Gateways: International Journal of Community Research and Engagement

Vol. 19, No. 2
October 2026

RESEARCH ARTICLE (PEER-REVIEWED)

Adapting the Carnegie Community Engagement Classification in Thailand: Lessons for Institutional Learning and Change

Sorasun Rungsiyanont^1,*, Narin Hiransuthikul ², Aworn O-paspatanakit ³, Nongyao Sripromsuk ³, Apirak Songrak³

¹ Srinakharinwirot University, Secretariat, The Association for University Social Engagement Thailand

² President, The Association for University Social Engagement Thailand

³ Board Committee, The Association for University Social Engagement Thailand

Corresponding author: Sorasun Rungsiyanont, sorasun@g.swu.ac.th

DOI: https://doi.org/10.5130/w1d8k740

Article History: Received 05/06/2026; Revised 30/08/2026; Accepted 09/09/2026; Published 10/2026

Abstract

Across multiple higher education systems, universities are increasingly expected to demonstrate public purpose alongside teaching and research. In Thailand, academic service to society is already a statutory function of higher education institutions. This study examines the early use of the Carnegie Elective Classification for Community Engagement in a pilot involving six Thai universities and considers how the framework supported institutional learning on university–society engagement.

Using a qualitative case study of a bounded developmental pilot, the analysis drew on institutional self-assessments, documentary materials, records from a kick-off event and four 2-day on-site sessions, written reflections and facilitator observations. Materials were organised by institution and source type, and a reflexive thematic analysis examined recurring opportunities, constraints, enabling conditions and differences in interpretation across a deliberately heterogeneous six-university cohort.

The Carnegie Elective Classification for Community Engagement framework was used as a guide. Findings include the following: existing engagement activity was extensive but often dispersed; the Carnegie framework served as a structured diagnostic and reflective tool; senior leadership support and coordinating capacity appeared to facilitate progress; peer exchange supported cross-institutional learning; evidence systems were a recurrent challenge; and the process reframed engagement as an institution-wide capacity rather than a collection of projects.

In this early pilot, the framework’s immediate value was evident in the institutional reflection and learning it prompted, while the possibility of external recognition remained relevant. The findings support contextual adaptation rather than literal transfer, but they do not establish sector-wide change or long-term outcomes.

Keywords

Thailand; Community Engagement; Institutional Learning; Carnegie Classification; Engaged University

Introduction

Internationally, community engagement in higher education is enacted through service-learning, community-based and participatory research, knowledge exchange, public scholarship, outreach and place-based partnerships. These practices sit within differently named traditions of civic engagement, the third mission of university, university social responsibility and engaged scholarship; they are not a single global model (Farnell 2020; Symaco & Tee 2019). Across multiple systems, international organisations, governments, funders and communities have nevertheless placed greater emphasis on universities demonstrating public purpose alongside teaching and research, including contributions to societal wellbeing, sustainability and regional development (Compagnucci & Spigarelli 2020; UNESCO 2022). In Thailand, the issue is not the introduction of a previously absent social role: academic service to society is already a statutory function of higher education institutions. The more specific question is how public-facing work is organised, enacted reciprocally, evidenced and sustained.

The terminology used for university relationships with society is neither uniform nor uncontested (Benneworth et al. 2018). In this article, university–society engagement is used as an umbrella term for relationships between universities and external actors. Community engagement is used in the narrower Carnegie sense: collaboration between higher education institutions and their larger communities for the mutually beneficial exchange of knowledge and resources in a context of partnership and reciprocity (American Council on Education n.d.; Bringle & Hatcher 2002). Outreach and academic service refer here to activities through which university knowledge or services are extended externally; these may be socially valuable but do not necessarily involve co-definition, shared authority or reciprocal knowledge exchange. Community engagement is more likely to be sustained when it is connected to institutional mission, governance, academic work, partnership processes, evidence systems and reward structures rather than depending only on episodic projects or individual commitment (Holland 1997, 2009; Saltmarsh & Johnson 2020).

