<abstract xmlns="http://www.w3.org/1999/xhtml">When monitoring industrial processes, a Statistical Process Control tool, such as a multivariate Hotelling <italic>T</italic>2 chart is frequently used to evaluate multiple quality characteristics. However, research into the use of <italic>T</italic>2 charts for survey fieldwork–essentially a production process in which data sets collected by means of interviews are produced–has been scant to date. In this study, using data from the eighth round of the European Social Survey in Belgium, we present a procedure for simultaneously monitoring six response quality indicators and identifying outliers: interviews with anomalous results. The procedure integrates Kernel Density Estimation (KDE) with a <italic>T</italic>2 chart, so that historical “in-control” data or reference to the assumption of a parametric distribution of the indicators is not required. In total, 75 outliers (4.25%) are iteratively removed, resulting in an in-control data set containing 1,691 interviews. The outliers are mainly characterized by having longer sequences of identical answers, a greater number of extreme answers, and against expectation, a lower item nonresponse rate. The procedure is validated by means of ten-fold cross-validation and comparison with the minimum covariance determinant algorithm as the criterion. By providing a method of obtaining in-control data, the present findings go some way toward a way to monitor response quality, identify problems, and provide rapid feedbacks <italic>during</italic> survey fieldwork.</abstract>

When monitoring industrial processes, a Statistical Process Control tool, such as a multivariate Hotelling T2 chart is frequently used to evaluate multiple quality characteristics. However, research into the use of T2 charts for survey fieldwork–essentially a production process in which data sets collected by means of interviews are produced–has been scant to date. In this study, using data from the eighth round of the European Social Survey in Belgium, we present a procedure for simultaneously monitoring six response quality indicators and identifying outliers: interviews with anomalous results. The procedure integrates Kernel Density Estimation (KDE) with a T2 chart, so that historical “in-control” data or reference to the assumption of a parametric distribution of the indicators is not required. In total, 75 outliers (4.25%) are iteratively removed, resulting in an in-control data set containing 1,691 interviews. The outliers are mainly characterized by having longer sequences of identical answers, a greater number of extreme answers, and against expectation, a lower item nonresponse rate. The procedure is validated by means of ten-fold cross-validation and comparison with the minimum covariance determinant algorithm as the criterion. By providing a method of obtaining in-control data, the present findings go some way toward a way to monitor response quality, identify problems, and provide rapid feedbacks during survey fieldwork.

Identifying Outliers in Response Quality Assessment by Using Multivariate Control Charts Based on Kernel Density Estimation

Journal of Official Statistics

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

When monitoring industrial processes, a Statistical Process Control tool, such as a multivariate Hotelling T2 chart is frequently used to evaluate multiple...

Identifying Outliers in Response Quality Assessment by Using Multivariate Control Charts Based on Kernel Density Estimation

Published Online: Mar 13, 2021

Page range: 97 - 119

Received: Jun 01, 2019

Accepted: Sep 01, 2020

DOI: https://doi.org/10.2478/jos-2021-0005

Keywords
Kernel density estimation, Hotelling chart, multivariate control charts, response quality, ten-fold cross-validation

© 2020 Jiayun Jin et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Identifying Outliers in Response Quality Assessment by Using Multivariate Control Charts Based on Kernel Density Estimation

Published Online: Mar 13, 2021

Page range: 97 - 119

Received: Jun 01, 2019

Accepted: Sep 01, 2020

DOI: https://doi.org/10.2478/jos-2021-0005

KeywordsKernel density estimation, Hotelling chart, multivariate control charts, response quality, ten-fold cross-validation

© 2020 Jiayun Jin et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Keywords
Kernel density estimation, Hotelling chart, multivariate control charts, response quality, ten-fold cross-validation