Academy of Education Journal

Vol. 17, No. 1, July 2026, Page: 405-413

ISSN: 1907-2341 (Print), ISSN: 2685-4031 (Online)

405

Elwin P. Zebua (A Discourse Analysis Of Students' Metaphorical....)

A Discourse Analysis Of Students' Metaphorical

Language Comprehension In Interpretative Listening

At The English Education Study Program Of

Universitas Nias

Elwin P. Zebua

English Education Study Program, Faculty of Teacher Training and Education, Universitas Nias (UNIAS),

Gunungsitoli, North Sumatra, Indonesia

Corresponding Author: e[email protected]

INFO ARTICLE

ABSTRACT

Article History:

Received: 28 May 2026

Revised: 19 June 2026

Accepted: 10 July 2026

Available Online: 28 July 2026

Interpretative listening asks learners to move beyond decoding surface

acoustic input and to reconstruct meaning that a speaker leaves implicit, a

demand that intensifies considerably when spoken discourse is organised

around metaphor. This study offers a discourse-analytic account of the

recurrent errors that second-semester students of the English Education

Study Program at Universitas Nias (UNIAS) produce when inferring

implicated, non-literal meaning during interpretative listening. Framed

within a qualitative descriptive design (Creswell and Poth, 2018), the study

was conducted from March to May 2026 with the entire cohort of 43 second-

semester students enrolled in the 2025/2026 academic year, engaged

through total sampling. Three complementary instruments generated the

data set: a Listening Comprehension Worksheet cross-referenced with an

Audio Transcript Matrix, Think-Aloud Protocols (TAPs) capturing real-time

inferential reasoning, and a Semi-Structured Interview Guide eliciting

retrospective metacognitive reflection. The three data sets were analysed

through discourse-analytic coding informed by relevance-theoretic and

conceptual-metaphor perspectives, and triangulated across instruments to

strengthen credibility. Six recurrent categories of inferencing error emerged:

literal-meaning fixation, source-domain mismapping, discourse-context

neglect, prosodic and paralinguistic cue oversight, first-language schema

overgeneralisation, and processing-load-induced shallow inferencing. These

patterns are traced to learners' as yet underdeveloped capacity to construct

the ad hoc, context-adjusted concepts that relevance theory holds to be

central to metaphor resolution, compounded by conceptual-metaphor

schemas transferred from Indonesian and Nias. The study closes with a

proposal to embed explicit metaphor-awareness instruction and reflective,

TAP-based listening tasks into the Course Learning Outcomes of

interpretative listening subjects under the Outcome-Based Education (OBE)

curriculum, so that listening pedagogy is more closely aligned with the

inferential demands of authentic spoken discourse.

Keywords:

interpretative listening;

metaphor comprehension;

inferencing errors;

discourse analysis;

EFL;

Outcome-Based Education

This is an open access article under CC BY-SA license

Academy of Education Journal

Vol. 17, No. 1, July 2026, Page: 405-413

ISSN: 1907-2341 (Print), ISSN: 2685-4031 (Online)

406

Elwin P. Zebua (A Discourse Analysis Of Students' Metaphorical....)

1. Introduction

Interpretative listening occupies a distinct position among second-language listening sub-

skills: it requires learners not merely to decode acoustic input but to reconstruct meaning that a

speaker has left implicit, presupposed, or figuratively encoded. Unlike literal comprehension,

which can be checked against a transcript, interpretative listening rests on inferencing — the

cognitively demanding process of deriving unstated propositional content from contextual,

pragmatic, and discourse-level cues (Goh & Vandergrift, 2021). This demand grows sharply

when spoken discourse is structured around metaphor, since figurative utterances routinely

require listeners to suppress a literal reading and construct, in real time, an alternative and

context-sensitive interpretation (Littlemore, 2019).

