scholarly journals Calibration Weighting for Nonresponse with Proxy Frame Variables (So that Unit Nonresponse Can Be Not Missing at Random)

2018 ◽  
Vol 34 (1) ◽  
pp. 107-120 ◽  
Author(s):  
Phillip S. Kott ◽  
Dan Liao
Keyword(s):  
Home Ownership ◽  
Mail Survey ◽  
Missing At Random ◽  
Energy Information Administration ◽  
Annual Income ◽  
Sampling Frame ◽  
Unit Nonresponse ◽  
Residential Energy Consumption Survey ◽  
Nationally Representative ◽  
Not Missing At Random

Abstract When adjusting for unit nonresponse in a survey, it is common to assume that the response/nonresponse mechanism is a function of variables known either for the entire sample before unit response or at the aggregate level for the frame or population. Often, however, some of the variables governing the response/nonresponse mechanism can only be proxied by variables on the frame while they are measured (more) accurately on the survey itself. For example, an address-based sampling frame may contain area-level estimates for the median annual income and the fraction home ownership in a Census block group, while a household’s annual income category and ownership status are reported on the survey itself for the housing units responding to the survey. A relatively new calibration-weighting technique allows a statistician to calibrate the sample using proxy variables while assuming the response/ nonresponse mechanism is a function of the analogous survey variables. We will demonstrate how this can be done with data from the Residential Energy Consumption Survey National Pilot, a nationally representative web-and-mail survey of American households sponsored by the U.S. Energy Information Administration.

Targeting Key Survey Variables at the Unit Nonresponse Treatment Stage

2020 ◽  
Author(s):  
David Haziza ◽  
Sixia Chen ◽  
Yimeng Gao
Keyword(s):  
Propensity Score ◽  
Simulation Study ◽  
Auxiliary Information ◽  
Missing At Random ◽  
Bias Reduction ◽  
Nonresponse Bias ◽  
Unit Nonresponse ◽  
Treatment Stage ◽  
Key Variables ◽  
Not Missing At Random

Abstract In the presence of nonresponse, unadjusted estimators are vulnerable to nonresponse bias when the characteristics of the respondents differ from those of the nonrespondents. To reduce the bias, it is common practice to postulate a nonresponse model linking the response indicators and a set of fully observed variables. Estimated response probabilities are obtained by fitting the selected model, which are then used to adjust the base weights. The resulting estimator, referred to as the propensity score-adjusted estimator, is consistent provided the nonresponse model is correctly specified. In this article, we propose a weighting procedure that may improve the efficiency of propensity score estimators for survey variables identified as key variables by making a more extensive use of the auxiliary information available at the nonresponse treatment stage. Results from a simulation study suggest that the proposed procedure performs well in terms of efficiency when the data are missing at random and also achieves an efficient bias reduction when the data are not missing at random. We further apply our proposed methods to 2017–2018 National Health Nutrition and Examination Survey.

scholarly journals Developing Calibration Weights and Standard-Error Estimates for a Survey of Drug-Related Emergency-Department Visits

2014 ◽  
Vol 30 (3) ◽  
pp. 521-532 ◽  
Author(s):  
Phillip S. Kott ◽  
C. Daniel Day
Keyword(s):  
Emergency Department ◽  
Error Estimates ◽  
Standard Error ◽  
Emergency Department Visits ◽  
Simple Random Sample ◽  
Statistical Efficiency ◽  
Sampling Frame ◽  
Unit Nonresponse ◽  
Department Size ◽  
Hospital Emergency Departments

Abstract This article describes a two-step calibration-weighting scheme for a stratified simple random sample of hospital emergency departments. The first step adjusts for unit nonresponse. The second increases the statistical efficiency of most estimators of interest. Both use a measure of emergency-department size and other useful auxiliary variables contained in the sampling frame. Although many survey variables are roughly a linear function of the measure of size, response is better modeled as a function of the log of that measure. Consequently the log of size is a calibration variable in the nonresponse-adjustment step, while the measure of size itself is a calibration variable in the second calibration step. Nonlinear calibration procedures are employed in both steps. We show with 2010 DAWN data that estimating variances as if a one-step calibration weighting routine had been used when there were in fact two steps can, after appropriately adjusting the finite-population correct in some sense, produce standard-error estimates that tend to be slightly conservative.

Using a National Register as a Sampling Frame from the Perspective of an Early Career Researcher

2019 ◽  
Vol 5 (1) ◽  
pp. 92-93
Author(s):  
A. Callanan
Keyword(s):  
Representative Sample ◽  
Target Population ◽  
Early Career ◽  
Sampling Frame ◽  
National Register ◽  
Advantages And Disadvantages ◽  
Early Career Researcher ◽  
Nationally Representative ◽  
General Dentists ◽  
Selection Of

Background: Selection of a sampling frame is a key component of conducting survey-based research. This article discusses the use of a national register, the Dental Register, as a sampling frame from the perspective of an early career researcher. Methods: While conducting a survey-based study of a nationally representative sample of general dentists in Ireland, I documented the difficulties I encountered while using a national register. As a research assistant and novice researcher, I recorded the advantages and disadvantages I discovered over the course of the project and its impact on the study. Conclusion: While using a national register has advantages such as a readily available sample of the target population, there are also inherent disadvantages depending on the manner in which records are recorded. Knowledge Transfer Statement: This article can be used as an informative guide to researchers in selecting a sampling frame, with particular emphasis on the use of a national register in selecting a nationally representative sample of dentists.

