Skip to main content
. Author manuscript; available in PMC: 2017 Mar 1.
Published in final edited form as: J Health Econ. 2016 Jan 29;46:70–89. doi: 10.1016/j.jhealeco.2016.01.005

Table A3.

Compositional changes (Panel sample)

Control
Diff/SD CCT
Diff/SD
Pre Post Pre Post
log(Expenditures) 8.493 8.550 0.057 8.234 8.510 0.396
Household size 4.251 4.007 −0.108 5.665 5.574 −0.044
Household members who are adult 0.683 0.700 0.064 0.461 0.473 0.075
Head is male 0.813 0.815 0.005 0.860 0.836 −0.067
Education of head
 None 0.0868 0.101 0.048 0.112 0.124 0.037
 Some primary 0.579 0.577 −0.004 0.620 0.625 0.010
 Some secondary 0.261 0.249 −0.028 0.241 0.226 −0.035
 Some tertiary 0.074 0.073 −0.004 0.027 0.025 −0.013
Native language of head
 Spanish 0.645 0.596 −0.101 0.299 0.330 0.065
 Quechua 0.264 0.297 0.074 0.665 0.610 −0.115
 Other 0.091 0.107 0.054 0.035 0.060 0.118
Marital status of head
 Married or cohabiting 0.746 0.732 −0.032 0.860 0.827 −0.091
 Widowed, divorced, or separated 0.199 0.219 0.049 0.127 0.152 0.072
 Single 0.055 0.049 −0.023 0.013 0.020 0.063
Respondent is male 0.278 0.251 −0.059 0.201 0.145 −0.148
District poverty 0.063 0.054 −0.257 0.108 0.094 −0.438
Community type
 Rural, 0–140 residents 0.148 0.169 0.058 0.202 0.207 0.012
 Rural, 141–400 residents 0.516 0.513 −0.006 0.703 0.662 −0.088
 Urban, 0–400 residents 0.156 0.168 0.033 0.0682 0.0915 0.085
 Urban, 401–4,000 residents 0.180 0.150 −0.080 0.027 0.039 0.067
Region
 Highlands 0.598 0.627 0.059 0.925 0.889 −0.125
 Rainforest 0.402 0.373 −0.059 0.075 0.111 0.125

Number of observations 3524 2508 1583 1431

Note: This table displays the pre-post changes for non-CCT and CCT households in the panel sample. The magnitude of the compositional changes are expressed as the difference in means divided by the pooled standard deviation (Diff/SD).