Metaphor is not a peripheral stylistic ornament in spoken discourse; it is a pervasive

organising device through which speakers structure abstract reasoning, evaluate experience,

and signal stance (Gibbs, 2017). For second-language listeners, however, metaphor

comprehension is disproportionately vulnerable to breakdown, since learners must

simultaneously process unfamiliar lexical and prosodic input, detect a shift away from literal

meaning, and map source-domain vocabulary onto an appropriate target-domain interpretation,

all within the constraints of real-time processing (Ifantidou & Plata, 2021). Where first-

language conceptual schemas diverge from those instantiated in English discourse, the

inferencing task becomes still more precarious. Despite the growing theoretical sophistication

of metaphor and inferencing research, comparatively little discourse-analytic work has

examined, in a granular and process-sensitive manner, precisely how and where Indonesian

EFL listeners' inferencing breaks down when confronted with metaphorical spoken input.

This gap carries direct pedagogical consequences for English Department curricula, where

interpretative listening courses frequently privilege literal comprehension checks over

diagnostic attention to the inferencing process itself. Without a discourse-analytic account of

the specific error types learners produce, listening instruction risks remaining generic,

targeting comprehension outcomes while leaving the underlying inferential mechanisms that

generate miscomprehension unaddressed. This study therefore analyses the common errors

made by second-semester English Education students at Universitas Nias (UNIAS) when

inferring unstated, metaphorically encoded meaning during interpretative listening, with the

aim of (1) identifying and categorising recurrent inferencing-error types, (2) explaining these

errors through conceptual-metaphor and relevance-theoretic frameworks, and (3) formulating

recommendations for embedding metaphor-aware inferencing instruction into the Outcome-

Based Education (OBE) curriculum governing listening courses at UNIAS.

The study is significant on three fronts. Theoretically, it contributes discourse-analytic,

process-oriented evidence to the still-developing literature on L2 metaphor comprehension.

Methodologically, it demonstrates how transcript-matrix worksheets, think-aloud protocols,

and retrospective interviews can be triangulated to expose inferencing processes that product-

only comprehension tests cannot reveal. Practically, it equips listening instructors with an

empirically grounded error taxonomy capable of directly informing course learning outcomes,

materials design, and formative assessment within an OBE framework.

Literature Review

Interpretative Listening and Inferencing in L2

Within Goh and Vandergrift's (2021) metacognitive account, inferencing occupies a

distinctly higher-order tier: it draws on, yet is irreducible to, the phonological decoding and

schema activation that underpin more basic listening operations, requiring learners to weld

fragmentary acoustic input to contextual and world knowledge in order to reconstruct what a

speaker means but does not overtly state. Aryadoust (2019, 2022) formalises this layered view

by distinguishing pre-comprehension, comprehension, and post-comprehension mechanisms in

Academy of Education Journal

Vol. 17, No. 1, July 2026, Page: 405-413

ISSN: 1907-2341 (Print), ISSN: 2685-4031 (Online)

407

Elwin P. Zebua (A Discourse Analysis Of Students' Metaphorical....)

L2 listening, situating inferencing within the post-comprehension stage, where listeners

connect propositions into a global mental representation through predominantly top-down

processing. Because interpretative tasks withhold explicit propositional content by design,

breakdowns at this stage are easily folded indiscriminately into a diffuse verdict of "poor

comprehension" rather than traced to their true locus in the inferencing process.

This imprecision obscures how inferencing performance is shaped by an interlocking set

of learner-internal factors. Bozorgian, Muhammadpour, and Mahmoudi (2022) show that

embedding metacognitive strategy instruction within listening practice measurably strengthens

learners' capacity to plan, monitor, and evaluate their own comprehension processes, while

Tanewong's (2018) comparative study of less-skilled EFL listeners over a full semester

similarly demonstrates that a structured metacognitive pedagogical sequence yields

measurable gains in listeners' problem-solving and inferencing behaviour — evidence that

positions inferencing as a trainable skill rather than a fixed aptitude. This layered

understanding carries direct methodological consequence: comprehension scores alone are too

coarse an instrument to reveal precisely where, within the unfolding of an utterance, an

inference goes astray, which is why process-sensitive data capable of tracking meaning

construction in real time are required (Aryadoust, 2019) — a methodological orientation that

underpins the transcript-matrix and think-aloud procedures adopted in this study.

