Process vs.
Product: Divergent Translanguaging Practices and Perceptions in Emergent
Bilinguals’ Digital Multimodal Composing
Congchao Hua1* and Yongli Liu2*
|
*Correspondence: |
|
1 City University of Macau,
China, Macau SAR 2
Xiangtan University, China |
Abstract
This study investigated
emergent Mandarin-English bilinguals’ practices and perceptions of
translanguaging in the process and product of digital multimodal composing
(DMC) in a content and language integrated learning (CLIL) context. Fifty-one
first-year undergraduates taking a CLIL course took a questionnaire survey and
submitted ten group videos. To account for potential individual differences,
linear mixed-effects models were built for the questionnaire responses. The
results showed that the participants’ translanguaging practices differed in the
process and product of DMC. Regarding their translanguaging practices, in the
process of DMC, they mainly translanguaged from
Mandarin to English and occasionally to Chinese dialects, which was most
evident during group discussion; in the product of DMC, they mainly translanguaged from English to Mandarin, with little use of
Chinese dialects. The analysis of the videos supported the questionnaire
results, revealing translanguaging mostly in subtitles and visuals. Concerning
perceptions of translanguaging, the participants generally had positive
attitudes and viewed translanguaging primarily as a way of learning rather than
a tool of building interpersonal relationships or expressing identity. Based on
these findings, pedagogical implications for fostering a more functionally diverse
approach to translanguaging in DMC are discussed.
Keywords:
translanguaging;
digital multimodal composing; emergent bilinguals; practices; perceptions
1. Introduction
With the prevalence of
digital technologies, digital multimodal composing (DMC) has become a common
practice for second-language (L2) teaching and
learning (Jiang & Hafner, 2025) in and out of the classroom. As an L2
pedagogy (Jiang, 2018; Zhang et al., 2023), DMC refers to textual practices
using digital tools to manoeuvre a variety of
semiotic modes such as visuals, words and sounds to communicate meaning
(Ainsworth et al., 2023; Jiang, 2018; Kessler, 2024). The various forms of DMC
include PowerPoint slides, podcasts, infographics, videos, and webpages, to
name the most common ones (Jiang & Hafner,
2025). In recent years, under the influence of multilingualism, attention has turned
to the translanguaging practices in L2 teaching and learning. Translanguaging,
defined as the deployment of one’s multilingual and multimodal repertoires in
meaning making (Li, 2018), offers a valuable new framework for researching DMC.
DMC provides abundant
opportunities for translanguaging practices (Jiang & Hafner, 2025; Wang,
2019). During DMC, L2 learners, as emergent bilinguals, strategically deploy
their linguistic and semiotic repertoires to support meaning-making, scaffold communication,
and construct knowledge. Such translanguaging practices create more inclusive
and flexible learning environments that not only enhance emergent bilinguals’
learning but also validate their linguistic identities.
Translanguaging in DMC
has been empirically proven to support both L2 development and content mastery
by enhancing learners’ engagement and scaffolding L2 use in meaningful contexts
(e.g., Li & Wang, 2024; Wang, 2024; Wu et al., 2025; Zhou & Pan, 2025). In addition, while
recent studies have documented L2 teachers’ translanguaging practices in class
(e.g., Li & Wang, 2024; Liu et al., 2024; Talley, 2025) and their
ambivalent attitudes to translanguaging (cf., Alrayes,
2024; Talley, 2025; Wang et al., 2025), there is a paucity of research on
emergent bilinguals’ own translanguaging practices and perceptions. Yet, for
any translanguaging pedagogy to be impactful in real classrooms, we must
understand it from both teachers’ and students’ perspectives, addressing not
just its theoretical potential, but also the needs, preferences, and challenges
that shape how translanguaging is adopted, adapted, and experienced by learners
in and out of the classroom.
Against this backdrop,
this study investigated Chinese L2 English learners’ practices and perceptions
of translanguaging in DMC in a content and language integrated learning (CLIL)
context, which offers particularly abundant opportunities for translanguaging.
