Latent variable modeling to develop a robust proxy for sensitive behaviors: application to latrine use behavior and its association with sanitation access in a middle-income country

Table 1 Background characteristics of the study sample

Background characteristic	Mean (SD) / Proportion	N
Age (in years)	23 (14)	251
Sex
Female	0.65	162
Male	0.35	89
Ethnicity
Afro-Ecuadorian	0.63	157
Mestizo and Other	0.13	33
Chachi	0.24	61
Educational Attainment
Less than primary school	0.25	63
Completed primary school	0.20	51
Less than secondary school	0.36	90
Completed secondary school	0.17	43
Missing	0.02	4
Payment Type Received for Employment
Solely cash	0.47	118
Cash and kind/ solely kind	0.06	16
Not paid	0.16	41
Not employed in the previous 12 months	0.30	76
Living in Household with Cement Walls
Yes	0.45	115
No	0.51	128
Missing	0.03	8
Living in an Asset-Deprived Household
Yes	0.61	152
No	0.39	98
Missing	< 0.01	1
Household Sanitation Access
Improved Access
Basic sanitation	0.53	132
Shared sanitation	0.33	118
Unimproved Access	0.05	13
No access to a sanitation facility	0.09	22
Missing	0.01	2

Median (SD) is reported for age. For the other covariates, the frequency distribution (in proportions) is reported. The corresponding number of individuals (N) is also shown

ISSN: 1471-2458