intersectional multilevel modelling (MAIHDA: Multilevel analysis of individual heterogeneity and discriminatory accuracy) to various aspects of social and economic outcomes of inequalities, and she is currently writing a book on the topic based on her experience of training other researchers on how to use this new approach to intersectional quantitative analysis.
The Statistical Advantages of MAIHDA for Estimating Intersectional Inequalities
Professor George Leckie
University of Bristol
Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) is a multilevel regression modeling approach, rooted in intersectionality theory, for examining social inequalities across intersections of multiple social identities (e.g., gender, ethnicity, social class). Proponents argue that MAIHDA’s predicted intersectional means are statistically superior to simple means from descriptive statistics or conventional regression models, but this claim remains largely untested. We derive and analyze analytical expressions to compare the variance and correlation of predicted intersection means from two MAIHDA-based predictors and simple mean calculations with the distribution of true but unknown population means. Additionally, we assess the bias, variance, and mean squared error when predicting the mean of a given intersection. Our findings show that MAIHDA-based means outperform simple means, particularly when using the predictor that decomposes intersectional means into additive and non-additive components of social identities. However, the relative advantage of each MAIHDA predictor depends on the nature of intersectional inequalities and intersection sizes. MAIHDA’s benefits are most pronounced when inequalities are subtle or when data on certain intersections, such as those for marginalized groups, are sparse—conditions common in practice, highlighting the practical significance of our findings.
Biography
George is a Professor of Social Statistics and Co-Director of the Centre for Multilevel Modelling (CMM) at the University of Bristol, UK. His research interests focus on the application and dissemination of multilevel modelling. He has applied this method to study the effects of institutional and geographical contexts on sociodemographic inequalities in individual educational and health outcomes. A long-standing interest of his has been examining school effects on student learning through school value-added models, as well as engaging in debates regarding the use of such data to inform school accountability and choice. A more recent interest has been the application of multilevel modelling, specifically MAIHDA, to study intersectional inequalities.
(Im)possible categorisation? Exploring the challenges and contributions of intersectional quantitative research from a mixed-race perspective
Dr Rhianna Garrett
Loughborough University
The UK has become a global hub of multiculturalism, containing an intricate tapestry of what it means to hold a racialised identity. Yet, mixed-race identities continue to be limited to colonial ‘race’ categories that do not reflect the identities of mixed-race populations in Britain today. This presentation highlights the key challenges involved in conducting intersectional quantitative research from a mixed-race perspective, while also articulating the benefits of incorporating multiracial considerations into quantitative intersectional research. To achieve this, the discussion will present key findings from a new collaborative paper that employs Multiple Correspondence Analysis (MCA) on the EVANS survey to uncover new empirical insights into mixed-race Britain. Dr Garrett aims to encourage scholars to work outside of their methodological comfort zones and continue to make multiracial matters matter.
Biography
Dr Rhianna Garrett is a multiple award-winning mixed methods researcher, consultant and community organiser at Loughborough University’s Department of Geography and Environment. Her work explores intersectional mixed-race identities in Britain, exposing how white supremacist monoracial norms shape society—an issue increasingly urgent amid the rise of far-right politics. As Global Coordinator for the Critical Mixed Race Studies Association’s (CMRSA) executive board, she also examines how ideas of race, ethnicity, and culture shift across different geopolitical contexts. Inspired by liberatory decolonial and feminist approaches, Rhianna draws on both creative qualitative methods and critical quantitative methods to centre lived experiences and reveal monoracism’s widespread influence on British conceptualisations of ‘race’.
How Doing Intersectional Quantitative Research questions statistical convention
Professor Niels Spierings
Radboud University
A broadly shared position is that intersectionality and doing quantitative research are not mutually exclusive. Frameworks have been developed on what kind of intersectionality fits quantitative research. Debate is on-going on what is minimally needed for quantitative research to be considered intersectional. And the umber of manuals and ‘how to’-guides is growing. Building on these developments in the field of doing intersectional quantitative research (DIQR), in this keynote I want think out loud on how DIQR raises issues about statistical conventions. Particularly, I aim to discuss how certain conventions and practices might be particularly problematic for DIQR, and what this means for the role (quantitative intersectionalist) academics have in the process of academic research.
Biography
Niels Spierings is full professor in sociology, processes of inclusion and exclusion, at Radboud University (Nijmegen, the Netherlands). His work focuses mainly on the position of ethno-religious minorities and LGBTQ+ people, including exlcusionary populist politics, in Western Europe. As part of his research, he likes to think and write about combining feminist, critical and intersectional insights into quantitative research. For instance, he wrote the chapter ‘Quantitative Intersectional Research: Approaches, Practices, and Needs’ in The Routledge international handbook of intersectionality studies.
