NWG ACCESS - Institutional barriers to study decisions - social inequality, admission procedures and their consequences

The empirical analyses are based on available information on selection procedures for all locally admission-restricted degree programmes between 2005 and 2018 and the selection criteria and limits for all centrally admission-restricted degree programmes since 1973. This institutional data is combined for the first time with longitudinal individual data on eligible students and students. In order to describe the effect of admission procedures on the composition of student cohorts, various reforms relating to admission requirements will be analysed longitudinally. Mediation and moderation analyses with pooled cross-sectional data will be carried out to analyse the specific relationship between admission criteria and choice of degree programme. The international comparison serves to determine the strength of the inequality effect of admission procedures for selected degree programmes. Counterfactual distribution analyses illustrate the effects of alternative admission regimes. The group adds an innovative perspective to empirical educational research and evaluates the inequality effect of admission procedures for higher education policy.

Subproject (SP) 1: Data preparation and collection
In SP1 (data preparation and collection), all group members prepare the institutional data for analysis. The data from the Higher Education Compass with information on local admission restrictions for 9,000 to 11,000 degree programmes per year between 2005 and 2018 are available to the project and are updated annually. This means that information is available on the existence of (local and central) admission procedures for over 126,000 degree programmes. The study programmes awarded in the ZV are available with the ZVS data up to winter semester 2017 and must first be updated. In addition, there is information on the local admission procedures carried out by the ZVS (or SfH) since 2000. In total, detailed information is therefore available on at least two selection criteria and limits from around 12,000 admission procedures.

Subproject 2: Theoretical modelling;

In SP2 (Theoretical modelling), the supervisor develops the central theoretical model (2.1) in cooperation with the doctoral students. The aim is to extend current RC models to include resource-sensitive institutional selection mechanisms as a context for rational educational decisions (Breen/Goldthorpe 1997; Esser 1999). In particular, the totality of the selection boundary and the gatekeeper role of admission procedures must be taken into account (Lohmann/Groh-Samberg 2010; Lucas 2017). In addition, not only current decision-making situations but also investment behaviour aimed at anticipating future selection limits play an important role here (Alon 2007; Breen et al. 2014). The theoretical work can be usefully combined with a research stay at the EUI in 2021/22. In a second step, the development of admission requirements in federal state award regulations over time will be analysed and contrasted with the uniform federal access law in terms of equal opportunities (2.2). The institutional changes in subject-specific social closure developed in this way are used in a further empirical project to simulate the distributional effects of changes in admission procedures over time on the composition of the labour force. Composition of the labour force over time (2.3)

For this purpose, admission profiles for each degree programme, transition matrices after graduation based on graduate surveys and NEPS SC5 data, as well as the (age- and occupation-specific) distribution of the working population in the microcensus are linked with each other. Based on this data, the possible influence of different admission procedures on the development of the subject-specific labour supply (e.g. doctors in rural areas) is then modelled using tested counterfactual distribution analyses (Cunha et al. 2006; Erikson et al. 2005; Jaeger/Karlsson 2018).

Sub-project 3 (Longitudinal study decisions)

The first doctoral project in SP3 (Longitudinal study decisions) investigates the longitudinal relationship between access conditions and decisions to study at university. Changes in selectivity following legal reforms, especially of admission

of admission procedures allow the (causal) inequality effect to be identified quasi-experimentally (3.1). In the analyses for 3.1, both ZVS and Hochschulkompass data can be used in conjunction with consecutive surveys of eligible students and the NEPS SC4 data to map the relationship between resources, HZB type, admission criteria and admission to degree programmes before and after the 1998, 2002, 2004 and (if already possible) 2019 reforms (e.g. with panel regressions, regression discontinuity analyses and DvD approaches according to Bernardi 2014; Betthäuser 2017; Rauscher 2015).

The effect of admission requirements on the study decisions of successive cohorts of eligible students is analysed over time using differentiating selection criteria within the same degree programmes (3.2). Finally, counterfactual analyses describe how changes in admission regimes can affect student diversity (3.3). Where 3.2 and 3.3 are concerned with describing the composition of the student body, the analyses - taking into account selectivity - can also draw on the information provided by surveys of first-year students, social surveys or student surveys. Admission criteria are mapped here using the ZVS dataset.

Sub-project 4: Inequality in the choice of degree programme

In subproject 4 (Inequality in the choice of degree programme), the second doctoral project investigates the relationship between admission requirements and the choice of degree programme by those eligible to study. Firstly, variations in the selectivity of degree programmes are described in cross-section (4.1). Then, the outlined decision-making process of students with higher education entrance qualifications is modelled in order to determine the effect of different resources on subject choice in the transition to higher education (4.2). Finally, changes in the level of admission barriers are analysed in order to quantify the effects of greater competition on social inequality (4.3). For all three work packages, the surveys of eligible students pooled over time and linked to the ZVS data from the waves before and after admission are analysed. Methodologically, mainly discrete choice models (logistic regressions) are used here to investigate the processes of degree programme choice (Temme 2007; Mood 2010). The focus is firstly on the identification of possible mediation effects of the different selection criteria in the explanation of resource effects on the choice of degree programme and secondly on possible moderation relationships of the selection boundaries within this context.

SP5: International comparison

The aim of SP 5 of the ‘ACCESS’ project is to conduct international comparative research into the role of institutional barriers in the reproduction of social inequality in higher education access. Against the background of initial simple analyses based on the ESS, which show that the correlation between social background and access to higher education varies between countries, the question arises as to whether or to what extent country differences in the extent of social inequality in access to higher education can be explained by differences in admission procedures (institutional hurdles). As the available data on country differences in admission systems proved to be either not reliable enough (OECD) or too superficial or geographically limited (EU Commission), we decided to collect the necessary data ourselves using a questionnaire. In the meantime, we have collected 33 interviews with experts in order to be able to analyse the typical admission routes to university on a comparative basis. In a first step, we used a qualitative analysis to reconstruct a typology of alternative admission regimes. In a second step, we correlate these regimes with access inequalities in the 33 European and other OECD countries. Counterfactual analyses complement these descriptive analyses in order to determine which patterns of social inequality could be observed under alternative admission criteria.