Declined quality? A poststructural policy analysis of the ‘quality problem’ in Taiwanese higher education

ABSTRACT

The tension between broadening university capacity and maintaining quality is a global phenomenon. While numerous studies have analysed the enactments of various policies that aimed to address declining university quality, few have examined the underlying beliefs that define what is conceptualised as a ‘quality problem’ within the policy proposals. Utilising Carol Bacchi’s ‘What’s the Problem Represented to be?’ (WPR) approach, this study provides a nuanced analysis of this issue, centring on an underexplored aspect: how the ‘quality’ is made to be a particular kind of problem within a specific policy. Specifically, it scrutinises an uncommon policy in Taiwan – proposing the reduction of university places as a solution to declining quality – to critically analyse what quality is established as ‘reality’ that set limits on how higher education can be thought. A central thesis of this analysis is that higher education system comes into existence through ongoing formation as specific kind of object rather than existing as a fixed entity. Understanding the construction of quality problem and the objectification of higher education requires a critical examination of the underlying assumptions and a detailed mapping of heterogeneous practices that create forms of authority for certain perceptions of ‘quality’ accepted as true.

KEYWORDS:

• Higher education
• quality
• university place
• WPR approach
• poststructural interview analysis
• Taiwan

Introduction

The tension between broadening university capacity and maintaining quality has become a global phenomenon (Altbach, Reisberg, and Rumbley 2019; Dicker et al. 2019; Marginson 2006, 2016; Martin and Stella 2007; Schendel and McCowan 2016). This issue is often seen as ‘an inevitable outcome when HE systems evolve from elite to mass and universal stages’ (Trow 1973, 6–7). It results from a significant increase in student enrolment, including those who may be academically under-prepared, and a consequent dilution of funding for teaching (Altbach, Reisberg, and Rumbley 2019; Marginson 2006). While numerous studies have explored how various HE policies, such as quality assurance framework and regulatory bodies, were enacted to address declining university quality (A. Y. C. Hou 2014; Dicker et al. 2019; Martin and Stella 2007; Mok 2000; Hsu 2018), few have examined the underlying beliefs that define what was conceptualised as a ‘quality problem’ within the policy proposals.

Drawing upon Bacchi’s (2009) ‘What’s the Problem Represented to be?’ (WPR) approach to policy analysis, this paper aims to challenge the conventional belief that the expansion of the HE system inevitably leads to reduced quality. Instead of adopting arguably uncritical views that the admission of academically under-prepared students or dilution of funding for teaching cause a declining quality, this study emphasises how the ‘quality-quantity tension’ is made to be a particular kind of problem within a specific policy (see Bacchi and Goodwin 2016, 17). This way of thinking creates an opening to tease out the ontological presupposition underpinning the represented problems (Bacchi 2009). Since the paper engages with Mol’s (1999) concept of ‘ontological politics’, it sees ‘things’ come into existence through ‘ongoing formation’ rather than being ‘fixed’. The attention is thus directed to the heterogeneous practices involved in the making of HE system, wherein the paper explores what availability of university places is accepted to make high-quality education sayable and what quality is constituted as ‘reality’ that shapes how HE ought to be thought.

Furthermore, this paper offers a novel analytical framework to highlight how governing takes place through problematisation(s) by integrating WPR questions with Poststructural Interview Analysis (PIA). Although the WPR approach has been increasingly applied to critically analyse policies (e.g. Han 2024; Hsu and Xu 2024; Mufic and Fejes 2022; Rahm 2022; Stacey 2017; Stacey et al. 2024; Woo 2023), scant studies using WPR approach have concurrently examined both ‘policy documents’ and ‘interview transcripts’ to probe the ‘truthful things to think and say’ from the ‘texts’. These ‘things said and thought’, as Bacchi and Bonham (2016) argue, are crucial for ‘understanding how provisional subjects, objects, and place are formed and deformed, as well as what gives authority to certain discursive practices’ (p. 118). This conjoined framework provides an important vehicle for interrogating how represented problems have (re)shaped the involved people said about place capacity and university quality, and their policy decisions in the management of university places, designed to either widen or limit access to university.

Taiwan presents a unique context in which to trouble the assumption that the expansion of the HE system inevitably leads to reduced quality through examining an ‘uncommon’ policy. From 1990 to the late 2000s, the number of university places has surged from 625,000 to over 1 million. Rather than being wholeheartedly applauded, this expansion was mismatched with the declining student population caused by an ultra-low birth rate, leading to nearly universal access to HE, with all students being offered places in universities (Chan and Lin 2015; Y. C. Hou et al. 2021; Xu 2022). These concerns were further exacerbated when numerous universities began to ease their admission requirements as a strategy to attract prospective students, deepening doubts about the university quality. Not only did the Ministry of Education (MOE) establish a quality assurance system to uphold educational quality amid the expansion of enrolment capacity (see also A. Y. C. Hou 2014; Hsu 2018), but it also took an unconventional step of reducing the number of university places by imposing the Conditional Standard of Developmental Enrolment and Resources for Tertiary Education (hereafter the Standard) in 2009 (Ministry of Education 2009a, 2009b, 2009c). This move is unusual in the global context of policy practice that public concerns about a deterioration in the quality of universities was met by a government strategy to ration university places.

