Inverse probability weighting for clustered nonresponse

Skinner, Chris J.; D’arrigo,

doi:10.1093/biomet/asr058

Cited by 36 publications

(44 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such a random effects model would allow testing of whether the impact of the exam scores varies across schools. However, Skinner and D'Arrigo (2012) show that basing weights of the type we construct in Section 4 on a random effects model can in fact be detrimental in bias terms.…”

Section: Pupil Responsementioning

confidence: 95%

Non-Response Biases in Surveys of Schoolchildren: The Case of the English Programme for International Student Assessment (PISA) samples

Micklewright¹,

Schnepf

Skinner

2012

Journal of the Royal Statistical Society Series A: Statistics in Society

View full text Add to dashboard Cite

We analyse response patterns to an important survey of school children, exploiting rich auxiliary information on respondents' and non-respondents' cognitive ability that is correlated both with response and the learning achievement that the survey aims to measure. The survey is the Programme for International Student Assessment (PISA), which sets response thresholds in an attempt to control data quality. We analyse the case of England for 2000 when response rates were deemed high enough by the PISA organisers to publish the results, and 2003, when response rates were a little lower and deemed of sufficient concern for the results not to be published. We construct weights that account for the pattern of non-response using two methods, propensity scores and the GREG estimator. There is clear evidence of biases, but there is no indication that the slightly higher response rates in 2000 were associated with higher quality data. This underlines the danger of using response rate thresholds as a guide to data quality.Key words: non-response, bias, school survey, data linkage, PISA. AcknowledgementsThis research was funded by ESRC grant RES-062-23-0458 ('Hierarchical analysis of unit non-response in sample surveys'). It builds on earlier work funded by the (then) Department for Education and Skills (DfES). We thank staff at DfES, the Office for National Statistics, and the Fischer Family Trust for their help in setting up and interpreting the data. We alone remain responsible for their use. We are grateful for comments to seminar and conference participants at DfES, the ESRC Research Methods Festival, the International Association for Research on Income and Wealth, the European University Institute, Southampton, and Bremen. Micklewright thanks the Institute for Research on Poverty at the University of Wisconsin-Madison for hospitality during a sabbatical visit when he worked on the paper. We thank the editors and referees for very helpful suggestions.2

show abstract

Section: Pupil Responsementioning

confidence: 95%

Non-Response Biases in Surveys of Schoolchildren: The Case of the English Programme for International Student Assessment (PISA) samples

Micklewright¹,

Schnepf

Skinner

2012

Journal of the Royal Statistical Society Series A: Statistics in Society

View full text Add to dashboard Cite

show abstract

“…Lettingδ denote the estimator of δ in (13) using the simple informative estimator, the 't-statistic' obtained by dividingδ by its bootstrap standard error will have a standard normal distribution under noninformativeness and may be used to test this assumption. Note that the validity of this null distribution does not depend on the model assumption in (11) but only on the general assumption in (2). Similarly, it is possible to construct a test from the t-statistic formed by dividing the two-step estimator of c in (9) by its bootstrap standard error.…”

Section: Estimation Of β and Testing Of Informativenessmentioning

confidence: 83%

“…Now we vary the quantity δ in (11) which governs the degree of informativeness. We repeat the following steps, again 10,000 times.…”

Section: Study 2 Based On Simple Informative Nonresponse Modelmentioning

confidence: 99%

See 1 more Smart Citation

Estimation of a cluster-level regression model under nonresponse within clusters

Lawson

Skinner

2017

METRON

View full text Add to dashboard Cite

When sample surveys are clustered and subject to non-response, it is possible to study cluster-level association between response rates and cluster-level quantities derived from survey variables. The existence of association may suggest informative nonresponse with possible biasing effects. In this paper, this problem is studied for the case where the aim is to fit a cluster-level regression model. Two possible underlying models for nonresponse with potential biasing effects are considered. Alternative estimators of regression coefficients under these models are proposed. The properties of these estimators are studied in two simulation studies and with real data from a survey of employees, where the clusters consist of workplaces.

show abstract

“…Yuan and Little (2007) considered a special case of CSNI where the response indicator depends on cluster-specific covariates. A few methods have been proposed in the context of survey sampling under CSNI in the presence of covariates that vary within cluster (Skinner and D'Arrigo, 2011;.…”

Section: Introductionmentioning

confidence: 99%

A Robust Calibration-Assisted Method for Linear Mixed Effects Model Under Cluster-Specific Nonignorable Missingness

Kwon¹,

Kim²,

Paik³

et al. 2018

STAT SINICA

View full text Add to dashboard Cite

We propose a method for linear mixed effects models when the covariates are completely observed but the outcome of interest is subject to missing under cluster-specific nonignorable (CSNI) missingness. Our strategy is to replace missing quantities in the full-data objective function with unbiased predictors derived from inverse probability weighting and calibration technique. The proposed approach can be applied to estimating equations or likelihood functions with modified E-step, and does not require numerical integration as previous methods. Unlike usual inverse probability weighting, the proposed method does not require correct specification of the response model as long as the CSNI assumption is correct, and renders inference under CSNI without a full distributional assumption. Consistency and asymptotic normality are shown with a consistent variance estimator. Simulation results and a real data example are presented.

show abstract

Inverse probability weighting for clustered nonresponse

Cited by 36 publications

References 20 publications

Non-Response Biases in Surveys of Schoolchildren: The Case of the English Programme for International Student Assessment (PISA) samples

Non-Response Biases in Surveys of Schoolchildren: The Case of the English Programme for International Student Assessment (PISA) samples

Estimation of a cluster-level regression model under nonresponse within clusters

A Robust Calibration-Assisted Method for Linear Mixed Effects Model Under Cluster-Specific Nonignorable Missingness

Contact Info

Product

Resources

About