Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 540 - Statistical Methods for Adolescent HIV Trials
Type: Topic Contributed
Date/Time: Thursday, August 6, 2020 : 1:00 PM to 2:50 PM
Sponsor: Biometrics Section
Abstract #313714
Title: Using Multiple Imputation to Account for Uncertainty in Summary Scores Derived from Likert Scale Survey Items
Author(s): Evan Kwiatkowski*
Companies: University of North Carolina At Chapel Hill
Keywords: incomplete data; missing at random; MICE; sequential regression multiple imputation

Readiness to provide Pre-Exposure Prophylaxis (PrEP) and factors that influence the adoption of PrEP services were assessed using a survey among providers and administrators of Title X-funded family planning clinics in the Southern US. Outcome measures were derived as the mean of related Likert scale survey items identified as having high internal consistency based on Cronbach’s Alpha. The relationship between Readiness for PrEP Implementation and factors that influence the adoption of PrEP services was analyzed using a covariate-adjusted linear mixed model that included a clinic-specific random intercept. Missing data was present for survey items contributing to outcome measures, which presented challenges for estimation and inference. Multiple imputation by chained equations was used to impute missing survey responses, which were then used to compute summary outcome measures from the imputed data. Survey item responses were imputed using full conditional specifications based on truncated regression models. Predictors used in each conditional specification were restricted to those variables which were determined to be related a priori based on the internal consistency results.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program