Sample Design, Sample Augmentation, and Estimation for Wave 2 of the NSHAP

Abstract

The inclusion of coresident partners enhanced the study by allowing the examination of how intimate, household relationships are related to health trajectories and by augmenting the size of the NSHAP sample size for this and future waves. The uncommon strategy of returning to Wave 1 nonrespondents reduced potential bias by ensuring that to the extent possible the whole of the original sample forms the basis for the field effort. NSHAP Wave 2 achieved its field objectives of consolidating the panel, recruiting their resident spouses or romantic partners, and converting a significant proportion of Wave 1 nonrespondents.

Changes in Prescription and Over-the-Counter Medication and Dietary Supplement Use Among Older Adults in the United States, 2005 vs 2011

“…As has been previously described, 26 a sample of 4400 individuals was drawn from households screened in 2004 (one individual per household), of whom 4017 were determined to be eligible and 3005 successfully interviewed between July 2005 and March 2006. The weighted response rate in this first wave (baseline), calculated using the American Association for Public Opinion Research’s response rate 2, was 75.5%.…”

Elder mistreatment predicts later physical and psychological health: Results from a national longitudinal study

“…For W1 in 2005-06, NSHAP obtained an overall response rate of 75.5%, resulting in a sample of 3,005 adults aged 57–85 (O’Muircheartaigh, Eckman, & Smith, 2009; Waite et al, 2014a). Five years later in 2010-11 NSHAP re-interviewed 2,261 surviving W1 respondents (as well as 161 W1 non-respondents and 955 spouses and cohabiting romantic partners) for W2 (O’Muircheartaigh, English, Pedlow, & Kwok 2014; Waite et al, 2014b). Data for both waves are collected by an interviewer-administered in-home questionnaire, an in-home biomeasure collection procedure, and a self-administered post-interview questionnaire (leave-behind questionnaire or LBQ).…”

“…First, we used the weights provided by NSHAP in all analyses to account for non-response at W2 (O’Muircheartaigh, Eckman, & Smith, 2009; O’Muircheartaigh et al, 2014). Second, to account for any other missing data 1 , we used -ice- in Stata 14 to multiply impute (m=5) missing values of the dependent and independent variables, and used -mi estimate- to analyze the imputed dataset.…”

“…These methods are appropriate for these variables with valid zeroes and long right tails in their distributions. All models are weighted for non-response and sample design (O’Muircheartaigh et al, 2009; O’Muircheartaigh et al, 2014), and include controls for gender, race, education, marital status, and age at W2. Additional control variables are added in a stepwise manner.…”

The power of a paired t-test with a covariate

“…For example, surveys of older adults may wish to find evidence of increasing or decreasing depression symptoms (see, e.g., O’Muircheartaigh, English, Pedlow, & Kwok, 2014; Payne, Hedberg, Kozloski, Dale, & McClintock, 2014). Given limited resources for any study, whether observational or experimental, it is important to maximize power for detecting such changes.…”