Transition to Long-Term Baccalaureate School in Switzerland: Governance, Tensions, and Justiﬁcations

: In Switzerland, the baccalaureate school is an important pathway to university education, and the aspirations of families and students to enter baccalaureate school have grown. However, vocational education and training (VET) remains the predominant educational pathway and has a strong lobby. We investigate how in this context, the transition from primary education to baccalaureate school is governed and justiﬁed at the cantonal level. We study how two Swiss cantons try to meet the ofﬁcial or unofﬁcial maximum baccalaureate quotas desired by educational policymakers through different selection procedures and admission criteria. Drawing on the Economics of Conventions, we conceptualize selection procedures as cantonal transition chains and show that the strategies, procedures, and instruments applied in governance are rooted in diverse principles of action. This causes tensions within cantons. Our analysis shows that agency and regulating effects in the governance of transitions must be understood as distributed among actors, technologies, and objects.


Transition to Baccalaureate School as a Royal Road to Traditional Universities
In Switzerland, as in many other countries, the baccalaureate school (Gymnasium) is an important pathway to university. In contrast to other certificates at upper secondary level, such as the specialized baccalaureate and vocational baccalaureate, the baccalaureate certificate (Gymnasiale Maturität) allows students to enter most areas of study at traditional universities (Universitäten), universities of applied sciences (Fachhochschulen), and universities of teacher education (Pädagogische Hochschulen) without any further admission procedure except an internship at universities of applied sciences. Today, almost all baccalaureate school graduates enter tertiary level education, and more than two thirds enter traditional universities [1]. Consequently, the baccalaureate school is considered the "royal road" to traditional universities [2]. Correspondingly, the aspirations of families and students to enter baccalaureate school have grown in recent decades [3].
However, the proportion of young people that attend baccalaureate school after lower secondary education differs significantly among the 26 Swiss cantons. This variation cannot be explained solely by differences in students' academic performances [4][5][6][7]. It can be assumed that these differences are the result of governance mechanisms: regulations and restrictions at the institutional level of the transition to baccalaureate school. Because the Swiss education system is highly federalist [8], selection procedures, admission criteria to baccalaureate school, and the numbers of study places available vary between cantons [1,9]. Some cantons officially pursue a maximum quota of students [1,6]; others strive to limit the places available unofficially without communicating this quota in official documents; still others operate without quotas. However, a quota and thus a restriction of places in the baccalaureate school as well as cantonal differences regarding the places available are at odds with societal demands for equal opportunities in education and for addressing the shortage of skilled labor. Therefore, baccalaureate quotas and selection procedures are frequent topics of dispute in public discourse, education policy, and science [10][11][12][13][14]. This indicates that selection processes and instruments, such as admission regulations and official and unofficial maximum quotas of baccalaureate students, must be publicly justifiable.
Baccalaureate school is offered in two versions: the long-term baccalaureate school of six years, which starts at lower secondary level after primary school, and the short-term baccalaureate school of four years, which begins at upper secondary level. In this article, we focus on the governance of selection to long-term baccalaureate school and policy attempts to regulate the proportion of students entering.
Educational transitions have mostly been studied with a focus on the individual that transitions and the factors that influence this transition. What remains largely unexplored is the institutional perspective on transitions, which considers the governance and institutional configurations that frame transitions. We aim to close this gap by investigating how educational provision, rules, processes of selection, and discourses influence the transition to long-term baccalaureate school and how specific selection procedures are implemented and legitimized by policymakers and other actors.
We conduct a comparative case study in two cantons that apply different selection procedures; moreover, one canton applies an official quota, the other an unofficial one. We investigate how selection procedures are regulated, organized, and justified by asking the following questions: • With which instruments, procedures and strategies do the actors responsible attempt to achieve the maximum selection quota for long-term baccalaureate school? • Which logics of action and justification underlie these governance mechanisms? • Which problems, tensions, and criticisms result from the governance of the selection procedure, and how can they be explained by cantonal contexts?
The article is structured as follows. Section 2 briefly describes the main features of the Swiss education system that are relevant to the topic. Section 3 outlines the state of research on educational transitions from primary to upper secondary level. In Section 4, we present the theoretical approach of economics of conventions and its suitability for our research topic, which is followed by a description of the data and methods used in Section 5. Results on the governance of selection to baccalaureate school in two cantons are presented in Section 6. Section 7 consists of a summary, discussion, and conclusion.