Utilising the WPR approach, this study aims to investigate four key issues: a) how the problem of declining university quality was constituted by the Standard; b) how this problematisation has come about; c) how the university place and its number were objectified as particular types of problems; and d) particularly in Taiwan, where there is an ongoing struggle to balance university provision and university quality, this study offers a way into identifying the underlying beliefs that constitute what ‘quality’ means within the Standard and what availability of university places was accepted as true to make a high-quality HE system sayable.

In what follows, I provide an overview of the quality issues in HE as a way into describing the expansion of student enrolment in Taiwan, and the resulting tension between university quality and the availability of university places. Following this, I discuss the theoretical and methodological approaches that guided the analysis of the Standard presented here. Utilising Bacchi’s (2009) analytical framework, the subsequent sections examine the problem representation constituted within the Standard, scrutinise the underlying presuppositions, map the power relations through a genealogical approach, and explore the produced effects of problem representation. The paper concludes by presenting additional insights into what kind of HE system is formed within the problematisation in Taiwan.

Quality problems and the HE expansion

The concept of quality in HE and its relationship with the expansion of student enrolment remain elusive despite widespread policy discourse and technology on the topic (Dicker et al. 2019; A. Y. C. Hou 2014; Marginson 2006; Mufic and Fejes 2022; Hsu 2018; Hsu and Xu 2024). Trow’s (1973) elite-mass-universal triptych – ‘elite’ HE systems as enrolling less than 15% of a nation’s given age group; ‘mass’ HE systems enrolling between 15 and 50%; and ‘universal’ systems enrolling over 50%. (p. 6–7) – provides an initial framework to conceptualise this relationship. For Trow, as HE systems expand to meet the growing demand for university education, which many see as a conduit to improved socio-economic status, the tension between quality and quantity inevitably escalates. Historically, university was accessible to a small, elite portion of the population, which allowed institutions to maintain high standards and selectivity. Yet, as access broadened, they faced pressures to accommodate a larger, more diverse student body. In trying to serve more students, institutions may find it difficult to maintain the same level of rigorous academic and administrative oversight, leading to variability in the quality of education across different programs and institutions (Altbach, Reisberg, and Rumbley 2019; Marginson 2016; Trow 1973).

While Trow’s analysis focuses on the prestige associated with elite HE institutions, other scholars have explored how neoliberal ideologies have redefined the dynamics between perceived quality and university capacity (e.g. Boden and Nedeva 2010; Mok 2000). In the context of market-oriented societal shifts, HE expansion has resulted in fiscal deficits due to the dilution of limited public investment across an increasing number of institutions and student cohorts, compounded by precarious and competitive funding mechanisms. Consequently, many universities have opted to lower admission standards or increase student-teacher ratios as tactics to maximise enrolment and reduce teaching costs (Connell 2013), sparking concerns about a potential decline in university quality. Moreover, from a human capital viewpoint, Chan and Lin (2015) argue that the focus in the quality-quantity discourse has transitioned from admission processes to outcomes, such as graduate employability and the alignment of education with labor market demands. The influx of graduates into a saturated job market, often resulting in underemployment or employment in unrelated fields, can undermine the value of HE investments, potentially affecting prospective students’ choices and altering public perceptions of the economic value of university education (Boden and Nedeva 2010). Building on this context, the following section provides an overview of HE expansion in Taiwan which provides important background for understanding the emergence of tension between university quality and the availability of university places.

Taiwan’s HE policies: problematics, problems, and problematisations

Taiwan possesses one of the world’s highest proportions of individuals holding university degrees within its population. According to Trow’s (1973) classification, the HE system in Taiwan can be identified as a ‘universal’ system as, in 2019, approximately 70% of those aged 25–34 had obtained a bachelor’s degree (Ministry of Education 2021). Its recent figures showed a domestic undergraduate enrolment of 1.24 million in a population of 23 million, across 124 universities, comprising 44 publicly and 82 privately-owned institutions (Ministry of Education 2019).

Historically, Taiwan’s earliest universities were small and exclusive, primarily serving young men – and some women – of the affluent classes. In the authoritarian epoch (1949–1987),¹ types of HE institutions were diversified with the establishment of military-specialised colleges, specialised junior colleges, and a few general universities. The general universities, seen as the pinnacle of the HE system, had limited places tightly controlled by the government. As such, university places were oversubscribed, with an excessively competitive university entrance examination, in which the acceptance rate prior to the 1980s was overall lower than 30 per cent (Education Reform Committee 1994a, 1994b).

This highly competitive system led schools to focus heavily on preparing students for university entrance examination, resulting in a widespread practice where thousands of students sought private tutoring after school hours to increase their chances of receiving offers from universities. The system faced criticism from educators, students, and parents alike, culminating in the 410 Educational Movement in 1994, which advocated for massifying the university provision (Law 2002). This movement, under the slogan ‘universal establishment of senior high school and universities (廣設高中大學)’, prompted a significant reform (Chan and Lin 2015; Law 2002; Xu 2022). In 1996, the MOE took decisive actions, including upgrading specialised junior colleges to university level, deregulating the private HE sector, and removing caps on university places, allowing institutions to expand their capacity and increase enrolments (Executive Yuan 1996).

A decade after the reform, the number of universities soared from 23 to 75 and enrolment rose from approximately 626,000 to 944,000 (Ministry of Education 2001). By 2009, this number had further expanded to 147 universities – nearly tripling since the reform began in 1994 (Ministry of Education 2022). As such, university access broadened, particularly for students from historically marginalised backgrounds. For instance, HE enrolment rates for males and females aged 18–21 saw a significant shift: in 1976, about 11% of males were enrolled, compared to less than 9% of females. By 2013, this gap was reversed, with female enrolment at 75% surpassing the male rate of 66% (Kuan 2016).

