What does the MAIHDA method explain?

Multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) is a new approach to quantitative intersectional modelling. Along with an outcome of interest, MAIHDA entails the use of two sets of independent variables. These include group demographics such as race, gender, and poverty status as well as strata which are constructs such as Black female poor, Black female wealthy, and White female poor. These constructs represent the combination of the demographic variables. To operationalize the approach, an initial random intercepts model with strata as a level 2 context is specified. Then, another model is specified that includes the strata as well as the demographic variables as level 1 fixed effects. As such, it is argued that MAIHDA uniquely identifies the additive and intersectional effects for any given outcome. In this paper we show that MAIHDA falls short of this promise: the strata are an individual-level composite variable not a level 2 context. Rather than being analogous to neighborhoods as contexts, strata are analogous to socio-economic status which is a combination of individual-level demographic variables, albeit often presented as a group-level characteristic. The result is that the demographic variables are inserted in both level 2 and 1. This duplication across the levels in MAIHDA means that there is a built-in collinearity across the levels and that the models are mis-specified and, therefore, redundant. We conclude that single-level models with the demographic variables and interactions or with the strata as fixed effects are still the more accurate models for quantitative intersectional analyses.

Multi-group data versus dual-side theory: On race contrasts and police-caused homicides.

Empirical evidence points to a persistent Black-White racial gap in police-caused homicides. Some scholarship treats the gap as denoting criminal justice exposure either in terms of involvement in crime or living in a high-crime context. By contrast, health scholarship typically points to the importance of racism including the attitudes, institutional practices, and overall structures that operate to privilege one group over another. Still, given the demographics of US society, the Black-White racial contrast overlooks the 25% of Americans who are neither Black nor White: Native Americans, Latinos, and Asians. The question of how the groups should be organized vis-a-vis the current Black-White model and theories arises. An answer is not straightforward. There is a rank-ordering to the groups' mortality rates as well as an exponential increase in the number of possible comparisons. In this paper we systematically review the literature on race and police-caused homicide with a particular focus on studies that attempt to move beyond the Black-White model. We find that studies on race and police-caused homicide either make no comparison between the groups, or, alternatively, use a White-non-White, a Black-non-Black, and/or a Black-Native American-Latino vs. White-Asian comparison. We use data on group-specific mortality rates to examine the strengths and limits of each of these practices. The limits are the selection of counterfactual gaps, the selection of smaller gaps, and/or the omission of larger gaps. To address these limits, we propose that a Black-Native American vs. Latino-White-Asian model best captures the higher and lower mortality rates in police-caused homicide data.

Assimilated or the boundary of Whiteness expanded? A boundary model of group belonging.

In 19th century U.S., European ethnics were considered dark-skinned and working-class. By the 1950s, they were re-categorized as White Americans. Assimilation theory suggests that Europeans' intergroup interactions such as socioeconomic attainments and intermarriage with Anglo-Saxon Whites led to their assimilation and racial re-categorization. The theory anticipates that class mobility translates into re-categorization into Whiteness such that non-Europeans' upward mobility will also propel them into Whiteness. Despite non-Europeans' successful intergroup interactions, their assimilation and belonging is still up for debate. If both Europeans and non-Europeans have participated in similar rates of intergroup interactions, which other factor has determined their differential assimilation outcomes? In response, we conceptualize a boundary model for understanding group belonging. To do so, we distinguish between symbolic and phenotype or somatic race. The former provides and attaches meanings to soma. This boundary model comprises the combined effects of individuals' intergroup interactions and majority groups' symbolic racial boundary expansion-contraction. Assimilation outcome occurs only when majority groups' boundary expands to recognize intergroup interactions as meaningful and to include newcomers as racial group insiders. We revisit the case of European ethnics to show that the symbolic boundary of Whiteness expanded to re-categorize them as assimilated Whites. Accordingly, we formulate four hypotheses about the possibilities of re-categorizing groups in and out of a master category such as Whiteness.

Experiences matter: A longitudinal study of individual-level sources of declining social trust in the United States.

The US has experienced a substantial decline in social trust in recent decades. Surprisingly few studies analyze whether individual-level explanations can account for this decrease. We use three-wave panel data from the General Social Survey (2006-2014) to study the effects of four possible individual-level sources of changes in social trust: job loss, social ties, income, and confidence in political institutions. Findings from fixed-effects linear regression models suggest that all but social ties matter. We then use 1973-2018 GSS data to predict trust based on observed values for unemployment, confidence in institutions, and satisfaction with income, versus an alternative counterfactual scenario in which the values of those three predictors are held constant at their mean levels in the early 1970s. Predicted values from these two scenarios differ substantially, suggesting that decreasing confidence in institutions and increasing unemployment scarring may explain about half of the observed decline in US social trust.

