Table 7. Regression analysis of the relationship between team structure and the impact factor of journals publishing coronavirus research in pre- and during COVID-19.
| Independent variables | Dependent variable—Source Normalized Impact per Paper | ||||
|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | |
| COVID-19 | 0.086*** (0.020) p = .000 | 0.088*** (0.019) p = .000 | 0.088*** (0.019) p = .000 | 0.064*** (0.024) p = .007 | 0.107*** (0.027) p = .000 |
| Authors China | -0.009 (0.013) p = .459 | 0.0012 (0.013) p = .921 | -0.023** (0.012) p = .044 | ||
| International Team | 0.069*** (0.012) p = .000 | 0.078*** (0.011) p = .000 | |||
| COVID-19 x Authors China | 0.062 (0.041) p = .127 | ||||
| COVID-19 x International Team | -0.046 (0.039) p = .235 | ||||
| N | 4,502 | 4,502 | 4,502 | 4,502 | 4,502 |
Estimates stem from ordinary least square model regression specifications with dependent variables being inverse hyperbolic sine transformed SNIP of a publication in the sample, and independent variables being the period of the publication (COVID-19 or pre-COVID-19) (column 1), whether the authors of the publication are from a Chinese institution (column 2), and whether the publication author team is international (column 3). In columns 4, 5, and 6 we include interaction terms of COVID-19 period and the team structure to assess whether there is a different relationship between team structure and SNIP of a publication pre and during-COVID-19.
Robust standard errors in parentheses.
*, **, *** denote statistical significance at p values of 0.1, 0.05 and 0.01.