However, substantial inequities in educational opportunity distribution persist across other demographics. Since most of the new and upgraded institutions are private universities, studies show that students from advantaged backgrounds tend to enrol in prestigious ‘public’ universities, whereas those from disadvantaged backgrounds, including low socioeconomic statuses and Aboriginal communities, often attend lower-ranking ‘private’ institutions (Chang and Lin 2015). Furthermore, dropout and suspension rates at these private universities are considerably higher, due to higher tuition costs and less family financial support (Control Yuan 2022).

Despite these ongoing inequities, the so-called ‘18-score incident’ contingently emerged as a critical point in Taiwan’s evolution of university places. Due to the population decline, the number of 18-year-olds dropped from 414,069 in 1981 to 322,912 in 2009 (National Development Council 2021), creating a surplus in recently-created university places and a new competition. Rather than students competing for university, universities needed to compete for students. In 2007, this shift was underscored when a student was admitted to a private university with just 18 out of a possible 500 points in the annual Advanced Subjects Test,² breaking the record of the lowest HE admission ever. This incident sparked widespread public debate about university admission standards.

Several newspapers connected the ‘18-score incident’ to perceived inadequacies in student readiness for university education, a decline in academic standards, and the 410 Education Reform (United Daily News 2007a, 2007b, 2007c). Criticism culminated in a statement from Wei-Chao Chen, who was at the time President of the National Taiwan University, condemned the 410 Education Reform for allegedly lowering educational quality and contributing to the 18-score incident:

I think [the 18-score incident] came from the widened student participation. The 410 Education Reform is an illusion of education equality. Don’t you think so? The consequence of increasing university provision is the deterioration of university quality.

(United Daily News 2007a)

Suddenly, a ‘problem’ with university provision was widely noticed in Taiwan. When asked about the ‘18-score incident’, former Kuomintang (KMT) presidential candidate Ying-Jeou Ma commented on the ‘the number of university places has been oversupplied in Taiwan’, and promised to prioritise this issue if elected (United Daily News 2007d, A6). Following the 2008 elections, the KMT won a decisive victory over the Democratic Progressive Party. Just a month into office, Education Minister Ray-Chen Cheng was questioned by the Educational and Cultural Committee of the Legislative Yuan about the government’s handling of the surplus of university places. During this session, legislator Chi-Bao Lai demanded immediate action from the Ministry of Education, exclaiming, ‘you need to deal with it! The university quality is dropping! Students don’t even need to study hard because the universities need them rather than students need universities … ’ (Yuan 2008, 142). Despite these discussions, declining enrolment continued to affect all universities. Many of the newer private universities were particularly hard hit, experiencing students shortages and, in some cases, even closures. As a result, many non-prestigious universities prioritised student recruitment, often loosening or entirely removing their academic entry requirements to attract more students.

This short history of HE policies illustrates how university provision is and has historically been shaped, perceived, and identified as specific types of problems. That is, the 410 Education Reform problematised the limited university places as not meeting student demand, proposing the massification as a solution. However, the problem concerning university places has been redefined. It appears that the ‘problem’ has become one of maintaining ‘quality’ amidst demographic changes, with policy solution centred on reducing university provision. While several studies have explored how the widened student participation and demographic change have raised concerns about declining university quality in Taiwan (e.g. A. Y. C. Hou 2014; Chan and Lin 2015; Hsu 2018), there is little understanding of what ‘decreasing quality’ specifically means, or, to put in a different way, what standards are used to define ‘high-quality’ education. Therefore, this paper seeks to address this gap by investigating the problem representation that led to the tension between university quality and the number of university places within the Standard.

Applying Bacchi’s WPR approach

In this analysis, I draw on Carol Bacchi’s (2009) ‘What’s the Problem Represented to be?’ (WPR) approach to policy analysis to explore the constitution of the ‘quality problem’ in Taiwan. The WPR approach has emerged as a robust conceptual tool that can be deployed in the analysis of policy texts, discourses, and processes (Hsu and Xu 2024; Han 2024; Mufic and Fejes 2022; Stacey et al. 2024; Stacey 2017; Rahm 2022; Woo 2023). Unlike conventional policy analysis approaches which proceed on the premise that policies address pre-existing problems, the WPR approach starts from a simple idea: ‘what we propose to do about something indicates what we think needs to change and hence what we think the “problem” is’ (Bacchi and Goodwin 2016, 16). A problem is never seen just as a problem because ‘the way in which the “problem” is represented carries all sorts of implications for how the issue is thought about and for how the people involved are treated, and are evoked to think about themselves’ (Bacchi, 2009, 1; emphasis in original).

From this point of view, the question of ‘how’ policy operates becomes vital to understand how governing is taking place. According to Bacchi (2009, 31), ‘we are governed through problematisations, rather than through policies’, signalling the importance of critically investigating problem representations. Inspired by Foucault’s work on power, knowledge, and discourses, the WPR approach enables analysts to challenge the taken-for-granted status of policy as a mode of governing (Bacchi and Goodwin 2016). That is, according to Bacchi’s (2012), this form of analysis can make visible what is possible to think, write and speak about; and how subjects are made available, shaped and classified within policies.

