A sequential explanatory mixed-methods study on the acceptance of a social robot for EFL speaking practice among Chinese primary school students: Insights from the Computers Are Social Actors (CASA) paradigm

Bin Zou; Chenghao Wang; Huimin He; Jinlong Li; Yiran Du

arxiv: 2604.12789 · v1 · submitted 2026-04-14 · 💻 cs.HC

A sequential explanatory mixed-methods study on the acceptance of a social robot for EFL speaking practice among Chinese primary school students: Insights from the Computers Are Social Actors (CASA) paradigm

Yiran Du , Jinlong Li , Huimin He , Chenghao Wang , Bin Zou This is my paper

Pith reviewed 2026-05-10 14:34 UTC · model grok-4.3

classification 💻 cs.HC

keywords social robot acceptanceEFL speakingprimary school studentsTechnology Acceptance ModelCASA paradigmperceived enjoymentanthropomorphismmixed methods

0 comments

The pith

Perceived enjoyment and ease of use are the strongest predictors of Chinese primary school students' acceptance of a social robot for English speaking practice, with social attributes like warmth enhancing enjoyment.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper explores what makes young Chinese learners willing to use a social robot for practicing English speaking. By combining the Technology Acceptance Model with the Computers Are Social Actors idea, it examines how both practical and social features affect their plans to use the robot. Data from 436 students' surveys and 12 interviews reveal that enjoyment and how easy the robot is to use matter most for acceptance. Social qualities such as warmth, looking human-like, and feeling socially present increase that enjoyment, while the robot seeming intelligent mainly affects how useful it seems. The work indicates that creating robots that engage kids emotionally and socially can better support motivation and confidence in learning to speak a new language.

Core claim

Integrating the Technology Acceptance Model and the Computers Are Social Actors paradigm, the study establishes that among Chinese primary school students, perceived enjoyment and ease of use are the strongest predictors of behavioral intention to use a social robot for EFL speaking practice. Social attributes including warmth, anthropomorphism, and social presence significantly boost enjoyment, whereas perceived intelligence influences perceived usefulness but not ease of use. These results underscore that emotional and social engagement play a central role in young learners' acceptance of educational robots.

What carries the argument

The integration of the Technology Acceptance Model (TAM) and the Computers Are Social Actors (CASA) paradigm, which models functional factors like enjoyment and ease of use alongside social factors like anthropomorphism to predict acceptance.

If this is right

Robots designed with more social warmth and human-like features will lead to higher enjoyment and greater acceptance among young EFL learners.
Emphasizing ease of use and fun interactions should take priority over solely improving the robot's intelligence for this age group.
Such designs can increase motivation and speaking confidence in foreign language practice.
Educational technology for children benefits from incorporating social presence to foster acceptance.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Designers of educational robots might achieve better results by focusing on creating enjoyable social experiences rather than maximizing technical intelligence alone.
These findings could extend to other subjects or languages if the emphasis on enjoyment holds across contexts.
Future work could test whether high reported acceptance translates into measurable improvements in actual speaking proficiency over time.

Load-bearing premise

That the Technology Acceptance Model and CASA paradigm apply directly to young Chinese primary school students without needing major changes for their age or culture, and that self-reported intention matches real usage of the robot.

What would settle it

A classroom study tracking actual robot usage and speaking improvement among students who score high or low on the acceptance model to see if the predicted relationships hold in practice.

Figures

Figures reproduced from arXiv: 2604.12789 by Bin Zou, Chenghao Wang, Huimin He, Jinlong Li, Yiran Du.

**Figure 1.** Figure 1: The Conceptual Model 2.3 The Roles of Perceived Enjoyment, Ease of Use, and Usefulness The Technology Acceptance Model (TAM) provides a widely validated framework for explaining how users develop intentions to adopt new technologies. It posits that individuals’ behavioural intention (BI) to use a system is primarily determined by their perceived usefulness (PU) and perceived ease of use (PEOU) (Davis, 1989… view at source ↗