The gender gap in commenting: Women are less likely than men to comment on (men's) published research.

Subtle gender dynamics in the publishing process involving collaboration, peer-review, readership, citation, and media coverage disadvantage women in academia. In this study we consider whether commenting on published work is also gendered. Using all the comments published over a 16-year period in PNAS (N = 869) and Science (N = 481), we find that there is a gender gap in the authorship of comment letters: women are less likely than men to comment on published academic research. This disparity is greater than gender differences in the publication of research articles. There is also a gendered pattern in commenting: women comment writers are relatively less likely to engage with men's research. If left unaddressed, these patterns in academic commenting could impede scholarly exchange between men and women and further marginalize women within the scientific community.

Does Reconciliation and Racial Justice Necessitate a Struggle against White Supremacy?

Several literatures including those focusing on settler colonialism, critical antiracism as well as ethnic studies and sociology more broadly often position racial injustice and genocide as a struggle against whiteness and white supremacy. Here I use my own positionality to illustrate what might be unseen in the current thinking about the meaning of what whiteness entails. Then I propose the preliminary workings of a nonbinary approach to thinking about racial justice and reconciliation that still centers the specific experiences of oppression but that does not also entail blaming a particular group as oppressor. While I focus on Canada and responsibility for Indigenous genocide and, to some extent, anti-Black racism, my hope is that the theoretical logic will also be of utility for thinking about moving forward on issues of racial justice and genocide in other contexts.

Social Change and the Gendered Division of Household Labor in Canada.

Housework is asymmetrically distributed by gender. This uneven allocation is an important indicator of inequality between women and men. The imbalance is closing, although exactly why remains uncertain. It is also unclear if the convergence has more to do with women's lives becoming more like men's, or whether it is because men are changing their practices on the home front. Using 30 years of nationally representative time use diary data, we explore three broad theoretical frameworks addressing social change-cultural, structural, and demographic-to examine how and why the gender dynamics around housework are shifting. We find that structural factors, and in particular women's engagement with paid work, have changed most sharply as drivers of greater symmetry in domestic labor, although changing cultural beliefs have contributed as well. Furthermore, there have been significant changes in men's behavior. One focal point for this domestic change is in men's and women's shifting practices around childcare. Intensive parenting, not just intensive mothering, has become more prevalent.

Le travail ménager n'est pas distribué de manière uniforme entre les sexes. Cette inégalité des rôles est un important indicateur de celle qui existe entre les femmes et les hommes. Ce déséquilibre s'atténue, même si la raison exacte en demeure incertaine. On ignore également si cette convergence se rapporte davantage au fait que la vie des femmes ressemble de plus en plus à celle des hommes ou au fait que les hommes changent d'attitude face au travail ménager. À partir d'un échantillonnage représentatif au niveau national de trente années de données journalières sur l'utilisation du temps, nous explorons trois vastes cadres théoriques - culturel, structurel et démographique - abordant les changements sociaux pour savoir comment et pourquoi se modifie la dynamique sexuelle autour du travail ménager. Nous avons découvert que les facteurs structurels, particulièrement la participation des femmes au travail rémunéré, ont évolué le plus nettement comme source d'une plus grande symétrie dans le travail ménager, même si une modification des croyances culturelles y a également joué un rôle. De plus, le comportement des hommes s'est modifié de manière significative. Cette évolution du travail ménager dépend en grande partie de celle qui touche la garde des enfants. On assiste aujourd'hui à une parentalité intensive, plutôt qu'à une simple maternité intensive.

Immigration, Discrimination, and Trust: A Simply Complex Relationship.

Many immigrants experience discrimination. In this paper we consider how discrimination affects their trust. We make a theoretical case for a formal mediation approach to studying the immigration, discrimination, and trust relationship. This approach shifts attention to the basic fact that the overall levels of discrimination experienced by different immigrant and native-born groups are not the same. We also build on previous empirical research by considering multiple forms of discrimination, multiple types of trust and multiple immigrant/native-born groups. Drawing on the 2013 Canadian General Social Survey data (N = 27,695) we analyze differences in three kinds of trust (generalized trust, trust in specific others, and political trust), and the role of perceived discrimination (ethnic, racial, any), between five immigrant-native groups (Canadian-born whites, Canadian-born people of color, foreign-born whites, foreign-born people of color, and Indigenous people). We find that perceived discrimination is more relevant to general trust and trust in specific others than to political trust. We also find that perceived discrimination explains more of the trust gap between racialized immigrants and the native-born than the gap between non-racialized immigrants and the native-born. The results illustrate that what appears to be a simple relationship is far more complex when attempting to explain group differences.