The WPR approach involves working backwards from policy texts to ‘read off’ the problem representation. In this form of analysis, it is not to try to identify the intentions behind a particular policy or program (Bacchi and Goodwin 2016). Rather, it aims to inquire into their implicit problematisation(s) and examine and reflect critically on the deep-seated assumptions upon which they are based. The analytical framework consists of six questions (adapted from Bacchi 2009, 48)³:

1. What is the ‘problem’ represented to be in a specific policy?

2. What presuppositions or assumptions underlie this representation of the ‘problem’?

3. How has this representation of the ‘problem’ come about?

4. What is left unproblematic in this problem representation? Where are the silences? Can the ‘problem’ be thought about differently?

5. What effects are produced by this representation of the ‘problem’?

6. How/where has this representation of the ‘problem’ been produced, disseminated and defended? How could it be questioned, disrupted and replaced?

In alignment with the WPR approach, the methodological framework is twofold. First, the WPR questions are employed for an in-depth examination of policy documents. Second, for the analysis of interview transcripts, the Poststructural Interview Analysis (PIA) is utilised to seek a nuanced exploration of the narratives and discourses articulated by the interviewees.

The corpus of documents: applying WPR’s questions

In order to critically analyse the 2009 Standard, the forms of questioning and analysis outlined in the WPR approach were applied to a corpus of documents:

the 2009 Standard, its amendments in 2011, 2014, 2015, 2017, and the related archives, such as newspaper articles, speeches, and policy-related notes, provided by the National Central Library, Parliamentary Library, and National Library of Public Information (see Table 1). According to Bacchi (2009), researchers can choose which question(s) to investigate, the order of investigation, or the format of investigation – question-by-question or integrated analysis. While all these questions could productively be applied to illuminate key aspects of the policy documents, in this article I discuss the findings from the WPR analysis in the following order: WPR question 1, 2, 3, and 5.

Table 1. Analysed documents.

Title	Author (Year of Publications)
The Six-year Scheme of Economic Development, vol. 3–4	Council for Economic Planning and Development (1979)
The First-Term Working Paper	Education Reform Committee (1994)
The Second-Term Working Paper
Premier requested the MOE to propose the resolution	Taiwan Apple Daily
The 7^th National Education Conference	Ministry of Education (1994)
The Review Guideline for University Expansion, Modification, and Student Admission	Ministry of Education (2001)
The Conditional Standard of Developmental Enrolments and Resources for Tertiary Education	Ministry of Education (2009a, 2011, 2014–15, 2017)
The Information Sheet for the Conditional Standard of Developmental Enrolments and Resources for Tertiary Education
The Review of the Educational Reform	Control Yuan (2005)
Legislative Yuan Gazette	Legislative Yuan (2005–15)
United Daily News: Education	United Daily News (2006–7)

Widening the ambit of WPR: interviews

This study used what Bacchi (2021) argues a ‘sister strategy’ of WPR approach – Poststructural Interview Analysis (PIA) – to scrutinise what was said in the transcripts of 19 interviews conducted with former education ministers, legislators, senior policy analysts, union officers, and university executives in Taiwan from December 2019 to March 2020. In a PIA account, objects and subjects are not essences but are in ‘ongoing formation’ (Mol 1999). In other words, subjects and objects, as Bacchi and Bonham (2016) argue, are enacted within the interview practices in ongoing formation of subjectification and objectification, and what is said represented the constituted ‘truth’ that shaped how object can be thought. Moreover, interviews were treated as sites within discursive practices, in which each interviewee constituted themselves as particular kinds of subjects with their ways of questioning, feelings, thoughts, and ethics. Simultaneously, I, as a researcher, have also been involved in ontological politics, making different or multiple versions of reality (Bacchi and Bonham 2016). The way I approached the material collected in interviews aimed to question how statements about ‘quality’ are possible for policy actors to make, or contests, or resist, and indicate how discursive practices contributed to the ongoing-formation of ‘HE’ in Taiwan.

Interview participants were selected based on experience in the policymaking processes during the development of the 2009 Standard or its amendment. Participant recruitment was approved by the University of Sydney Human Research Ethics Committee and completed through two approaches. First, numerous policy documents released in Taiwan named key policymakers and stakeholders with their positions and backgrounds. Therefore, I primarily contacted them via email with the outline of my research proposal and Participant Information Statement. Second, my background as an alumnus of one of the universities specialised in educational research and teacher education, as well as previous employment in the government sector enabled me to have a wide range of social networks. I was hence supported by the senior public servants and university executives to contact other potential participants. Moreover, most of the interviewees proactively gave their suggestions regarding other potential participants or helped me to liaise with recommended policy actors. All participants gave informed consent prior to interview, and pseudonyms have been used, in line with the ethics protocol, to protect their anonymity. Table 2 includes an overview of the list of interviewees, their pseudonyms, roles and expertise.

Table 2. List of interview participants.

