The Impact of Better Access to Improved Water Sources on Health, Schooling, and Water Collection of Girls and Boys in Rural Zambia

Figures & data

Figure 1. Timeline of the project. Source: JICA (2014).

Figure 2. Procedure of site selection and sampling. Source: JICA (2014).

Table 1. The main outcome variables (baseline)

	Project sites	Control sites	Diff.	Project sites	Control sites	Diff.
	(A)	(B)	(C)	(D)	(E)	(F)
Water collection	Panel households
No. of observations	424	205
Distance to the nearest water source (km)	0.454	0.548	−0.095*
Water volume carried yesterday (liters)	64.1	63.6	0.568
Number of trips over the last 1 week	23.0	22.5	0.505
Diarrhea incidence (Individual level)	Pre-school age children (0–6)			School-age children (7–18)
No. of observations	505	269		751	325
Diarrhea incidence (=1)	0.055	0.048	0.007	0.013	0.015	−0.002
School attendance (Individual level)	School-age girls (7–18)			School-age boys (7–18)
No. of observations	382	158		369	167
School attendance (=1)	0.791	0.747	0.044	0.843	0.772	0.070*
Time allocation (Individual level)	School-age girls (7–18)			School-age boys (7–18)
No. of observations	364	155		359	166
Fetching water	0.89	1.15	−0.26**	0.68	0.75	−0.07
Collecting firewood	0.08	0.11	−0.03	0.09	0.08	0.01
Grazing/taking care of animals	0.00	0.00	0.00	0.01	0.02	0.00
Washing/cleaning	0.53	0.65	−0.12	0.54	0.51	0.03
Cooking	0.16	0.27	−0.11**	0.07	0.12	−0.06**
Taking care of children/elders	0.09	0.13	−0.04	0.03	0.07	−0.03
Shopping	0.03	0.01	0.02	0.04	0.02	0.02
Collecting grass	0.04	0.00	0.04	0.05	0.02	0.02
Other household chores	1.08	0.79	0.29*	0.86	0.74	0.12
Income generating activities	0.52	0.53	−0.01	0.52	0.47	0.06
School and homework	4.06	3.98	0.08	4.32	4.43	−0.11
Travel/walking/commuting	0.60	0.25	0.34***	0.44	0.27	0.17*
Social/community activities	0.16	0.23	−0.07	0.08	0.26	−0.18***
Religious activities	0.07	0.02	0.05	0.05	0.05	0.01
Eating/drinking	1.31	1.34	−0.03	1.24	1.36	−0.11*
Leisure/relaxing/resting	2.50	2.41	0.10	2.69	2.67	0.02
Sleeping	1.62	1.49	0.13	1.49	1.66	−0.16*
Other	1.27	1.65	−0.37	1.79	1.52	0.27

Notes: The sample is confined to children in the households that were surveyed in both rounds.

Incidence of diarrhea symptoms over the last two weeks reported by the respondents of the sampled households. Time allocation is measured in hours.

t-test or Fisher's exact test results are shown: * Significant at 10 per cent, ** Significant at 5 per cent, *** Significant at 1 per cent.

Table 2. Impact on fetching water (household level)

Dependent variable: Water volume (liters) and number of trips	Distance to the nearest water source			Water volume carried yesterday			Number of trips over the last 1 week
	Project	Project and distance		Project	Project and distance		Project	Project and distance
		(km)	(min.)		(km)	(min.)		(km)	(min.)
	(A)	(B)	(C)	(D)	(E)	(F)	(G)	(H)	(I)
Project and year dummy variables
Project site * Year 2013 (=1)	−0.147**	−0.240***	−0.191***	3.791	6.119	6.975	2.357	3.017	3.835**
	(0.064)	(0.069)	(0.069)	(5.941)	(5.837)	(6.001)	(1.699)	(1.866)	(1.820)
Project site * Year 2013		0.398***	0.010***		−9.934	−0.703***		−2.816	−0.326**
* Distance (km/min.)		(0.063)	(0.003)		(6.299)	(0.263)		(2.926)	(0.130)
Year 2013 (=1)	−0.149***	−0.154***	−0.151***	−15.221***	−15.092***	−15.064***	−4.006***	−3.970***	−3.934***
	(0.045)	(0.044)	(0.045)	(4.956)	(4.960)	(4.972)	(1.396)	(1.393)	(1.398)
Individual characteristics	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Household characteristics	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Village characteristics	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Village fixed effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
R sq.	0.380	0.393	0.384	0.236	0.238	0.239	0.152	0.152	0.157
No. of observations	1258	1258	1258	1258	1258	1258	1258	1258	1258

Notes: Village-level cluster-adjusted standard errors are in parentheses: * Significant at 10 per cent, ** Significant at 5 per cent, *** Significant at 1 per cent.

