The Gap-Closing Estimand: A Causal Approach to Study Interventions That Close Disparities Across Social Categories

Disparities across race, gender, and class are important targets of descriptive research. But rather than only describe disparities, research would ideally inform interventions to close those gaps. The gap-closing estimand quantifies how much a gap (e.g., incomes by race) would close if we intervened to equalize a treatment (e.g., access to college). Drawing on causal decomposition analyses, this type of research question yields several benefits. First, gap-closing estimands place categories like race in a causal framework without making them play the role of the treatment (which is philosophically fraught for non-manipulable variables). Second, gap-closing estimands empower researchers to study disparities using new statistical and machine learning estimators designed for causal effects. Third, gap-closing estimands can directly inform policy: if we sampled from the population and actually changed treatment assignments, how much could we close gaps in outcomes? I provide open-source software (the R package gapclosing) to support these methods.

Estimation and sensitivity analysis for causal decomposition in health disparity research

In the field of disparities research, there has been growing interest in developing a counterfactual-based decomposition analysis to identify underlying mediating mechanisms that help reduce disparities in populations. Despite rapid development in the area, most prior studies have been limited to regression-based methods, undermining the possibility of addressing complex models with multiple mediators and/or heterogeneous effects. We propose a novel estimation method that effectively addresses complex models. Moreover, we develop a sensitivity analysis for possible violations of an identification assumption. The proposed method and sensitivity analysis are demonstrated with data from the Midlife Development in the US study to investigate the degree to which disparities in cardiovascular health at the intersection of race and gender would be reduced if the distributions of education and perceived discrimination were the same across intersectional groups.

Recurrent Multinomial Models for Categorical Sequences

This paper presents a model of recurrent multinomial sequences. Though there exists a quite considerable literature on modeling autocorrelation in numerical data and sequences of categorical outcomes, there is currently no systematic method of modeling patterns of recurrence in categorical sequences. This paper develops a means of discovering recurrent patterns by employing a more restrictive Markov assumption. The resulting model, which I call the recurrent multinomial model, provides a parsimonious representation of recurrent sequences, enabling the investigation of recurrences on longer time scales than existing models. The utility of recurrent multinomial models is demonstrated by applying them to the case of conversational turn-taking in meetings of the Federal Open Market Committee (FOMC). Analyses are effectively able to discover norms around turn-reclaiming, participation, and suppression and to evaluate how these norms vary throughout the course of the meeting.

And the Rest is History: Measuring the Scope and Recall of Wikipedia’s Coverage of Three Women’s Movement Subgroups

Narrating history is perpetually contested, shaping and reshaping how nations and people understand both their pasts and the current moment. Measuring and evaluating the scope of histories is methodologically challenging. In this paper we provide a general approach and a specific method to measure historical recall. Operationalizing historical information as one or more word phrases, we use the phrase-mining RAKE algorithm on a collection of primary historical documents to extract first-person historical evidence, and then measure recall via phrases present on contemporary Wikipedia, taken to represent a publicly-accessible summary of existing knowledge on virtually any historical topic. We demonstrate this method using women's movements in the United States as a case study of a debated historical field. We found that issues important to working-class elements of the movement were less likely to be covered on Wikipedia compared to other subsections of the movement. Combining this method with a qualitative analysis of select articles, we identified a typology of mechanisms leading to historical omissions: paucity, restrictive paradigms, and categorical narrowness. Our approach, we conclude, can be used to both evaluate the recall of a body of history and to actively intervene in enlarging the scope of our histories and historical knowledge.

A Joint Model for Longitudinal and Time-to-event Data in Social and Life Course Research: Employment Status and Time to Retirement

Longitudinal studies including a time-to-event outcome in social research often use a form of event history analysis to analyse the influence of time-varying endogenous covariates on the time-to-event outcome. Many standard event history models however assume the covariates of interest to be exogenous and inclusion of an endogenous covariate may lead to bias. Although such bias can be dealt with by using joint models for longitudinal and time-to-event outcomes, these types of models are underused in social research. In order to fill this gap in the social science modelling toolkit, we introduce a novel Bayesian joint model in which a multinomial longitudinal outcome is modelled simultaneously with a time-to-event outcome. The methodological novelty of this model is that it concerns a correlated random effects association structure that includes a multinomial longitudinal outcome. We show the use of the joint model on Danish labour market data and compare the joint model to a standard event history model. The joint model has three advantages over a standard survival model. It decreases bias, allows us to explore the relation between exogenous covariates and the longitudinal outcome and can be flexibly extended with multiple time-to-event and longitudinal outcomes.