Pseudonym	Background	Involvement in the Standard
Shih-Fong	Former Education Minister	Implementation at governmental level
Chia-Kang	Former Education Minister	Implementation at governmental level
Hao-Wei	University President	Consultation
Shih-Ming	University executive	Consultation
Zheng-Hong	Former deputy Education Minister	Implementation at governmental level
Xiang-Ying	University executive	Implementation at university level
Shih-Ying	University executive	Implementation at university level
Ya-Ting	University executive	Consultation
Ming-Hong	University executive	Implementation at university level
Zi-Hong	University executive	Implementation at university level
Yu-Rui	Former deputy Education Minister	Implementation at governmental level
Yi-Yang	University vice President	Implementation at university level
Ciao-Deng	University vice President	Implementation at university level
Zi-Wei	Former Minister	Implementation at governmental level
Chi-Hao	Policymaker of the enrolment Standards	Implementation at governmental level
Chi-Wen	Policymaker of the enrolment Standards	Implementation at governmental level
Jing-Ming	Policymaker of the enrolment Standards	Implementation at governmental level
Kai-Siang	University President	Consultation
Xin-Ru	Former chair of a university union	Consultation

Analysing the documents and interview texts

The project generated an extensive volume of text for analysis. My focus was on the content of the texts to map what kinds of ‘object’ and ‘subject’ became possible, and I reflected on how their status can be questioned and disrupted. This means, unlike the conventional document and interview analysis methodologies that aims to explore what is the ‘truth’, the data analysis centres on understanding how ‘”what is said” could be said – how they are considered to be legitimate or ‘truthful’ things to say’ (Bacchi and Goodwin 2016, 116; emphasis in original). Therefore, my attention was directed to the multitudes of practices that render ‘what is said’ as ‘sayable’ and ‘within the true’ (Bonham and Bacchi 2017).

What is the problem represented to be in the 2009 standard?

The first entry for the WPR analysis begins by working backwards from postulated solutions through the policy texts to examine how propose solutions produce the ‘problem’ in policy proposals (Bacchi 2009). According to two key policy documents, the reduction in the number of university places was proposed as a ‘solution’ to the decline in university quality, as the following statements show:

The Standard aims to maintain the basic quality of university education by limiting the university provision, leading to a competitive selection in university examination.

(Ministry of Education 2009b, 1)

The Standard is an attempt to correct the historical policy mistake of [410] education reform that allowed universities to increase their places by imposing a cap on the number of university places.

(Yuan 2013, 110)

Therefore, the problem is represented to be as follows: First, there is an imbalance between the university provision and student population, resulting in a decline in university quality. Second, the 410 Education Reform is positioned as an historical mistake that caused this imbalance. As previously mentioned, the aim of this reform was to transition the HE system from an elite to a mass stage by broadening university capacity, thereby widening the access to HE system, and then reducing the pressure on students during the examination preparation. These two policy problem representations are intertwined within the 2009 Standard, establishing a cause-and-effect relationship: the universal HE system, resulting from the 410 Education Reform, fails to align with the declining student population, ultimately leading to deteriorating university quality as nearly universal HE access is achieved with every student securing a university place.

Assumptions embedded in the problem of ‘deteriorating university quality’

WPR Question 2 asks which presuppositions or assumptions underlie this representation of the problem (Bacchi 2009, 4), in this case that of ‘deteriorating university quality’. Implicit in the statements is an assumption that when the university places are oversupplied, admission selection mechanism fails to filter out unprepared students from the HE system, leading to deteriorating university quality.

The cap on university provision is intended to prevent such incident from recurring… the policy aims to ensure that student selection functions properly. (Legislative Yuan 2009, 232)

This statement implies an assumption that the university quality is intrinsically linked to, and shaped by, the university admission process. It suggests that the number of students receiving offers should be appropriate. Therefore, low-performing students, such as the one with 18 score, ought to be excluded from universities. Jing-Ming, a director of the MOE, shared his thoughts:

I believe that by reducing the university provision, the quality issue could be addressed as students would be required to perform well for their university places in a rebuilt competitive university admission.

(Jing-Ming)

It can be argued that by ‘excluding’ certain sections of student population from universities, the HE system would revert to an elite age, subsequently improving the quality of university education. This policy solution is underpinned by a deep-seated assumption that the problem of declining university quality is rooted in universal access to HE system – place capacity itself is constituted as a problem that affects quality.

How the problem of deteriorating university quality has come to be?

Question 3 in Bacchi’s (2009) framework inquires, ‘How has this representation of the “problem” come about?’ (p. 10; emphasis in original). This question aims to discern how we have got ‘here’ from ‘there’, with ‘here’ (Bacchi 2020) in this stance signalling the dominance of ‘high-quality HE system’ characterised by a ‘low acceptance rate’. The emphasis in this genealogical examination is not on some linear narrative or ‘any search for origins of easily traceable evolution of a policy’ (Bacchi and Goodwin 2016, 22), but on ‘the plurality of factors and influences that happened along the way’ (Bacchi 2020, p. 85). That plurality of factors comprises a discursive practice that constructed by the forms of authority for certain knowledges.

Turning first to the domain of ‘correct the historical mistake of [410] education reform’ (Yuan 2013, 110), it implies that universal university access, which was constituted as a solution in the 410 Education Reform has been re-constituted as a problem that damaged the quality of university education. One key policy document also stated

In the 1990s, the government did not adequately consider how universal access to HE could impact the quality of the HE system. This accessibility allowed under-prepared students to enrol in universities, potentially affecting the overall academic standard (Control Yuan 2005, 14).

It is important to take ‘things said’ (i.e. under-prepared students should be excluded from universities) as a way into inquiring what knowledges exclude the universal access to HE from the regime of truth. Here an interview with a former commission member of the 1994 Civic Educational Reform, and later worked as a university executive at a private university, Xiang-Ying, is useful:

…before the HE system was expanded, the only thing students and their families were taught was to be successful in the university admission … as we all heard the phrase ‘nothing is more honourable than learning’ (萬般皆下品，惟有讀書高), right? I think it shaped our mind from the deep. So teachers and parents told their students and kids, hey! you don’t need to worry about other things but just focus on study, grades, and to be accepted by a university, and then your life would be changed.