**Figure 3.** Figure 3: Students interacting with the social robot Lvbao [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗

read the original abstract

This study investigates Chinese primary school students' acceptance of a social robot for English-as-a-foreign-language (EFL) speaking practice through a sequential explanatory mixed-methods design. Integrating the Technology Acceptance Model (TAM) and the Computers Are Social Actors (CASA) paradigm, the research explores both functional and social factors influencing learners' behavioural intention to use the robot. Quantitative data from 436 students were analysed using structural equation modelling, followed by qualitative interviews with twelve students to interpret the findings. Results show that perceived enjoyment and ease of use are the strongest predictors of acceptance, while social attributes such as warmth, anthropomorphism, and social presence significantly enhance enjoyment. Perceived intelligence affects usefulness but not ease of use. The findings suggest that emotional and social engagement are central to young learners' acceptance of educational robots, highlighting the importance of designing socially intelligent technologies that promote motivation and speaking confidence in EFL learning contexts.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper adds targeted data on what drives Chinese primary kids to accept a social robot for EFL practice but rests on self-reports whose validity for this age group is not clearly established.

read the letter

The main thing to know is that enjoyment and ease of use turned out as the strongest predictors of acceptance, while social traits like warmth and anthropomorphism mainly worked by increasing enjoyment, and intelligence linked more to usefulness. The study ran a survey of 436 students analyzed with SEM and followed up with 12 interviews to unpack the numbers. That mixed-methods setup on an under-studied group and task is the useful part. It takes established TAM and CASA ideas and shows how the predictor weights look in this specific setting, which gives designers some concrete signals on what to emphasize for young learners. The sample size is respectable and the sequential design helps explain the quantitative patterns. The soft spots sit mostly in the measurement. Primary school students can struggle with abstract survey items on social presence or anthropomorphism, and classroom data collection raises the chance of social desirability answers or simple misunderstanding, especially with EFL language barriers. Treating behavioral intention as the acceptance outcome is standard but leaves a gap when there is no check against actual robot use or speaking practice over time. The abstract does not include model fit details or variable definitions, which makes it harder to judge how cleanly the SEM supports the claims. This work is mainly for people building or studying educational robots for language learning with young children in similar contexts. It supplies one more empirical anchor point even if the measures need tightening. It deserves peer review because the approach is sensible and the population focus fills a gap, though referees will likely push on validating the constructs for this age and adding some behavioral outcome data.

Referee Report

2 major / 2 minor

Summary. The manuscript reports a sequential explanatory mixed-methods study on Chinese primary school students' acceptance of a social robot for EFL speaking practice. It integrates the Technology Acceptance Model (TAM) and Computers Are Social Actors (CASA) paradigm, analyzing survey data from 436 students via structural equation modeling (SEM) followed by 12 qualitative interviews. Key claims are that perceived enjoyment and ease of use are the strongest predictors of behavioral intention to use the robot, social attributes (warmth, anthropomorphism, social presence) enhance enjoyment, and perceived intelligence affects usefulness but not ease of use.

Significance. If the measurement instruments prove valid for this age group and self-reported intentions align with actual usage, the work offers useful empirical insights into how social and emotional factors shape young learners' acceptance of educational robots. The mixed-methods design and focus on primary-school EFL contexts add value to the TAM/CASA literature, which has been dominated by adult or general-education samples.

major comments (2)

[Abstract and §4] Abstract and §4 (Quantitative Results): The central claims rest on SEM results from 436 responses, yet the manuscript provides no model-fit statistics (CFI, RMSEA, SRMR, χ²/df), no information on data exclusion or missing-value handling, and no full operational definitions or item lists for the latent variables. Without these, the reported path coefficients and the assertion that enjoyment and ease of use are the strongest predictors cannot be evaluated.
[§3 and §5] §3 (Methodology) and §5 (Discussion): The study applies unmodified TAM and CASA constructs to 8–12-year-old Chinese EFL learners. No pilot validation, cognitive interviewing, or measurement-invariance checks are described to confirm that children interpret items on anthropomorphism, social presence, or warmth in the intended way. The qualitative phase interprets rather than independently validates the quantitative model against objective usage logs or observed behavior, leaving the weakest assumption (self-reported intention as proxy for acceptance) untested.

minor comments (2)

[Abstract and §4] The abstract and results section should explicitly state the software, estimator (e.g., ML, WLSMV), and handling of ordinal Likert data used in the SEM.
[§4] Table or figure captions for the structural model should include standardized vs. unstandardized coefficients and significance levels for all paths.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the constructive and detailed feedback, which highlights important areas for improving transparency and methodological rigor. We address each major comment below and indicate the revisions we will make.

