There is an app for that? The impact of community knowledge workers in Uganda

Figures & data

Figure 1. Sampling areas used.

Table 1. Sample design.

	North		East		Total
	Gulu/Oyam (T)	Gulu (C)	Kapchorwa (T)	Sironko (C)	T	C	Total
2005/2006	294	136	36	52	330	188	518
2009/2010	1147	136	842	52	1989	188	2177
2011/2012	133	142	173	158	306	300	606
Total	1574	414	1051	262	2625	676	3301

Table 2. Means of key variables.

	Control	Treatment	Test statistic	p-value
household size	6.13	6.49	−1.36	.174
rural (%)	76.50	83.63	3.63	.057
male headed (%)	77.04	71.64	1.60	.205
age head	42.12	43.17	−0.80	.421
head no formal education (%)	41.81	41.33	−0.09	.985
no of rooms in house	4.24	3.99	−0.13	.198
land size	2.92	4.08	−3.04	.002

Note: A t-test was used to test the equality of continuous variables; chi-squared test was used in case of proportions.

Figure 2. Change in proportion of farmers reporting growing crop attributed to community knowledge workers.

Table 3. The effect of CKW on maize productivity.

	Dependent variable: productivity
	(1)	(2)	(3)	(4)
treat	116.039**	86.002*	144.867***	174.793***
	(45.447)	(50.500)	(51.832)	(56.866)
mid				169.505**
				(66.447)
post		5.111	219.726***	180.884***
		(59.172)	(52.252)	(56.930)
east	385.710***	497.201***	355.947***	477.575***
	(50.741)	(22.622)	(35.466)	(20.635)
treat:mid				−87.820
				−74.888
treat:post		29.878	−28.786	−58.886
		−68.415	−65.532	−72.173
Constant	381.020***	295.451***	182.773***	133.842***
	(48.746)	(49.640)	(44.721)	(47.713)
Observations	389	1711	668	1,990
R-squared	0.143	0.232	0.265	0.248
Adj. R-squared	0.138	0.230	0.261	0.246
Residual std. error	448.172	455.162	395.329	438.248
F statistic	32.183***	128.694***	59.889***	109.270***

Note: Standard errors are in parentheses.

*p < .10, **p < .05, ***p < .01.

Table 4. The effect of CKW on maize on the share of maize harvest sold.

	Dependent variable: share sold
	(1)	(2)	(3)	(4)
treat	−4.322	4.684	10.509**	12.054***
	(3.032)	(3.636)	(5.159)	(4.215)
mid				0.000
				(5.103)
post		20.834***	13.874***	20.523***
		(4.008)	(5.121)	(4.052)
east	1.619	−12.075***	0.561	−11.510***
	(3.237)	(1.091)	(2.707)	(1.070)
treat:mid				−7.552
				(5.585)
treat:post		−9.517**	−14.871**	−16.865***
		(4.345)	(5.941)	(4.852)
Constant	38.029***	26.742***	24.893***	26.660***
	(3.155)	(3.571)	(4.427)	(3.612)
Observations	372	1,827	526	1,981
R-squared	0.006	0.097	0.017	0.092
Adj. R-squared	0.001	0.095	0.009	0.089
Residual std. error	29.165	22.841	28.234	23.107
F statistic	1.171	48.969***	2.199*	33.265***

Note: Standard errors are in parentheses.

*p < .10, **p < .05, ***p < .01.

Table 5. The effect of CKW on the price received for maize.

	Dependent variable: price received
	(1)	(2)	(3)	(4)
treat	91.831***	−9.505	18.435	1.344
	(24.982)	(11.649)	(16.652)	(17.398)
mid				207.898***
				(17.106)
post		188.757***	359.635***	395.171***
		(18.290)	(19.316)	(19.667)
east	100.538***	−37.272***	50.511***	−33.802***
	(26.879)	(8.308)	(14.044)	(7.267)
treat:mid				−11.203
				(20.536)
treat:post		85.510***	67.651***	75.060***
		(23.383)	(24.209)	(25.554)
Constant	513.112***	426.892***	190.699***	217.896***
	(26.761)	(10.893)	(14.013)	(14.206)
Observations	346	2,181	793	2,628
R-squared	0.067	0.173	0.618	0.306
Adj. R-squared	0.061	0.171	0.616	0.305
Residual std. error	230.471	187.905	165.096	174.436
F statistic	12.240***	113.788***	319.280***	192.893***

Note: Standard errors are in parentheses.

*p < .10, **p < .05, ***p < .01.

Table 6. DD regressions with additional controls.

	Dependent variable
	prodacre	Sellshare	price	sellshare	Price
	(1)	(2)	(3)	(4)	(5)	(6)
treat	141.284***	11.492**	16.582	11.926**	15.718	−4.071
	(51.881)	(5.171)	(16.717)	(5.319)	(17.456)	(33.775)
post	183.807***	12.178**	344.579***	12.291**	344.858***	289.161***
	(53.738)	(5.242)	(20.450)	(5.375)	(21.244)	(36.720)
east	346.243***	0.626	39.560***	0.916	39.432***	33.921*
	(36.770)	(2.830)	(14.446)	(2.877)	(14.973)	(20.419)
hhsize	6.697	0.1	5.404**	0.058	5.356**	8.352***
	(5.720)	(0.454)	(2.124)	(0.459)	(2.187)	(3.225)
agehead	−2.147*	−0.150	0.166	−0.147	0.182	0.426
	(1.130)	(0.094)	(0.433)	(0.095)	(0.447)	(0.672)
femhead	3.780	−12.274***	−6.475	−11.670***	−8.201	−10.443
	(42.873)	(3.564)	(15.566)	(3.614)	(16.280)	(26.759)
prim	−64.176*	−1.395	−33.056**	−1.496	−32.982**	−15.845
	(35.525)	(2.822)	(14.273)	(2.835)	(14.622)	(19.830)
hasshoes	59.68	3.124	25.884*	3.015	25.325*	31.634
	(38.119)	(3.118)	(14.717)	(3.142)	(15.169)	(22.400)
hasclothes	159.457***	5.383	−9.783	5.176	−8.855	5.514
	(52.657)	(4.156)	(18.215)	(4.236)	(18.924)	(31.334)
land				0.068	−0.242	−0.329
				(0.093)	(0.551)	(0.620)
sellshare						0.759**
						(0.338)
treat:post	−47.65	−19.079***	68.272***	−19.520***	69.407***	93.667**
	(66.242)	(5.972)	(24.537)	(6.112)	(25.257)	(40.776)
Constant	136.365	29.398***	181.917***	29.332***	181.631***	134.512**
	(86.980)	(7.173)	(30.544)	(7.238)	(31.418)	(52.639)
Observations	658	515	790	510	765	470
R-squared	0.295	0.065	0.63	0.062	0.626	0.526
Adj. R-squared	0.284	0.047	0.625	0.041	0.620	0.513
F statistic	27.082***	3.523***	132.483***	2.969***	114.464***	42.239***

Note: Standard errors are in parentheses.

*p < .10, **p < .05, ***p < .01.

Table 7. Treatment intensity.

	Dependent variable
	Prodacre	sellshare	Price
	(1)	(2)	(3)
freq	−6.463	−0.505	1.025
	(7.227)	(0.369)	(3.408)
post	−56.621	−17.167***	285.078***
	(52.414)	(2.861)	(30.298)
Observations	338	431	369
R-squared	0.027	0.233	0.463
Adj. R-squared	0.009	0.096	0.163
F statistic	1.642	27.024***	55.977***

Note: Standard errors are in parentheses.

*p < .10, **p < .05, ***p < .01.