1,721,133 research outputs found
On the definition of a confounder
No full textThe causal inference literature has provided a clear formal definition of confounding expressed in terms of counterfactual independence. The causal inference literature has not, however, produced a clear formal definition of a confounder, as it has given priority to the concept of confounding over that of a confounder. We consider a number of candidate definitions arising from various more informal statements made in the literature. We consider the properties satisfied by each candidate definition, principally focusing on (i) whether under the candidate definition control for all "confounders" suffices to control for "confounding" and (ii) whether each confounder in some context helps eliminate or reduce confounding bias. Several of the candidate definitions do not have these two properties. Only one candidate definition of those considered satisfies both properties. We propose that a "confounder" be defined as a pre-exposure covariate C for which there exists a set of other covariates X such that effect of the exposure on the outcome is unconfounded conditional on (X,C) but such that for no proper subset of (X,C) is the effect of the exposure on the outcome unconfounded given the subset. A variable that helps reduce bias but not eliminate bias we propose referring to as a "surrogate confounder"
A new criterion for confounder selection
No full textWe propose a new criterion for confounder selection when the underlying causal structure is unknown and only limited knowledge is available. We assume all covariates being considered are pretreatment variables and that for each covariate it is known (i) whether the covariate is a cause of treatment, and (ii) whether the covariate is a cause of the outcome. The causal relationships the covariates have with one another is assumed unknown. We propose that control be made for any covariate that is either a cause of treatment or of the outcome or both. We show that irrespective of the actual underlying causal structure, if any subset of the observed covariates suffices to control for confounding then the set of covariates chosen by our criterion will also suffice. We show that other, commonly used, criteria for confounding control do not have this property. We use formal theory concerning causal diagrams to prove our result but the application of the result does not rely on familiarity with causal diagrams. An investigator simply need ask, “Is the covariate a cause of the treatment?” and “Is the covariate a cause of the outcome?” If the answer to either question is “yes” then the covariate is included for confounder control. We discuss some additional covariate selection results that preserve unconfoundedness and that may be of interest when used with our criterion
A complete graphical criterion for the adjustment formula in mediation analysis
No full textVarious assumptions have been used in the literature to identify natural direct and indirect effects in mediation analysis. These effects are of interest because they allow for effect decomposition of a total effect into a direct and indirect effect even in the presence of interactions or non-linear models. In this paper, we consider the relation and interpretation of various identification assumptions in terms of causal diagrams interpreted as a set of non-parametric structural equations. We show that for such causal diagrams, two sets of assumptions for identification that have been described in the literature are in fact equivalent in the sense that if either set of assumptions holds for all models inducing a particular causal diagram, then the other set of assumptions will also hold for all models inducing that diagram. We moreover build on prior work concerning a complete graphical identification criterion for covariate adjustment for total effects to provide a complete graphical criterion for using covariate adjustment to identify natural direct and indirect effects. Finally, we show that this criterion is equivalent to the two sets of independence assumptions used previously for mediation analysis
Assessing religion and spirituality in a cross-cultural sample: development of religion and spirituality items for the Global Flourishing Study
No full textThe Global Flourishing Survey (GFS) was initiated to provide an open-access, longitudinal study of health and well-being using a participant panel from 22 countries around the world. At the core of the GFS are questions on religion and spirituality—notoriously difficult to assess in a cross-cultural context. Additionally, the longitudinal aspect will allow for tests of within-person change over time. In developing the religion and spirituality items, we received suggestions and feedback from over 130 scholars, and the items underwent several rounds of peer-review by experts in the field. The preliminary survey items were also made publicly available to gather more feedback. Experts at Gallup then translated candidate items to ensure consistency across languages/cultures. Here, we present the results of cognitive interviews of 230 participants in 22 religiously diverse countries regarding the efficiency, efficacy, and difficulty of our candidate items. In the spirit of open science, we wish to share our findings from the interviews and provide recommendations regarding Likert scale usage, item specificity, assessment of God representations, and inclusivity when assessing religion and spirituality across cultures. In this, we aim to assist other researchers and support confidence in the reliability and validity of GFS data when it becomes publicly available
Childhood predictors of adults’ belief in god, gods, and spiritual forces across 22 countries