The Carnegie Elective Classification for Community Engagement (CE), referred to hereafter as the Carnegie CE Classification, is a voluntary institutional self-study and recognition framework first piloted in 2005 and offered in successive classification cycles since 2006. It is one of the Carnegie Elective Classifications administered within the Carnegie Classification system (American Council on Education n.d.). Its influence derives from two connected functions: it recognises demonstrated institution-wide commitment to community engagement, and it requires institutions to examine evidence across mission, leadership, curriculum, partnerships, assessment, infrastructure and reward systems. The framework was designed to promote inquiry, reflection and continuing institutional development rather than to rank institutions (Driscoll 2008; Saltmarsh & Johnson 2020). External recognition and institutional learning should therefore be treated as complementary, although their relative importance may vary by institution and stage of participation.

Cross-national use of a framework developed in United States higher education often requires contextual adaptation rather than literal transfer (Dolowitz & Marsh 2000). The relevant differences should not be collapsed into a single category such as ‘non-Western and middle-income systems’. Cultural and historical contexts can shape meanings of community, reciprocity, authority and public purpose; governance and accountability arrangements can shape institutional decision-making and evidence requirements; and resource conditions can shape administrative and data capacity. International relevance therefore depends on whether the framework’s underlying principles can be interpreted and translated within each higher education system without assuming institutional equivalence.

Thailand provides a distinctive setting for examining this translation. The higher education system includes autonomous and non-autonomous public universities, Rajabhat universities, Rajamangala universities of technology, private institutions and community colleges, with different histories, governance arrangements, resource bases, student profiles and territorial mandates (Crocco & Pitiyanuwat 2023). Academic service is one of the statutory functions of higher education institutions, while the Higher Education Act also requires institutions to enable participation in the development of communities, rural society and disadvantaged groups (Higher Education Act B.E. 2562 2019). Thai scholarship has discussed related work through university social responsibility, service-learning and community-oriented development, while also noting the importance of trust, duration and the integration of social responsibility into core university activity (Denpaiboon et al. 2018; Symaco & Tee 2019).

Institutional differentiation has become more explicit through the Reinventing University agenda and the Ministerial Regulation on the Classification of Higher Education Institutions B.E. 2564 (2021). The regulation permits strategic grouping around global and frontier research; technology and innovation; area-based or other community development; religious wisdom and moral development; professional and specialist workforce development; and other ministerially designated purposes. These profiles do not eliminate universities’ shared statutory obligations, but they change the institutional meaning and relative priority of engagement. A Rajabhat university, whose legal mandate centres local development, and a research-intensive university may both undertake reciprocal community engagement while locating it differently within mission, incentives, evidence systems and academic identity (Rajabhat University Act B.E. 2547 2004). The policy context therefore makes institutional type analytically relevant rather than a background descriptor.

The Carnegie pilot was not designed to persuade every Thai university to seek formal classification or to replace Thai policy with a United States standard. Its dual rationale was narrower: to test whether the framework could support structured institutional self-study in a heterogeneous Thai cohort and to examine what contextual translation would be necessary if individual institutions later chose to pursue recognition. Formal classification remained a possible institutional objective, but it was neither a sector-wide prescription nor an outcome evaluated in this study.

Against this background, the pilot addressed a bounded institutional question: how might six participating Thai universities use the Carnegie CE Classification to examine the coherence, reciprocity, documentation and sustainability of their existing university–society engagement? The framework was approached both as a possible route to recognition for institutions that might later choose it and as a structured opportunity for self-study. The present analysis focuses on the latter because the pilot generated process evidence about reflection and organisational learning, not evidence of completed classification, sector-wide adoption or improved community outcomes.

This article examines the early use of the Carnegie CE Classification in a multi-university pilot involving six Thai institutions, additional observing institutions, national stakeholders and international advisers. It explores how participants interpreted the framework, what institutional opportunities and constraints were documented and what the process suggests about moving from dispersed activity towards more durable institutional support for engagement.

The article argues more cautiously that within this early cohort, the framework’s immediate significance lay in the questions it prompted institutions to ask about the mission, coordination, partnership reciprocity, evidence and continuity. The pilot offers context-dependent insights into adaptation, but it does not establish that institutional redesign or improved community outcomes had occurred.

Conceptual Framework

This study brings three perspectives into dialogue: the engaged university as a normative account of public purpose; the Carnegie CE Classification as an operational framework for institutional self-study and recognition; and the Quadruple Helix scholarship as a contextual account of cross-sector innovation relationships. These perspectives are potentially complementary, but not conceptually interchangeable. The first foregrounds reciprocity and the public purposes of academic work, the second translates institutional commitment into documentary questions and the third maps relations among universities, government, industry and civil society.