(Xiang-Ying)

Rather than explaining why Xiang-Ying said what she said, I focused on how the phrase ‘nothing is more honourable than learning’ has produced what was possible to say about the HE system.

Such phrase is deeply rooted in the Confucian knowledge (see also Yang 2022). For Confucius, the ideal society is ordered by four occupations – scholar-officials, farmers, labours, and merchants, and the first is supposed to know how to self-cultivate and lead the country. People’s rights were defined by their status. Becoming a member of the scholar-officials in Imperial China was very appealing because only scholar-officials were free from tax, labour, and military service. This Confucian mentality was integrated with the Song dynasty’s governing mechanism of Imperial examination – Keju (科舉) – that was the only way of joining the scholar-officials. The pursuit of this prestigious role therefore largely shaped the parent-child relationship, as children were expected to strive for academic excellence for being selected as scholar-officials, thus bringing honour to their family.

‘What is said’ is concerned with discourse as knowledge that shapes a certain number of university places to be thought as true. As Confucian mentality places much emphasis on effort in academic achievement, students are under great pressure to study hard and pursue academic success. Being selected from the university entrance examination remained to shape the ultimate goal of education (Marginson 2011; Yang 2022), and arguably, constitute how HE system ought to be. This powerful Confucian knowledge of ‘nothing is more honourable than learning’ constructed what ‘HE’ was said as true, that is, only highly academically-achieved students can access to universities. As such, when the imbalance between university provision and student population produced a universal HE system, which indicated a malfunctioning student selection – the categorisation of a high quality HE system was challenged.

Turning a genealogical lens onto the other aspect of the 410 Education Reform to show how the idea of ‘universal access to HE’ contingently emerged. My exploration commences with a key document – the First Working Paper of the Education Reform Committee, as it elucidates how the universal access to HE was understood as in the true.

The purpose of education reform is not only about the examination-centric education, which is underpinned by the Confucian tradition, but also such system represents a political control … school and HE systems were predominantly used as an instrument for cultivating political loyalty (Education Reform Committee 1994a, 17).

Indeed, prior to the end of forty-year-KMT-rule regime, competitive university admission was the effective governing apparatus that shaped how people thought about the access to HE (see also Law 2002). Only limited sections of population who ‘studied hard’ and ‘were competitive’ deserved university education.

The 7th National Education Conference of 1994 can be seen as a significant turning point, as highlighted in its policy statement:

The education system was too obsolete to effectively equip our population with work-related skills. It is imperative for the government to create more opportunities for our young generation, not only to relieve their stress from the examination but also, more importantly, to ensure they are highly skilled and productive for our economic and national development.

(Ministry of Education 1994, 60)

This statement underscores the belief that university degree equalled higher-level skills and was recognised as a means to boost individual and national productivity. It demonstrates that the linkage between individuals and the job market was established by the ‘university degree’, wherein degree holders were formed to be more skilled and productive. In this regard, HE system was objectified as a provider of professionally productive works, rather than reserving for those who performed well in the examination.

Effects of ‘deteriorating university quality’ representations

Keeping the genealogical examinations in mind, the study shifts to explore the effects that follow from particular problem representation (Question 5) that specifically focuses on ‘discursive effects’.

According to Bacchi (2009), the notion of discursive effects refers to how the terms of reference established by a particular problem representation sets limits on what can be thought and was said. In line with this theorising, I focus on the represented problem ‘deteriorated university quality’, as a form of object, and interrogate how this objectification works to make the notion of the exclusive HE system into a regime of truth within the Standard. As Foucault (1977) argues, a regime of truth refers to the mechanisms which enable one to distinguish true from false statements, as well as a system of ordered procedures for the production, regulation, circulation and functioning of true statements. In this case, the notion of ‘university quality’ was particularly perceived in conjunction with the apparatus of ‘university entrance examination’, reflecting the multitudes of practices (e.g. processes, procedures, and apparatuses) involved in the production of ‘truth’ (Bacchi and Bonham 2016; Bonham and Bacchi 2017). Put simply, the university entrance examination – a mechanism essentially decides the student population – produced the limit on ‘what can be said’ about high-quality education.

In one interview, when I asked how the university entrance examination influenced university quality, the former Education Minister Shih-Fong responded

The acceptance rate was approximately 20 per cent in the 1970s and 1980s … since the 1990s, it was up due to the surge in the new establishments of universities or college upgrade. And now the score does not matter for students if all they want is just to study at any university, right? Because there is an imbalance [between university provision and student population] … Then what we see is all students are admitted by universities … I would say they did not prepare enough. Then the quality is so poor, at least compared with my age. (Shih-Fong)

Shih-Fong’s interview transcript shows an ethical dimension at play: the more students admitted into the HE system, the more likely the quality of university education would suffer. This perspective on the acceptance rate was not solely based on the proportion of applicants who failed the admission examination, but rather on the notion that the rate itself represented a measure of the system’s competitiveness and, by extension, its quality. It enabled everyone to judge whether the HE system was competitive or not because the rate embodied ‘quality’ in a quantitative and measurable form. As the acceptance rate increased, it signalled that the HE system was becoming less exclusive and, therefore, of lower quality.

