Corruption red flags in public procurement: new evidence from Italian calls for tenders

Decarolis, Francesco; Giorgiantonio, Cristina

doi:10.1140/epjds/s13688-022-00325-x

EPJ Data Science

Table 7 Estimates for the small and large models—verification data

From: Corruption red flags in public procurement: new evidence from Italian calls for tenders

	OLS		LASSO		Ridge
	Corruption risk	Corruption risk	Corruption risk	Corruption risk	Corruption risk	Corruption risk
Design-Build	−0.008	−0.009	0.000	0.000	−0.001	−0.000
Design-Build	[0.008]	[0.008]
Urgency	−0.005	−0.004	−0.003	−0.001	−0.003	−0.002
Urgency	[0.005]	[0.005]
Negotiated	−0.003	−0.000	0.000	0.000	0.000	0.000
Negotiated	[0.013]	[0.013]
Negotiated-No Tender	0.007	0.008	0.004	0.000	0.004	0.002
Negotiated-No Tender	[0.009]	[0.013]
Price Only—w. ABA	0.012	0.010	0.000	0.000	−0.004	−0.005
Price Only—w. ABA	[0.009]	[0.009]
Scoring Rule (MEAT)	0.011	−0.005	0.008	0.000	0.008	0.003
Scoring Rule (MEAT)	[0.009]	[0.013]
Open Tender Days	0.009	0.008	0.011	0.010	0.012	0.009
Open Tender Days	[0.010]	[0.010]
Open Tender Day V.	0.021^∗∗	0.021^∗∗	0.020	0.016	0.017	0.012
Open Tender Day V.	[0.008]	[0.008]
Missing Tender Call		−0.076		0.000		−0.002
Missing Tender Call		[0.047]
Page Count		0.002		0.000		0.002
Page Count		[0.009]
Word Count		0.009		0.008		0.007
Word Count		[0.008]
Legality Protocols		−0.003		0.000		−0.002
Legality Protocols		[0.011]
Local Regulations		−0.015		−0.006		−0.006
Local Regulations		[0.011]
Negotiated-No T/N		−0.003		0.000		0.001
Negotiated-No T/N		[0.014]
Sole Source Forbidden		−0.010^∗		−0.007		−0.006
Sole Source Forbidden		[0.005]
MEAT-Qual. Score		0.024^∗		0.015		0.010
MEAT-Qual. Score		[0.013]
Firm List Preference		0.007		0.003		0.003
Firm List Preference		[0.007]
Firm Other Preference		−0.005		0.000		−0.000
Firm Other Preference		[0.008]
Documents Verificat		−0.008		−0.005		−0.002
Documents Verificat		[0.017]
Worksite Verificat		0.011		0.000		0.000
Worksite Verificat		[0.015]
Ad Hoc Subcontract		−0.067^∗∗∗		−0.004		−0.004
Ad Hoc Subcontract		[0.012]
No Subcontr to Bid		0.068^∗∗∗		0.000		−0.001
No Subcontr to Bid		[0.014]
Contact Points Out		0.007		0.002		0.004
Contact Points Out		[0.009]
DV-Hours Share		0.008		0.005		0.005
DV-Hours Share		[0.011]
WV-Hours Share		0.012		0.009		0.008
WV-Hours Share		[0.007]
Observations	3195	3195	3195	3195	3195	3195
Adj R2	0.028	0.029
MSE	0.356	0.356	0.129	0.128	0.128	0.128
False Positive	649	643	699	696	750	730
False Negative	630	623	625	628	622	620
Precision	0.255	0.257	0.245	0.246	0.235	0.240
Recall	0.261	0.263	0.266	0.265	0.270	0.271
F Measure	0.258	0.260	0.255	0.255	0.251	0.254
Threshold	0.196	0.201	0.176	0.177	0.175	0.175

^∗ p<0.1, ^∗∗ p<0.05, ^∗∗∗ p<0.01. All specifications include year and region fixed effects. All the choices involving standard error, validation, and MSE calculations are identical to those reported in the note to Table 6.

Back to article page