NSDUH 2020 Editing And Imputation Report

This report details the editing and statistical imputation procedures that were applied to respondent data for the 2020 National Survey on Drug Use and Health (NSDUH). Logical editing resolves inconsistencies or ambiguous data based just on a respondent’s answers to other questions in the survey. Statistical imputation uses mathematical techniques to make educated guesses for missing values in the data. Statistical imputation also usually involves some randomness to preserve the natural variability in the data.

This report begins with an introduction in Chapter 1, including a discussion of changes from the 2019 to 2020 survey. Chapter 2 describes the procedures and general principles for editing the NSDUH Data. Chapter 3 describes the general imputation procedures used in NSDUH. The remaining chapters describe the editing and imputation for different types of NSDUH variables in 2020: front-end and back-end demographics in chapters 4 and 5 respectively; substance use in chapter 6; special drugs and substance use disorder in chapter 7; additional substance use in chapter 8, including treatment and emerging issues; substance use risk and protective factors in chapter 9; physical and mental health in chapters 10; roster variables in chapter 11; income in chapter 12; health insurance in chapter 13; and pair variables in chapter 14.

The document is available as a PDF.