The concept of the engaged university provides the normative foundation of the study. It positions public purpose not as an activity added to teaching and research, but as a way of connecting academic knowledge and institutional resources with societal questions through reciprocal relationships (Boyer 1996; Watson et al. 2011). This perspective is not politically neutral. Critical scholarship asks whose knowledge is recognised, who defines the problem, how authority is shared, and whether engagement changes academic practices or merely relabels outreach (Mtawa et al. 2016; Saltmarsh & Hartley 2011). Durable engagement therefore concerns not only institutional structures but also the quality and distribution of power within university-community relationships.

The Carnegie CE Classification provides the operational lens. Its documentation framework asks institutions to examine relationships among leadership, strategic planning, curriculum, partnerships, assessment, infrastructure and academic reward systems. This makes engagement visible as an institution-wide concern and can support collective reflection. A tension nevertheless remains between relational quality and documentary visibility: evidence requirements can illuminate reciprocity and continuity, but they can also privilege what universities are able to record and may encourage compliance-oriented accounts if community voices are not involved. In this study, institutional embedding refers to durable integration across university domains, not to the centralisation or bureaucratisation of community relationships (Holland 1997, 2009; Saltmarsh & Johnson 2020).

Quadruple Helix approaches extend the university–industry–government model by adding the media- and culture-based public or civil society as a fourth sphere in knowledge and innovation systems (Carayannis & Campbell 2009). This perspective helps situate Thai policy emphases on area-based development, social innovation and cross-sector collaboration. It can also sit uneasily beside the engaged-university perspective: an innovation ecosystem may value communities as contributors to economic or policy innovation, whereas community engagement requires attention to mutual benefit, shared authority and community-defined value. Because the pilot did not systematically map all four spheres or their power relations, the Quadruple Helix is used to frame context rather than to generate or test the themes.

Together, the perspectives frame community engagement as a cross-institutional capacity while preserving three analytic tensions: recognition versus learning, auditable evidence versus relational quality and innovation-system priorities versus community-defined public value. International applications demonstrate that these tensions are resolved contextually rather than by simple transfer. The Irish pilot identified the cultural specificity of a United States-derived instrument and used a community of practice to interpret it; the European TEFCE (Towards a European Framework for Community Engagement in Higher Education) work developed a participatory self-reflection toolbox for diverse university settings; and the Canadian pilot adapted both process and language through collaborative work (McIlrath et al. 2021; O’Brien et al. 2022; University of British Columbia 2022). These experiences inform the Thai analysis without presuming equivalent outcomes.

Research Objectives

This study had three interrelated objectives: first, to examine how the Carnegie CE Classification was interpreted and used within a pilot involving six Thai universities; second, to identify the institutional opportunities, constraints and organisational conditions documented during the pilot; and third, to derive practical implications for the continued development of university–society engagement within the participating institutions and the Thai policy context. Possible relevance to other systems is considered cautiously in the Discussion and Implications section rather than treated as an empirical objective.

Study Design and Methodology

This study used a qualitative case study of a bounded developmental pilot to examine the early use of the Carnegie CE Classification across six Thai universities. The immediate impetus was a practical concern shared by participating institutions and national university-engagement stakeholders: substantial public-facing activity was visible, but there was no common process for examining its reciprocity, institutional support, documentation and continuity. The case was the shared pilot process, with each university treated as an embedded institutional setting. The study did not test an intervention or evaluate completed organisational change; it examined a learning process comprising self-assessment, documentary review, peer exchange and reflection on contextual fit.

The core pilot cohort comprised Chulalongkorn University, Chiang Mai University, Srinakharinwirot University, Suranaree University of Technology, Uttaradit Rajabhat University and Rajamangala University of Technology Srivijaya. The six joined as an initial willing cohort for exploratory work; they were not selected through probability sampling or a documented maximum-variation matrix. Additional observing institutions, national policy stakeholders and international advisers participated in selected activities but were not treated as equivalent embedded cases. The absence of a formal comparative selection matrix limits claims about how far the cohort represents the Thai sector.

