Suchen Sie hier über ein Suchformular im Vorlesungsverzeichnis der Leuphana.


Analyse fehlender Werte in R (Seminar)

Dozent/in: Jan-Bennet Voltmer

Einzeltermin | Fr, 09.07.2021, 14:15 - Fr, 09.07.2021, 19:45 | Online-Veranstaltung
Einzeltermin | Sa, 10.07.2021, 10:15 - Sa, 10.07.2021, 19:45 | Online-Veranstaltung
Einzeltermin | So, 11.07.2021, 10:15 - So, 11.07.2021, 19:45 | Online-Veranstaltung

Inhalt: "Obviously the best way to treat missing data is not to have them." (Orchard and Woodbury, 1972, p. 697) "Though there is a lot of truth in this statement, Orchard and Woodbury realized the impossibility of attaining this ideal in practice. The prevailing scientific practice is to downplay the missing data. [...] Missing data are there, whether we like it or not. In the social sciences, it is nearly inevitable that some respondents will refuse to participate or to answer certain questions. In medical studies, attrition of patients is very common. The theory, methodology and software for handling incomplete data problems have been vastly expanded and refined over the last decades." (van Buuren, 2018) In this seminar, participants will get in touch with different strategies to deal with missing data: Deleting the data, deterministic imputation, probabilistic imputation, and how to diagnose the "success" of the chosen strategy. Participants will learn how to create multiple imputed datasets using chained equations with the mice-package in R, which might be the most advanced method to look into the black box called "missing data". Within the seminar participants will learn how to do this in R from scratch, so data reading, manipulation, and output in R will of course also be part of the seminar.