scholarly journals Responsiveness-informed multiple imputation and inverse probability-weighting in cohort studies with missing data that are non-monotone or not missing at random

2016 ◽  
Vol 27 (2) ◽  
pp. 352-363 ◽  
Author(s):  
James C Doidge
Keyword(s):  
Missing Data ◽  
Data Collection ◽  
Cohort Studies ◽  
Multiple Imputation ◽  
Missing At Random ◽  
Inverse Probability Weighting ◽  
Probability Weighting ◽  
Inverse Probability ◽  
Not Missing At Random ◽  
Over Time

Population-based cohort studies are invaluable to health research because of the breadth of data collection over time, and the representativeness of their samples. However, they are especially prone to missing data, which can compromise the validity of analyses when data are not missing at random. Having many waves of data collection presents opportunity for participants’ responsiveness to be observed over time, which may be informative about missing data mechanisms and thus useful as an auxiliary variable. Modern approaches to handling missing data such as multiple imputation and maximum likelihood can be difficult to implement with the large numbers of auxiliary variables and large amounts of non-monotone missing data that occur in cohort studies. Inverse probability-weighting can be easier to implement but conventional wisdom has stated that it cannot be applied to non-monotone missing data. This paper describes two methods of applying inverse probability-weighting to non-monotone missing data, and explores the potential value of including measures of responsiveness in either inverse probability-weighting or multiple imputation. Simulation studies are used to compare methods and demonstrate that responsiveness in longitudinal studies can be used to mitigate bias induced by missing data, even when data are not missing at random.

scholarly journals Low assets and financial stressors associated with higher depression during COVID-19 in a nationally representative sample of US adults

2020 ◽  
pp. jech-2020-215213
Author(s):  
Catherine K Ettman ◽  
Salma M Abdalla ◽  
Gregory H Cohen ◽  
Laura Sampson ◽  
Patrick M Vivier ◽  
...  
Keyword(s):  
Representative Sample ◽  
Home Ownership ◽  
Well Being ◽  
Period 2 ◽  
Time Period ◽  
Patient Health ◽  
The Usa ◽  
Nationally Representative ◽  
Health And Well Being ◽  
Probable Depression

BackgroundCOVID-19 and related containment policies have caused or heightened financial stressors for many in the USA. We assessed the relation between assets, financial stressors and probable depression during the COVID-19 pandemic.MethodsBetween 31 March 2020 and 13 April 2020, we surveyed a probability-based, nationally representative sample of US adults ages 18 and older using the COVID-19 and Life stressors Impact on Mental Health and Well-being survey (n=1441). We calculated the prevalence of probable depression using the Patient Health Questionnaire-9 (cut-off ≥10) and exposure to financial stressors by financial, physical and social assets categories (household income, household savings, home ownership, educational attainment and marital status). We estimated adjusted ORs and predicted probabilities of probable depression across assets categories and COVID-19 financial stressor exposure groups.ResultsWe found that (1) 40% of US adults experienced COVID-19-related financial stressors during this time period; (2) low assets (OR: 3.0, 95% CI 2.1 to 4.2) and COVID-19 financial stressor exposure (OR: 2.8, 95% CI 2.1 to 3.9) were each associated with higher odds of probable depression; and (3) among persons with low assets and high COVID-19 financial stressors, 42.7% had probable depression; and among persons with high assets and low COVID-19 financial stressors, 11.1% had probable depression. Persons with high assets and high COVID-19 financial stressors had a similar prevalence of probable depression (33.5%) as persons with low assets and low COVID-19 financial stressors (33.5%). The more assets a person had, the lower the level of probable depression.ConclusionPopulations with low assets are bearing a greater burden of mental illness during the COVID-19 pandemic.

scholarly journals Examining Validity-Related Correlations and Their Differences Under Assumption Violations

2019 ◽  
Vol 80 (2) ◽  
pp. 389-398
Author(s):  
Tenko Raykov ◽  
Abdullah A. Al-Qataee ◽  
Dimiter M. Dimitrov
Keyword(s):  
Latent Variable ◽  
Social Research ◽  
Tertiary Education ◽  
Model Fitting ◽  
Interval Estimation ◽  
Missing At Random ◽  
Entrance Examination ◽  
Data Sets ◽  
Modeling Methodology ◽  
Not Missing At Random

A procedure for evaluation of validity related coefficients and their differences is discussed, which is applicable when one or more frequently used assumptions in empirical educational, behavioral and social research are violated. The method is developed within the framework of the latent variable modeling methodology and accomplishes point and interval estimation of convergent and discriminant correlations as well as differences between them in cases of incomplete data sets with data not missing at random, nonnormality, and clustering effects. The procedure uses the full information maximum likelihood approach to model fitting and parameter estimation, does not assume availability of multiple indicators for underlying latent constructs, includes auxiliary variables, and accounts for within-group correlations on main response variables resulting from nesting effects involving studied respondents. The outlined procedure is illustrated on empirical data from a study using tertiary education entrance examination measures.

Sign in / Sign up

Export Citation Format

Share Document