Hopefully, the findings may provide additional empirical evidence for
translanguaging pedagogies and inform L2 teachers about learners’ perspectives
for implementing translanguaging pedagogies.
2. Literature Review
As a linguistic
practice shaping DMC processes and products, translanguaging enables emergent
bilinguals to capitalize on their full linguistic and semiotic repertoires to
navigate the multifaceted demands of DMC. Understanding how translanguaging
facilitates DMC and promotes learning requires examining not only how emergent
bilinguals strategically deploy translanguaging practices but also their
perceptions of these practices.
2.1 Translanguaging in DMC
DMC creates unique
spaces for translanguaging, where emergent bilinguals are not confined to the
L2 they are learning but draw on all available linguistic and semiotic
resources to construct meaning, collaborate, and express identities.
The translanguaging
spaces afforded by DMC can be both interactional and performative (Ho &
Tai, 2024). Interactional spaces arise as learners discuss, negotiate, and
collaborate during the process of a DMC task, while performative spaces emerge
in the final product, where emergent bilinguals construct and display their
multiple identities. At the same time, Garciá
et al. (2017) identify three key elements of translanguaging pedagogies:
stance, design, and shifts. Stance concerns teachers’ and students’ perceptions
of translanguaging; design involves how teachers plan learning activities to
create translanguaging spaces; and shifts refer to the dynamic movement between
linguistic and semiotic resources during interaction.
While Ho and Tai’s
(2024) classification necessitates the investigation of both the process and
product of DMC, Garciá et al.’s (2017) three
elements highlight what constitutes an effective translanguaging pedagogy.
However, few studies have specifically examined how they interrelate in DMC
contexts, especially from the student’s perspective in both the process and
product of DMC. To fully understand how translanguaging facilitates L2
learning, it is necessary to examine both the shifts and stance of emergent
bilinguals in the process and product of DMC.
2.2 Emergent bilinguals’ translanguaging practices
Studies
on emergent bilinguals’ translanguaging practices have explored both the
process and product of L2 learning tasks. Concerning the product, Ho (2022)
interviewed a group of undergraduates at a university in Hong Kong and analyzed
the video they produced, finding that during the DMC task they skillfully
deployed a wide range of linguistic and semiotic resources, subject knowledge,
and learning experiences to facilitate meaning expression and enhance the
content of their video. As regards the process of learning, Wang (2016)
observed L2 Chinese lessons in a New Zealand university, reporting that the L2
Chinese learners, just like their teachers, initiated translanguaging as a
“co-constructed dialogic approach” (p. 1) to enhance the communicativeness of
the class.
In
a review study, Pacheco et al. (2022) summarize the functions of emergent
bilinguals’ translanguaging practices in DMC, highlighting the necessity to
differentiate such functions in the process and product of DMC. According to
Pacheco et al. (2022), in the process of DMC, translanguaging helps emergent
bilinguals to collect information and prepare scripts (reported by 47% of the
studies reviewed, hereafter indicated by percentages), discuss and share
information and ideas (34%), and interact with communities out of class (31%).
In the product of DMC, translanguaging serves a different set of functions: it
helps emergent bilinguals to show their identities as multilingual speakers
(74%), convey refined meanings (34%), and engage diverse audiences (23%).
In
a recent study, Wang (2024) acknowledged Pacheco et al.’s process-product model
and explored beginning L2 Chinese learners’ translanguaging in
self-introduction videos. Despite the participants’ limited Chinese
proficiency, Wang found that they actively utilized diverse semiotic resources
to express their identities and improve their videos’ form and content. In
addition, their videos also demonstrated translanguaging practices in the
process of the DMC project, as some videos featured the students’ family
members introducing themselves in Chinese, which they had learned from the
student before making the video. Thus, Wang (2024) highlights that
translanguaging in this DMC project not only promoted academic achievements but
also consolidated the participants’ relationships with their family members.
Notably, Wang (2024) did not specifically explore her participants’
translanguaging practices in the process of the DMC project but only made
inferences from the product of the DMC project, thus leaving an incomplete
picture of the emergent bilinguals’ translanguaging practices.
