Value of statistical life year in extreme poverty: a randomized experiment of measurement methods in rural Burkina Faso

Table 6 Determinants of the selected row

Selected row	(1)	(2)	(3)	(4)
Large range	− 2.94*** (0.31)	− 2.93*** (0.31)	− 2.97*** (0.31)	− 2.77*** (0.41)
WTA	3.268*** (0.34)	3.26*** (0.34)	3.24*** (0.33)	3.13*** (0.46)
Interaction of WTA and large range	− 0.547 (0.44)	− 0.56 (0.44)	− 0.52 (0.44)	− 1.06* (0.60)
Wealth Index		0.14* (0.08)	0.05 (0.08)	0.06 (0.11)
Female			− 1.07*** (0.23)	− 1.17*** (0.32)
Age			− 0.024** (0.01)	− 0.02 (0.02)
Married			0.70*** (0.26)	0.65* (0.36)
Self-reported health (1–5 scale)				− 0.38** (0.16)
Cigarette expenditure				− 0.18*** (0.06)
Constant	9.91*** (0.24)	9.91*** (0.24)	11.28*** (0.71)	12.72*** (1.27)
Observations	2907	2907	2904	1554
R-squared	0.119	0.120	0.135	0.130

OLS regression results are presented. The dependent variable is the row selected by the respondents. Large Range is a dummy variable that equals to 1 if the covered range of the choice list is large. WTA is a dummy variable that equals to 1 if the choice list concerns payments to accept an increased risk. Wealth index is calculated at the household level, but vast majority of households have only one family member interviewed. Age is age in years, female is a binary indicator for female, and married is a binary indicator for being currently married. Self-reported health is measured on a 1–5 Likert Scale with 5 being the highest health status and 1 being the lowest. Cigarette expenditure is measured by the self-reported total household expenditure on cigarette in the past 7 days in PPP adjusted 2017 international $; measured only for half of the participants. Robust standard errors in parentheses: ***p < 0.01, **p < 0.05, *p < 0.1

ISSN: 1478-7954