Moreover, a set of discursive practices creates the forms of authority for certain knowledge, which make ‘object’ understood in a specific way (Bacchi and Goodwin 2016). Two materials are useful for understanding how the idea of a massified HE system was problematised: one is a public speech given by the premier of the Executive Yuan Chao-Shiuan Liu when the 18-score incident occurred, and the other is an interview transcript of Shih-Ming, who was a director of the student admission office at a top university.

Liu: we can see now is that all the students whatever their readiness for university education are all accepted by our HE system … and I think we have the responsibility to ensure our HE system can provide high-quality education and training (Taiwan Apple Daily 2008).

Shih-Ming: nowadays it is abnormal if a student did not receive any place … and I am not satisfied with this system as university is not a school for everyone (Shih-Ming).

Liu’s phrase ‘all the students whatever their readiness for university education are all accepted’ reflects a believed truth that universal access to HE means low student readiness for university education. Simultaneously, Shih-Ming’s use of ‘abnormal’, shows that ‘universal access to HE’ was set against the constituted truth that the HE system ought to be selective. Such a regime of truth arranged the ‘student population’ into two categories: those who are well-prepared for university education, and those who are unprepared for university education as determined by the examination; it also objectified the nature of HE system in Taiwan through a set of discursive practices that created a binary between high-quality and declining-quality HE systems. Arguably, the high acceptance rate was thought abnormal because it was incompatible with the notion of high-quality HE system.

It is important to highlight that policy and its constituted problem(s) are made in power relations (Bacchi 2009). As Taylor (2014) argues, power relations do not make everything subservient, but create a field that allows power to be exercised ‘in the very interplay of force and resistance’ within the network (p. 24). The following cases demonstrate the discursive practice of resistance in response to claims that universal access to HE was said to deteriorate the university quality.

University programs should be designed to cater to national development, particularly in meeting workforce demands. In a globally competitive market, our economy depends on highly-skilled graduates (Yuan 2013, 408).

When discussing the reduction of university places across the higher education system, I think it’s important to consider how university education contributes to our society… I don’t believe that ‘college students are everywhere’ (大學生到處都是) is a negative thing because it means our students are aiming to achieve higher employability and our population is well-competitive compared with other countries. (Yu-Rui)

The belief that university degree can boost individual productivity and national competitiveness represents the ‘processes implemented for conducting others’ (Bacchi and Goodwin 2016, 31). As a university president, Kai-Siang, noted:

Our main goal is to equip our students for success in securing high-paying jobs. This focus is irrespective of their past academic performance … We focus on aligning our recruitment strategies with the current demands of the job market, rather than solely prioritizing performance in university entrance exams. So, test performance is not our main consideration.

(Kai-Siang)

Clearly, the notion of ‘university education for economic purposes’ represents a counter-discourse. This perspective redefines the role of the HE system. Rather than perpetuating traditional divisions between scholar-officials and other classes, the focus shifts towards fuelling economic growth (Connell 2013). University education is no longer just about academic distinction; it can be seen as a pathway for students to develop into highly-skilled graduates, ready for the job market. This shift has also established a vital connection between universities and economic advancement, emphasising the role of universities in producing job-ready, skilled professionals (see also Boden and Nedeva 2010). Consequently, this has led to a normalisation, where the measure of educational quality is increasingly linked to how well university education prepares students for future careers and aligns with job market demands.

Discussion

The tension between broadening university capacity and maintaining quality has become a pressing issue in many HE systems worldwide (Hsu 2018; Martin and Stella 2007). This tension is often seen as an inevitable outcome when HE systems evolve from elite to mass and universal stages (Trow 1973). It results from a significant increase in student enrolment, including those who may be academically under-prepared, and a consequent dilution of funding for teaching, especially under the market-oriented agenda of social transformation (Altbach, Reisberg, and Rumbley 2019; Boden and Nedeva 2010). However, defining ‘decreasing quality’ is challenging due to its amorphous nature, making it difficult to set a clear standard for what constitutes low quality. By examining the Standards designed to address the declining university quality, this study has offered a way into identifying what underlying beliefs that define what ‘higher quality’ university education was accepted as true in the condition of a specific number of university places.

As specified above, the universal access to HE was thought to be a result of a decline in university quality, revealing that broadened university capacity and high quality were constructed as binary, and opposite, concepts within the problem representation of the Standard. With the emphasis of WPR’s Questions 1 and 2, the analysis has shown that the universal access to HE was constituted as a quality problem under the condition of ‘acceptance rate of university admission’. While the postulated proposal was to reduce the number of university places, the notion of quality university education was embedded in the calculation of how many students were offered a places through the entrance examination. The near-universal ability of students to receive a university place was problematised, highlighting that universal access to HE – as more likely to admit under-prepared students – was perceived as an ‘abnormal HE system’. This led to the ‘true statement’ that university students should be academically high-achievers and universities should serve as learning sites for a limited, well-prepared population who succeeded in the selection process.

