Analysis of Complex Survey Samples

By Thomas Lumley. 2004

2,390 citations recorded by the source corpus.

I present software for analysing complex survey samples in R. The sampling scheme can be explicitly described or represented by replication weights. Variance estimation uses either replication or linearisation.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/619ff8e52ce13be69cef8c478768ec060396312e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

Difficulties in solving context-based PISA mathematics tasks: An analysis of students’ errors (2014, methodology): “Because these responses are nested within students, a chi-square with a Rao-Scott adjustment for clustered data in the R survey package was used (Lumley, 2004, 2012).”
Imputation with the R Package VIM (2016, methodology): “In addition, objects from the package survey (Lumley 2016, 2004) can be imputed.”
From start to finish: a framework for the production of small area official statistics (2018, background): “In particular, the section on complex survey designs includes packages, like survey (Lumley, 2004) and sampling (Tillé and Matei, 2012), that can be used for point and variance estimation of direct estimators of means, totals, ratios and quantiles under complex survey designs.”
Fake news on Twitter during the 2016 U.S. presidential election (2019, methodology): “Confidence intervals for weighted survey data are calculated using the survey package in R (37); confidence intervals for our Twitter panel are computed using the approximate method in (38).”
Proxy Pattern-Mixture Analysis for Survey Nonresponse (2011, methodology): “Design-based estimators were computed using the “survey” routines in R, which estimate variances using Taylor series linearizations (Lumley 2004).”
Prognostic score–based balance measures for propensity score methods in comparative effectiveness research (2013, methodology): “Effect estimates were calculated using the R survey package [21].”
Visualizing Fit and Lack of Fit in Complex Regression Models with Predictor Effect Plots and Partial Residuals (2018, methodology): “In the effects package for R, we have included methods for linear, multivariate linear, and generalized linear models fit by the standard lm and glm functions and by the svyglm function in the survey package (Lumley 2004); linear models fit by generalized least squares using the gls function in the ”
Images and Misinformation in Political Groups: Evidence from WhatsApp in India (2020, methodology): “This characterization is similar to the qualitative characterization by (Hemsley and Snyder [2018]).”
ipw: An R Package for Inverse Probability Weighting (2011, methodology): “Weighting the original observations by the stabilized weights (7) to adjust for confounding, we can fit the MSM, estimating the marginal causal effect of A on Y , Ya = β0 + β1a, (8) which can be done as follows, using a robust standard error estimate from the survey package (Lumley 2004):”
lavaan.survey: An R Package for Complex Survey Analysis of Structural Equation Models (2014, methodology): “6.2.2) notes that where r n is a term that converges to zero as the sample size increases (see also Lumley 2004, 3-4).”
Teachers Are People Too: Examining the Racial Bias of Teachers Compared to Other American Adults (2020, methodology): “Instead, we use Taylor-series linearization for the computation of variances (see Lumley, 2004).”