The Swiss Case: Controversial Transition Quotas and Admission Regulations in a Highly Federalist, Differentiated, Employment-Centered Education System
The Swiss education system has several distinctive features. Firstly, its education system is strongly federalist, and consequently the 26 Swiss cantons have a high degree of autonomy over their education systems [8]. Therefore, it is an excellent place to study the governance of transitions, since it allows comparative studies that otherwise would have to be conducted internationally [15,16]. Studies here can thus show how different ways of governing transitions in different local contexts can lead to different outcomes and tensions.
Secondly, the Swiss education system is characterized by early differentiation and tracking [17]: all cantons group students by ability after just six years of primary school, when the transition to lower secondary education takes place. About half the cantons offer long-term baccalaureate schools, which is the most demanding track. However, the number of baccalaureate schools and their selection procedures and admission criteria vary [9]. In some cantons, the selection procedure consists of an entrance examination in various subjects organized by the baccalaureate school and usually averaged with primary school grades. In other cantons, the allocation to baccalaureate school is based on primary school grades, primary school teachers' assessments, and parental preference.
Thirdly, vocational education and training (VET) predominates within the Swiss education system [18]. A strong VET lobby strives to lead high-performing students into vocational education after lower secondary education [18], and indeed two thirds of young people enter VET after compulsory school [19]. The Swiss education system therefore can be classified as employment centered [20] (p. 48). This explains to a certain extent why the proportion of students entering long-term baccalaureate school after primary school is controversial, as these students are diverted from the VET track after compulsory education. Moreover, the overall proportion of baccalaureates among upper-secondary school-leavers in Switzerland is just 22% [21], only half that of the other German-speaking countries, Germany and Austria [22,23].
This however is in contradiction with the international trend of young people delaying vocational specialization and rather remaining in general forms of education [20] (p. 51). Additionally, ever more parents have undergone education at university level [3] and prefer that their children also choose an academic path [24]. The demand for higher education is reinforced by a trend towards academization of society as a whole [3]. In this context, the low proportion of baccalaureate students in Switzerland once more points to mechanisms of governance being used to regulate the selection procedure to baccalaureate school.
In conclusion, Switzerland can be classified as a highly differentiated [17,25] education system with restrictive selection procedures to baccalaureate school. It is thus comparable with other countries with similar features, such as Germany and Austria [26]. Nevertheless, it is of general interest how selection processes in educational transitions are governed by admission regulations and quotas, and how this is justified.

Governance of Transition to Lower Secondary Education and the Role of Maximum Transition Quotas: The State of Research
Educational transitions have mostly been studied with a focus on the individual that transitions and the factors that influence this transition. This has been conducted from a life course perspective with rites of passage [27] and status passages [28], with a focus on and problematization of unemployment and early school leaving [29,30], on transition to employment [31], and on support that might be needed [32]. Furthermore, considerable attention has been given to how access to different school types and their outcomes depends on social origin, migration background, gender [33][34][35], and class-based educational decision-making [5,36].
However, little empirical research has examined governance mechanisms and selection procedures into lower secondary education. At a European level, educational transitions have been studied in the Governance of Educational Trajectories in Europe (GOETE) Project [37]. This project examined how educational transitions and pathways are regulated in eight EU countries and how educational decisions are made at various policy levels [26,38]. It studied how countries seek to improve access to education and reduce disadvantage with a multilevel analysis at European, national, regional, and local levels, and with a focus on the institutional and discursive opportunity structures that frame educational transitions in the eight countries. However, the study did not address forms of the restriction of places or its justification in public education, which are key issues in our paper.
For Switzerland, Künzle provides an overview of official and unofficial quotas for the various performance tracks in the German-speaking cantons [6]. Hofstetter shows which instruments and allocation practices teachers in the Canton of Fribourg use to keep quotas for various requirement profiles stable [39]. Quotas communicated by the authorities and legitimized by previous experience are crucial in this process. The "maneuvering mass of unclear selection cases" [39] (p. 237) serves to fill the places available at the lower secondary level. Institutional and organizational needs and discourses influence the arguments of teachers and school administrators.
Hasse and Schmidt investigated how allocation agreements are reached between teachers and parents in the transition from primary to lower secondary school in the context of cantonal selection quotas [40,41]. Formal accounts, such as test scores and grades, and collective accounts of shared patterns of justification seem to be crucial, and institutionalized practices and routines seem to prevent parents from refusing to consent.
Neuenschwander emphasizes the importance of cantonal governance in defining not only the procedures, responsibilities of actors, and selection criteria, but also the maximum transition quotas. Thus, allocation results are affected not only by students' individual performances but also by cantonal governance [42,43]. Until now, hardly any research has examined the role of governance in baccalaureate school selection, and especially in entrance examinations. One exception is the study by Bauer [44], who identifies "grey zones and fuzzy logics" that are widely neglected by political actors in a canton where students are selected by entrance examination. The study indicates the contradictory and paradoxical logics of action in the selection procedure.
The state of research shows that, firstly, official and unofficial maximum selection quotas to baccalaureate school are highly important for configuring and implementing selection procedures. Secondly, the proportion of baccalaureate students varies between cantons [45]. Thirdly, the number of students attending baccalaureate school is a controversial topic in education policy in Switzerland. Public figures in policy, business, and academia hold conflicting views. Some criticize the admission of only a limited number of students and call for abolishing all restrictions so that children from socially disadvantaged classes gain access to academic education [10]. Others emphasize the integrative power of VET [11][12][13] and advocate reducing baccalaureate school quotas to lead high-achieving students into VET, reduce youth unemployment, strengthen the VET system, and maintain a high level of performance at baccalaureate schools [14]. The diversity of these grounds indicates the high importance of justifications for transition quotas and their restriction. Thus, to study educational transitions from a governance perspective, quota regulations, selection instruments, and actors' justifications must all be taken into account.