Distance is measured in kilometers in Columns (B), (E), and (H), and measured in minutes in Columns (C), (F), and (I).

Table 3. Impact on incidence of diarrhea symptoms

Dependent variable:	Diarrhea incidence over the last two weeks (=1)
	Pre-school age children			School-age children
	Project	Project and distance		Project	Project and distance
		(km)	(min.)		(km)	(min.)
	(A)	(B)	(C)	(D)	(E)	(F)
Project and year dummy variables
Project site * Year 2013 (=1)	−0.049*	−0.067**	−0.064**	−0.001	0.002	0.001
	(0.028)	(0.029)	(0.028)	(0.013)	(0.014)	(0.013)
Project site * Year 2013		0.073	0.003*		−0.011	−0.001
* Distance (km/min.)		(0.051)	(0.002)		(0.010)	(0.001)
Year 2013 (=1)	0.024	0.022	0.023	−0.009	−0.008	−0.008
	(0.025)	(0.025)	(0.025)	(0.010)	(0.010)	(0.010)
Individual characteristics	Yes	Yes	Yes	Yes	Yes	Yes
Household characteristics	Yes	Yes	Yes	Yes	Yes	Yes
Village characteristics	Yes	Yes	Yes	Yes	Yes	Yes
Village fixed effects	Yes	Yes	Yes	Yes	Yes	Yes
R sq.	0.108	0.110	0.110	0.052	0.052	0.052
No. of observations	1457	1457	1457	2103	2103	2103

Note: Village-level cluster-adjusted standard errors are in parentheses: *Significant at 10 per cent, **Significant at 5 per cent, ***Significant at 1 per cent.

Distance is measured in kilometers in Columns (B) and (E), and measured in minutes in Columns (C) and (F).

Table 4. Impact on school attendance and time spent on schooling and homework

Dependent variable:	School attendance (=1)						Time spent on schooling and homework per day (hours)
	School-age girls			School-age boys			School-age girls			School-age boys
	Project	Project and distance		Project	Project and distance		Project	Project and distance		Project	Project and distance
		(km)	(min.)		(km)	(min.)		(km)	(min.)		(km)	(min.)
	(A)	(B)	(C)	(D)	(E)	(F)	(G)	(H)	(I)	(J)	(K)	(L)
Project and year dummy variables
Project site * Year 2013 (=1)	−0.034	−0.058	−0.039	−0.009	0.007	0.032	−0.805	−1.188*	−0.909	0.169	0.076	0.214
	(0.061)	(0.071)	(0.069)	(0.066)	(0.066)	(0.070)	(0.616)	(0.642)	(0.648)	(0.551)	(0.632)	(0.600)
Project site * Year 2013		0.090	0.001		−0.071	−0.010		1.420*	0.019		0.416	−0.011
* Distance (km/min.)		(0.123)	(0.005)		(0.114)	(0.008)		(0.790)	(0.039)		(1.113)	(0.061)
Year 2013 (=1)	−0.053	−0.053	−0.053	−0.052	−0.052	−0.050	0.205	0.200	0.202	−0.589	−0.593	−0.586
	(0.051)	(0.051)	(0.051)	(0.058)	(0.058)	(0.058)	(0.565)	(0.563)	(0.565)	(0.447)	(0.448)	(0.446)
Individual characteristics	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Household characteristics	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Village characteristics	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Village fixed effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
R sq.	0.285	0.286	0.285	0.275	0.275	0.277	0.322	0.326	0.323	0.338	0.338	0.338
No. of observations	1049	1049	1049	1054	1054	1054	1000	1000	1000	1014	1014	1014

Note: Village-level cluster-adjusted standard errors are in parentheses: * Significant at 10 per cent, ** Significant at 5 per cent, *** Significant at 1 per cent.

Distance is measured in kilometers in Columns (B), (E), (H), and (K), and measured in minutes in Columns (C), (F), (I), and (L).

