Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.

Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Integrating Reported Reasons for Missingness in Pretest-Posttest Research Designs

Paul J Dizona, University of Nebraska - Lincoln

Abstract

Missing data is a common challenge to any researcher in almost any field of research. In particular, human participants in research do not always respond or return for assessments leaving the researcher to rely on missing data methods. The most common methods (i.e., Multiple Imputation and Full Information Maximum Likelihood) assume that the missingness falls under the MAR mechanism (Enders, 2010). The National Research Council (2010) made recommendations for addressing missing data including recording the reason participants dropout of a study whenever possible. Beyond making a case for MAR, it is unclear how a researcher would go about integrating these reasons for dropout into their statistical analysis.The purpose of this study is to develop and evaluate a method to incorporate these reasons for dropout. In order to accomplish this, two studies were conducted. In Study 1, a clinical trial that recorded the reasons for dropout was evaluated using a sensitivity analysis to compare estimates of a treatment effect. In Study 2, a simulation study was performed that included the reason for dropout to compare the proposed method with the standard approaches to missingness.

Subject Area

Educational psychology|Psychology|Information science|Statistics

Recommended Citation

Dizona, Paul J, "Integrating Reported Reasons for Missingness in Pretest-Posttest Research Designs" (2022). ETD collection for University of Nebraska-Lincoln. AAI29215837.
https://digitalcommons.unl.edu/dissertations/AAI29215837

Share

COinS