Doing Governance with Plural Logics of Action: Economics of Conventions
The cantonal governance of annual selection to long-term baccalaureate school is a complex situation of coordination of various actors involved in synchronous and diachronic processes. Maximum transition quotas, the number of teachers and school facilities available, and selection regulations must all be considered. The actors involved need to coordinate their actions to manage the annual selection procedure and justify the outcome to educational authorities, the public, and students and their families.
To examine this coordination, we draw on the educational governance approach [46][47][48], which proposes an extended actor model with a multitude of actors (policymakers, administration, school management, teachers, and parents), coordination of action across different levels of political responsibilities, and constellations of actions embedded in institutions and structures. This conception of governance is compatible with various theoretical approaches [49], particularly with social practice theories [50]. From a practice-theoretical perspective, governance is understood as doing governance [50]. Therefore, the focus is on how coordination of action between various actors is realized, processed, regulated, negotiated, and performed; how rules, routines, norms, and values stabilize practices and how they are changed and brought to existence by practices; and how artefacts and objects intervene in practices and are involved in actions.
The economics of conventions (EC) is a praxeological sociology [51] that offers a framework for integrating the conceptual requirements of the educational governance perspective as well as the methodological prerequisites of social practice theory. Rooted in French sociology [51,52], it integrates pragmatic and institutional perspectives and bridges the semantic and the pragmatic level of action. In contrast to Kussau and Brüsemeister [53], the EC assumes that actors are not bound to certain logics of action and justification; for instance, because of their profession. Instead, the assumption is that individuals have agency and reflexive and moral competences [51,52] and can evaluate the situational appropriateness of a range of plural logics of action.
However, because situations of coordination are characterized by uncertainty, ambiguity, and contingency, the rationality of actors is limited [54]. It is uncertain how many students want to enter baccalaureate school or whether they meet performance requirements. It is a complex task to bring the number of aspiring baccalaureate students together with the local conditions in schools, such as numbers of classes and teachers and financial resources.
To manage these governance processes, actors base their evaluations, decisions, and justifications on conventions: culturally established principles of action and evaluation. These conventions are historically established, legitimated logics of action and orders of justification and worth [52] employed to pursue common goods, such as efficiency, equality, and community. Several conventions have been reconstructed that have proven to be empirically significant for the coordination of education [55][56][57][58]. Table 1 lists those that are relevant for the analysis in this article. EC considers practical situations of governance as being characterized by a plurality of co-existing conventions that provide alternative and legitimate logics of action [50]. By taking this plurality into account, the assumption is rejected that only certain logics of action apply in specific social contexts (e.g., scientific logics in science). Therefore, coordination means mediating a plurality of competing interests, handling criticism and disputes, and finding solutions. For example, it is conceivable that when actors are doing governance of the selection, the civic convention of expecting equal access to the baccalaureate school might clash with the market convention regarding financial resources for teachers and the industrial convention that requires efficient implementation. In such cases, compromises can be made. If a selection procedure is criticized, it can be subjected to a reality test with the relevant convention. If the test is successful, the procedure and its legitimacy is strengthened.
Conventions manifest themselves both cognitively in schemes, routines, and norms, and materially in objects. Such cognitive and material forms include selection criteria, estimates of transition quotas, funding mechanisms, examination and assessment procedures, real estate, teaching staff, and information brochures. The EC terms these items form investments, because they put knowledge and values into forms [59]. To coordinate, evaluate, and justify in situations, actors draw on such cognitive formats and objects. Investing in forms means sacrificing alternatives [60] but provides "relief and reliance on the established" [51] (p. 297) in return. As a result, conventions gain social, temporal, and spatial scope, freeing up agency and power to achieve coordination for the common good across levels of action, usually referred to as micro, meso, and macro [51,61,62]. The methodological and analytical focus is on "situations" of governance, cognitive formats, constellations of objects, coordination requirements, institutional arrangements, such as organizations, people, and concepts [51] (p. 374f.), and still considers the agency of actors in these situations [63] (p. 239). EC refers to configurations of situations, including their material and cognitive forms, actors, and conventions, as dispositives [64] We conceptualize governance of selection to baccalaureate school as a chain of synchronous and diachronic situations of coordination of action among various actors. Borrowing from the ideas of global value chains [65], statistical chains [66], and trans-sequential analysis [67], such transition chains consist of series of situations in which the actors involved shape, format, and transform the "production of selection." The term statistical chain has been used to describe a series of stations involving various forms of practice, logics of action and conventions, and groups of actors that produce statistics [51]. We apply the same logic to transitions. Trans-sequential means that the linkage between the situations in the transition chain is not simply linear, that situations are characterized by several local interlocking sequentialities, where objects such as statistics are brought in, back, and forth, are formed, construct value, and reshape the situation [67]. The conventions funding these situations can be transmitted from situation to situation. However, as other actors, objects, and coordination requirements come into play, constellations of conventions can also change.
From this theoretical perspective, we examine the doing of governance of the baccalaureate school selection as a complex network of actions with synchronously and diachronically interlinked coordination situations based on multiple conventions.

Data and Methods: A Comparative Case Study in Two Swiss Cantons
This study compares two Swiss cantons that differ in two criteria: (1) policies for restricting the number of places at baccalaureate schools and (2) admission regulations and selection procedures. To ensure that personal data are protected, the cantons are not named, all data are anonymized, and the exact number of baccalaureate schools as well as certain sources are not disclosed in the text or in Table 2 below. Upon reasonable request, a list of the sources and references not made public can be obtained from the authors. In Canton A, students are allocated by the primary school, and a maximum quota is officially imposed. In Canton B, students are selected through an entrance examination, the results of which are averaged with marks from primary school. In addition, students must pass a probationary period of several months. Canton B also aims for a maximum quota, but this is not officially documented. The cantons share similarities in the socioeconomic profile of the population and economic structure, but they differ in population sizes and numbers of baccalaureate schools.
In both cantons, about 30% of the overall population and 40-50% of the immigrant population hold a university qualification. The cantons differ in their overall baccalaureate quota, in the transition quota to long-term baccalaureate school, and in their development over the previous 15-20 years. As in the rest of Switzerland [70], the number of students is rising in both cantons for demographic reasons (Documents A6 & B14, anonymized).
For this article, we analyzed documents such as cantonal laws and regulations concerning selection, evaluation forms, information brochures, and reports. We conducted six problem-centered online or in-person interviews [71] with representatives of the education administration and those responsible for the entrance examination. The data were analyzed with theory-oriented coding [72][73][74] following grounded theory [75]. The analysis was conducted in line with EC premises, with a focus on conventions, logics of actions, orders of worth, situations, actors, form investments, reality tests, justifications, criticisms, and compromises relevant to the governance of the transition to baccalaureate school.

