Chapter 5: Population Inferences and Variance Estimation for NAEP Data

1992 ◽  
Vol 17 (2) ◽  
pp. 175-190 ◽  
Author(s):  
Eugene G. Johnson ◽  
Keith F. Rust
Keyword(s):  
Variance Estimation ◽  
Estimation Procedure ◽  
Sample Design ◽  
Sampling Weights ◽  
Sampling Variability ◽  
Complex Sample ◽  
The Stability ◽  
Replication Procedure ◽  
Variance Estimates ◽  
Complex Sample Design

In the National Assessment of Educational Progress (NAEP), population inferences and variance estimation are based on a randomization-based perspective where the link between the observed data and the population quantities of interest is given by the distribution of potential values of estimates over repeated samples from the same population using the identical sample design. Because NAEP uses a complex sample design, many of the assumptions underlying traditional statistical analyses are violated, and, consequently, analysis procedures must be adjusted to appropriately handle the structure of the sample. In this article, we discuss the use of sampling weights in deriving population estimates and consider the effect of nonresponse and undercoverage on those estimates. We also discuss the estimation of sampling variability from complex sample surveys, concentrating on the jackknife repeated replication procedure—the variance estimation procedure used by NAEP—and address the use of a simple approximation to sampling variability. Finally, we discuss measures of the stability of variance estimates.

How to use replicate weights in health survey analysis using the National Nutrition and Physical Activity Survey as an example

2019 ◽  
Vol 22 (18) ◽  
pp. 3315-3326
Author(s):  
Carole L Birrell ◽  
David G Steel ◽  
Marijka J Batterham ◽  
Ankur Arya
Keyword(s):  
Physical Activity ◽  
Health Survey ◽  
Large Scale ◽  
Sample Design ◽  
Sampling Weights ◽  
Complex Sample ◽  
Unbiased Estimates ◽  
Nutrition And Physical Activity ◽  
Activity Survey ◽  
Complex Sample Design

AbstractObjective:To conduct nutrition-related analyses on large-scale health surveys, two aspects of the survey must be incorporated into the analysis: the sampling weights and the sample design; a practice which is not always observed. The present paper compares three analyses: (1) unweighted; (2) weighted but not accounting for the complex sample design; and (3) weighted and accounting for the complex design using replicate weights.Design:Descriptive statistics are computed and a logistic regression investigation of being overweight/obese is conducted using Stata.Setting:Cross-sectional health survey with complex sample design where replicate weights are supplied rather than the variables containing sample design information.Participants:Responding adults from the National Nutrition and Physical Activity Survey (NNPAS) part of the Australian Health Survey (2011–2013).Results:Unweighted analysis produces biased estimates and incorrect estimates of se. Adjusting for the sampling weights gives unbiased estimates but incorrect se estimates. Incorporating both the sampling weights and the sample design results in unbiased estimates and the correct se estimates. This can affect interpretation; for example, the incorrect estimate of the OR for being a current smoker in the unweighted analysis was 1·20 (95 % CI 1·06, 1·37), t= 2·89, P = 0·004, suggesting a statistically significant relationship with being overweight/obese. When the sampling weights and complex sample design are correctly incorporated, the results are no longer statistically significant: OR = 1·06 (95 % CI 0·89, 1·27), t = 0·71, P = 0·480.Conclusions:Correct incorporation of the sampling weights and sample design is crucial for valid inference from survey data.

Complex Sample Design Effects and Health Insurance Variance Estimation

2009 ◽  
Vol 43 (2) ◽  
pp. 346-366 ◽  
Author(s):  
ROBERT B. NIELSEN ◽  
MICHAEL DAVERN ◽  
ARTHUR JONES ◽  
JOHN L. BOIES
Keyword(s):  
Health Insurance ◽  
Variance Estimation ◽  
Sample Design ◽  
Design Effects ◽  
Complex Sample ◽  
Complex Sample Design

scholarly journals Synthetic Multiple-Imputation Procedure for Multistage Complex Samples

2016 ◽  
Vol 32 (1) ◽  
pp. 231-256 ◽  
Author(s):  
Hanzhi Zhou ◽  
Michael R. Elliott ◽  
Trivellore E. Raghunathan
Keyword(s):  
Multiple Imputation ◽  
Survey Design ◽  
Efficient Estimation ◽  
Sample Design ◽  
Sampling Weights ◽  
Design Data ◽  
Item Level ◽  
Complex Sample ◽  
Analytic Strategy ◽  
Modeling Strategy

Abstract Multiple imputation (MI) is commonly used when item-level missing data are present. However, MI requires that survey design information be built into the imputation models. For multistage stratified clustered designs, this requires dummy variables to represent strata as well as primary sampling units (PSUs) nested within each stratum in the imputation model. Such a modeling strategy is not only operationally burdensome but also inferentially inefficient when there are many strata in the sample design. Complexity only increases when sampling weights need to be modeled. This article develops a generalpurpose analytic strategy for population inference from complex sample designs with item-level missingness. In a simulation study, the proposed procedures demonstrate efficient estimation and good coverage properties. We also consider an application to accommodate missing body mass index (BMI) data in the analysis of BMI percentiles using National Health and Nutrition Examination Survey (NHANES) III data. We argue that the proposed methods offer an easy-to-implement solution to problems that are not well-handled by current MI techniques. Note that, while the proposed method borrows from the MI framework to develop its inferential methods, it is not designed as an alternative strategy to release multiply imputed datasets for complex sample design data, but rather as an analytic strategy in and of itself.

scholarly journals kg_nchs: A command for Korn–Graubard confidence intervals and National Center for Health Statistics’ Data Presentation Standards for Proportions

2019 ◽  
Vol 19 (3) ◽  
pp. 510-522
Author(s):  
Brian W. Ward
Keyword(s):  
Confidence Interval ◽  
Degrees Of Freedom ◽  
Statistical System ◽  
Health Statistics ◽  
Sample Design ◽  
Data Presentation ◽  
Federal Statistical ◽  
Complex Sample ◽  
The U.S ◽  
Complex Sample Design

In August 2017, the National Center for Health Statistics (NCHS), part of the U.S. Federal Statistical System, published new standards for determining the reliability of proportions estimated using their data. These standards require one to take the Korn–Graubard confidence interval (CI), CI widths, sample size, and degrees of freedom to assess reliability of a proportion and determine whether it can be presented. The assessment itself involves determining whether several conditions are met. In this article, I present kg_nchs, a postestimation command that is used following svy: proportion. It allows Stata users to a) calculate the Korn–Graubard CI and associated statistics used in applying the NCHS presentation standards for proportions and b) display a series of three dichotomous flags that show whether the standards are met. I provide empirical examples to show how kg_nchs can be used to easily apply the standards and prevent Stata users from needing to perform manual calculations. While developed for NCHS survey data, this command can also be used with data that stem from any survey with a complex sample design.

scholarly journals Assessing usual dietary intake in complex sample design surveys: the National Dietary Survey

2013 ◽  
Vol 47 ◽  
pp. 171s-176s
Author(s):  
Flávia dos Santos Barbosa ◽  
Rosely Sichieri ◽  
Washington Leite Junger
Keyword(s):  
Dietary Intake ◽  
Sample Design ◽  
Dietary Survey ◽  
Complex Sample ◽  
Complex Sample Design
Sign in / Sign up

Export Citation Format

Share Document