read point-by-point responses

Referee: [Abstract and §4] Abstract and §4 (Quantitative Results): The central claims rest on SEM results from 436 responses, yet the manuscript provides no model-fit statistics (CFI, RMSEA, SRMR, χ²/df), no information on data exclusion or missing-value handling, and no full operational definitions or item lists for the latent variables. Without these, the reported path coefficients and the assertion that enjoyment and ease of use are the strongest predictors cannot be evaluated.

Authors: We agree that these details are necessary for readers to fully evaluate the SEM results. In the revised manuscript, we will expand §4 to report the model-fit statistics (including CFI, RMSEA, SRMR, and χ²/df), describe the data exclusion criteria and missing-value handling procedures (e.g., listwise deletion thresholds and use of full-information maximum likelihood), and add an appendix with the complete item lists, sources, and operational definitions for all latent variables. These additions will directly support the claims about the strongest predictors. revision: yes
Referee: [§3 and §5] §3 (Methodology) and §5 (Discussion): The study applies unmodified TAM and CASA constructs to 8–12-year-old Chinese EFL learners. No pilot validation, cognitive interviewing, or measurement-invariance checks are described to confirm that children interpret items on anthropomorphism, social presence, or warmth in the intended way. The qualitative phase interprets rather than independently validates the quantitative model against objective usage logs or observed behavior, leaving the weakest assumption (self-reported intention as proxy for acceptance) untested.

Authors: We acknowledge that the manuscript does not describe formal pilot validation, cognitive interviewing, or measurement-invariance testing for this age group. In the revision, we will expand §3 to detail the pre-testing steps taken for item clarity with a small sample of similar-aged students and add a limitations paragraph in §5 explicitly discussing the risks of differential interpretation of social constructs by children, along with a call for future invariance checks. The qualitative interviews were designed to explain the quantitative patterns rather than to provide independent behavioral validation; we will clarify this scope and note the reliance on self-reported intentions as a limitation. revision: partial

standing simulated objections not resolved

The study design relies on self-reported behavioral intention without accompanying objective usage logs or observed behavior; addressing this fully would require a new longitudinal data collection effort beyond the scope of the current manuscript.

Circularity Check

0 steps flagged

No circularity: purely empirical mixed-methods analysis

full rationale

The paper reports a sequential explanatory mixed-methods study applying the external Technology Acceptance Model and CASA paradigm to survey data from 436 primary students via structural equation modeling, followed by 12 interviews. No equations, derivations, or predictive claims are present that reduce to inputs by construction, fitted parameters renamed as predictions, or self-citation chains. Central results (e.g., enjoyment and ease of use as strongest predictors) are direct outputs of standard statistical analysis on collected data, with no self-definitional loops or ansatz smuggling. The study is self-contained against external benchmarks and draws on established frameworks without load-bearing self-references.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on the direct applicability of TAM and CASA constructs to young Chinese learners without cultural or developmental adjustment, plus the assumption that survey-based intention measures validly capture acceptance.

axioms (2)

domain assumption The Technology Acceptance Model (TAM) constructs apply to social robots in educational contexts for children.
Integrated into the study design without testing alternatives or cultural adaptations.
domain assumption Self-reported measures of behavioral intention reflect actual acceptance and use.
Standard in TAM research but not independently validated in this study.

pith-pipeline@v0.9.0 · 5482 in / 1298 out tokens · 55313 ms · 2026-05-10T14:34:28.719032+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

20 extracted references · 20 canonical work pages

[1]

Introduction Social robots, physically embodied agents capable of socially meaningful interaction—are increasingly used in education to promote communication and learning (Breazeal et al., 2016; Mavrogiannis et al., 2023). Their physical presence and expressive behaviours enable natural interaction, making them particularly effective for language learning...

work page 2016
[2]

and the Computers Are Social Actors (CASA) paradigm (Reeves & Nass, 1996; Xu et al., 2022). While TAM focuses on perceptions of usefulness, ease of use, and enjoyment, CASA highlights how social attributes such as social presence, anthropomorphism, warmth, and intelligence influence users’ perceptions (Gambino et al., 2020). Adopting a sequential explanat...

