Family research is influenced by the systemic nature of the family itself, so that missing data could be found at different levels (i.e., item, respondent, dyad). The aim of the study is to give family researchers a step-by-step description of the procedures used to analyze the amount of missingness and the mechanisms causing the missingness at the different levels featuring family data. Examples from two family datasets were provided and both individual and relational auxiliary variables related to the missingness were examined. The largest amount of missingness was found at the respondent level and, specifically, for the father’s role. Regarding the missingness mechanism, missing completely at random (MCAR) was found for both dyad and respondent level missingness, whereas missing at random (MAR) could be hypothesized for missing data at the item level. The complexities inherent in family research levels and in family research planning, as well as future steps were discussed.

Tagliabue, S., Donato, S., Missing data in family research: Examining different levels of missingness, <<TPM. TESTING, PSYCHOMETRICS, METHODOLOGY IN APPLIED PSYCHOLOGY>>, 2015; (22): 199-217. [doi:10.4473/TPM22.2.3] [http://hdl.handle.net/10807/71193]

Missing data in family research: Examining different levels of missingness

Tagliabue, Semira;Donato, Silvia
2015

Abstract

Family research is influenced by the systemic nature of the family itself, so that missing data could be found at different levels (i.e., item, respondent, dyad). The aim of the study is to give family researchers a step-by-step description of the procedures used to analyze the amount of missingness and the mechanisms causing the missingness at the different levels featuring family data. Examples from two family datasets were provided and both individual and relational auxiliary variables related to the missingness were examined. The largest amount of missingness was found at the respondent level and, specifically, for the father’s role. Regarding the missingness mechanism, missing completely at random (MCAR) was found for both dyad and respondent level missingness, whereas missing at random (MAR) could be hypothesized for missing data at the item level. The complexities inherent in family research levels and in family research planning, as well as future steps were discussed.
Inglese
Tagliabue, S., Donato, S., Missing data in family research: Examining different levels of missingness, <<TPM. TESTING, PSYCHOMETRICS, METHODOLOGY IN APPLIED PSYCHOLOGY>>, 2015; (22): 199-217. [doi:10.4473/TPM22.2.3] [http://hdl.handle.net/10807/71193]
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10807/71193
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 8
  • ???jsp.display-item.citation.isi??? 8
social impact