Transition Chains from Primary to Long-Term Baccalaureate School: Governance, Actors, and Justifications
We conceptualize and analyze the doing of governance of the two selection procedures to baccalaureate school as a chain of coordination situations: a transition chain. Which instruments, procedures, and strategies do the actors use to achieve the desired maximum quota? Which logics of action and justification underlie these governance mechanisms? Which problems, conflicts, criticisms, and policy reactions result from the governance of the selection procedure and how can they be explained by cantonal contexts?
In both cantons, according to the actors involved, the overall goal is to select the "right" students: those able to pass baccalaureate school. This should protect them from disruptions in their educational career. The primary concern is for the well-being and personal development of adolescents, which refers to the common good of the domestic convention (see Table 1). As we show, it has a different scope and power in both cantons. The citations used in the following chapters have been translated to English by the authors.

Canton A: Highly Monitored Allocation by Primary School Teachers and the Problem of the Rising Quota
In Canton A, transition from primary to long-term baccalaureate school has been based on a teacher's recommendation for 30 years. Investments in the domestic convention to protect pupils from disruptions can be found in almost all situations of the transition chain: We want to allocate children to the right school and not just say, based on an exam: "These children will go to baccalaureate school, 15% will drop out, and they should see for themselves what to do next." Instead, the goal is to have a very low quota of dropouts so that there are no career disruptions for the children (representative of the transition committee Canton A).
6.1.1. Characteristics of the Transition Chain Figure 1 illustrates the selection procedure of Canton A, which consists of many interlinked situations that are not merely linear but trans-sequential [67]. The transition chain is activated by various actors, mainly members of the transition committee, teachers, and parents. Every year, selection is characterized by the challenge of fitting the large number of aspiring students to the maximum transition quota desired by educational policymakers. The actors tackle this challenge with the following strategies, instruments, and procedures.
The maximum transition quota to baccalaureate school that educational policymakers strive for is officially communicated and documented. We consider this as an investment in a cognitive form that influences and frames the transition chain into baccalaureate school in Canton A.
The transition committee conducts an annual monitoring of the selection procedure and publishes a monitoring report for the Cantonal Education Council (box on the left, p). Every year, the report analyzes the development of the transition quotas of each municipality, the number of cases for which no agreement was reached between teachers and parents in the allocation meeting (Figure 1, box h), the dropouts from baccalaureate school during the first year, and other key figures of the previous year's selection. Based on the monitoring, the annual report compares actual with desired transition quotas and discusses discrepancies and their possible causes. The report names municipalities that allocate particularly large numbers of children to baccalaureate schools. We interpret this strategy as an act of "naming and shaming" [76]. It puts criticism of a high transition quota and 'in form' and places it on record in the annual monitoring report. The monitoring of the transition and its formatting in an annual report with statistics, key figures, and analyses can be considered a governance strategy based on industrial logics of action, such as plannability, measurability, and scientific expertise. This instrument provides the industrial convention with a broad scope because the monitoring observes various situations of the transition chain, compares them with the desired results, and provides a basis for communication of the transition committee with school headmasters and teachers.
Every year, based on the results of the monitoring, the transition committee sends a letter to all primary school teachers involved in the selection procedure ( Figure 1, box b). It contains recommendations for the next school year's allocation. This letter increases the industrial convention's scope and reaches teachers' workplaces. We assume that this letter has a regulating effect on teachers' practice because it formats the constant monitoring by the transition committee and makes it visible and tangible to teachers.
Furthermore, the transition committee conducts annual instruction events concerning the selection procedure (Figure 1, box d), in which the target transition quota is communicated and visualized. The transition committee shows bar charts in which the range of the target quota is highlighted in blue. This shows the years when the quota was above target. In another chart, values below the target quota are blue, and values above the target quota are red. These charts depict ideas of accepted transition quotas and make them comparable.
The labelling of these events as "instructional" implies that teachers receive practical recommendations. The presentation of target quotas appears as such, just as is the presentation and explanation of evaluation documents for the allocation of the students. It is conceivable that these visualizations and recommendations will influence teachers' selection practice. At instruction events advice is also provided on how to conduct conversations with parents, including how to resolve conflicts and show understanding for parents' feelings. Domestic values of interpersonal exchange and mutual appreciation (see Table 1) are emphasized.
Teachers inform the transition committee of the provisional allocation of students (Figure 1, box e) to tracks at the lower secondary level before the allocation meeting with the parents. If certain schools or municipalities have allocated particularly high numbers of students to baccalaureate school, the transition committee contacts the schools' headmaster and, if necessary, the teachers (Figure 1, box f) to recommend that these be re-evaluated. Whereas this close monitoring and contacting can again be interpreted as a domestic logic of coordination, the close monitoring of transition quotas as a strategy of the industrial convention also motivates this contact.
Finally, the allocation meeting between the teacher and parents (Figure 1 box g) takes place when, based on an overall assessment, a decision is made on the allocation of the child to a performance track at the lower secondary level. The conversation and joint discussion about the child's future points to the values of the domestic convention: according to the transition committee, the meeting is supposed to be a co-operative dialogue in which "a common path is followed" and "a joint decision is reached" (Document A1, anonymized). Interpersonal closeness and exchange, as well as following a common path, are values that point to community as a common good of the domestic convention [77] (p. 416).
In this allocation meeting, the teachers have various instruments at their disposal. One of them is the regulation that students must achieve an average school grade over 5 in certain subjects (with 6 being the highest and 1 the lowest) to be allocated to baccalaureate school. This measure was introduced a few years ago. It can be considered an investment in an industrial form on which teachers base their reasoning during the allocation meeting with the parents. However, the teacher can also allocate students who do not attain an average grade of 5 to baccalaureate school by including social and personal qualities. This can be considered as a compromise between the industrial and the domestic convention.
Standardized student evaluation documents are important forms for teachers and parents at the allocation meeting; it is on their basis that teachers justify students' allocation to lower secondary level. These evaluation documents and the parents' obligation to fill them out may prevent parents from objecting to teachers' decisions. These evaluation documents can also be interpreted as a compromise between the industrial and the domestic convention. In addition to subject-related grades, they also include domestic qualities of students, such as respectful, appreciative behavior and willingness to help others. Consequently, the representative of the transition committee emphasizes that the selection procedure is about assessing the "overall package" of the students and their personalities, which also points to the domestic convention [56] (p. 23). This person denounces selection procedures that rely on overly accurate test-based measurement as the only relevant allocation criterion: In our selection procedure we look at performance, but also at interdisciplinary competence and presumed development of the child. That is all part of the package. And it cannot be that being a baccalaureate school student is dependent on whether their grade is three hundredths higher or lower. I think we lead a school system to the point of absurdity if we claim that a student with a grade of 5.17 does not belong to the baccalaureate school, but one with 5.2 does (representative transition committee, Canton A).
If no agreement can be reached between the teacher and parents at the allocation meeting ( Figure 1, box h), the student can take an aptitude test offered by the transition committee (Figure 1 box j). However, this test is very difficult and is rarely passed (Document A2, anonymized). Because this test evaluates students on subject-related competencies, it can be considered as a format of the industrial convention. The prospect of the test facilitates coordination at the allocation meeting: the parents know about its high level of difficulty (Document A2, anonymized) and thus may be inclined to agree with a teacher's decision. Nevertheless, a compromise with the domestic convention also seems to be at work here: if the result of the aptitude test is not clear cut, the transition committee can decide whether the student can enter a long-term baccalaureate school. This allocation is based on an interview with the parents and their child and on evaluation documents provided by the teacher. In accordance with the domestic convention, debates are centered on what is best for the student's well-being: "We discuss intensively; we give serious thought to those children and try to find the best solution for them" (representative of the transition committee).
Six months after the transition procedure, an official feedback event takes place at baccalaureate schools (Figure 1 box o). The committee presents the transition results and evaluates the previous year's allocation (Document A3, anonymized). Afterwards, baccalaureate school teachers meet individually with the primary school teachers responsible for the previous year's selection and report which students had difficulties with baccalaureate school requirements over the previous six months. We assume that the anticipation of this feedback event has a regulating effect on the primary school teachers' allocation practice. In this situation of the transition chain, industrial and domestic values are again incorporated in a compromise.