Table 5. Impact on time spent on fetching water and washing/cleaning/cooking

Dependent variable:	Time spent on fetching water per day (hours)						Time spent on water-related household chores^a per day (hours)
	School-age girls			School-age boys			School-age girls			School-age boys
	Project	Project and distance		Project	Project and distance		Project	Project and distance		Project	Project and distance
		(km)	(min.)		(km)	(min.)		(km)	(min.)		(km)	(min.)
	(A)	(B)	(C)	(D)	(E)	(F)	(G)	(H)	(I)	(J)	(K)	(L)
Project and year dummy variables
Project site * Year 2013 (=1)	0.245	0.272	0.232	0.179	0.195	0.203	0.518**	0.609**	0.508*	0.197	0.223	0.174
	(0.197)	(0.210)	(0.207)	(0.127)	(0.140)	(0.136)	(0.257)	(0.269)	(0.266)	(0.162)	(0.174)	(0.171)
Project site * Year 2013		−0.099	0.002		−0.074	−0.006		−0.339	0.002		−0.113	0.006
* Distance (km/min.)		(0.234)	(0.013)		(0.204)	(0.012)		(0.320)	(0.020)		(0.295)	(0.015)
Year 2013 (=1)	−0.690***	−0.690***	−0.690***	−0.455***	−0.455***	−0.454***	−1.061***	−1.060***	−1.061***	−0.712***	−0.711***	−0.713***
	(0.192)	(0.192)	(0.192)	(0.109)	(0.109)	(0.109)	(0.243)	(0.242)	(0.243)	(0.130)	(0.131)	(0.131)
Individual characteristics	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Household characteristics	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Village characteristics	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Village fixed effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
R sq.	0.244	0.244	0.244	0.235	0.235	0.235	0.232	0.233	0.232	0.243	0.243	0.243
No. of observations	1000	1000	1000	1014	1014	1014	1000	1000	1000	1014	1014	1014

Note: Village-level cluster-adjusted standard errors are in parentheses: * Significant at 10 per cent, ** Significant at 5 per cent, *** Significant at 1 per cent.

Distance is measured in kilometers in Columns (B), (E), (H), and (K), and measured in minutes in Columns (C), (F), (I), and (L).

^aWater-related household chores comprise fetching water and washing/cleaning/cooking.

Table 6. Impact on time spent by female adults

Dependent variable:	Time spent on fetching water per day (hours)			Time spent on water-related household chores^a per day (hours)
	Adult women			Adult women
	Project	Project and distance		Project	Project and distance
		(km)	(min.)		(km)	(min.)
	(A)	(B)	(C)	(D)	(E)	(F)
Project and year dummy variables
Project site * Year 2013 (=1)	−0.166	−0.192	−0.225	−0.375*	−0.512**	−0.565**
	(0.158)	(0.168)	(0.167)	(0.224)	(0.241)	(0.235)
Project site * Year 2013		0.107	0.012		0.562*	0.040***
* Distance (km/min.)		(0.170)	(0.009)		(0.320)	(0.013)
Year 2013 (=1)	−0.735***	−0.736***	−0.737***	−0.876***	−0.883***	−0.883***
	(0.133)	(0.133)	(0.133)	(0.180)	(0.180)	(0.179)
Individual characteristics	Yes	Yes	Yes	Yes	Yes	Yes
Household characteristics	Yes	Yes	Yes	Yes	Yes	Yes
Village characteristics	Yes	Yes	Yes	Yes	Yes	Yes
Village fixed effects	Yes	Yes	Yes	Yes	Yes	Yes
R sq.	0.302	0.303	0.303	0.287	0.288	0.290
No. of observations	1419	1419	1419	1419	1419	1419

Note: Village-level cluster-adjusted standard errors are in parentheses: * Significant at 10 per cent, ** Significant at 5 per cent, *** Significant at 1 per cent.

Distance is measured in kilometers in Columns (B) and (E), and measured in minutes in Columns (C) and (F).

^aWater-related household chores comprise fetching water and washing/cleaning/cooking.

Table

	Project sites	Control sites	Difference
	(A)	(B)	(C)
Individual characteristics	n = 2211	n = 1059
Female (%)	51.7	50.0	1.8
Age	21.0	20.7	0.2
Working age women aged 19–59	n = 453	n = 217
Education (years of schooling) completed	5.501	4.779	0.722***
Crop farmers (%)	79.2	77.4	1.8
Fishery workers (%)	0.0	0.5	−0.5
Traders/retail shopkeepers (%)	4.9	6.9	−2.1
Working age men aged 19–59	n = 417	n = 212
Education (years of schooling) completed	6.940	6.717	0.223
Crop farmers (%)	72.4	71.7	0.7
Fishery workers (%)	4.8	3.3	1.5
Traders/retail shopkeepers (%)	3.1	3.8	−0.7
Household characteristics	n = 424	n = 205
Female headed household (%)	20.0	20.0	0.05
Age of the head	43.5	42.3	1.20
Household size	5.215	5.166	0.049
Ratio of dependents to household size	0.450	0.452	−0.002
Monthly consumption per capita (thousand ZMK)	163.2	143.3	19.9
Agricultural land value (million ZMK)	3.399	4.729	−1.330
Value of productive assets (million ZMK)	0.533	0.476	0.058
Value of durable assets (million ZMK)	0.870	0.891	−0.021
Village characteristics	n = 63	n = 31
Population (households)	98.2	97.6	0.61
Population (individuals)	480.9	439.8	41.0
Land area (ha)	141.4	98.4	42.9
Flat villages (%)	31.7	38.7	−7.0
Slightly sloping villages (%)	38.1	35.5	2.6
Moderately sloping villages (%)	28.6	22.6	6.0
Steeply sloping/hilly villages (%)	1.6	3.2	−1.6
Average assets per household (million ZMK)	5.940	7.071	−1.131
Distance to district center (km)	45.4	36.4	8.9
Distance to town center (km)	26.5	15.8	10.7
Distance to market (km)	12.3	13.2	−0.9
Distance to government primary school (km)	2.0	2.7	−0.7
Distance to government secondary school (km)	29.2	30.8	−1.6
Distance to rural health center (km)	7.3	9.5	−2.2