The cohort was nevertheless institutionally heterogeneous. It included two large comprehensive, research-intensive universities (Chulalongkorn and Chiang Mai); an urban comprehensive university with strong professional-education traditions (Srinakharinwirot); a specialised science and technology university (Suranaree University of Technology); a Rajabhat university with a statutory local-development mandate (Uttaradit Rajabhat); and a Rajamangala university oriented towards applied technology, professional preparation and regional development (Rajamangala University of Technology Srivijaya; Rajabhat University Act B.E. 2547 2004; Rajamangala University of Technology Act B.E. 2548 2005). These descriptors locate different starting mandates; they are not used as rankings or as proxies for engagement performance.

The retained project records place the pilot within 2025–2026 and establish the following sequence: online alignment meetings with representatives associated with the Carnegie process; an on-site kick-off event comprising an expert lecture, a classification-criteria workshop and a joint university session; four subsequent 2-day on-site sessions combining academic input and structured reflection; and institutional synthesis of development priorities. Approximately 30 participants attended each on-site event, principally from the six core universities and member institutions of the national university–community engagement network. Attendance varied by event. The analytic records did not preserve sufficiently consistent calendar dates for every activity; the chronology is therefore reported without assigning unverified month-by-month dates.

At the time of analysis, the pilot activities described above had occurred, while formal Carnegie Classification, implementation of institutional action plans and longer-term evaluation remained outside the completed study stage. National policy stakeholders contributed policy context and participated in selected discussions; international advisers supported interpretation of the Carnegie process; observing institutions joined selected learning activities; and the six core university teams undertook the self-assessment work on which the embedded-case analysis rests. Their involvement is therefore differentiated rather than combined under a single category of participant.

The study was conducted from an insider-facilitator position rather than one of detached observation. Facilitator observations formed part of the corpus, and the organising and analytic work was connected to the developmental initiative. This position provided access to institutional deliberations but also created a risk of interpreting participation as endorsement or development intentions as achieved change. To address that risk, claims were checked across self-assessments, institutional documents, workshop records and written reflections; negative and divergent interpretations were retained; and conclusions were restricted to the recorded pilot process.

The process involved four connected activities. First, each core university undertook self-assessment across broad Carnegie domains, including mission, leadership, curriculum, partnerships, coordinating infrastructure and evidence. Second, institutional teams reviewed strategic plans, policies, reports and engagement records to identify alignment and gaps. Third, workshops and reflective consultations enabled participants to compare interpretations of terms and criteria. Broad support was recorded for reciprocity, mission alignment and stronger evidence, while differences concerned terminology, the feasibility of institution-wide documentation and the mapping of United States-derived criteria onto distributed Thai structures. Fourth, teams identified priorities for continued development and possible future readiness for formal classification. National stakeholders and observing institutions contributed to discussion and peer learning but did not supply equivalent institutional case materials.

The analytic corpus comprised the six institutional self-assessment sets and internal documents used in the pilot, together with records from the kick-off and four 2-day on-site sessions, meeting records, written reflections and facilitator observation notes. These materials were generated for institutional development rather than through a uniform interview protocol. Convergence across source types was used to strengthen interpretation, but document volume and statement frequency were not treated as performance measures. Because records were not consistently verbatim, direct quotations could not be reproduced reliably; instead, the Findings section uses concrete documentary and process evidence and avoids reconstructed participant speech.

The analysis followed a reflexive thematic process informed by Braun and Clarke (2006) and the transparency principles described by Nowell et al. (2017). Materials were organised by institution and source type, read iteratively, and coded for interpretations of the framework, existing arrangements, constraints, enabling conditions, evidence practices, and peer learning. Coding combined deductive attention to Carnegie domains with inductive issues arising from the pilot. Candidate themes were compared across institutions and sources, tested against divergent material, and refined around the distinction between activity and durable institutional support. Participants contributed to interpretation during workshops and through institutional self-assessments and reflections, but the available records do not document participant involvement in formal coding or a separate member-checking exercise. The result is therefore presented as an interpretive thematic synthesis, not as a co-produced analysis, an inter-rater estimate or a prevalence study.

The study sought analytic rather than statistical generalisation (Yin 2018). It does not claim representativeness across Thai universities. Instead, it identifies propositions and implementation questions arising from an early six-university learning process that may be considered in subsequent Thai initiatives and examined, rather than presumed, in other contexts. The pilot activities were undertaken as institutional development, and the analysis reports no identifiable individual or community-partner data. The final submission should nevertheless specify the applicable ethics approval or exemption and the permissions governing research use of workshop and reflective materials – these details were not available in the manuscript reviewed for this revision.