Problems, Tensions, and Criticisms: Rising Quotas and Criticism of VET Representatives
The selection procedure in Canton A provokes problems and criticisms that are rooted in cantonal conditions. The compromises described above between the domestic and the industrial conventions lead to close observation of teachers' allocation practices. As a result, according to the representative of the transition committee, teachers evaluate and allocate the students so accurately that most of them can keep up well at baccalaureate school. In recent years, the transition quota has risen slightly, and more students are entering baccalaureate school than are desired by educational policymakers. The reasons for this are the high level of education amongst the canton's population and high immigration of academically qualified families in the past decade, which has led to higher educational aspirations and higher levels of achievement among students ( Table 2). This has led policymakers to criticize the selection procedure and its governance, or lack thereof, even though the quota of students entering baccalaureate school is still noticeably lower than in other cantons with populations of similar educational status. Representatives of VET, trade, and industry argue that they would lose skilled students if too many of them entered long-term baccalaureate school. This prompted the cantonal education minister to reason: "The current governance of the transition is not working as well as we wish. Hands are needed on the steering wheel" (Document A4, anonymized). To gain more control over the quota of baccalaureate students, some actors are currently proposing the reintroduction of an entrance examination; the previous one was abolished 30 years ago (Document A5, anonymized). However, the domestic convention is of great importance in the selection procedure of this canton; consequently, the proposal has currently garnered little support. The representative of the transition committee reacted to these demands by pointing out that establishing and monitoring such a test-based selection would require a large financial investment. However, the canton has been affected by considerable austerity measures (Document A5, anonymized) that have affected the education sector in recent years. This makes political consensus on large financial investments highly unlikely. Additionally, the representative of the transition committee pointed out that the transition quota is regulated to a great extent by the many existing governance instruments (Figure 1).
Furthermore, the representative of the transition committee relied on a scientific study, a format of industrial convention, which proves the satisfaction and "extremely high acceptance" of the selection procedure by all actors involved. This study can be interpreted as another compromise between the domestic and the industrial convention: domestic qualities, such as appreciation and trust, are tested by the industrial format of scientific expertise.

