Table 2.

Settlement-level regression results predicting two corruption risk indicators. For both dependent variables, the first columns (1) and (3) correspond to the base model, predicting corruption risk using only control variables, and the second columns (2) and (4) show results, when the social network features are included. Note that all features are standardized with mean 0 and standard deviation 1.

dependent variable:	% closed or single bid		average CRI
	(1)	(2)	(3)	(4)
fragmentation		0.263***		0.207**
(bonding social capital)		(0.097)		(0.092)
diversity		−0.553***		−0.551***
(bridging social capital)		(0.176)		(0.168)
income/capita	−0.262	−0.277*	−0.075	−0.096
	(0.169)	(0.162)	(0.161)	(0.155)
N contracts (log)	−0.313*	−0.314*	−0.685***	−0.697***
	(0.171)	(0.165)	(0.162)	(0.158)
population (log)	−0.180	0.020	0.118	0.335**
	(0.143)	(0.166)	(0.136)	(0.159)
rate iWiW use	0.045	0.037	0.122	0.107
	(0.137)	(0.132)	(0.130)	(0.126)
mayor victory margin	0.278***	0.255***	0.303***	0.281***
	(0.089)	(0.086)	(0.085)	(0.082)
% high school grads	0.166	0.374*	−0.176	0.040
	(0.190)	(0.199)	(0.181)	(0.190)
distance to Budapest	−0.021	−0.198*	0.061	−0.112
	(0.104)	(0.112)	(0.099)	(0.107)
share of pop. inactive	−0.797***	−0.805***	−0.716***	−0.754***
	(0.229)	(0.229)	(0.218)	(0.219)
unemployment rate	0.239**	0.262**	0.299***	0.320***
	(0.118)	(0.113)	(0.112)	(0.108)
% population 60+	0.501***	0.491***	0.500***	0.503***
	(0.163)	(0.158)	(0.155)	(0.151)
has university	0.351	0.294	0.431**	0.352*
	(0.220)	(0.221)	(0.210)	(0.211)
constant	1.245*	1.206*	2.779***	2.790***
	(0.725)	(0.702)	(0.689)	(0.671)
observations	169	169	169	169
adjusted R2	0.163	0.230	0.183	0.243
F statistic	3.967***	4.859***	4.419***	5.142***

Significance thresholds: *p < 0.1; **p < 0.05; ***p < 0.01.