work page 1996
[3]

Literature Review and Hypothesis Development 2.1 Social Robot for Language Learning A social robot is a physically embodied agent designed to interact and communicate with humans in a socially meaningful way (Breazeal et al., 2016). Unlike virtual agents or chatbots, social robots possess physical presence, expressive behaviours, and social cues that enab...

work page 2016
[4]

to explain Chinese primary school students’ acceptance of a social robot for English-speaking practice (see Figure 1). According to TAM (Venkatesh, 2000), users’ behavioural intention to use a technology (BI) is influenced by their perceptions of its usefulness (PU), ease of use (PEOU), and enjoyment (PE). Complementing this framework, the CASA paradigm p...

work page 2000
[5]

It posits that individuals’ behavioural intention (BI) to use a system is primarily determined by their perceived usefulness (PU) and perceived ease of use (PEOU) (Davis, 1989)

The Conceptual Model 2.3 The Roles of Perceived Enjoyment, Ease of Use, and Usefulness The Technology Acceptance Model (TAM) provides a widely validated framework for explaining how users develop intentions to adopt new technologies. It posits that individuals’ behavioural intention (BI) to use a system is primarily determined by their perceived usefulnes...

work page 1989
[6]

extends this perspective by suggesting that individuals unconsciously apply social rules and expectations to technologies that display social cues such as speech, gesture, and personality. In the context of social robots, this paradigm highlights how users’ perceptions of the robot’s social attributes can shape their emotional engagement and acceptance (X...

work page 2022
[7]

It could greet learners, introduce speaking activities, ask and answer questions, and offer immediate encouragement or corrective feedback

The 3D model, printing process, and completed Lvbao Lvbao was designed to simulate natural spoken communication through its integrated speech recognition, text-to-speech, and cloud-based dialogue system. It could greet learners, introduce speaking activities, ask and answer questions, and offer immediate encouragement or corrective feedback. The robot’s f...

work page 2023
[8]

Each construct comprised four items measured on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree)

and the Computers Are Social Actors (CASA) paradigm (Deng & Yan, 2025; Song et al., 2024; Zheng et al., 2023). Each construct comprised four items measured on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). All items were written in clear, age-appropriate language suitable for primary school students (Y . Du,

work page 2025
[9]

Can you tell me more about that?

Constructs and Measurement Items Construct and Definition Measurement Item (English) Measurement Item (Chinese) Perceived Usefulness (PU) Definition: The degree to which using the robot enhances learners’ English-speaking performance PU1 Using Lvbao improves my English-speaking skills PU1 使⽤绿宝能提⾼我的英语⼝语能⼒ PU2 Lvbao helps me practise speaking English more e...

work page 2025
[10]

Prior to analysis, the dataset was screened for missing values, outliers, and normality

to examine the hypothesised relationships among constructs from the integrated TAM and CASA framework. Prior to analysis, the dataset was screened for missing values, outliers, and normality. Descriptive statistics, including means, standard deviations, skewness, and kurtosis, were calculated to evaluate participants’ perceptions of the social robot. The ...

work page 2006
[11]

Cohen’s Kappa reached 0.82, indicating strong reliability (J

to calculate inter-coder agreement. Cohen’s Kappa reached 0.82, indicating strong reliability (J. Cohen, 1960). Discrepancies were discussed and resolved through consensus before finalising the coding scheme. Integration of quantitative and qualitative findings was carried out during interpretation to provide a comprehensive understanding of students’ acc...

work page 1960
[12]

The mean values ranged from M = 3.88 to M = 4.24 on a five-point Likert scale, indicating generally positive perceptions of the social robot among the participants

Results 4.1 Quantitative Results 4.1.1 Descriptive Statistics and Normality Table 4 presents the descriptive statistics for the eight constructs examined in this study, including Perceived Usefulness (PU), Perceived Ease of Use (PEOU), Perceived Enjoyment (PE), Behavioural Intention (BI), Perceived Social Presence (PSP), Perceived Anthropomorphism (PA), P...

work page 2016
[13]