Summary
In Canton A, actors attempt to achieve the target transition quota to baccalaureate school by putting into practice a dispositive consisting of strategies, instruments, and procedures at various stages of the transition chain. In almost all situations, the logics of the domestic and industrial conventions come into play and are interwoven in compromises that stabilize them both. The actors involved are heavily invested in domestic values of interpersonal closeness and community: the transition chain is characterized by various situations in which individual and personal guidance, conversations, and personal feedback are aimed at individual actors: primary school teachers, students, parents, and baccalaureate school teachers. This domestic logic corresponds with the objective of the selection procedure expressed by the actors involved: to prioritize the well-being of the children and to protect them from suffering.
This dispositive of governance is also enabled by instruments and forms of the industrial convention: standardized monitoring and reporting of the transition procedure as well as the standardized aptitude test for those students where no agreement can be reached between teachers and parents based on domestic logics of action. The comprehensive monitoring of the transition procedure and the key figures that give form to data allow the close observation of a teacher's allocation practice through "monitored coordination" [76] and the "naming and shaming" [76] of individual municipalities. We expect strategies, such as visualization, naming and shaming, instruction, and feedback events for primary school teachers, to have a substantial impact on their allocation practice and thus be successful in minimizing the transition rate to baccalaureate school. Because the average education level in the canton is high, in recent years the transition quota to baccalaureate school has risen slowly above the target. This trend has been criticized by figures in VET, trade, and industry, who complain that they are losing talented apprentices. Such criticism puts educational policymakers in Canton A under pressure.

Canton B: Adjustment of the Baccalaureate Quota and Tensions with Equal Opportunities
In Canton B, transition from primary to long-term baccalaureate school has been based on an entrance examination and the score averaged with the student's primary school grades for several decades. In addition, students must successfully pass a probationary period of a few months. As in Canton A, the actors want to ensure that students' educational careers are not disrupted and that a student is "comfortable" in the new school (representative of entrance examination Canton B). These values represent the common good of the domestic convention. However, we show that the domestic convention has considerably less scope and power in Canton B.

Characteristics of the Transition Chain
In contrast to Canton A, Canton B has no official maximum quota for the transition to long-term baccalaureate school. Hence, not all actors interviewed were willing to confirm its existence. However, some of the actors did admit that the transition procedure is adjusted to achieve a politically desired maximum quota: "There's a political will that no more than X% of the cohort should transition to longterm baccalaureate school. I mean, you [the authors] came here and you knew that the quota is adjusted somewhere. Of course, it's adjusted somewhere. So why don't they just publicly say the maximum quota is X%?" (representative entrance examination, Canton B).
Given the challenge of keeping the transition rate at a certain percentage without revealing to the public how it is achieved, we will show along the situations depicted in Figure 2 how the transition chain is put into practice and adjusted to result in a stable annual transition rate to baccalaureate school. We will show how the unofficial quota sought by educational policymakers can be interpreted as a cognitive form that affects the actors. The entrance examination has been centralized and standardized in Canton B for over 10 years. All students take the same examination at the same time in all the canton's baccalaureate schools. Every year, it has the same, high level of difficulty. When writing the exam (Figure 2, box b), the teachers deliberately incorporate leeway in the assessment scale, which later allows grades to be adjusted to scores. In one subject, the examiners are even instructed to preset the average score to fail to avoid high scores that could result in too many successful examinations.
Canton B has a much larger population than Canton A and therefore in absolute numbers many more students interested in entering long-term baccalaureate school (Table 2). Coordinating this centralized and standardized examination is a major challenge in the transition chain and requires major investments in and usage of forms and strategies of the industrial convention: the entrance examination must be safely transmitted (Figure 2, box c) through an internal server on the due date to the responsible actor, who sends it to a printing office. Then, thousands of copies must be printed and delivered under tight security to all baccalaureate schools, which must coordinate hundreds of teachers to supervise and grade the examinations.
After the exam (Figure 2, box d), the scores are determined (box e) and averaged with preliminary grades from primary schools (boxes e, a). Then, the main coordination of the examination results and quota adjustment (box f) takes place, and a statistician calculates the resulting quota. If it is above what is sought by educational policymakers and what is possible with the rooms and teachers available at baccalaureate schools, the evaluation procedure is adjusted.
The person responsible for the entrance examination reminds the subject supervisors and the examiners that if they abide by their assessment scale, the transition quota will be too high. The subject supervisors, in agreement with the assessing teachers, are then free to adjust the assessment scale so that more points are needed to pass the exam. Because the subject supervisors and examining teachers are aware of the target quota and availability of classrooms and usual class sizes, they usually adjust the scale with the leeway introduced during the writing of the exam. Afterwards, the transition quota is recalculated. If it is deemed acceptable, the evaluation is complete, and the results are made public.
Referring to the increasing number of students aspiring to enter baccalaureate school, the representative of the examination procedure legitimizes these adjustments as "the only possibility we have, because actually, at some point, there are only as many places as there are". This justification by seating and classroom capacity with a focus on functional long-term planning indicates industrial logics of action. The involvement of a statistical expert to calculate the quota and adjust the grades until the target quota is reached can also be interpreted as strategies of the industrial convention. Whereas Canton A's procedure rejects the allocation of students based on hundredths of a grade, Canton B's procedure relies on adjustments made within this tiny range. The form of the evaluation leeway and the fine adjustments described above make it possible to keep the transition quota at the level set by educational policymakers.
The second step of the selection procedure is the probationary period at baccalaureate school ( Figure 2, box g). Students who have passed the entrance examination must achieve a certain level of academic performance over a few months of probation. This additional investment in industrial logics by relying on performance grades is not only a test for the students to prove their skills but also a reality test for the entrance examination to see if it did in fact select the best-performing students.

