EDA for HLM: Visualization when Probabilistic Inference Fails

  title={EDA for HLM: Visualization when Probabilistic Inference Fails},
  author={Jake Bowers and Katherine W. Drake},
  journal={Political Analysis},
  pages={301 - 326}
Nearly all hierarchical linear models presented to political science audiences are estimated using maximum likelihood under a repeated sampling interpretation of the results of hypothesis tests. Maximum likelihood estimators have excellent asymptotic properties but less than ideal small sample properties. Multilevel models common in political science have relatively large samples of units like individuals nested within relatively small samples of units like countries. Often these level-2… Expand
Using Graphs Instead of Tables to Improve the Presentation of Empirical Results in Political Science
When political scientists present empirical results, they are much more likely to use tables rather than graphs, despite the fact that the latter greatly increases the clarity of presentation andExpand
Exploratory Data Analysis
The philosophical justification for EDA is presented in terms of C.S. Pierce's concept of abduction and the recognition of a broad range of analytic needs that arise throughout the research process. Expand
Using Graphs Instead of Tables in Political Science
When political scientists present empirical results, they are much more likely to use tables than graphs, despite the fact that graphs greatly increases the clarity of presentation and makes itExpand
Estimating the Impact of State Policies and Institutions with Mixed-Level Data
Researchers are often interested in the effects of state policies and institutions on individual behavior or other outcomes in sub-state-level observational units, such as election results in stateExpand
Multilevel Models for the Analysis of Comparative Survey Data: Common Problems and Some Solutions
An overview over the application of mixed models (multilevel models) to comparative survey data where the context units of interest are countries is provided, with a focus on the small-N problem, influential cases (outliers) and the issue of omitted variables at the country level. Expand
Experimental evaluation of students' performance in judging statistical visualizations
Constant misinterpretation of the p-value and the absence of an assessment of the underlying assumptions of statistical models have led to alarmingly low replicability in, among others, the fieldsExpand
Empirical Strategies for Various Manifestations of Multilevel Data
Equivalent separate-subsample (two-step) and pooled-sample (one-step) strategies exist for any multilevel-modeling task, but their relative practicality and efficacy depend on dataset dimensions andExpand
Regression Analysis of Country Effects Using Multilevel Data: A Cautionary Tale
Cross-national differences in outcomes are often analysed using regression analysis of multilevel country datasets, examples of which include the ECHP, ESS, EU-SILC, EVS, ISSP, and SHARE. We reviewExpand
Problems Using Aggregate Data to Infer Individual Behavior: Evidence from Law, Finance, and Ownership Concentration
Cross-country comparisons in finance use two distinct approaches. Some analyze country averages; others analyze the underlying firm-level data. The influential finding of an inverse relation betweenExpand
Context and the Economic Vote: A Multilevel Analysis
Voters use observed economic performance to infer the competence of incumbent politicians. These economic perceptions enter the voter's utility calculations modified by a weight that is minimizedExpand


Cross-Level Inference
In the last few years, new disputes have erupted over the use of group averages from census areas or voting districts to draw inferences about individual social behaviour. Social scientists, policyExpand
Estimation and inference in econometrics
A theme of the text is the use of artificial regressions for estimation, reference, and specification testing of nonlinear models, including diagnostic tests for parameter constancy, serial correlation, heteroscedasticity, and other types of mis-specification. Expand
▪ Abstract Bayesian data analysis relies on Bayes' Theorem, using data to update prior beliefs about parameters. In this review I introduce and contrast Bayesian analysis with conventionalExpand
Robustness issues in multilevel regression analysis
A multilevel problem concerns a population with a hierarchical structure. A sample from such a population can be described as a multistage sample. First, a sample of higher level units is drawn (e.g.Expand
Exploratory Data Analysis for Complex Models
This article proposes an approach to unify exploratory data analysis with more formal statistical methods based on probability models, developed in the context of examples from fields including psychology, medicine, and social science. Expand
Modeling Multilevel Data Structures
data are becoming quite common in political science and provide numerous opportunities for theory testing and development. Unfortunately this type of data typically generates a number of statisticalExpand
Practical implications of modes of statistical inference for causal effects and the critical role of the assignment mechanism.
  • D. Rubin
  • Computer Science, Medicine
  • Biometrics
  • 1991
A fundamental conclusion is that in nonrandomized studies, sensitivity of inference to the assignment mechanism is the dominant issue, and it cannot be avoided by changing modes of inference, for instance, by changing from randomization-based to Bayesian methods. Expand
A Procedure for Robust Estimation and Inference in Linear Regression
A procedure which should fit many purposes reasonably well for robust regression estimators and the reluctance to use the straightforward inference based on asymptotics is proposed. Expand
Estimating Regression Models in Which the Dependent Variable Is Based on Estimates
Researchers often use as dependent variables quantities estimated from auxiliary data sets. Estimated dependent variable (EDV) models arise, for example, in studies where counties or states are theExpand
Unifying Political Methodology: The Likelihood Theory of Statistical Inference
One of the hallmarks of the development of political science as a discipline has been the creation of new methodologies by scholars within the discipline--methodologies that are well-suited to theExpand