Background: Non-response bias is a major problem for the validity of survey answers. Analyses of reasons for non-response in surveys with nursing home residents (NHR) are scarce., Aim and Method: The aim of this study was to develop reliable and valid categories which represent reasons for non-response in nursing homes using the "Prozessmodell induktiver Kategorienbildung" following Mayring (2010). Furthermore, characteristics of unit non-responder and responder were compared., Results: Forming of categories of reasons for non-response was conducted with 522 NHR, longitudinally. Four major categories were identified (general refusal, health reasons, accessibility, excessive demand) and further 17 subcategories. Unit-non-responder and responder did not differ in age and sex; however follow-up non-responder showed differences in marital status, training qualification, and cognitive status., Conclusions: The presented scheme of categories can be used to assess reasons for non-response in nursing homes. Standardised assessment of reasons for non-response may contribute to higher response rates in this setting.