IMPUTING MISSING DATA IN ORDINAL VARIABLES: A COMPARISON OF METHODS UNDER DIFFERENT PROPORTIONAL ODDS ASSUMPTIONS
| dc.rights.license | CC-BY-NC-ND | |
| dc.contributor.advisor | Luijken, Kim | |
| dc.contributor.author | Coşkun, Valat | |
| dc.date.accessioned | 2024-01-01T01:04:16Z | |
| dc.date.available | 2024-01-01T01:04:16Z | |
| dc.date.issued | 2024 | |
| dc.identifier.uri | https://studenttheses.uu.nl/handle/20.500.12932/45773 | |
| dc.description.sponsorship | Utrecht University | |
| dc.language.iso | EN | |
| dc.subject | The current study aims to investigate the impact of violations to the assumption of proportional odds in imputation models for missing data in ordinal variables. We performed a proof-of-concept Monte Carlo simulation study focusing on the methods used for the imputation of ordinal categorical variables and compared the bias in a substantive analysis between the different imputation methods. | |
| dc.title | IMPUTING MISSING DATA IN ORDINAL VARIABLES: A COMPARISON OF METHODS UNDER DIFFERENT PROPORTIONAL ODDS ASSUMPTIONS | |
| dc.type.content | Master Thesis | |
| dc.rights.accessrights | Open Access | |
| dc.subject.courseuu | Epidemiology | |
| dc.thesis.id | 26122 |