Missing Data: A Unified Taxonomy Guided by Conditional Independence

Marco Doretti; Sara Geneletti; Elena Stanghellini

doi:10.1111/insr.12242

Missing Data: A Unified Taxonomy Guided by Conditional Independence

International Statistical Review ◽

10.1111/insr.12242 ◽

2017 ◽

Vol 86 (2) ◽

pp. 189-204 ◽

Cited By ~ 1

Author(s):

Marco Doretti ◽

Sara Geneletti ◽

Elena Stanghellini

Keyword(s):

Missing Data ◽

Conditional Independence

Download Full-text

Lattice conditional independence models for contingency tables with non-monotone missing data patterns

Journal of Statistical Planning and Inference ◽

10.1016/s0378-3758(99)00003-8 ◽

1999 ◽

Vol 79 (2) ◽

pp. 259-287 ◽

Cited By ~ 2

Author(s):

Michael D. Perlman ◽

Lang Wu

Keyword(s):

Missing Data ◽

Conditional Independence ◽

Contingency Tables

Download Full-text

Item Imputation Without Specifying Scale Structure

Methodology ◽

10.1027/1614-2241/a000004 ◽

2010 ◽

Vol 6 (1) ◽

pp. 31-36 ◽

Cited By ~ 21

Author(s):

Stef van Buuren

Keyword(s):

Missing Data ◽

Conditional Independence ◽

R Package ◽

Scale Structure ◽

Promising Alternative ◽

Fully Conditional Specification ◽

Conditional Specification ◽

Number Of Factors

Imputation of incomplete questionnaire items should preserve the structure among items and the correlations between scales. This paper explores the use of fully conditional specification (FCS) to impute missing data in questionnaire items. FCS is particularly attractive for items because it does not require (1) a specification of the number of factors or classes, (2) a specification of which item belongs to which scale, and (3) assumptions about conditional independence among items. Imputation models can be specified using standard features of the R package MICE 1.16. A limited simulation shows that MICE outperforms two-way imputation with respect to Cronbach’s α and the correlations between scales. We conclude that FCS is a promising alternative for imputing incomplete questionnaire items.

Download Full-text