An Empirical Comparison of Four Generalized Trust Scales: Test–Retest Reliability, Measurement Invariance, Predictive Validity, and Replicability

The Stranger Face Trust scale (SFT) and Imaginary Stranger Trust scale (IST) are two new self-report measures of generalized trust that assess trust in strangers—both real and imaginary—across four trust domains. Prior research has established the reliability and validity of SFT and IST, but a number of measurement validation tests remain. Across three separate studies, I assess the test–retest reliability, measurement invariance, predictive validity, and replicability of SFT and IST, with the misanthropy scale (MST) and generalized social trust scale (GST) serving as benchmarks. First, tests of internal consistency, test–retest reliability, and longitudinal measurement invariance established that all four generalized trust scales were acceptably reliable, with SFT and IST yielding greater overall reliability than MST and GST. Second, tests of multiple group measurement invariance revealed that SFT and IST were equivalent across gender, race, education, and age groups, while MST and GST were non-equivalent across the same sociodemographic groups. Third, an investment game established the predictive validity of SFT and MST, with IST and GST yielding poor predictive validity. Fourth, tests of factor structure and measurement invariance indicated that all four generalized trust scales replicated across samples. The present findings bolster the validity, reliability, and measurement equivalence of SFT and IST, while illustrating the compromised validity and measurement non-equivalence of MST and GST. Implications for the measurement of generalized trust are discussed.

Analyzing factorial survey data with structural equation models

In factorial survey designs, respondents evaluate multiple short descriptions of social objects (vignettes) that experimentally vary different levels of attributes of interest. Analytical methods (including individual-level regression analysis and multilevel models) estimate the weights (or utilities) assigned to the levels of the different attributes by participants to arrive at an overall response to the vignettes. In the current paper, we explain how data from factorial surveys can be analyzed in a structural equation modeling framework using an approach called structural equation modeling for within-subject experiments. We review the use of factorial surveys in social science research, discuss typically used methods to analyze factorial survey data, introduce the structural equation modeling for within-subject experiments approach, and present an empirical illustration of the proposed method. We conclude by describing several extensions, providing some practical recommendations, and discussing potential limitations.

A Rolling Panel Model of Cohort, Period, and Aging Effects for the Analysis of the General Social Survey

The linear dependence of age, period, and birth cohort is a challenge for the analysis of social change. With either repeated cross-sectional data or conventional panel data, raw change cannot be decomposed into over-time differences that are attributable to the effects of common experiences of alternative birth cohorts, features of the periods under observation, and the cumulation of lifecourse aging. This article proposes a rolling panel model for cohort, period, and aging effects, suggested by and tuned to the treble panel data collected for the General Social Survey from 2006 through 2014. While the model does not offer a general solution for the identification of the classical age-period-cohort accounting model, it yields warranted interpretations under plausible assumptions that are reasonable for many outcomes of interest. In particular, if aging effects can be assumed to be invariant over the course of an observation interval, and if separate panel samples of the full age distribution overlap within the same observation interval, then period and aging effects can be parameterized and interpreted separately, adjusted for cohort differences that pulse through the same observation interval. The estimated cohort effects during the observation interval are then interpretable as effects during the observation interval of entangled period and cumulated aging differences from before the observation interval.

Using Inverse Probability Weighting to Address Post-Outcome Collider Bias

We consider the problem of bias arising from conditioning on a post-outcome collider. We illustrate this with reference to Elwert and Winship (2014) but we go beyond their study to investigate the extent to which inverse probability weighting might offer solutions. We use linear models to derive expressions for the bias arising in different kinds of post-outcome confounding, and we show the specific situations in which inverse probability weighting will allow us to obtain estimates that are consistent or, if not consistent, less biased than those obtained via ordinary least squares regression.

Method Bias Mechanisms and Procedural Remedies

As a crucial method in organizational and social behavior research, self-report surveys must manage method bias. Method biases are distorted scores in survey response, distorted variance in variables, and distorted relational estimates between variables caused by method designs. Studies on method bias have focused on post hoc statistical control, but integrated analyses of the sociopsychological mechanism of method bias are lacking. This review proposes a framework for method bias and offers a relatively complete and detailed review of the sociopsychological and statistical mechanisms of four main types of method bias and their procedural remedies. This review proposes “reduce, remove, and rectify” as a guideline for researchers in survey design to address method bias. Finally, this review presents two directions for future methodology research.