Does Weighting for Nonresponse Increase the Variance of Survey Means

By Roderick J. A. Little, Sonya Vartivarian. 2004

312 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/917560b6a7fbdba46c24ae90c40d84444cdae68d

How researchers see this paper

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

Citations in the corpus

Marginal Mean Weighting Through Stratification: Adjustment for Selection Bias in Multilevel Data (2010, methodology): “For example, to adjust for nonresponse in surveys, Little and Vartivarian (2003, 2005) suggested computing a propensity score for response and creating adjustment cells based on this score.”
Nonresponse weighting adjustment using estimated response probability (2007, background): “Some authors, including Rosenbaum (1987), Robins, Rotnitzky & Zhao (1994), and Little & Vartivarian (2005), noted that the estimator using the estimated response probability can be mor efficient than the estimator using the true response probability, but they did not fully explain when and why this…”
Proxy Pattern-Mixture Analysis for Survey Nonresponse (2011, background): “This approach is attractive since it gives appropriate credit to the availability of good predictors of Y in the auxiliary data as well as a high response rate, and arguably good prediction of the survey outcomes is a key feature of good covariates; in particular, conditioning on a covariate Z that ”
On the Construction of Imputation Classes in Surveys (2007, background): “This point was also noted by Little (1986) and Little & Vartivarian (2005).”
Attrition in developmental psychology (2017, background): “It is preferable to choose auxiliary variables by intentionally collecting variables that are theoretically important to the response variables as well as to other key variables that might also contain missingness (Little & Vartivarian, 2005).”
A Review of Hot Deck Imputation for Survey Non‐response (2010, background): “Table 1, from Little & Vartivarian (2005), summarizes the effect of high or low levels of these associations on bias and variance of the estimated mean of Y .”
ANALYZING ESTABLISHMENT NONRESPONSE USING AN INTERPRETABLE REGRESSION TREE MODEL WITH LINKED ADMINISTRATIVE DATA (2012, methodology): “Further, adjusting for large amounts of nonresponse may induce more variance in the estimator [Little and Vartivarian (2005)], as well as a loss of confidence in the data by the stake-holders.”
NICE DSU Technical Support Document 18: Methods for population-adjusted indirect comparisons in submissions to NICE (2016, background): “This approximate ESS is only accurate if the weights are fixed and known, or if they are uncorrelated with outcome – neither of which is true here; as such, this approximation is likely to be an underestimate of the true ESS.(40) However, small effective sample sizes are an indication that the weigh”
Does more balanced survey response imply less non‐response bias? (2016, background): “…is unaffected by the inclusion of auxiliary variables that do not relate to variables of interest nor to non-response, but for the precision of estimators it is a crucial choice that has been much debated in the literature; see for example Little and Vartivarian (2005) and Huber et al. (2013).”