Metaphor Comprehension: Conceptual and Relevance-Theoretic Perspectives

Gibbs's (2017) embodied elaboration of conceptual metaphor theory holds that

metaphorical understanding is grounded in recurring patterns of bodily and experiential

structure, such that comprehending a metaphor involves activating an entire conceptual

domain rather than decoding an isolated figure of speech. From a complementary pragmatic

angle, relevance theory argues that metaphorical utterances are understood through the same

inferential mechanism that governs loose or approximate language use more generally:

listeners adjust the meaning of a lexically encoded sense and broaden it into an ad hoc,

occasion-specific concept that satisfies their expectation of optimal relevance, rather than

retrieving a fixed figurative meaning from memory (Wilson & Carston, 2007). Ifantidou and

Plata (2021) extend this account to second-language use specifically, characterising metaphor

as a kind of mental shortcut that L2 learners rely on to circumvent the semantic constraints

imposed by a limited target-language vocabulary — a description strikingly consistent with the

shallow, effort-minimising inferencing observed among the participants of the present study.

Individual variation compounds this picture. Littlemore's (2019) research demonstrates

that L2 listeners differ substantially in metaphor-processing efficiency depending on working-

memory capacity, exposure, and the degree of overlap between first- and second-language

conceptual systems — variation that is highly consequential for real-time interpretative

listening, where processing time cannot be recovered once an utterance has passed.

Discourse-Analytic Approaches to Spoken Metaphor

Gee's (2018) discourse-analytic framework foregrounds situated meaning and cultural

models, treating utterances as interpretable only against the discourse and cultural context in

which they are embedded — an orientation directly applicable to metaphor, whose resolution

likewise depends on co-textual and situational cues rather than on lexical decoding alone.

Methodologically, the metaphor-identification procedures consolidated in Nacey, Dorst,

Krennmayr, and Reijnierse (2019) provide a systematic basis for locating and classifying

metaphorical expressions within authentic spoken transcripts, enabling the present study to

align specific inferencing errors with precisely identified metaphorical units in the discourse

rather than treating figurative language impressionistically.

Academy of Education Journal

Vol. 17, No. 1, July 2026, Page: 405-413

ISSN: 1907-2341 (Print), ISSN: 2685-4031 (Online)

408

Elwin P. Zebua (A Discourse Analysis Of Students' Metaphorical....)

Errors and Breakdown in L2 Metaphor Comprehension

Empirical evidence on L2 figurative comprehension indicates that miscomprehension

rarely stems from a single source. Littlemore, Krennmayr, Turner, and Turner's (2011) corpus-

based investigation of second-language writing found that a substantial proportion of

metaphorical expressions readily understood by native speakers proved genuinely difficult for

second-language users, a gap subsequently confirmed across multiple experimental paradigms

in the broader field, as documented in a recent bibliometric synthesis of five decades of

metaphor-processing research (Yang et al., 2023, Humanities and Social Sciences

Communications). This body of work implies that untrained inferencing, left to default

processing strategies, is systematically prone to error unless learners receive explicit attention

to conceptual domains and cross-linguistic contrast.

Outcome-Based Education (OBE) and Constructive Alignment

Biggs and Tang's (2015) principle of constructive alignment holds that intended learning

outcomes, teaching-learning activities, and assessment tasks must be deliberately interlocked

so that what is taught and assessed directly serves the stated outcome — a principle that now

underlies Outcome-Based Education as implemented across Indonesian higher education

through AUN-QA (2020) quality-assurance standards and the Merdeka Belajar-Kampus

Merdeka policy framework (Kemendikbud, 2020). Recent studies of English-language

programmes in Indonesia confirm that constructive alignment remains unevenly realised in

practice: Wahyuni, Fitriyah, and Hasanah (2023) report that the implementation of Merdeka-

Belajar-based curricula at English departments continues to privilege broad comprehension

descriptors over sub-skill-specific outcomes, a pattern echoed in a recent review of OBE

alignment practice in Indonesian English-language education programmes more broadly.

Within this architecture, a Course Learning Outcome for interpretative listening that specifies

only general comprehension, without explicit reference to inferencing accuracy or metaphor

awareness, cannot be considered constructively aligned with the actual cognitive demands the

skill entails — underscoring the need for outcome statements informed by discourse-analytic

error evidence such as that generated in this study.