Findings

Across the six participating universities, the Carnegie CE Classification was discussed both as a possible recognition framework and as a structured prompt for self-study. The evidence for this interpretation came from convergence among the six self-assessment sets, internal documentary review, records from the kick-off and four on-site sessions, written reflections, and facilitator observations. These sources provided clearer evidence of diagnostic and learning functions than of external recognition or completed organisational change. Direct quotations are not presented because the records were not consistently verbatim. Six themes emerged from the analysis.

Existing engagement was extensive but institutionally dispersed

Across the six self-assessment sets, institutional documents identified multiple forms of pre-existing public-facing work, including community-based research, professional outreach, health and education services, area-based development, local enterprise support, and student volunteering. Workshop and meeting records repeatedly returned to the difficulty of assembling these dispersed activities into an institution-wide account because faculties, centres, and projects used different records and reporting routes. This triangulation supports the finding of extensive but dispersed activity – it does not establish the quality or reciprocity of every listed initiative.

Institutional mandate altered the context in which this issue arose. For the Rajabhat and Rajamangala settings, local or applied development was closely connected to the formal institutional identity. For comprehensive and specialised universities, public-facing activity sat alongside different research, professional and technological priorities. The records nevertheless did not provide a sufficiently standardised comparison to conclude that one institutional type had stronger reciprocity, better infrastructure or greater readiness. The finding is therefore cross-case but not type-blind: common questions appeared across distinct mandates, while the configuration and priority of those questions remained institution-specific.

The coexistence of extensive activity and limited institution-wide support suggests a strong bottom-up tradition sustained by professional missions, local relationships, committed individuals and project funding. At the same time, reliance on these arrangements made continuity and institutional visibility vulnerable. The study cannot determine whether formal embedding would increase the quantity, quality or community impact of engagement because it includes no comparison institutions or pre–post outcomes. It can conclude only that participants identified coordination, incentives, documentation and continuity as unresolved institutional questions.

The Carnegie framework functioned as an institutional mirror

A second theme concerned the diagnostic role of the framework. Self-assessment shifted attention from counting activities towards examining the arrangements that supported them, including leadership, mission alignment, curriculum, partnership reciprocity, assessment, coordinating infrastructure and academic recognition. The framework did not itself create these arrangements; it made their presence, absence and unevenness more discussable across institutional teams.

The workshop records and written reflections indicated broad support for using the framework to connect previously separate programs within a wider institutional account. The same sources recorded differences in how reciprocity should be evidenced, how much central documentation was feasible and how responsibility should be distributed across faculties and central units. The analysis retained these differences rather than converting them into consensus. Their recurrence across source types supports the interpretation that use of the framework involved negotiation rather than simple adoption.

The pilot therefore suggests that the Carnegie CE Classification can function as a structured reflective tool before formal classification is pursued. This conclusion concerns the questions and comparisons prompted during the pilot; it does not demonstrate that the resulting priorities were subsequently implemented.

Leadership support and coordinating capacity appeared to facilitate progress

Progress reported in institutional reflections varied across the six settings. Workshop and facilitator records associated visible senior sponsorship and internal coordinating capacity with greater ability to convene cross-faculty contributors, locate evidence and connect the work with strategic agendas. No common progress scale or outcome measure was applied, and the small heterogeneous cohort did not support comparison by institutional type. These conditions are therefore interpreted as perceived enabling factors rather than demonstrated causal determinants.

Designated units, experienced coordinators or existing engagement offices helped convene contributors, locate dispersed records and interpret framework criteria. Coordination should not, however, be equated with centralisation. A coordinating structure can connect distributed expertise and consolidate evidence while ownership remains with faculties, academic and professional staff, students and community partners. This hub-and-network interpretation is consistent with distributed leadership research, which describes engaged institutions as coordinating functions across formal and informal sources of expertise (Liang & Sandmann 2015).

Where responsibility was unclear, the work of interpreting criteria and assembling evidence remained dependent on individual champions. The finding is not that decentralisation itself is problematic, but that distributed activity requires an explicit mechanism for connection, accountability and succession. The framework alone could not supply leadership, staff time or data capacity.