Through Questions 3 and 5, the analysis has also mapped the heterogeneous procedures and practices at work in the production of the HE system in Taiwan. Two different kinds of HE system were formed through the enactment of relations within discursive practice regarding the number of university places. On the one side, the HE system was thought of as an educational stage that only provided limited opportunities and remained exclusive, where students must study diligently and outperform the majority of their peers in examinations. This type of HE system was accepted as a true educational stage when only a certain proportion of students received university offers. In this regard, the low acceptance rate operated as a technique to make ‘high quality HE system’ sayable and thinkable. On the other side lies the argument that university capacity should be widened to serve the economic demands, emphasising that a population can only become ‘highly-skilled’ primarily through university education. In the 410 Educational Reform, the limited proportion of the population with access to universities was problematised, as such type of HE system was perceived as a barrier to economic growth. In this regime of truth, quality university education was not defined by the low acceptance rate. Instead, it was constituted by the ability to enhance individual employability and the collective productivity for economic competitiveness, and most importantly, the HE system must be capable of accommodating as many students as possible.

As a final point, it is critical to argue that the elite/mass/universal HE system should not be viewed as a ‘naturalised’ linear progression, but rather as an object under ‘ongoing formation’ (see also Bonham and Bacchi 2017; Mol 1999). Conventionally, elite/mass/universal HE system is understood as a progressive evolution (Altbach, Reisberg, and Rumbley 2019; Trow 1973), and is characterised by an assumption that elite HE equals ‘high quality’ due to its ‘competitive acceptance rate’ (Marginson 2011; Yang 2022). As the transition from elite to mass and universal stages occurs, the quality issue inevitably emerges, owing to the acceptance of academically under-prepared students. However, this study challenges this taken-for-granted assumption through a form of ‘deconstructing’ analysis (Bacchi and Goodwin 2016), suggesting that HE system should be perceived as emergent and continuously shaped by and in ongoing interactions with discourses and practices. Hence, the relationship between quality and capacity is not fixed, but is constituted by the encounter of different networks of relations. This study has clearly shown that two discourses – ‘nothing is more honourable’ and ‘university education serves for economic growth’ – have coexisted in defining what ‘numbers of university places’ was accepted as true. Within this, relations of contestation and resistance reject each other, yet do not eliminate one another. In other words, different kinds of HE system, underpinned by either a Confucian-oriented or an economic-driven mentality, remain within the network of practices. Their struggle and contest form a web of power relations that are categorised as different provisional objects of HE system and have produced what is accepted as ‘high quality’ university education by imposing a law of truth on the involved people, enabling them to recognise and attach themselves to the objectified university system in Taiwan.

Conclusion

Analyses such as the one presented in this article are crucial as they show how policy responses to problems serve not only as solutions but also as mechanisms for governing the involved people through problematisation (e.g. Rahm 2022; Stacey et al. 2024; Woo 2023). Unlike conventional research that primarily focuses on the effectiveness of ‘solutions’ in addressing ‘quality problems’ (e.g. A. Y. C. Hou 2014; Schendel and McCowan 2016; Hsu 2018), this study challenges the taken-for-granted assumption that the expansion of the HE system inevitably reduces the educational quality. Crucially, it has revealed that problem representation constructs the HE system as a specific kind of object for thought, sets limits on how quality is considered as truthful statements, and influences the ways governing takes place.

This way of analysis represents a significant contribution by demonstrating that quality-related HE policies, which at first glance seem designed to enhance or maintain university standards, produce unexpected discursive ‘effects’ on subjects, such as, categorising students as either well-prepared or underprepared, or between high and low employability (e.g. Boden and Nedeva 2010). Through a genealogical examination, this paper has also highlighted that how the involved people said about the quality problem and what was accepted as good-quality HE system are not fixed but have evolved over time. For instance, the universal HE access underpinned by an economic growth-focused mentality, which was once considered a truthful statement where ‘high-quality’ predominantly revolved around the ideas of economic growth, was later problematised within the Standard. Last, analysis tools of the WPR questions and PIA have provided the field of HE studies with open ontological perspectives: Rather than perceiving HE system as a static entity defined solely by enrolment numbers (Altbach, Reisberg, and Rumbley 2019; Marginson 2006, 2016; Trow 1973), it is important to view HE system as a provisional, and often contested, object that comes into existence through ongoing formation, thereby exploring how ‘reality’, such as declined quality, is made in and through discursive practices.

Ethical approval

The study was approved by the University of Sydney Human Research Ethics Committee.

Acknowledgments

The author would like to thank the interviewees who generously dedicated their time to participate in this project. I am grateful to the reviewers for their insightful comments on earlier drafts of this article. My sincere thanks go to Professor Nicole Mockler, Professor Susan Goodwin, and Dr. Matthew A.M. Thomas for their helpful feedback and encouragement.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Ren-Hao Xu

Ren-Hao Xu is a Lecturer at the Graduate School of Education, the University of Western Australia. He is a policy sociologist with research interests in higher education, international and comparative education, and governance. Ren-Hao’s research is published in the Journal of Education Policy, Higher Education Policy, Using Social Theory in Higher Education (Springer Nature, 2024), and Australian Universities (Sydney University Press, 2022).

Notes

1. The KMT authoritarian epoch specifically refers to the over 38-year-long consecutive martial law period between 20 May 1949 and 14 July 1987 (戒嚴時期).

2. Advanced Subjects Test was one of two major university entrance examination in Taiwan from 2002 to 2021. It consisted of 11 subjects with each full score of 100. A final-year high school student can select between 5 and 6 subjects during the examination as their portfolios for university applications.

3. Bacchi’s (2009) WPR approach is subtly distinct from Bacchi and Goodwin’s (2016)version. The 2009 framework comprises six interrelated questions, while the 2016 version includes these six questions along with an additional undertaking. Furthermore, the descriptions of the six questions slightly differ between the two versions.