Table 4. Node level regression for collaboration tie degrees.

	Collaboration tie degrees within Rocom
Variable	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8	Model 9	Model 10	Model 11
LnTenure	.42***	−.25
LnPapers		.74***	.61***	.61***	.61***	.59***	.60***	.51***	.50***
LnCitations			−.12	−.13	−.08	−.10	−.15			.34***	.29***
Gender				.02	.03	.10	.01
Professor					−.12	−.07
Associate prof.						.11
Prestige							.06		.03		.10
R2	.17***	.28***	.27***	.27***	.28***	.29**	.27**	.26***	.26***	.11***	.12***
Adjusted R2	.17***	.27***	.26***	.26***	.27***	.27**	.26**	.26***	.26***	.11***	.11***

Note. Standardized Coefficients reported, *p<.05, **p<.01, ***p<.001. The node level regression models were performed using 20.000 random permutation. The most important academic titles within the Romanian higher education system are those of a ‘Professor’ and ‘Associate Professor’ (in Romanian ‘Conferentiar’). In our regression models, we created dummy variables for both titles. We also created a dummy aggregated variable called ‘Prestige’, where an ego has 1 if Professor or Associate Professor, and 0 otherwise.