Student dropout remains a major concern in many higher education systems. Based on the classic models developed by Spady and Tinto on the link between social and academic integration and dropout, we propose a refined model to explain dropout intentions – relating to dropout from higher education and dropout from a specific study programme – that more strongly emphasises individual background characteristics (e.g. gender, social origin, and immigration background) and, in addition, considers students’ satisfaction with the institutional support structures.

In theory-oriented sociology of education, the influential Spady–Tinto models of student dropout (Spady 1971; Tinto 1975, 1993) serve as useful concepts in the explanation of HE dropout. These models specifically focus on the academic and social integration of students, and presume that these are linked to a certain commitment towards students’ academic and social environment in educational institutions, as well as having an impact on university dropout as a negative outcome. This theoretical framework has been applied by many scholars and – in its classical or extended versions – proven to still be an adequate tool in the explanation of dropout (e.g. Berger and Braxton 1998; Nicoletti 2019).

While major attention in dropout research has been devoted to integration and commitment (e.g. Piepenburg and Beckmann 2021), many were institutions’ support with regard to their roles in students’ dropout intentions (e.g. Rumberger et al. 1990; Yair, Rotem, and Shustak 2020). Given that many dropout studies are based on administrative and/or single-institution data sources, different types of dropout such as leaving the entire HE system, a HE institution or a specific study programme are often not adequately distinguished (for exceptions, see e.g. Meggiolaro, Giraldo, and Clerici 2017; Rodríguez-Gómez et al. 2016).

The innovative potential of revisiting the Spady–Tinto approach with regard to Luxembourg relates to its nationally and culturally diverse HE system, which reflects a case where the effects of the internationalisation of HE can be observed in a magnified way. First, the student population is highly heterogeneous due to a large number of international students – about 50 percent (2019) – and students who received their university entrance certificate in Luxembourg, but have an immigrant background (more than 20 percent). This interesting HE context allows the relationship between immigrant background and dropout intention to be investigated, given the common assumption that immigrant students’ minority status hampers them in developing a sense of belonging within the HE environment as compared to majority group students (Hurtado and Carter 1997). By contrast, international students’ persistence has either been framed in the light of integration obstacles or, conversely, as a product of being highly motivated, and determined to succeed within the host HE context. Second, the University of Luxembourg as a key HE institution is young (it was founded in 2003). It was launched taking into account different international HE systems, and follows an international, interdisciplinary, and multilingual research university approach (Harmsen and Powell 2018).

Vis-à-vis the indicated research desiderata, our main research questions in relation to the conceptual considerations are how individual characteristics (social origin, gender, immigrant backgrounds) impact dropout intentions – regarding dropout from HE and dropout from a specific study programme – via aspects of commitment and integration, and what role institutional support plays in this.

The analysis is based on the Luxembourgish Eurostudent VII data. Eurostudent is a recurring, cross-sectional survey covering HE students, which is conducted in many European countries. Major themes covered relate to students’ living, socioeconomic, and study conditions. The study is run by a European consortium that is managed by the German Center for Higher Education Research and Science Studies (DZHW) in Hanover, Germany (e.g. DZHW 2018). Students in Luxembourg participated for the first time in 2019 (Eurostudent VII). The population covered includes all students who were enrolled in HE programmes in Luxembourg in the academic year 2018/2019 consisting of students from the first to the final year of their studies. The population includes students in (vocationally oriented) short-cycle tertiary study programmes (BTS), classified as level 5 of the International Standard Classification for Education (ISCED), Bachelor students (ISCED level 6), Master students (ISCED level 7) and PhD candidates who are all attending doctoral education at the University of Luxembourg (ISCED level 8). Although the latter are not part of the Eurostudent population, they are included in the Luxembourgish survey. The net sample, after data cleaning (i.e. exclusion of incomplete surveys), amounts to 871 students. Measurements. We distinguish between two different types of dropout intentions. Students responses were dichotomised to generate two dependent binary variables: seriously considering dropping out of one’s study programme (9% of the sample) and of HE entirely (13.5% of the sample). The measurements for academic and social integration, as well as commitment and institutional support, have been developed within the Eurostudent framework (Eurostudent Consortium 2021), and are based on previous scales of the National Educational Panel Study (NEPS) in Germany (Dahm and Lauterbach 2016). Analytical approach. As the theory-driven hypothetical model at the core of our analysis resembles the Spady–Tinto approach with its complex interdependencies between explanatory factors of dropout, structural equation modelling (SEM) appears to be the appropriate data analysis technique. The SEM approach seems to be especially useful in fitting the processual nature of the conceptual background of the Spady–Tinto approach and to adequately analyse the mediating functions of the Spady-Tinto core factors of commitment and integration. We estimated structural equation models using Mplus 7.3 (Muthén and Muthén 1998–2014). As some of the variables are categorical dependent variables, we used the weighted least square mean and variance adjusted (WLSMV) estimator mode for an adequate estimation of path coefficients.