Problems, Tensions, and Criticisms: Inequalities in Access to Baccalaureate School
This selection procedure of an exam, preliminary grades, and a probationary period has repeatedly been criticized based on the civic convention. This criticism is expressed in recurrent parliamentary interventions referring to inequality of access. The government mandated the educational administration, supported by scientific studies and project teams, to evaluate and answer these interventions. As in Canton A, actors in Canton B rely on scientific evidence and thus on a format of the industrial convention to justify the selection procedure. These studies and expert opinions subject the selection procedure to a civic reality test that in some situations leads to adaptations of the procedure and its instruments.
A significant redesign of the selection procedure was initiated at the turn of the millennium. Improving the "quality" of the entrance examination and eliminating regional inequalities in the selection conditions (Document B2, anonymized) relied on strategies of the industrial convention. The examination had hitherto been designed individually by each school; this process was unified, standardized, and centralized, a change that was supposed to achieve "comparability" of results and "reliability" in the assessment of the "talent potential" of students (Document B1, anonymized).
In addition, the examination procedure itself was subjected to a scientific test (Document B3, anonymized). For several years, the entrance examination was supplemented by a second test on a trial basis that analyzed non-subject-related cognitive abilities but had no bearing on entry to a baccalaureate school. In a scientific evaluation, these test results were compared to those of the subject-specific entrance examination results and success in the probationary period. Analysis then determined whether groups identified by such characteristics as gender and native language performed better on this test than on the actual entrance examination. This would point to weaknesses in the examination's ability to select "gifted" students. In this case, incorporation of the second test within the entrance examination could mitigate the influences of origin, gender, and unequal preparation in schools and private preparatory courses (Figure 2, green box, i). However, the results of the comparison were fuzzy and not clear-cut, and the second test was never implemented (Document B11, anonymized).
Representatives in parliament also demanded the introduction of a potential test for all students one year before the entrance examination to identify the truly "gifted" rather than the socially privileged (Document B4, anonymized). Another proposal called for har-monizing exam preparation in the municipalities through state provision (Document B12, anonymized). Both efforts failed to gain political acceptance.
In recent years, the selection procedure has repeatedly been called into question as scientific studies and statistics have shown inequalities of access. Sociospatial inequalities according to place of residence (Document B5, anonymized) and gender inequalities (Document B6, anonymized) have been criticized. One political initiative proposed that children's names be anonymized to counteract unequal treatment according to nationality and gender (Document B13, anonymized). One proposal even called for the abolition of the entrance examination altogether (Document B7, anonymized). The entrance examination was criticized for the short time frame it gives students to prove their potential. "Equality of opportunity" and "exhaustion of reserves of talent" would be better ensured by longer observation and by teacher assessment, according to critics. In response to these demands, the probationary period was extended to six months. However, the entrance examination is so firmly anchored in Canton B's tradition that no majority could be found for its abolition.
The probationary period as part of the transition chain was also put to a reality test by scientific procedures (Document B8, anonymized). Triggered by very unequal dropout rates between baccalaureate schools, political authorities commissioned another study. The analyses indicated a correlation between dropout rates, native language, and social origin, and schools with many foreign-language children from socially disadvantaged milieus were urged to improve their integrative efforts (Document B8, anonymized). Whereas in Canton A the aim of educational policy is to completely avoid dropouts from baccalaureate school, in Canton B they are part of the selection procedure during the probation period, but these should not be excessive or unduly affect socially disadvantaged groups. Additionally, because the quota of baccalaureate students is kept low by the entrance examination, welloff families spend a lot of time and money on private preparatory courses for the exam, thus circumventing education policy efforts to achieve social justice in access to baccalaureate school (Document B9&10, anonymized).

Summary
In Canton B, as in Canton A, actors seem to be concerned with protecting students from disruptions to their educational pathway and suffering; however, the domestic convention is scarcely identifiable in the transition chain. The governance mechanisms, strategies, and investments that are implemented in the transition chain mostly point to the industrial convention. The canton-wide standardized entrance examination for thousands of students results in a testing system that reflects industrial values, such as planning, functionality, measurement accuracy, and performance grades. The probationary period is another supplementary situation in the transition chain that is dominated by the industrial convention. Figure 2 also shows that, compared to Canton A, the transition chain in Canton B includes fewer situations of coordination where the transition rate can be adjusted to achieve the unofficial target quota. In Canton B, the entrance examination keeps the transition rate within the desired range with relatively little effort and high efficiency, even with increasing numbers of applications.
At the same time, educational policymakers rely on civic justifications and demand that the selection procedure be implemented in a manner that ensures equal access. Consequently, the selection procedure has repeatedly been criticized and scrutinized with such industrial formats as scientific studies and supplementary test constructs, thus subjecting the selection procedure to civic reality tests. Although the entrance examination has been adapted several times following these studies, civic criticism continues to be voiced.
As in Canton A, the population in Canton B has a high proportion of academics who are inclined to send their children to baccalaureate school. However, keeping the quota of baccalaureate students low with the entrance examination encourages wealthy aspirational parents to pay for more private tuition and exam preparation courses. Such developments frustrate public policy initiatives to equalize access to baccalaureate school (Document B9&10, anonymized).