As shown in Table 5, all standardised factor loadings exceeded 0.70 (p < .001), indicating that each item was a strong indicator of its corresponding latent construct (Kline, 2016)

Descriptive Statistics of the Constructs Construct M SD Skewness Kurtosis Perceived Usefulness (PU) 4.12 0.58 −0.41 −0.36 Perceived Ease of Use (PEOU) 4.08 0.61 −0.35 −0.44 Perceived Enjoyment (PE) 4.24 0.54 −0.52 −0.22 Behavioural Intention (BI) 4.17 0.63 −0.46 −0.31 Perceived Social Presence (PSP) 4.09 0.57 −0.39 −0.37 Perceived Anthropomorphism (PA) 3....

work page 2016
[14]

Model Fit Indices for the Measurement and Structural Models Fit Index Threshold Measurement Model Structural Model χ²/df < 3.00 2.41 2.58 CFI ≥ 0.90 0.96 0.95 TLI ≥ 0.90 0.95 0.94 RMSEA ≤ 0.08 0.06 0.06 SRMR ≤ 0.08 0.04 0.05 4.1.3 The Structural Model The structural model was tested to examine the hypothesised relationships among the constructs derived fr...

work page 2022
[15]

Discussion 5.1 The Impact of Perceived Enjoyment, Ease of Use, and Usefulness The findings reaffirm the Technology Acceptance Model’s (TAM) emphasis on enjoyment and ease of use as key predictors of technology acceptance (Davis, 1989; Davis & Granić, 2024). Perceived enjoyment had the strongest effect on both usefulness and behavioural intention, indicati...

work page 1989
[16]

smartness

Conclusion In conclusion, this study provides a comprehensive understanding of Chinese primary school students’ acceptance of a social robot for EFL speaking practice by integrating the Technology Acceptance Model (TAM) and the Computers Are Social Actors (CASA) paradigm. Quantitative and qualitative findings revealed that perceived enjoyment and ease of ...

work page doi:10.1007/s10639-023-11959-3 2024
[17]

https://doi.org/10.3390/educsci14080880 Huang, G., & Moore, R. K. (2023). Using social robots for language learning: Are we there yet? 3(1), 208–230. https://doi.org/10.1515/jccall-2023-0013 Johal, W. (2020). Research Trends in Social Robots for Learning. Current Robotics Reports, 1(3), 75–83. https://doi.org/10.1007/s43154-020-00008-3 Kim, Y ., & Sundar,...

work page doi:10.3390/educsci14080880 2023
[18]

https://doi.org/10.3390/info16010029 Lee, J., Kim, J., & Choi, J. Y . (2019). The adoption of virtual reality devices: The technology acceptance model integrating enjoyment, social interaction, and strength of the social ties. Telematics and Informatics, 39, 37–48. https://doi.org/10.1016/j.tele.2018.12.006 Lei, L., & Qin, J. (2022). Research in foreign l...

work page doi:10.3390/info16010029 2019
[19]

L., Darvin, R., & Ma, C

https://doi.org/10.3390/brainsci11091126 Liu, G. L., Darvin, R., & Ma, C. (2024). Exploring AI-mediated informal digital learning of English (AI-IDLE): A mixed-method investigation of Chinese EFL learners’ AI adoption and experiences. Computer Assisted Language Learning, 1–29. https://doi.org/10.1080/09588221.2024.2310288 Liu, G., & Ma, C. (2023). Measuri...

work page doi:10.3390/brainsci11091126 2024
[20]

Can you tell me more about that?

https://doi.org/10.3389/frobt.2018.00114 Plowman, L., & McPake, J. (2013). Seven Myths About Young Children and Technology. Childhood Education, 89(1), 27–33. https://doi.org/10.1080/00094056.2013.757490 R Core Team. (2025). R: A language and environment for statistical computing (Version 4.4.3) [Computer software]. R Foundation for Statistical Computing....

work page doi:10.3389/frobt.2018.00114 2018

[1] [1]

Introduction Social robots, physically embodied agents capable of socially meaningful interaction—are increasingly used in education to promote communication and learning (Breazeal et al., 2016; Mavrogiannis et al., 2023). Their physical presence and expressive behaviours enable natural interaction, making them particularly effective for language learning...