2. Method

Research Design

This study employed a qualitative descriptive design (Creswell & Poth, 2018), chosen for

its capacity to produce a detailed, context-sensitive account of learners' inferencing processes

without imposing predetermined categorical hypotheses. This design was considered

appropriate because the aim of the study was not to test a theory-driven model of listening

comprehension, but rather to capture, in naturalistic detail, how meaning was actively

constructed, misconstrued, or abandoned as learners engaged in interpretative listening tasks.

Setting, Time, and Participants

The study was conducted at the English Education Study Program, Faculty of Teacher

Training and Education (FKIP), Universitas Nias (UNIAS), from March to May 2026.

Participants comprised the entire cohort of 43 second-semester students enrolled in the

2025/2026 academic year who were concurrently registered in a Listening or Interpretative

Listening course, engaged through total (census) sampling rather than a drawn subset. This

approach was considered appropriate given that second-semester students constitute a

relatively homogeneous group in terms of academic exposure — having completed

foundational listening coursework but not yet advanced listening or discourse-analysis subjects

— thereby providing a consistent developmental baseline against which variation in

inferencing behaviour could be meaningfully examined. Involving the complete cohort, rather

than a sample drawn from it, further allowed the study to capture the full range of inferencing

Academy of Education Journal

Vol. 17, No. 1, July 2026, Page: 405-413

ISSN: 1907-2341 (Print), ISSN: 2685-4031 (Online)

409

Elwin P. Zebua (A Discourse Analysis Of Students' Metaphorical....)

strategies and difficulties present within this population, minimising the risk that atypical or

idiosyncratic cases would be lost through sampling.

Data Collection Instruments

Three complementary instruments were employed to capture both the product and the

process of inferencing. First, the Listening Comprehension Worksheet and Audio Transcript

Matrix required participants to listen to short authentic or semi-authentic spoken discourse

segments containing identified metaphorical expressions and to record their inferred meaning

on a structured worksheet; these responses were then cross-referenced against a transcript

matrix aligning each metaphorical unit, its surrounding co-text, and the participant's stated

inference, enabling systematic localisation of where an inference departed from the intended

meaning.

Second, Think-Aloud Protocols (TAPs) were used to elicit a subset of participants' real-

time verbalisation of their reasoning while completing selected listening segments, producing

a process-level record of how literal input was — or was not — reinterpreted figuratively as it

unfolded. This instrument follows established guidance on the reactivity and veridicality of

verbal-report data in second-language research (Bowles & Leow, 2005; Yang, Hu, Que, &

Fan, 2025), which cautions that TAP data are most reliable when paired with a second, non-

concurrent data source, a requirement met here by the worksheet and interview instruments.

Third, a Semi-Structured Interview Guide was administered in retrospective sessions

following the listening and think-aloud tasks, giving participants the opportunity to reflect on

and elaborate the reasoning behind particular inferences that may not have been fully

articulated during the think-aloud process, and allowing verification of patterns observed in

the worksheet and TAP data.

Together, these three instruments enabled data triangulation across concurrent (TAPs),

product-based (worksheet/matrix), and retrospective (interview) sources of evidence, an

arrangement consistent with the broader methodological recommendation that comprehension-

outcome data alone cannot reveal the process by which an inference is constructed or

abandoned (Aryadoust, 2019).

Data Collection Procedure

Data collection proceeded sequentially yet in an integrated fashion for each participant

within a single research session, so as to preserve the temporal and cognitive continuity

between the listening event and its introspective account. The procedure began with the

administration of the Listening Comprehension Worksheet, completed while participants

listened to the audio material segment by segment; participants' written responses were

subsequently aligned with the Audio Transcript Matrix to establish a precise record of which

linguistic and acoustic cues corresponded to each response. Immediately following (or, where

segment length permitted, concurrently with) the listening task, participants verbalised their

thoughts through the Think-Aloud Protocol, generating a real-time account of the inferencing

process as it unfolded. Finally, upon completion of the listening and think-aloud tasks, each

participant took part in a retrospective session guided by the Semi-Structured Interview Guide,

in which they were invited to elaborate on, clarify, or reconstruct the reasoning behind specific

inferences — particularly those left ambiguous or underarticulated in the think-aloud data.