Cross-university exchange supported institutional learning

Workshops and shared reflection enabled institutional teams to compare interpretations, discuss obstacles and identify practices that might be adapted locally. The data support describing these exchanges as a learning resource, but not claiming that they accelerated institutional development because no rate of change was measured.

The community-of-practice dynamic was relevant because institutions encountered overlapping concerns about terminology, data systems, staff recognition and relationships with teaching and research. The Canadian pilot similarly used a staged process to test and adapt the United States framework and established a national community of practice (American Council on Education 2024; University of British Columbia 2022). This comparison supports the value of collaborative adaptation as a design principle; it does not establish that peer learning had identical effects in Canada and Thailand.

The Thai pilot therefore indicates that collaborative networks can complement institutional self-study by providing shared interpretive and capacity-building space. Whether networked approaches are more effective than compliance-based approaches requires comparative evaluation and cannot be determined from this pilot alone.

Evidence systems were a recurrent challenge

Evidence generation and data systems recurred across self-assessment, workshop and reflection materials. Teams could identify narratives of partnership activity and social contribution but reported difficulty locating comparable records of outcomes, partnership continuity, resource inputs or activity across units. The convergence of this issue across source types supports describing it as recurrent. Because barriers were not counted through a uniform instrument, it is not described as the most frequent or most important challenge.

This difficulty matters because the Carnegie framework requires evidence not only of activity but of institutional processes, partnership reciprocity, assessment and outcomes. Narrative construction can help institutions assemble a coherent account of distributed work, as documented in the Canadian pilot’s institution-wide data collection and co-construction of narratives (University of British Columbia 2022). In the Thai pilot, however, community partners and students were not systematically recruited as an analytic sample. The study therefore cannot independently verify how they experienced reciprocity or mutual benefit.

The pilot therefore indicates a need for both information infrastructure and human capacity to document, interpret, connect and communicate engagement across dispersed institutional settings. Such capacity should support learning and accountability without reducing reciprocal relationships to centrally reported activity counts.

From project activity to institution-wide learning

Taken together, the themes indicate that the pilot reframed engagement from a collection of worthwhile activities into an institution-wide learning question. Participating teams considered how leadership, incentives, curriculum, partnership practices, evidence and coordination might interact to support continuity. The process did not demonstrate completed system change, but it identified areas in which institutions could direct subsequent work.

Accordingly, the immediate value documented in this six-university pilot lay in structured self-examination and the articulation of development priorities. Recognition remained a legitimate institutional motivation, but evidence of internal redesign, sustainability, and community benefit lies beyond the temporal and evidential scope of the study. The six themes and the types of process evidence supporting them are summarised in Table 1.

**Table 1. Summary of key findings and indicative evidence from the Thai pilot initiative**
Key themes	Indicative evidence from participating universities	Institutional implication
Existing engagement was extensive but institutionally dispersed	Self-assessments and institutional documents described multiple outreach, service, community-based research and area-based initiatives distributed across faculties and units.	Distributed activity requires facilitative coordination, continuity and strategic connection without displacing local ownership.
The Carnegie framework supported structured diagnosis and reflection	Self-assessment and workshop records identified questions about governance, documentation, reciprocity and support systems, alongside differences in interpretation.	Structured reflection can identify development priorities; implementation requires follow-up evidence.
Leadership support and coordinating capacity appeared to facilitate progress	Process records associated executive endorsement and coordinating capacity with participation, evidence gathering and planning; no independent progress measure was used.	Leadership, staff time, and facilitative coordination are plausible enabling conditions rather than demonstrated causal effects.
Cross-university exchange supported institutional learning	Workshop and reflection records documented comparison of interpretations, shared challenges and potentially adaptable practices.	Communities of practice warrant further evaluation as mechanisms for collaborative adaptation and capacity-building.
Evidence systems were a recurrent challenge	Institutional narratives were not always linked to accessible records of partnership continuity, reciprocal outcomes and activity across units.	Information infrastructure and human capacity are needed for learning and accountability, alongside community-partner evidence.
The process reframed engagement as an institution-wide learning question	Self-assessment linked engagement activities with questions about the mission, leadership, curriculum, partnerships, incentives, and evidence.	The pilot identified priorities for continued institutional development but did not evaluate redesign or sustainability outcomes.

