Skip to main content
Fig. 5 | BMC Medical Research Methodology

Fig. 5

From: Missing-data analysis: socio- demographic, clinical and lifestyle determinants of low response rate on self- reported psychological and nutrition related multi- item instruments in the context of the ATTICA epidemiological study

Fig. 5

Beta- coefficient and standard error with regard to the effect of the Zung’s Self- Rating Depression Scale on the participants’ likelihood of being hypertensive, after adjusting for age and sex, under three different scenarios concerning the number of missing items in the total scale and 7 different missing data handling methods; the ATTICA epidemiological study. Notes: a Original dataset: N = 1988 participants (90.6%) have missing data in the total score of the Zung’s Self- Rating Depression scale, bN = 945 participants have missing data in 1–7 items of the Zung’s Self- Rating Depression scale and (C) N = 1043 participants have missing data in 8+ items of the Zung’s Self- Rating Depression scale. CCA = Complete case analysis (based on 206 participants), SMI = Score mean imputation, IMI = Item mean imputation, SRI = Stochastic regression imputation, PMI = Person mean imputation, MI = Multiple imputation

Back to article page