No full textReligion is an integral part of everyday life for billions of people, yet little is known about the developmental antecedents of religious belief outside of Western cultures. Using data from over 200,000 individuals across 22 countries, we evaluate several childhood predictors of belief in God, gods, and spiritual forces (Belief in God) in adulthood. We hypothesized that these childhood experiences, personal attributes, and familial or social circumstances would have meaningful and varied associations with Belief in God as adults, with the strength of these associations differing by country, reflecting diverse cultural influences. Most candidate predictors (e.g., parental marital status, childhood socioeconomic status, abuse, being an outsider, and immigration) were associated with Belief in God in some countries but with substantial variation. However, when pooled across countries, only childhood religious service attendance, birth cohort, and gender were significant predictors. Yet there was important variation even for these predictors, and no predictor had a consistent association across all countries. Though this cross-sectional design is limited in allowing causal inference, results provide insights into early-life experiences that might contribute to adults’ Belief in God. The heterogeneity of results highlights the importance of considering any childhood predictor within its social and cultural context.</p
Recommended from our members
Outcome-wide Epidemiology
No full textThe author proposes that epidemiologic studies should more often assess the associations of a single exposure with multiple outcomes simultaneously. Such “outcome-wide epidemiology” will be especially important for exposures that may be beneficial for some outcomes but harmful for others. Outcome-wide epidemiology may also be helpful in prioritizing public health recommendations. Methodologically, the conduct of outcome-wide epidemiology will generally be more straightforward than recent proposals for exposure-wide epidemiologic studies, in which the associations between a single outcome and many exposures are assessed simultaneously. Such exposure-wide studies are likely to be subject to numerous biases because of the inability to make simultaneous confounding control and because exposures are likely to affect, and mediate the effects of, other exposures. These problems simplify considerably in an outcome-wide approach when a single exposure is being considered. Moreover, outcome-wide approaches will generally be more useful than exposure-wide approaches in shaping public health recommendations.Version of Recor
Recommended from our members
The Role of Religious Communities and Practices in Mitigating the Impact of Racism, White Supremacy, and Discrimination on Health
No full textThe overarching goals of this dissertation are: 1) to examine the extent to which practices of religion/spirituality act as a buffer against the negative impacts of experiences of discrimination on health; and 2) to examine the ways that religious communities can act as a site of intervention to influence negative racial attitudes/beliefs/practices among members of predominantly White churches. The dissertation accepts as a premise that experiences of racism negatively impact health and that churches can be effective sites of intervention for health promotion.
The first paper examines the relationship between experiences of discrimination and psychological well-being, and the extent to which aspects of religion/spirituality buffer that relationship. Using the longitudinal dataset of Midlife in the U.S., this paper suggests that meditation/chanting daily or more, service attendance weekly or more, and attendance at other church/temple activities weekly or more can have a buffering effect on the ways that discrimination harms psychological well-being.
The second paper examines the ways that church engagement in racial justice contributes to individual involvement in racial justice activities. Using three timepoints of cross-sectional data from two different predominantly White religious denominations, the study suggests that there is a stronger association between hearing a sermon on racial justice or attending a workshop, training, or study group at one’s church and individual-level involvement in racial justice than between those individual-level factors and attending a church that either invited a guest preacher of a different racial/ethnic identity from one’s congregation or worshipping jointly with a congregation made up of people primarily of a different racial/ethnic identity from one’s congregation.
The final paper evaluates a teach-in on white supremacy within a predominantly White liberal religious denomination in the U.S., and whether this teach-in was associated with increased self-efficacy to address racism and increased feelings of belonging in one’s own church. Results suggest that participants in the teach-in did not feel increased self-efficacy to a level of statistical significance. Feelings of connectedness increased for White participants but decreased for participants of color. Qualitative findings revealed a mixed reaction to the use of the term “white supremacy” in the framing of the teach-in.Social and Behavioral Science
Recommended from our members
Statistical Methods for High-Dimensional Data in Genetic Epidemiology
No full textRecent technological advancements have enabled us to collect an unprecedented amount of genetic epidemiological data. The overarching goal of these genetic epidemiology studies is to uncover the underlying biological mechanisms so that improved strategies for disease prevention and management can be developed. To efficiently analyze and interpret high-dimensional biological data, it is imperative to develop novel statistical methods as conventional statistical methods are generally not applicable or are inefficient. In this dissertation, we introduce three novel, powerful and computationally efficient kernel machine set-based association tests for analyzing high-throughput genetic epidemiological data. In the first chapter, we construct a test for identifying common genetic variants that are predictive of a time-to-event outcome. In the second chapter, we develop a test for identifying gene-environment interactions for common genetic variants. In the third chapter, we propose a test for identifying gene-environment interactions for rare genetic variants
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