In
short, the empirical studies reviewed above, although drawing on data from both
the process and product of DMC, generally have not differentiated the functions
translanguaging might serve in the process and product of DMC.
2.3 Emergent bilinguals’
perceptions of translanguaging
Studies
on emergent bilinguals’ perceptions of translanguaging have mostly focused on
the process of in-class learning, reporting mixed attitudes.
Some
of these studies explored perceptions of teachers’ translanguaging practices in
class. For example, Wang (2016) reports that more than half of the L2 Chinese
students surveyed expressed an affirmative view on multilingual instruction.
This finding is echoed by Zhou and Mann (2021), who conducted an action
research study on teachers’ translanguaging strategies and students’ attitudes,
reporting that the students acknowledged the benefits of their teachers’
carefully planned translanguaging practices in promoting learning.
Other
studies examined emergent bilinguals’ perceptions of their own translanguaging
practices in class. For instance, Trinh (2025) surveyed Vietnamese students’
translanguaging practices and perceptions, finding that these students viewed
their own translanguaging practices as a beneficial tool to achieve better
communication and better learning environments in class while acknowledging
potential confusion and over-reliance on L1. However, Wang (2024) reports
challenges felt by beginning level L2 Chinese learners in a New Zealand
university, such as unpredictable workload and insufficient technological
skills.
While
investigating emergent bilinguals’ perceptions of both their teachers’ and
peers translanguaging practices, Wang and Shen (2023) report that L2 learners
were positive about their teachers’ translanguaging practices for the benefits
of facilitating learning but concerned about their peers’ translanguaging
practices out of a monolingual ideology.
These
studies have revealed emergent bilinguals’ ambivalent perceptions of
translanguaging practices in class: positive about their teachers’
translanguaging but doubtful about their peers’ translanguaging. Little
attention has been paid to either translanguaging practices in out-of-class
contexts or those manifested in learning products (e.g., DMC artifacts).
In
conclusion, empirical studies have found that emergent bilinguals can
skillfully engage in translanguaging practices to enhance communication in the
process of DMC (e.g., Wang, 2016) and express themselves in the product of DMC
(e.g., Ho, 2022; Wang, 2024). Moreover, their perceptions of their own
translanguaging practices in the learning process in class are mixed, with both
positive and negative feelings (e.g., Trinh, 2025; Wang, 2024; Wang & Shen,
2023). While these findings help us to understand translanguaging among
emergent bilinguals, two major research gaps remain: 1) there is a lack of
research on emergent bilinguals’ practices and perceptions of translanguaging
in the process of DMC, especially out of class; 2) most empirical studies have
focused on either process or product, practices or perceptions, with limited
attention to how the two dimensions interact, thus failing to differentiate the
functions of translanguaging in the process and product of meaning-making.
Understanding out-of-class translanguaging and the interaction between the
product/process and practices/perceptions dimensions is crucial for unpacking
how translanguaging practices evolve during DMC and are manifested in the final
artifact.
Therefore,
adopting Ho and Tai’s (2024) division of translanguaging spaces and Garciá et al.’s (2017) translanguaging elements, this study
investigated the practices (shifts) and perceptions (stance) of translanguaging
among emergent bilinguals, both in the process (interactional) and product
(performative) of a DMC project in a CLIL context.
The
specific research questions are:
1) How do Chinese L2 English learners deploy
translanguaging in the process and product of their DMC project?
2) How do Chinese L2 English learners perceive the
functions of translanguaging in the process and product of their DMC project?
3. Research Method
This
study adopted a mixed-methods design, combining a questionnaire survey that
investigated the participants’ practices and perceptions of translanguaging in
DMC with a qualitative analysis of their videos as evidence of translanguaging
practices.
3.1
Participants
The
participants were 51 first-year undergraduates majoring in Business English or
International Trade at a university in South China. There were 20 boys and 31
girls, all aged 17-19 at the time of the investigation. Their English
proficiency level was intermediate, as all had just passed the National College
Entrance Examination.