This fixed sequence (worksheet-TAP, then retrospective interview) was maintained across

participants to ensure procedural consistency and to minimise memory decay between the

listening event and its retrospective elaboration.

Data Analysis Procedure

Data were analysed through discourse-analytic coding informed by Gee's (2018) situated-

meaning framework, combined with a systematic qualitative coding process following

Saldaña's (2021) coding manual and operationalised through the interactive analytic model of

Academy of Education Journal

Vol. 17, No. 1, July 2026, Page: 405-413

ISSN: 1907-2341 (Print), ISSN: 2685-4031 (Online)

410

Elwin P. Zebua (A Discourse Analysis Of Students' Metaphorical....)

Miles and Huberman (1994) — data condensation, data display, and conclusion

drawing/verification. Open coding was first applied to worksheet responses, TAP transcripts,

and interview data to identify recurrent points of inferencing breakdown; axial coding then

grouped these instances into discrete error categories, each verified against the corresponding

metaphorical unit identified in the audio transcript matrix following the identification

procedures of Nacey et al. (2019). Condensed units were organised into episode-based data

displays linking worksheet responses, TAP segments, and interview explanations for each

listening item, allowing the researcher to trace how inferences were initiated, sustained,

revised, or discontinued across participants and listening segments. Source triangulation across

the three instruments, together with participant verification during interviews, was used to

establish credibility, consistent with contemporary trustworthiness criteria for qualitative

thematic analysis (Nowell, Norris, White, & Moules, 2017).

Trustworthiness

The trustworthiness of this study was established following the criteria proposed by

Lincoln and Guba (1985): credibility, transferability, dependability, and confirmability.

Credibility was strengthened through methodological triangulation across the

Worksheet/Matrix, Think-Aloud Protocols, and Semi-Structured Interviews (Denzin, 1978),

and through member checking, in which selected participants reviewed the researcher's

interpretation of their think-aloud and interview data to confirm that it accurately reflected

their intended meaning. Transferability was supported through a thick, contextualised

description of the research setting, participants' academic background, listening materials, and

procedural steps, so that readers may judge the applicability of the findings to comparable

settings, notwithstanding that statistical generalisation was not the aim of this qualitative

descriptive study. Dependability was addressed through a systematic audit trail comprising

raw TAP recordings and transcripts, annotated worksheets, interview recordings and

transcripts, and analytic memos documenting decisions made during data condensation and

display. Confirmability, finally, was pursued by grounding all interpretive claims in verifiable

excerpts from the three data sources and by maintaining reflexive notes distinguishing

participants' inferencing patterns from the researcher's own preconceptions.

3. Result and Discussion

Categories of Inferencing Error

Cross-instrument analysis identified six recurrent categories of inferencing error among

participants when interpreting metaphorically encoded, unstated meaning, as follows:

Literal-meaning fixation, occurred when participants remained anchored to the literal,

surface-level sense of an utterance and failed to trigger a figurative reinterpretation even when

co-textual cues signalled non-literal intent.

Source-domain mismapping, arose when participants correctly recognised that an

expression was non-literal but mapped it onto an inappropriate conceptual source domain,

producing a plausible yet discourse-inconsistent inference.

Discourse-context, which neglect described inferences constructed from an isolated

utterance without integration of preceding or following co-text, resulting in interpretations

disconnected from the speaker's cumulative discourse intention.

Prosodic and paralinguistic, appeared when participants failed to register stress,

intonation, or pausing that would ordinarily signal a shift into figurative or ironic meaning,

defaulting instead to a neutral literal reading.

First-language schema over-generalization which occurred when participants applied

Indonesian- or Nias-based conceptual metaphor schemas to English input, producing

Academy of Education Journal

Vol. 17, No. 1, July 2026, Page: 405-413

ISSN: 1907-2341 (Print), ISSN: 2685-4031 (Online)

411

Elwin P. Zebua (A Discourse Analysis Of Students' Metaphorical....)

inferences consistent with L1 figurative convention but inconsistent with the target-language

discourse; and

Processing-load-induced shallow inferencing, particularly evident in the TAP data,

described instances in which participants abandoned effortful inferencing mid-utterance under

real-time pressure and defaulted to a minimally elaborated or guessed interpretation.

