Four Essays on University Economics - KOPS - Universität Konstanz

More documents

Recommendations

Info

3.3.3 Legal Situation Return of foreign students In the period between 1990 and 2004, German migration laws obliged foreign students to return to their home countries after graduation. Even though some students might have been forced to return by these laws, many students found a way to stay. The enforcement of these laws was, thus, not too strict. Graduates, who married a German spouse, were allowed to stay in Germany. Unfortu- nately, we do not know, when students in our sample married and what nationalities their spouses are. However, many unmarried students also stayed in Germany. Migration agencies made the decision whether graduates were allowed to stay on a case-by-case basis. It was common practice, however, that graduates who found adequate employment in Ger- many or who planned further education in Germany were allowed to stay. Furthermore, we know of no informal rules which guided migration agencies’ decisions. In the years 1993, 1997 and 2000 migration laws in Germany were changed. In 1993 the right of asylum was restricted, in 1997 the legal position of foreigners living in Germany was improved and in 2000 the green card was introduced and conditions for naturalization were changed. Even though these changes were not explicitly aimed at students, they might have affected their return decision. Apart from the change in 1993, the changes should have had a positive effect on the propensity to stay. Since 2005 foreign students are allowed to stay in Germany after graduation to search for employment. Accordingly, we expect that more students choose to stay than before. 3.4 Data and Empirical Strategy For our analysis we use data of 2,436 students from 76 countries who were sponsored by the KAAD and who finished their study period in Germany between 1990 and 2005. The largest group of students came from Poland (359 students), followed by China (165 students) and Romania (134 students). We perform a duration analysis to investigate which determinants have an impact on the decision of these graduates whether and when to return to their home countries after they finished their studies in Germany. We assume that at the end of the study period in the host country a graduate decides whether to stay or to return and that he or she revises this decision regularly. A graduate who did not 59
Return of foreign students return immediately after graduation can decide to return at any time later. Our data are interval censored. We know in which months the students in our sample graduated and returned. Therefore, we use a discrete time model. The hazard rate can, then, be interpreted as the conditional probability that a graduate returns in a particular month given that he or she did not return before. For the functional form of the hazard rate, we choose the complementary log-log transformation. This is the discrete time representation of a continuous time proportional hazards model. The data is reorganized in person period format: For every student in our sample, we created as many data rows as there are months he or she stayed in Germany after the end of the study period. We include fixed and time varying covariates in our regression. To account for duration dependence we model the baseline hazard as a cubic polynomial. This specification allows for non-monotonicity of the baseline hazard and has a significantly better fit than a quadratic polynomial. Because of the large number of time periods, we did not include dummies for every time period. To test for potential unobserved heterogeneity (“frailty”), we estimated the model in- cluding normally distributed unobserved heterogeneity. The likelihood ratio tests indicate that there is statistically significant frailty of this type in some of our sub-samples. In the following, we briefly describe the variables we use in the duration analysis (for de- scriptive statistics see Table 3.8, for variable descriptions see Table 3.15 in the Appendix). 3.4.1 Dependent Variable The dependent variable is a dummy for the decision to stay in Germany. Since our data is restricted to the period between January 1987 and August 2005, the spells of some graduates are right censored. The return date is missing for 39.24 % of the students in our sample. The differences between the three programs are quite large (see Table 3.3). Given the aim of the KAAD and the legal situation in Germany, the number of students who decided to stay - at least for some more months after graduation - is surprisingly high in all three programs. Our identifying assumption is that all students, for whom the return date is missing, are still in Germany in August 2005. In principle, it might be the case that some students whose return dates are missing did return but did not report to the KAAD. Variables, 60
Page 1 and 2:
Four Essay
Page 3 and 4:
Contents 3.1 Introduction . . . . .
Page 5 and 6:
List of Figures 1.1 Total economic
Page 7 and 8:
List of Tables 3.6 Summary: Signifi
Page 9 and 10:
Summary average quality of research
Page 11 and 12:
Chapter 1 How Did Economic Research
Page 13 and 14:
Economic research in Switzerland fo
Page 15 and 16: Economic research in Switzerland pu
Page 17 and 18: Economic research in Switzerland Fi
Page 19 and 20: Economic research in Switzerland si
Page 21 and 22: 0,45 0,05 Economic research in Swit
Page 23 and 24: Economic research in Switzerland (2
Page 25 and 26: Wirtschaftspolitik 5 (3), 363 - 375
Page 27 and 28: 2.1 Introduction Research Productiv
Page 29 and 30: Research Productivity in Business E
Page 45 and 46: 2.7 Appendix Sample: university dat
Page 55 and 56: Chapter 3 How Can Scholarship Insti
Page 57 and 58: Return of foreign students scholars
Page 59 and 60: Connections in the Home Country Ret
Page 61 and 62: Return of foreign students tegrate
Page 63 and 64: to finance the first period of thei
Page 65: Return of foreign students Table 3.
Page 69 and 70: children. Return of foreign student
Page 71 and 72: Return of foreign students A low pa
Page 73 and 74: Return of foreign students trading
Page 75 and 76: Return of foreign students Table 3.
Page 77 and 78: higher return rates might result. R
Page 79 and 80: 3.7 Appendix Return of foreign stud
Page 87 and 88: Bibliography Return of foreign stud
Page 89 and 90: Return of foreign students Park, J.
Page 91 and 92: 4.1 Einleitung Determinanten des be
Page 93 and 94: 4.3.1 Mögliche Einflussfaktoren De
Page 95 and 96: Determinanten des beruflichen Erfol
Page 101 and 102: teraktionseffekt). Determinanten de
Page 105 and 106: Frauen gibt. Determinanten des beru
Page 107 and 108: 4.5.2 Einfluss des Einkommens auf d
Page 111 and 112: nimmt ihre Zufriedenheit mit ihren
Page 117 and 118:
Literaturverzeichnis Determinanten
Page 119 and 120:
General Bibliography Arora, A., P.
Page 121 and 122:
General Bibliography Combes, P.-P.
Page 123 and 124:
General Bibliography Guimarães, P.
Page 125 and 126:
General Bibliography Martin, P. L.
Page 127 and 128:
General Bibliography Vieira, J. A.
Page 129 and 130:
Zusammenfassung sitäten sind jedoc
Page 131 and 132:
Zusammenfassung ligt, die ihr Studi
Page 133 and 134:
Erklärung Ich versichere hiermit,
show all

Four Essays on University Economics - KOPS - Universität Konstanz

Create successful ePaper yourself

Delete template?

Save as template?