Fig. 1

Estimates of the probability of dying between ages 15 and 50 (\(_{35}q_{15}\)) from SSH data using different methods in Cambodia, Senegal and Zimbabwe. Note: one set of direct estimates is calculated for periods of 0–6, 7–13, and 14–20 completed years before the survey (solid lines). Another set is calculated for the period 0–4 completed years before the survey (squares) to compare with indirect estimates derived from recent changes in proportions (triangles). The y-axis scale differs between panels of the figure