Note: t-test or Fisher's exact test results are shown: * Significant at 10 per cent, ** Significant at 5 per cent, *** Significant at 1 per cent.[AQ]

Monthly consumption per capita is adjusted by using adult equivalence scales and measured in real terms at the price level of 2012. Assets per household include the value of residence, residential and agricultural land, and productive/durable assets. 1USD was approximately 5200ZMK as of June 2012.

Table B1. Number of households in each district

	Before (2012)			After (2013)			Attrition rate (%)
	Project sites	Control sites	All	Project sites	Control sites	All
	(A)	(B)	(C)	(D)	(E)	(F)	(G)
District
Milenge	168	40	208	150	35	185	11.1
Mwense	192	96	288	151	85	236	18.1
Nchelenge	144	112	256	128	86	214	16.4
Total	504	248	752	429	206	635	15.6

Table B2. Summary statistics

2012	Mean	s.d.	Min.	Max.
	(A)	(B)	(C)	(D)
Individual characteristics	n = 1850
Female (=1)	0.506	(0.500)	0	1
Age	8.3	(5.25)	0	18
Household characteristics	n = 629
Female headed household (=1)	0.200	(0.401)	0	1
Age of the head	43.1	(13.65)	19	84
Highest education (years) among females	5.124	(2.928)	0	12
Highest education (years) among males	6.870	(2.818)	0	12
Household size	5.199	(2.367)	1	15
Ratio of dependents to household size	0.451	(0.246)	0	1
Monthly consumption per capita (thousand ZMK)	156.7	(193.9)	16.01	2780
Value of productive/durable assets (million ZMK)	1.391	(1.923)	0.005	29.40
Surveyed in June (=1)	0.647	(0.478)	0	1
Surveyed in July (=1)	0.353	(0.478)	0	1
Village characteristics	n = 94
Population	467.3	(524.9)	48	3360
Average assets per household (million ZMK)	6.318	(5.213)	1.448	29.30
	Mean	s.d.	Min.	Max.
2013	(A)	(B)	(C)	(D)
Individual characteristics	n = 1710
Female (=1)	0.504	(0.500)	0	1
Age	8.7	(5.11)	0	18
Household characteristics	n = 629
Project site (=1)	0.674	(0.469)	0	1
Project site * Borehole user (=1)	0.531	(0.499)	0	1
Distance to the new borehole (km)^a	0.203	(0.211)	0	1
Walking time to the new borehole (min)^a	3.737	(3.933)	0	30
Female headed household (=1)	0.196	(0.397)	0	1
Age of the head	44.0	(13.68)	18	85
Highest education (years) among females	4.990	(2.949)	0	12
Highest education (years) among males	6.776	(2.889)	0	12
Household size	5.402	(2.351)	1	15
Ratio of dependents to household size	0.451	(0.235)	0	1
Monthly consumption per capita (thousand ZMK)	174.2	(310.6)	4.020	3825
Value of durable/productive assets (million ZMK)	1.656	(2.825)	0.005	44.30
Surveyed in June (=1)	0.822	(0.383)	0	1
Surveyed in August (=1)	0.178	(0.383)	0	1
Village characteristics	n = 94
Population	482.5	(488.0)	80	3360
Average assets per household (million ZMK)	7.951	(6.766)	1.676	55.06

Note: The sample is confined to children in the households that were surveyed in both rounds.

Monthly consumption per capita is adjusted by using adult equivalence scales and measured in real terms at the price level of 2012. Assets per household include the value of residence, residential and agricultural land, and productive/durable assets. 1USD was worth approximately 5200ZMK as of June 2012.

^aThe figures are calculated based on the information about only borehole users (n = 334).

Japan International Cooperation Agency. (2014). Basic research for impact evaluation trial on grant aid: A case from rural groundwater development project in Zambia. Tokyo: JICA.

 Google Scholar