Higher-order structures of local collaboration networks are associated with individual scientific productivity

EPJ Data Science

Table 3 Fixed-effects Poisson regressions regarding the association between \(\beta _{0} \) and scientific productivity

Productivity(ln)	(1)	(2)	(3)	(4)	(5)
\(\beta _{0}\)	0.094***	0.164***	0.157***	0.139***	0.116***
\(\beta _{0}\)	(0.00006)	(0.00013)	(0.00014)	(0.00015)	(0.00017)
\(\beta _{0}^{2}\)		−0.004***	−0.004***	−0.003***	−0.004***
\(\beta _{0}^{2}\)		(0.00001)	(0.00001)	(0.00001)	(0.00001)
Network density			−0.255***	−0.198***	−0.026***
Network density			(0.00059)	(0.00059)	(0.00060)
Average tie strength (log)			0.262***	0.101***	0.071***
Average tie strength (log)			(0.00023)	(0.00033)	(0.00025)
Collaborative strength				0.128***	0.240***
Collaborative strength				(0.00068)	(0.00074)
Career age (log)				−0.003***	−0.026***
Career age (log)				(0.00014)	(0.00031)
\(\text{Productivity}_{\overleftarrow{t}}\)(log)				0.232***	0.020***
\(\text{Productivity}_{\overleftarrow{t}}\)(log)				(0.00035)	(0.00018)
Year fixed effects	N	N	N	N	Y
Individual fixed effects	N	N	N	N	Y
Network size effects	N	N	N	N	Y
Log pseudolikelihood	−44,919,517	−44,482,073	−43,456,407	−37,342,694	−34,025,920
R²	0.0848	0.0937	0.1146	0.1428	0.2189
Observations	27,786,774	27,786,774	27,786,774	24,000,967	24,000,967
Individuals	3,785,807	3,785,807	3,785,807	3,785,807	3,785,807

Notes: Robust standard errors in parentheses; *** p-value < 0.001, ** p-value < 0.01, * p-value < 0.05; 3,785,807 observations dropped when we control for \(\text{Productivity}_{\overleftarrow{t}}\).