EPJ Data Science

Table 3 Logistic regression models for GitHub vs. Penumbra outcome

From: The penumbra of open source: projects outside of centralized platforms are longer maintained, more academic and more collaborative

	Model 1			Model 2
	\(e^{\beta}\)	p	CI	\(e^{\beta}\)	p	CI
Constant	0.188	<0.001	[0.156,0.225]	0.435	<0.001	[0.364,0.520]
Language (vs. JavaScript)
Bourne (Again) Shell	3.478	<0.001	[3.162,3.826]
C	4.065	<0.001	[3.671,4.502]
C#	1.589	<0.001	[1.444,1.750]
C++	5.636	<0.001	[5.184,6.127]
C/C++ Header	5.829	<0.001	[5.103,6.657]
Java	2.192	<0.001	[2.070,2.321]
Jupyter Notebook	2.722	<0.001	[2.459,3.012]
OTHER	2.124	<0.001	[2.023,2.230]
PHP	2.524	<0.001	[2.323,2.743]
Python	2.804	<0.001	[2.651,2.965]
TeX	30.641	<0.001	[25.348,37.040]
TypeScript	1.078	0.187	[0.964,1.205]
Vuejs Component	4.940	<0.001	[4.331,5.635]
Files	1.000	<0.001	[1.000,1.000]	1.000	<0.001	[1.000,1.000]
Commits	1.001	<0.001	[1.001,1.001]	1.001	<0.001	[1.001,1.001]
Average editors per file	3.337	<0.001	[3.002,3.709]	3.328	<0.001	[3.002,3.689]
Average message length	1.002	<0.001	[1.001,1.002]	1.002	<0.001	[1.001,1.003]
Burstiness	1.059	<0.001	[1.055,1.063]	1.058	<0.001	[1.053,1.062]
Average interevent time [h]	1.000	<0.001	[1.000,1.000]	1.000	<0.001	[1.000,1.000]
Branches	0.971	<0.001	[0.965,0.976]	0.961	<0.001	[0.955,0.966]
Lead workload	0.461	<0.001	[0.407,0.522]	0.433	<0.001	[0.384,0.489]
Committers	0.945	<0.001	[0.936,0.955]	0.940	<0.001	[0.931,0.950]
Effective team size	1.010	0.516	[0.980,1.040]	1.016	0.315	[0.985,1.047]
-2LL	84,478.844			89,788.631
Pseudo-R2	0.100			0.044

Back to article page