work page 2016

[2] [2]

and the Computers Are Social Actors (CASA) paradigm (Reeves & Nass, 1996; Xu et al., 2022). While TAM focuses on perceptions of usefulness, ease of use, and enjoyment, CASA highlights how social attributes such as social presence, anthropomorphism, warmth, and intelligence influence users’ perceptions (Gambino et al., 2020). Adopting a sequential explanat...

work page 1996

[3] [3]

Literature Review and Hypothesis Development 2.1 Social Robot for Language Learning A social robot is a physically embodied agent designed to interact and communicate with humans in a socially meaningful way (Breazeal et al., 2016). Unlike virtual agents or chatbots, social robots possess physical presence, expressive behaviours, and social cues that enab...

work page 2016

[4] [4]

to explain Chinese primary school students’ acceptance of a social robot for English-speaking practice (see Figure 1). According to TAM (Venkatesh, 2000), users’ behavioural intention to use a technology (BI) is influenced by their perceptions of its usefulness (PU), ease of use (PEOU), and enjoyment (PE). Complementing this framework, the CASA paradigm p...

work page 2000

[5] [5]

It posits that individuals’ behavioural intention (BI) to use a system is primarily determined by their perceived usefulness (PU) and perceived ease of use (PEOU) (Davis, 1989)

The Conceptual Model 2.3 The Roles of Perceived Enjoyment, Ease of Use, and Usefulness The Technology Acceptance Model (TAM) provides a widely validated framework for explaining how users develop intentions to adopt new technologies. It posits that individuals’ behavioural intention (BI) to use a system is primarily determined by their perceived usefulnes...

work page 1989

[6] [6]

extends this perspective by suggesting that individuals unconsciously apply social rules and expectations to technologies that display social cues such as speech, gesture, and personality. In the context of social robots, this paradigm highlights how users’ perceptions of the robot’s social attributes can shape their emotional engagement and acceptance (X...

work page 2022

[7] [7]

It could greet learners, introduce speaking activities, ask and answer questions, and offer immediate encouragement or corrective feedback

The 3D model, printing process, and completed Lvbao Lvbao was designed to simulate natural spoken communication through its integrated speech recognition, text-to-speech, and cloud-based dialogue system. It could greet learners, introduce speaking activities, ask and answer questions, and offer immediate encouragement or corrective feedback. The robot’s f...

work page 2023

[8] [8]

Each construct comprised four items measured on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree)

and the Computers Are Social Actors (CASA) paradigm (Deng & Yan, 2025; Song et al., 2024; Zheng et al., 2023). Each construct comprised four items measured on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). All items were written in clear, age-appropriate language suitable for primary school students (Y . Du,

work page 2025

[9] [9]

Can you tell me more about that?

Constructs and Measurement Items Construct and Definition Measurement Item (English) Measurement Item (Chinese) Perceived Usefulness (PU) Definition: The degree to which using the robot enhances learners’ English-speaking performance PU1 Using Lvbao improves my English-speaking skills PU1 使⽤绿宝能提⾼我的英语⼝语能⼒ PU2 Lvbao helps me practise speaking English more e...

work page 2025

[10] [10]

Prior to analysis, the dataset was screened for missing values, outliers, and normality

to examine the hypothesised relationships among constructs from the integrated TAM and CASA framework. Prior to analysis, the dataset was screened for missing values, outliers, and normality. Descriptive statistics, including means, standard deviations, skewness, and kurtosis, were calculated to evaluate participants’ perceptions of the social robot. The ...

work page 2006

[11] [11]

Cohen’s Kappa reached 0.82, indicating strong reliability (J

to calculate inter-coder agreement. Cohen’s Kappa reached 0.82, indicating strong reliability (J. Cohen, 1960). Discrepancies were discussed and resolved through consensus before finalising the coding scheme. Integration of quantitative and qualitative findings was carried out during interpretation to provide a comprehensive understanding of students’ acc...

work page 1960

[12] [12]

The mean values ranged from M = 3.88 to M = 4.24 on a five-point Likert scale, indicating generally positive perceptions of the social robot among the participants

Results 4.1 Quantitative Results 4.1.1 Descriptive Statistics and Normality Table 4 presents the descriptive statistics for the eight constructs examined in this study, including Perceived Usefulness (PU), Perceived Ease of Use (PEOU), Perceived Enjoyment (PE), Behavioural Intention (BI), Perceived Social Presence (PSP), Perceived Anthropomorphism (PA), P...