Representations of race and skin tone in medical textbook imagery.

Although a large literature has documented racial inequities in health care delivery, there continues to be debate about the potential sources of these inequities. Preliminary research suggests that racial inequities are embedded in the curricular edification of physicians and patients. We investigate this hypothesis by considering whether the race and skin tone depicted in images in textbooks assigned at top medical schools reflects the diversity of the U.S. POPULATION: We analyzed 4146 images from Atlas of Human Anatomy, Bates' Guide to Physical Examination & History Taking, Clinically Oriented Anatomy, and Gray's Anatomy for Students by coding race (White, Black, and Person of Color) and skin tone (light, medium, and dark) at the textbook, chapter, and topic level. While the textbooks approximate the racial distribution of the U.S. population - 62.5% White, 20.4% Black, and 17.0% Person of Color - the skin tones represented - 74.5% light, 21% medium, and 4.5% dark - overrepresent light skin tone and underrepresent dark skin tone. There is also an absence of skin tone diversity at the chapter and topic level. Even though medical texts often have overall proportional racial representation this is not the case for skin tone. Furthermore, racial minorities are still often absent at the topic level. These omissions may provide one route through which bias enters medical treatment.

Canadian University Acknowledgment of Indigenous Lands, Treaties, and Peoples.

At many Canadian universities it is now common to publicly acknowledge Indigenous lands, treaties, and peoples. Yet, this practice has yet to be considered as a subject of scholarly inquiry. How does this practice vary and why? In this paper we describe the content and practice of acknowledgment, linking this content to treaty relationships (or lack thereof). We show that acknowledgment tends to be one of five general types: of land and title (British Columbia), of specific treaties and political relationships (Prairies), of multiculturalism and heterogeneity (Ontario), of no practice (most of Quebec), and of people, territory, and openness to doing more (Atlantic). Based on these results, we conclude that the fluidity of acknowledgment as a practice, including changing meanings depending on the positionality of the acknowledger, need to be taken into account. Plusieurs universités Canadien pratique une reconnaissance des territoires, des traités, et des peoples autochtone en publique. Cette pratique, cependant, n'a jamais été considérée comme une enquête savante. Dans ce projet nous regardons comment les reconnaissances varie par institution et pourquoi. Nous trouvons qu'il y a un lien entre le contenu des reconnaissances et les relations traité. On démontre cinq forme des reconnaissances: territoire et titre (Colombie britannique); traité spécifique and les relations politiques (Prairies); multiculturalisme et hétérogénéité (Ontario); l'absence (la majorité du Québec); et des peoples, territoire et volonté a plus faire (Atlantique). Nous concluons que la fluidité de la reconnaissance, comme pratique, est fluide et doit prendre en considération la position de la personne qui le fait.

We trust in government, just not in yours: race, partisanship, and political trust, 1958-2012.

Although it is generally accepted that political trust is reflective of satisfaction with the performance of the incumbent administration, this is only considered true for White Americans. Because their trust reflects a larger discontent with the political system, Black Americans, it is held, do not respond in the same way in the short term. This argument has yet to be tested with over-time data. Time matters. Not only does the race gap in trust change over time but the impact of partisanship and political winning is, by definition, time-dependent. The results of an analysis of the 1958-2012 American National Election Studies data show that Black Americans and White Americans are equally likely to tie short-term performance to trust in government. However, the relationship between partisanship and political trust and, therefore, system discontent, clearly differs for the two groups. Aggregate models that do not take race-partisan sub-group differences into account will therefore be misleading.

Hypersegregation in the twenty-first century.

We used metropolitan-level data from the 2000 U.S. census to analyze the hypersegregation of four groups from whites: blacks, Hispanics, Asians, and Native Americans. While blacks were hypersegregated in 29 metropolitan areas and Hispanics were hypersegregated in 2, Asians and Native Americans were not hypersegregated in any. There were declines in the number of metropolitan areas with black hypersegregation, although levels of segregation experienced by blacks remained significantly higher than those of the other groups, even after a number of factors were controlled. Indeed, although socioeconomic differences among the groups explain some of the difference in residential patterns more generally, they have little association with hypersegregation in particular, indicating the overarching salience of race in shaping residential patterns in these highly divided metropolitan areas.