Governance of Transition to Baccalaureate School: Summary, Discussion, and Conclusions
Educational transitions have mostly been studied with a focus on the individual that transitions and the factors that influence this transition. What remains largely unexplored is the institutional perspective on transitions that incorporates the governance and institutional configurations that frame transitions. We contributed to closing this gap by investigating how educational provision, rules, processes of selection, and policy discourses influence the transition chain to baccalaureate school and how it is realized and legitimized by representatives of educational administration and policymakers. We focused on the asyet-unaddressed issue of restriction of places and the consequent tensions and justifications in public education. This adds to existing work on governance of educational transitions in Europe [26,37,38].
We studied how two Swiss cantons, one with an official maximum quota and the other with an unofficial one for long-term baccalaureate students, implement and justify their selection procedures of either an allocation recommendation by primary school teachers or an entrance examination. We investigated which instruments, procedures, and strategies were part of the doing of governance to achieve the target quota of long-term baccalaureate students, which justifications underlie the governance mechanisms, and which problems, tensions, criticisms, and policy reactions, rooted in cantonal contexts, resulted from the selection procedures.
Our analyses showed that the selection of candidates for baccalaureate school can be conceptualized as a chain of synchronic and diachronic situations of coordination of action. Following the concepts of the global value chain or statistical chain [51], we speak of a transition chain. This chain produces and performs transitions through the interplay among various actors in politics, administration, and schools; conventions; cognitive formats for official and unofficial quotas and discourses; objects such as statistics and reports; and technologies for monitoring and testing. In various situations along the transition chain, actors rely on plural conventions to underpin their strategies, procedures, and instruments. Therefore, transition chains can be interpreted as historical compromises that address and adapt to emerging and potential future criticisms but are constantly confronted with new tensions. Furthermore, transition chains are trans-sequential in that the situations of the transition do not simply occur one after another but are interconnected and influence each other [65]. Objects, such as statistics and regulations, contribute to this interdependence and to the construction of value, travel between situations, and reshape them [65]. The conventions underlying these situations can remain or change from situation to situation.
We showed that, despite pursuing the same aim of selecting students that can successfully complete baccalaureate school to prevent them from disrupting their educational careers, the relevance, scope, and power of the domestic convention and its governance mechanisms differed substantially between cantonal transition chains. Despite the differing relevance of the domestic convention, neither of the cantons can forgo investment in and usage of strategies, forms, and instruments of the industrial convention-monitoring in case A and testing in case B-to avoid exceeding the maximum quota of baccalaureate students targeted by educational policy. As Diaz-Bone and Horvath state, "it is evident that the industrial convention is of outstanding importance for any modern institution, situation or 'world' in which numerical information is generated and used on a wider basis" [78] (p. 222). Furthermore, the comprehensive monitoring of the transition procedure, the dissemination of key figures, and the close observation of teachers' allocation practice or grading of exams are strategies of the industrial convention and result in a "monitored coordination" [76] which enables processes of "naming and shaming" [76]. This makes it possible to influence the selection procedure and the resulting transition quota. The results further showed that the official or unofficial maximum quotas, in the sense of governance by numbers [79][80][81], have a strong regulating effect on the governance of transitions.
Transition chains and selection procedures can further be considered as dispositives, consisting of actors, form investments, justifications, and technologies that have a regulating effect on the transition quota. Accordingly, following Landri [82], we can state that agency is distributed among actors, technologies, and objects. Doing governance requires the coordination of an "assemblage of humans and non-humans" [82] (p. 607). One of these assemblages is the entrance examination, which we consider as a "technology of government" [83]; the examination regulates the governance of the transition to baccalaureate school and generates data during the selection procedure. We extend Diaz-Bone and Horvath's conception [78] of three data worlds-of official statistics, big data, and civic data-by proposing an additional test-data world. This test-based format of governance has gained tremendous international importance in recent decades, particularly in education [83,84].
Both cantonal transition chains and their respective selection procedures are not equally transparent to the public [66]. In Canton A, the maximum quota desired by educational policymakers is publicly known, documented, and communicated to all actors involved. Thus, primary school teachers must assume that their allocation quota to baccalaureate school will be disclosed and criticized, especially if it is not in the range of the target quota. In line with existing research [39][40][41], we assume that the strategy of "naming and shaming" [76] and thus of being held accountable for allocation decisions exerts pressure on teachers and promotes a kind of self-governance [66] in favor of the target quota. In Canton B, the processes and instruments of the transition chain are opaque, and the quota set by educational policymakers remains unofficial. Because preliminary grades are averaged with the result of the entrance examination, primary school teachers give higher grades to students that are interested in taking the exam so that they have a higher chance of passing (Document B11, anonymized). Thus, in different types of transition chains, the selection procedure has an impact on the assessment practices of primary school teachers.
In this paper, we shed light on institutional factors of educational transitions in Switzerland and showed how actors are doing and justifying governance of the selection to baccalaureate school. The EC perspective, which integrates structure and agency, allowed us to take the agency of actors seriously and investigate their interplay with conventions and policy discourses, institutional regulations, cognitive formats, objects, and technologies. In this context, EC provides an excellent framework to show that selection procedures are characterized by a plurality of justifications, and are not just challenged by, for instance, questions of equal opportunities.
With the concept of transition chains, we offer a way to conceptualize and investigate educational transitions from a governance perspective, which may be fruitful for further research on the topic of governance of selection processes to VET, higher education, and further education. Transition chains can be applied to a variety of national and subnational contexts. For our investigation, certain local conditions proved to be explanatory for the governance of transition chains and the resulting tensions. However, the number of local, regional, and national factors that influence transition chains has yet to be determined empirically and can be extended by further research.

Institutional Review Board Statement:
Ethical review and approval were waived for this study, because it is an organizational analysis based on official documents and interviews with persons holding public positions giving in-formation about political and administrative processes. Since no children, minors or individuals under disability have been included in the study, the research project had not to be subjected to the ethics review of the School of Education FHNW.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The datasets for this study are not available to the public.