Discussion and Implications

The findings indicate that in this early pilot, the Carnegie CE Classification functioned simultaneously as a possible recognition framework and as a developmental instrument. The available process evidence was strongest for its diagnostic role: institutional teams used its questions to identify gaps, clarify responsibilities, examine evidence practices and reconsider relationships between engagement and mission. Recognition should not be dismissed as secondary for all institutions; rather, organisational learning was the aspect most directly observable within the study period.

This distinction matters because extensive activity does not, by itself, establish durable institutional support. Where public-facing work depends on individuals, temporary funding or disconnected administrative arrangements, continuity and institutional learning may remain vulnerable (Holland 1997, 2009). The term institutional embedding is used here to mean integration into mission, governance, academic work, partnership processes, evidence systems and support structures – not confinement within a central unit. A key policy question is therefore whether universities possess the relationships and internal capacity needed to sustain reciprocal engagement strategically.

The findings support viewing community engagement as a distributed organisational capacity rather than as a portfolio owned by one office. Senior endorsement can legitimise the work and allocate resources, while coordinating infrastructure can connect evidence and expertise across units. At the same time, distributed leadership is needed to preserve academic, professional, student and community ownership (Liang & Sandmann 2015). The implication is not that centralisation is inherently beneficial, but that institutional architecture must connect rather than displace locally situated relationships.

A second contribution concerns contextual transfer. The Carnegie framework was developed within the historical, linguistic, governance and accountability traditions of United States higher education. Its institutional unit of analysis, documentary expectations, recognition logic and vocabulary cannot be assumed to map directly onto Thai academic-service traditions or differentiated university mandates. The Irish pilot similarly found that cultural specificity required collective interpretation rather than literal transfer, while European work responded by embedding multiple stakeholders in contextual self-reflection (McIlrath et al. 2021; O’Brien et al. 2022). The Thai pilot adds a setting in which community-oriented obligations already exist in law but do not necessarily use Carnegie terminology or evidence conventions.

Three limitations of direct transfer are particularly important. First, English-language distinctions among service, outreach, engagement and reciprocity do not correspond automatically to Thai organisational categories. Second, the framework may privilege institution-wide documentary capacity, disadvantaging relationship-rich but administratively dispersed work. Third, a university-led self-study can reproduce an institutional viewpoint unless community partners and students participate in defining evidence and judging mutual benefit. These are not reasons to reject the framework; they define conditions under which adaptation must remain critical rather than procedural.

Institutional differentiation further limits uniform policy prescriptions. A common national framework might support shared language and learning, but it should not impose identical structures or indicators on research-intensive, technological, professionally oriented, Rajabhat, Rajamangala, community-college and private institutions. The 2021 strategic grouping policy explicitly recognises differentiated purposes. In this setting, the more defensible policy use of Carnegie-derived questions is as a flexible developmental reference whose application is interpreted against institutional mandate and community context, not as a single national template or ranking mechanism.

Accordingly, the study supports adaptation rather than exact institutional transfer. The relevant question is whether principles such as reciprocity, mission alignment, evidence-informed practice and durable commitment can be translated into locally meaningful structures without erasing differences in mandate, language, or community relationships. The Canadian pilot developed a national adaptation through collaborative work, and the Irish and European experiences likewise treated contextual interpretation and stakeholder participation as substantive design requirements (American Council on Education 2024; McIlrath et al. 2021; O’Brien et al. 2022; University of British Columbia 2022). These comparisons identify recurring transfer questions; they do not establish identical processes or outcomes.

The Thai context also demonstrates why university–society engagement and community engagement should not be used interchangeably. The former serves in this article as the wider umbrella covering academic service, regional innovation, policy contribution, place-based development and cross-sector collaboration. The latter is reserved for relationships characterised by partnership, reciprocity and mutual benefit. Quadruple Helix thinking helps locate universities among government, industry and civil society actors (Carayannis & Campbell 2009), but the present data do not permit evaluation of the quality or outcomes of all such relationships.

Peer networks provided an interpretive space in which institutions could compare approaches and avoid working through every question in isolation. The Canadian pilot likewise combined institutional self-study with a national community of practice (University of British Columbia 2022). These experiences justify further testing of communities of practice and collaborative capacity-building, but the Thai pilot did not compare networked and competitive models or measure whether peer exchange accelerated organisational change.