Discourse-Analytic and Theoretical Explanation

Interpreted through relevance theory, most identified errors reflect a failure to construct

the ad hoc concept that would satisfy optimal relevance within the given discourse context

(Wilson & Carston, 2007): rather than adjusting the literal concept encoded by a word to fit

the speaker's communicative intention, participants frequently defaulted to the encoded literal

meaning itself, consistent with the literal-meaning-fixation category. Source-domain

mismapping and first-language schema overgeneralisation, in turn, align closely with Gibbs's

(2017) embodied account of conceptual metaphor, indicating that participants' source-domain

activation was shaped more strongly by entrenched first-language conceptual structures than

by the target-language discourse cues available to them. Processing-load-induced shallow

inferencing corroborates Littlemore's (2019) finding that working-memory constraints

disproportionately affect real-time L2 metaphor resolution, and echoes Ifantidou and Plata's

(2021) characterisation of metaphor as a mental shortcut: when processing capacity was

exhausted by unfamiliar lexis co-occurring with a metaphorical shift, participants in the

present study's TAP data explicitly abandoned inferential effort rather than pursuing a fuller

reinterpretation.

Metacognitive Awareness from Interview Data

Interview data revealed that participants were often aware, after the fact, that their

inference had been implausible, yet could not identify at which point in the discourse the

figurative shift had occurred — suggesting a gap not in general listening motivation but in

explicit metacognitive monitoring during the inferencing process itself. This finding supports

the broader contention that targeted metacognitive instruction, rather than repeated exposure

alone, is required to strengthen inferencing accuracy (Goh & Vandergrift, 2021; Bozorgian et

al., 2022).

Implications for the OBE Curriculum

Applying Biggs and Tang's (2015) constructive alignment principle, these findings

suggest that Course Learning Outcomes for interpretative listening subjects should explicitly

specify inferencing accuracy in figurative discourse as a measurable outcome, rather than

subsuming it within a general comprehension descriptor. Concretely, this entails: (1) revising

Course Learning Outcomes to state that students can accurately infer speaker-implicated

meaning from metaphorical spoken discourse; (2) embedding transcript-matrix-based listening

tasks and structured think-aloud practice as regular formative activities targeting each

identified error category; and (3) designing assessment rubrics that award credit not only for

correct final inferences but for evidence of context-integrated, metacognitively monitored

reasoning, in line with AUN-QA (2020) and Merdeka Belajar-Kampus Merdeka

(Kemendikbud, 2020) requirements that assessment be demonstrably aligned with stated

learning outcomes.

4. Conclusion

This study identified six recurrent categories of inferencing error — literal-meaning

fixation, source-domain mismapping, discourse-context neglect, prosodic cue oversight, first-

language schema overgeneralisation, and processing-load-induced shallow inferencing —

produced by second-semester English Education students at UNIAS when interpreting

metaphorically encoded, unstated meaning in spoken discourse. Analysed through conceptual-

Academy of Education Journal

Vol. 17, No. 1, July 2026, Page: 405-413

ISSN: 1907-2341 (Print), ISSN: 2685-4031 (Online)

412

Elwin P. Zebua (A Discourse Analysis Of Students' Metaphorical....)

metaphor and relevance-theoretic lenses, these errors indicate that miscomprehension in

interpretative listening is rarely a matter of generalised listening weakness, but is instead

traceable to specific, identifiable breakdowns in figurative and contextual inferencing. The

study recommends that these findings be translated into explicit, constructively aligned

Course Learning Outcomes within the OBE curriculum, supported by transcript-matrix

listening tasks, structured think-aloud practice, and metacognitively oriented assessment.

Limitations of the study include its relatively small, single-institution participant pool and its

focus on a limited range of discourse genres; future research might extend this error taxonomy

across broader genres and proficiency levels, and evaluate the instructional impact of

metaphor-aware listening interventions through classroom-based or quasi-experimental

designs.

5. References

Aryadoust, V. (2019). An integrated cognitive theory of comprehension. International

Journal of Listening, 33(2), 71-100. https://doi.org/10.1080/10904018.2018.1522636

Aryadoust, V. (2022). A review of comprehension subskills in listening assessment and

diagnosis. Language Testing, 39(4), 549-580.

