. 2019 Feb 7;14(2):e0211656. doi: 10.1371/journal.pone.0211656

Table 6. Decomposition of the extent to which five variables (Animal rights, Anthropocentrism, Lay utilitarian, Non-concern, and Trust in current animal welfare legislation) mediate the effect of the Animal protection orientation on Animal welfare-friendly meat consumption (N = 976)^A^,^B.

		Confounding (%)
Animal protection (variable of interest that is decomposed)
Reduced (unadjusted model)	-0.0264^***
Full model (adjusted model)	-0.004^n.s.
Difference (summary of confounding)	-0.022^***	83.5
Components of difference
Animal rights	-0.0073	27.92
Anthropocentrism	0.0016	-5.98
Lay utilitarian	-0.0018	6.73
Non-concern about animal welfare	-0.0109	41.32
Trust in current animal welfare legislation	-0.0035	13.55

* p<0.05

** p<0.01

*** p<0.001; n.s: not significant at the 0.05 level

^A Decomposition based on the khb method outlined by Breen, Karlson, & Holm, A. (2013) and implemented in Stata’s khb package.

^B Control variables included (using the concomitant command in the khb package) were: gender, age, household type (single adult, two adults (no children), and household with children), educational level, and living area.

Table 6. Decomposition of the extent to which five variables (Animal rights, Anthropocentrism, Lay utilitarian, Non-concern, and Trust in current animal welfare legislation) mediate the effect of the Animal protection orientation on Animal welfare-friendly meat consumption (N = 976)A,B.

Table 6. Decomposition of the extent to which five variables (Animal rights, Anthropocentrism, Lay utilitarian, Non-concern, and Trust in current animal welfare legislation) mediate the effect of the Animal protection orientation on Animal welfare-friendly meat consumption (N = 976)^A^,^B.