Abstracts
Part II of the annotated bibliography covers methodological discussions from an empirical point of view, including reviewing and implementing specific methods.
This article is a systematic review of 16 published studies using latent class analysis, latent profile analysis and clustering algorithms, which the authors identified as methods that group individuals’ characteristics following the intersectionality principles. The review revealed that most articles applying these methods, called by the authors ‘person-centred’, identified four classes/clusters by combining indicators of identities, (social) positions and processes such as stigmatisation and discrimination. Health-related research commonly uses these outcomes for sociodemographic comparisons or as predictors in linear (regression) models. The paper discusses theoretical and methodological rationales behind using these techniques, particularly how well they ‘translate’ the core tenets of intersectionality.
This study compares the performance of two-level multilevel models (MLM) and standard fixed-effects regression estimated via ordinary least squares (OLS) to study intersectionality. As illustrated by the authors, both methods have potential drawbacks. OLS regressions with interaction terms for dummy variables risk identifying statistically significant interactions by chance, especially with smaller sample sizes. On the other hand, MLM models that include dummy variables for each interaction may shrink effects towards the mean due to the uncertainty of their estimates. Additionally, MLM assumes that level-two residuals (the difference between estimated and predicted values) are independent and identically distributed. The performance of MLM and OLS is compared via simulations under different conditions, including those that violate the assumptions for implementing MLM. Simulations showed that MLM outperforms OLS regressions, but shrinkage on MLM may wrongly identify some intersectional effects. The article closes with guidance to enhance the robustness of MLM for the study of intersectional effects for predicting outcomes.
In this article, the authors explore the application of the intersectional MAIHDA (multilevel analysis of individual heterogeneity and discriminatory accuracy) previously introduced by the authors as a strategy to apply MLM models for studying intercategorical inequalities. MAIHDA treats intersecting categorisations as strata, which among other advantages, allows it to better deal with higher dimensionality (combinations of multiple categories). Framed as a response to a critique of the approach, the authors refer to simulations and empirical applications, emphasising MAIHDA’s advantages over single-level approaches. The study complements Bell et al (2019) by providing further explanations on interpreting fixed effects regression coefficients and highlighting the advantages of using multilevel models for analysing high-dimensional interactions
In this book chapter, the author starts by exploring ‘why’ and ‘how’ intersectionality is studied quantitatively using survey data. Dubrow suggests that despite the common limitations of surveys, such as small sample sizes for intersecting categories, hypotheses on cumulative and group-specific disadvantages can be quantitatively tested. Dubrow tested intersectional hypotheses by calculating the international socioeconomic index for combinations of intersecting categories and utilising unpaired t-tests. The simple, yet effective, statistical analysis is a call to action for social scientists, especially those interested in social stratification, to innovate when implementing quantitative methodologies that better reflect the realities of individuals’ lived experiences.
Gross and Goldan highlight the challenges and opportunities presented by applying intersectionality in quantitative studies, contrasting it with qualitative approaches. The authors offer a schematic introduction to OLS, MLM and MAIHDA, including reflections on integrating intersectionality into quantitative research methodologies and evaluating their effectiveness in capturing complex social dynamics. The paper calls for a deeper understanding of intersectionality among quantitative researchers and advocates for improved methodological practices to better address social inequalities.
Mahendran et al explored the application of statistical methods to study intersectionality in health inequalities. Descriptive and causal methods were reviewed, including cross-classification (individuals are assigned to groups based on the combination of two or more variables), decision tree algorithms (non-parametric methods that set decision rules for data segmentation) and regression with interactions and MAIHDA. Using simulations, the authors test how well the prevalence of an intersection-specific dependent variable is estimated by each method. Findings vary depending on sample sizes for simulated scenarios, but broadly point at MAIHDA, followed by conditional inference trees and chi-square automatic interaction detector as the better methods for smaller samples.
In this article, the authors investigate the use of regression models, including OLS with interaction terms, separate interaction regressions for different contexts, and effect decomposition, as well as two-level MLM. Taking a feminist approach to intersectionality, Scott and Siltanen evaluate how context, openness and the complexity of the structuring of inequality are operationalised by each approach. Although the authors acknowledge the inherent limitation of predictive/causal models, they concluded that using the proposed feminist-inspired criteria MLM better aligns with feminist literature on intersectionality.
