Table 3. Multinomial logit regression results for Ph.D. career choices.
| Base outcome: Academia | Industry | Public Administration | ||||
|---|---|---|---|---|---|---|
| I | II | III | IV | V | VI | |
| Female | 0.690*** | 0.609*** | 0.606*** | 1.662*** | 1.464** | 1.471** |
| (0.086) | (0.076) | (0.075) | (0.322) | (0.284) | (0.283) | |
| Ph.D. demographic and predetermined characteristics | ||||||
| EU-15 nationality | 0.808* | 0.841 | 0.878 | 0.609** | 0.643* | 0.662* |
| (0.091) | (0.097) | (0.102) | (0.140) | (0.150) | (0.155) | |
| Non-EU-15 nationality | 0.511*** | 0.528*** | 0.572*** | 0.357*** | 0.365*** | 0.376*** |
| (0.074) | (0.078) | (0.086) | (0.106) | (0.110) | (0.113) | |
| Age | 0.987 | 0.961** | 0.946*** | 1.103*** | 1.067** | 1.057* |
| (0.019) | (0.018) | (0.018) | (0.034) | (0.034) | (0.034) | |
| Worked prior to Ph.D. | 2.614*** | 2.625*** | 2.198*** | 0.971 | 0.962 | 0.900 |
| (0.394) | (0.398) | (0.334) | (0.297) | (0.302) | (0.288) | |
| Ph.D. publication output | ||||||
| # of publications during Ph.D. | 0.534*** | 0.505*** | 0.522*** | 0.507*** | ||
| (0.036) | (0.034) | (0.061) | (0.060) | |||
| Ph.D. involvement in applied projects | ||||||
| Involved in applied projects during Ph.D. | 3.121*** | 1.783** | ||||
| (0.413) | (0.499) | |||||
| Supervisor publication output | ||||||
| # of publications | 1.104 | 1.098 | 0.979 | 0.988 | ||
| (0.067) | (0.069) | (0.107) | (0.112) | |||
| Supervisor involvement in applied projects | ||||||
| Had patents granted | 1.059 | 0.751 | ||||
| (0.138) | (0.175) | |||||
| Involved in EU projects with industrial partners | 0.840 | 1.241 | ||||
| (0.104) | (0.269) | |||||
* p<0.1
** p<0.05
*** p<0.01.
Coefficients are relative risk ratios. N = 2,345. Standard errors clustered around supervisors are in parentheses. Controls include labor market characteristics at graduation, university-research field fixed effects, and graduation-year fixed effects.