work page 2016

[13] [13]

As shown in Table 5, all standardised factor loadings exceeded 0.70 (p < .001), indicating that each item was a strong indicator of its corresponding latent construct (Kline, 2016)

Descriptive Statistics of the Constructs Construct M SD Skewness Kurtosis Perceived Usefulness (PU) 4.12 0.58 −0.41 −0.36 Perceived Ease of Use (PEOU) 4.08 0.61 −0.35 −0.44 Perceived Enjoyment (PE) 4.24 0.54 −0.52 −0.22 Behavioural Intention (BI) 4.17 0.63 −0.46 −0.31 Perceived Social Presence (PSP) 4.09 0.57 −0.39 −0.37 Perceived Anthropomorphism (PA) 3....

work page 2016

[14] [14]

Model Fit Indices for the Measurement and Structural Models Fit Index Threshold Measurement Model Structural Model χ²/df < 3.00 2.41 2.58 CFI ≥ 0.90 0.96 0.95 TLI ≥ 0.90 0.95 0.94 RMSEA ≤ 0.08 0.06 0.06 SRMR ≤ 0.08 0.04 0.05 4.1.3 The Structural Model The structural model was tested to examine the hypothesised relationships among the constructs derived fr...

work page 2022

[15] [15]

Discussion 5.1 The Impact of Perceived Enjoyment, Ease of Use, and Usefulness The findings reaffirm the Technology Acceptance Model’s (TAM) emphasis on enjoyment and ease of use as key predictors of technology acceptance (Davis, 1989; Davis & Granić, 2024). Perceived enjoyment had the strongest effect on both usefulness and behavioural intention, indicati...

work page 1989

[16] [16]

smartness

Conclusion In conclusion, this study provides a comprehensive understanding of Chinese primary school students’ acceptance of a social robot for EFL speaking practice by integrating the Technology Acceptance Model (TAM) and the Computers Are Social Actors (CASA) paradigm. Quantitative and qualitative findings revealed that perceived enjoyment and ease of ...

work page doi:10.1007/s10639-023-11959-3 2024

[17] [17]

https://doi.org/10.3390/educsci14080880 Huang, G., & Moore, R. K. (2023). Using social robots for language learning: Are we there yet? 3(1), 208–230. https://doi.org/10.1515/jccall-2023-0013 Johal, W. (2020). Research Trends in Social Robots for Learning. Current Robotics Reports, 1(3), 75–83. https://doi.org/10.1007/s43154-020-00008-3 Kim, Y ., & Sundar,...

work page doi:10.3390/educsci14080880 2023

[18] [18]

https://doi.org/10.3390/info16010029 Lee, J., Kim, J., & Choi, J. Y . (2019). The adoption of virtual reality devices: The technology acceptance model integrating enjoyment, social interaction, and strength of the social ties. Telematics and Informatics, 39, 37–48. https://doi.org/10.1016/j.tele.2018.12.006 Lei, L., & Qin, J. (2022). Research in foreign l...

work page doi:10.3390/info16010029 2019

[19] [19]

L., Darvin, R., & Ma, C

https://doi.org/10.3390/brainsci11091126 Liu, G. L., Darvin, R., & Ma, C. (2024). Exploring AI-mediated informal digital learning of English (AI-IDLE): A mixed-method investigation of Chinese EFL learners’ AI adoption and experiences. Computer Assisted Language Learning, 1–29. https://doi.org/10.1080/09588221.2024.2310288 Liu, G., & Ma, C. (2023). Measuri...

work page doi:10.3390/brainsci11091126 2024

[20] [20]

Can you tell me more about that?

https://doi.org/10.3389/frobt.2018.00114 Plowman, L., & McPake, J. (2013). Seven Myths About Young Children and Technology. Childhood Education, 89(1), 27–33. https://doi.org/10.1080/00094056.2013.757490 R Core Team. (2025). R: A language and environment for statistical computing (Version 4.4.3) [Computer software]. R Foundation for Statistical Computing....

work page doi:10.3389/frobt.2018.00114 2018