AUN-QA. (2020). Guide to AUN-QA Assessment at Programme Level (Version 4.0).

ASEAN University Network.

Biggs, J., & Tang, C. (2015). Teaching for Quality Learning at University (4th ed.). Open

University Press.

Bowles, M. A., & Leow, R. P. (2005). Reactivity and type of verbal report in SLA research

methodology. Studies in Second Language Acquisition, 27(3), 415-440.

Bozorgian, H., Muhammadpour, M., & Mahmoudi, E. (2022). Embedding L2 listening

homework in metacognitive intervention: Listening comprehension development.

Frontiers in Education, 7, 819308. https://doi.org/10.3389/feduc.2022.819308

Creswell, J. W., & Poth, C. N. (2018). Qualitative Inquiry and Research Design: Choosing

Among Five Approaches (4th ed.). SAGE Publications.

Denzin, N. K. (1978). The Research Act: A Theoretical Introduction to Sociological

Methods. McGraw-Hill.

Gee, J. P. (2018). An Introduction to Discourse Analysis: Theory and Method (5th ed.).

Routledge.

Gibbs, R. W., Jr. (2017). Metaphor Wars: Conceptual Metaphors in Human Life. Cambridge

University Press.

Goh, C. C. M., & Vandergrift, L. (2021). Teaching and Learning Second Language

Listening: Metacognition in Action (2nd ed.). Routledge.

Ifantidou, E., & Plata, S. (2021). Relevance and metaphor understanding in a second

language. Journal of Pragmatics, 179, 197-211.

Kemendikbud. (2020). Buku Panduan Merdeka Belajar-Kampus Merdeka. Direktorat

Jenderal Pendidikan Tinggi.

Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic Inquiry. SAGE Publications.

Academy of Education Journal

Vol. 17, No. 1, July 2026, Page: 405-413

ISSN: 1907-2341 (Print), ISSN: 2685-4031 (Online)

413

Elwin P. Zebua (A Discourse Analysis Of Students' Metaphorical....)

Littlemore, J. (2019). Metaphors in the Mind: Sources of Variation in Embodied Metaphor.

Cambridge University Press.

Littlemore, J., Krennmayr, T., Turner, J., & Turner, S. (2011). An investigation into

metaphor use at different levels of second language writing. Applied Linguistics, 35(2),

117-144.

Miles, M. B., Huberman, A. M., & Saldaña, J. (2014). Qualitative Data Analysis: A Methods

Sourcebook (3rd ed.). SAGE Publications.

Nacey, S., Dorst, A. G., Krennmayr, T., & Reijnierse, W. G. (Eds.). (2019). Metaphor

Identification in Multiple Languages: MIPVU around the World. John Benjamins.

Nowell, L. S., Norris, J. M., White, D. E., & Moules, N. J. (2017). Thematic analysis:

Striving to meet the trustworthiness criteria. International Journal of Qualitative

Methods, 16(1), 1-13.

Saldaña, J. (2021). The Coding Manual for Qualitative Researchers (4th ed.). SAGE

Publications.

Tanewong, S. (2018). A comparative study of metacognitive pedagogical sequencing and

conventional listening instruction in an EFL setting. PASAA, 55, 96-136.

Wahyuni, S., Fitriyah, I., & Hasanah, I. (2023). The implementation of Merdeka Belajar

Curriculum at English department of Indonesian universities. Journal of English

Education and Linguistics Studies, 10(2), 307-332.

https://doi.org/10.30762/jeels.v10i2.1249

Wilson, D., & Carston, R. (2007). A unitary approach to lexical pragmatics: Relevance,

inference and ad hoc concepts. In N. Burton-Roberts (Ed.), Pragmatics (pp. 230-259).

Palgrave Macmillan.

Yang, C., Hu, G., Que, K., & Fan, N. (2025). Comparing eye-tracking and verbal reports in

L2 reading process research: Three qualitative studies. Journal of Eye Movement

Research, 19(1), 1-24. https://doi.org/10.3390/jemr19010002