In this article, Schudde examines the methodological complexities of integrating intersectionality into quantitative research within the field of education. Schudde critiques the reliance on secondary survey data, which often limits researchers’ ability to capture the full spectrum of intersectional identities, and the limitations of grouped results to account for unequal experience. The literature review is focused on varying effects (‘heterogeneous effects’) of identities and context in higher education. The author found that most papers in the field used regression models with main effects and a minority used interaction terms. Another salient method is propensity score models, which estimate the probability of an outcome based on the combination of observed characteristics. More widely, the author identifies significant challenges, including the difficulty of justifying complex models in academic publishing and the limitations of existing datasets.
This chapter explores the multiple ways to understand the link between statistical methods and operationalisations of intersectionality. The author overviews some of the statistical methods recently used to implement intersectionality divided into three groups. The first (‘minimalist’) views the methodological aspects of the study of intersectionality as merely of a technical nature. A step forward is to take a theoretically informed (‘reflective’) approach in which interest moves to the link between theorisations and methods, considering social contexts. A third approach, referred to as ‘radical’, focuses on the innovative use of quantitative methods which better fit specific forms of intersectionality. Ultimately, Spielrings calls for openness in access to relevant datasets and knowledge for progressing the quantitative exploration of intersectional studies.
Categories
Introduction to the GQIA project
Audio (5:27)
Transcript (slighly ammended)
Hello everyone, I’m Adrian Leguina, the Principal Investigator for the British Academy Talent Development Award “A new paradigm of quantitative intersectional analysis using geometric data analysis” or GQIA. What I want to do here is to tell you what this project is about in simple words.
My background is in statistics and sociology. And in particular, my research is part of something called Bourdieusian class analysis, which is interested in the intersection of the economic cultural and social dimensions of class inequality. There we use a family of techniques called geometric data analysis (GDA) which is what we traditionally used to construct the social space, a multidimensional representation of class inequalities, which represents positions within a social structure in more general terms. So, my discomfort with this type of analysis [or to be more precise, critique] is that we rarely say something about ethnicity and gender and many other social divisions.
So, methodologically, this is a problem, because the assumption that many scholars in the field make is that the social space, cultural, economic, social and symbolic capital, are a reflection of all of our conditions of existence, including ethnicity, gender and so on. My problem is I’m not fully convinced by that because when I see or when I’ve done myself this type of analysis, we… [are not acknowledging that] many other relevant social divisions impact our social positions within social structures.
My idea is to use techniques and some of the insight from class analysis alongside intersectional analysis, in particular the big wealth of knowledge that hasn’t been necessarily translated, in my view, accurately into quantitative methods. To understand how multiple intersecting social divisions are interrelated, the project involves learning about the use of statistics for intersectional analysis and understanding the foundations of intersectionality and the connections between [intersectionality] and Bourdieusian class analysis to come up with something that should hopefully provide a more accurate and a quite rich representation of intersectionality in the UK.
What I’m doing at the moment is to explore the literature alongside existing datasets that contain rich information, not only about social class, but also about ethnicity and gender, and build this multidimensional representation of intersecting inequalities by making use of advanced techniques of geometric data analysis. For example, one is called multiple factor analysis. So the goal of the project is to provide a space for interdisciplinary dialogue, which has the ambition of becoming a new way of studying inequalities.
The award has a developmental and dissemination strategy which includes multiple strands, this blog being one of them, the production of audiovisual material targeted to students, researchers and anyone interested in these issues and the organization of a workshop alongside other academic outputs.
Welcome everyone and please do get in touch if this is something that is of your interest and please stay updated on the progress in this blog.
By Adrian Leguina and Rhianna Garrett
Part I of the annotated bibliography offers an overview of the conceptualisation and operationalisation of intersectionality from a quantitative standpoint. The sources presented here establish the groundwork for understanding the statistical implications and limitations for developing analytical strategies that translate intersectionality into empirical research.
This list is not exhaustive and only provides a sample of influential work from various disciplines and traditions published over the last two decades. There is some overlap between them, so when using the annotated bibliography, students, lecturers, and researchers can choose those more closely related to their interests.
In this paper, Bauer explores the application of intersectionality theory in population health research. The issue identified by the paper is the lack of theoretical models addressing health and disease inequalities at different intersections of identity, social position, oppression, and privilege. Challenges highlighted by the author include translating theories into methods, the value of intersecting characteristics, distinguishing between intersecting identities, social positions, policies and social structures, and understanding the appropriate scale for interactions. The paper includes an overview of methods used to describe and test intersectional hypotheses
In this article, Bowleg discusses the measurement, analysis, and interpretation challenges found in researching Black lesbians in psychological research. By referring to the way intersectionality challenges the conventional additive assumption (e.g., Black + Lesbian + Woman), the author makes a compelling case for the limitations of traditional methods (such as ANOVA with interaction terms) in accounting for the multiplicative nature of intersectionality. Bowleg emphasises that the interpretation of findings within their social and historical context becomes crucial for intersectionality researchers.