The Spady–Tinto approach – particularly the Tinto (1975) model – could be reproduced based on current Luxembourgish data. The paths of the structural equation model resembled the hypothetical scenario derived from the concepts of Spady (1971) and Tinto (1975, 1993) on the linkages between individual characteristics, initial commitment to HE, academic and social integration, commitment (during studies), and dropout intention. Non-intuitive findings only relate to prior performance (secondary education) as an individual characteristic that showed no effect on any of the factors, as well as the lack of a link between initial commitment and the social integration regarding fellow students. Initial commitment seems to be important for social integration regarding teaching staff and faculty contacts than with regard to social relationships with fellow students. The perceived institutional support, a factor we emphasised refining the Spady–Tinto approach, showed significant positive effects on academic integration, social integration (fellow students, docents), and commitment. Perceiving support by the institution with regard to study support services, learning facilities, balancing studies, family, paid jobs, and future career prospects, appears to have a key role in the prevention of dropout. Considering the two different types of dropout intentions, our study focused on, the mechanisms behind the intention to dropout from a study programme and the intention to entirely leave the HE system seem to be the same. Study commitment appears to be a strong predictor of both intentions. Differences only relate to the mediating function of the Spady-Tinto factors of commitment and integration in the socioeconomic factors-dropout intentions link.

Berger, J. B., and J. M. Braxton. 1998. “Revising Tinto's Interactionalist Theory of Student Departure through Theory Elaboration: Examining the Role of Organizational Attributes in the Persistence Process.” Research in Higher Education 39 (2): 103–119. Clerici, R., Giraldo, A., and S. Meggiolaro. 2015. “The determinants of academic outcomes in a competing risks approach: evidence from Italy.” Studies in Higher Education 40 (9): 1535–1549. Dahm, G., and O. Lauterbach. 2016. “Measuring Students’ Social and Academic Integration—Assessment of the Operationalization in the National Educational Panel Study.” In Methodological Issues of Longitudinal Survey edited by H.-P. Blossfeld, J. von Maurice, M. Bayer, and J. Skopek, 313–329. Wiesbaden: Springer VS. DZHW (German Center for Higher Education Research and Science Studies) (Ed.). 2018. Social and Economic Conditions of Student Life in Europe. EUROSTUDENT VI 2016–2018. Synopsis of Indicators. Bielefeld: wbv. Eurostudent Consortium (Ed.). 2021. Eurostudent VII. Questionnaire Handbook. Hanover: DZHW. Harmsen, R., and J. J. W. Powell. 2018. “Higher Education Systems and Institutions, Luxembourg.” Encyclopedia of International Higher Education Systems and Institutions, https://doi. org/10.1007/1978-1094-1017-9553-1001_1398-1001 Hurtado, S., and D. F. Carter. 1997. “Effects of College Transition and Perceptions of the Campus Racial Climate on Latino College Students' Sense of Belonging.” Sociology of Education 70 (4): 324-345. Nicoletti, M. 2019. “Revisiting the Tinto's Theoretical Dropout Model.” Higher Education Studies 9 (3): 52-64. Piepenburg, J. G., and J. Beckmann. 2021. “The Relevance of Social and Academic Integration for Students’ Dropout Decisions. Evidence from a Factorial Survey in Germany.” European Journal of Higher Education (early online): 1-22, https://doi.org/10.1080/21568235.2021.1930089. Rodríguez-Gómez, D., Meneses, J., Gairín, J., Feixas, M., and J. L. Muñoz. 2016. “They have gone, and now what? Understanding re-enrolment patterns in the Catalan public higher education system.” Higher Education Research & Development 35 (4): 815–828. Rumberger, R. W., R. Ghatak, G. Poulos, P. L. Ritter, and S. M. Dornbusch. 1990. “Family Influences on Dropout Behavior in one California High School.” Sociology of Education 63 (4): 283-299. Spady, W. G. 1971. “Dropouts from Higher Education: Toward an Empirical Model.” Interchange 2: 38-62. Tinto, V. 1975. “Dropout from Higher Education: A Theoretical Synthesis of Recent Research.” Review of Educational Research 45 (1): 89-125. Tinto, V. 1993. Leaving College: Rethinking the Causes of Student Attrition. Chicago: University of Chicago Press. Yair, G., N. Rotem, and E. Shustak. 2020. “The Riddle of the Existential Dropout: Lessons from an Institutional Study of Student Attrition.” European Journal of Higher Education 10 (4): 436-453.