Evidence systems remained central because institutional narratives and relationship histories were not always linked to accessible records of inputs, partnership continuity, reciprocal outcomes or learning. The self-study process can help create an institution-level narrative that combines quantitative and qualitative evidence, but this benefit should not be assumed from the Thai data without longitudinal follow-up. Evidence work also requires human resources: people must document, interpret, connect and communicate information from dispersed units. The aim should be to make public value more legible within governance while retaining community partners’ accounts of what constituted value.

Limitations and Future Research

Several limitations should be acknowledged. First, the six universities formed a voluntary early cohort and were not selected to represent the Thai higher education sector. Willingness to participate may have made readiness and enabling conditions more visible than resistance. The cohort included different institutional types, but the data were not sufficiently standardised for type-level comparisons. Second, materials were generated primarily by institutional participants during a developmental process. Students and community partners were not systematically sampled, so reciprocity and mutual benefit could not be independently assessed. Third, self-assessment and workshop records may reflect social desirability, uneven participation, resource differences, and variation in documentation capacity. Records were not consistently verbatim, preventing reliable direct quotation, and they did not retain exact dates for every event. Fourth, the insider-facilitator position of the authors offered access but may have influenced interpretation despite cross-source checking and attention to divergent material. Fifth, political and macroenvironmental influences were not systematically investigated. Finally, the early, non-comparative design did not assess sustained organisational change or outcomes for communities, universities, or regions. It cannot establish causal effects, rates of change, or whether institutional embedding improved engagement quality. Longitudinal research should include community partners and students in both interpretation and outcome assessment.

Overall, the six-university pilot suggests that the Carnegie framework can prompt more systematic questions about how engagement is defined, governed, supported, rewarded, evidenced and sustained. The findings do not establish that these questions produced reform or improved community outcomes. They show that a classification framework can also be used as a structured learning process, while recognition remains one legitimate institutional purpose.

Conclusion

This study examined the early use of the Carnegie CE Classification in a pilot involving six Thai universities. The framework served both as a possible route to recognition and as a structured prompt for institutional self-study. Within the study period, the clearest evidence concerned reflection and organisational learning: participants identified questions about leadership, coordination, partnership reciprocity, academic integration and evidence systems. The study did not establish completed institutional change.

The pilot further shows why counts of outreach, service, and locally responsive projects are insufficient to demonstrate community engagement in the Carnegie sense. Reciprocity and mutual benefit concern how relationships are conducted, while institutional embedding concerns whether those relationships are supported through mission, governance, incentives, coordination and evidence practices. Both dimensions require attention. Because community partners and students were not systematically included in the analytic sample, the present study reports institutional interpretations rather than a complete assessment of relationship quality.

More broadly, the study highlights contextual adaptation when internationally recognised frameworks travel across higher education systems. The Thai and documented Canadian pilot processes both treated adaptation as collaborative interpretation rather than literal transfer (American Council on Education 2024; University of British Columbia 2022). Their relevance depends on how reciprocity, mission integration, accountability and public purpose are interpreted within local histories, governance arrangements, resources and institutional cultures.

For policymakers and university leaders, the pilot identifies questions and directions rather than proven effects. Potential priorities include senior sponsorship, facilitative coordinating infrastructure, recognition of engaged academic work, evidence systems combining quantitative and qualitative accounts, systematic inclusion of community-partner perspectives and peer networks for shared learning. These priorities should be adapted to institutional mandate and evaluated. The study does not recommend that every Thai institution seek Carnegie Classification or that a Carnegie-derived framework become a uniform national ranking.

Ultimately, the significance documented in this pilot lies in how the Carnegie framework helped six universities reframe an institutional question: not simply whether public-facing activities existed, but how reciprocal engagement could be connected, supported, evidenced and sustained. The resulting lessons are grounded in an early Thai pilot and should be tested through wider participation, community-partner evidence and longitudinal follow-up before being extended to the sector or to other systems.

Author Statement on the Use of AI

This article was adapted from a research and project grant funding proposal originally developed by the Association. It was then structured into a journal-style format, originally written in Thai. AI was used to translate, edit and refine the text, improving language flow and coherence. The authors have reviewed, edited and approved the final text and accept full responsibility for its content.