Choo and Ferree explore the implications of practising intersectionality from a theoretical and methodological point of view. Three ways of understanding intersectionality are identified by the authors: group (experiences of intersectionally marginalised groups), process (multiple oppressions and their relation to power) and system-centred (focused on social structures ) approaches. Referring to previous sociological research, the paper illustrates how intersectionality shapes research design and methodological choices. The authors propose that intersectionality should be used more widely to inform key sociological issues
In this article, Covarrubias quantitatively analyses the educational outcomes of Mexican-origin people in the United States of America. The author approaches the analysis of large-scale survey data emphasising the importance of looking at race, class and gender intersectionally for understanding educational outcomes. Analytically, this is done by disaggregating data by educational attainment at race, class, gender, and citizenship status categories, revealing systemic disparities and challenges faced by Chicana/o students. Covarrubias’ exemplary work contributes to the quantitative inquiry of inequalities from intersectional and critical race theory perspectives.
In this special issue introductory article, the authors present “QuantCrit,” a framework that bridges critical race theory (CRT) with statistics. QuantCrit critically engages with white supremacy and its role in the creation of quantitative methods and its perpetuation of the existence of objective and neutral research. The authors present a detailed historical overview of developments in the field, introduce articles on the issue and close with an invitation for continuing deracialising statistics. Special issue articles are recommended for those interested in CRT and QuantCrit, and more widely interested in a critical overview of quantitative methods from a CRT perspective.
In this thought-provoking article, Gillborn critically examines the use of surveys and statistics in education research, particularly concerning race and class. Particularly relevant is the critique of ‘old’ statistical assumptions and practices that can perpetuate discrimination against minoritised groups and the challenge to the tendency to blame inequalities on identities rather than the social processes that assign them value. Gillborn argues that statistical data can perpetuate racial oppression if not interpreted contextually and calls for reflexivity and critical engagement in research methodologies.
Hancock examines research on race, gender and class across disciplines and proposes to think about intersectionality as a research paradigm. To do so, the author presents empirical research standards for intersectionality, which are the answers to six questions that motivate the author’s endeavours: How many categories, their relationship, their conceptualisation, their political makeup, level of analysis and methodological wisdom. Hancock’s work emphasises the need for developing methodologies that account for the multiple dimensions explored by intersectional research.
McCall, L. (2005). The Complexity of Intersectionality. Signs, 30(3), 1771–1800.
McCall highlights the complexities faced by researchers when studying intersectionality. The author acknowledges that traditional single-axis analyses (such as examining gender or race alone) fall short of capturing the multifaceted nature of identity. Intersectionality demands a more nuanced approach regarding the use of analytical categories, which McCall synthesises in three: Anti-categorical (oriented to the study of the experiences lived by neglected intersecting social groups), intracategorical (the study inequalities within different social groups) and intercategorical (to use existing analytical categories to study inequalities between different social groups). Researchers must critically consider questions of measurement, context, and methodology to fully understand how intersecting identities shape individuals’ experiences.
In this paper, the authors guide designing intersectional research, focusing on key aspects of intersectionality to consider designing research, including its focus on power structures, relationality, complexity, comparison, deconstruction and context. The paper discusses qualitative, comparative, and quantitative strategies, calling for a more explicit link between theory, epistemology and methodology. The authors are motivated by the unlocked potential intersectionality offers to sociology and wider social sciences.
Spierings addresses the underrepresentation of quantitative methods in gender studies. The article introduces the ‘diversity continuum’ to assess their takes on diversity (no v. infinite) and type of diversity. Addressing this, the author suggests, can enhance our understanding of differences and similarities among individuals, help communication across fields and facilitate scientific knowledge. This article presents both negative and positive views of quantitative methods in feminist research, emphasising their potential contributions and highlighting good practices for understanding gender dynamics. The article is also recommended for those more widely interested in a critical overview of quantitative methods from a feminist perspective.
In this paper, Warner provides a “best practices guide” for applying intersectionality empirically. Some of the key points include the necessity for identifying the identities to empirically study, the implications of focusing on single dimensions of identity (‘master’) and/or ‘emerging’ or compound categories that emerge by combining multiple identities, and the necessity for treating identity as a process occurring within social structures and specific contexts. In sum, this article offers practical insights for integrating intersectionality into psychological research which apply more widely to other disciplines, including methodological considerations.
Part II of the annotated bibliography will cover discussions from an empirical point of view, including the review and implementation of specific methods such as regressions and clustering.