Both
the business English majors and international trade majors were required to
take the business English course, which aimed to improve their English
proficiency through learning texts on entrepreneurship, business management,
trade policies, and intercultural differences. Therefore, the course provided a
typical content and language integrated learning (CLIL) context. The videos
analyzed in this study were from a course assignment requiring the participants
to introduce a certain aspect of the Chinese culture to imaginary international
audiences.
3.2
Questionnaire and Videos
The
instrument adopted in this study was a questionnaire on practices and
perceptions of translanguaging in the process and product of DMC. The
participants’ videos were also analyzed to examine their translanguaging
practices in the product of DMC.
The
questionnaire was developed based on Pacheco et al.’s (2022) framework of the
process and product of translanguaging in DMC. It consisted of two sections,
totaling 34 items. Section 1 contained 18 items eliciting information about
frequencies of language use in the process and product of DMC. These 18 items
were in the form of 4-point Likert scales, with 1=never, 2=sometimes, 3=often,
4=always.
Section
2 contained 16 items, focusing on perceptions of the usefulness and functions
of translanguaging in DMC. Four functions were listed for process and four for
product. Apart from the three functions of translanguaging listed by Pacheco et
al. (2022) for the process of DMC (i.e., accessing information, collaborating
with peers, interacting with communities) and three for the product of DMC
(i.e., constructing identity, enhancing meaning, engaging audience), “learning”
was listed as an additional function for both process and product, as it is a
valued function of translanguaging mentioned by different groups of emergent
bilinguals (e.g., Wang & Shen, 2023; Zhou & Mann, 2021). These 16 items
were in the form of 5-point Likert scales, with 1=strongly disagree,
2=disagree, 3=not sure, 4=agree, 5=strongly agree.
The
reliabilities (Cronbach’s alpha values) of the overall questionnaire, Section
1, and Section 2 were .90, .67, and .96, suggesting acceptable to high
reliabilities.
In
addition, ten group videos were collected and analyzed to identify the
participants’ translanguaging practices in the product of DMC. Following the
requirements of the assignment, in each of the ten videos, the participants
introduced a cultural phenomenon or a cultural product that is unique to China.
The specific topics covered in the videos included: herbal tea, din sum (2
videos), bowl cakes, casserole porridge, double-layer milk custard, lion dance
(2 videos), cross talk, and Cantonese opera. Each video lasted about five
minutes, incorporating three components: subtitle, image, and voice-over.
3.3
Procedure
The
study lasted 15 weeks, with the procedures visualized in Table 1.
Table 1. Research procedures
|
Week |
DMC Activities |
|
1 |
DMC project assignment; grouping; consultation |
|
2-13 |
DMC project preparation; consultation |
|
14 |
Video submission |
|
15 |
Questionnaire survey |
In
the first week, the teacher assigned the video task and informed the
participants of the requirements for the assignment: the video should target at
international audiences, the content should focus on one cultural aspect or
artefact, the video should be about five minutes long with subtitles, and there
should be collaboration between the members in each 3-4-person group.
Then
the participants had 12 weeks to complete the group video project. During Weeks
2 to 13, they were allowed to consult the teacher or give the teacher feedback
whenever they felt necessary. They submitted their videos in Week 14 and took
the questionnaire survey online in Week 15.
3.4
Data Analysis
The
responses to the questionnaire were converted to scores for practices and
perceptions of translanguaging in the process and product of DMC, and the
scores were then analyzed statistically using SPSS (29.0). To account for
individual differences, linear mixed-effects models (LMMs) were built to make
comparisons.
An
overall LMM was first built to explore the participants’ translanguaging
practices in the process and product of DMC. The dependent variable was score
(1-4), the fixed factors were language (English, Mandarin, dialect), phase of
DMC (process, product), and their interaction. Then two additional LMMs were
built for process and product. The dependent variable was score, the fixed
factors were language, process activity (information search, group discussion,
real-life communication) or product component (image, subtitle, voice-over) and
their interaction.
Perceptions
were analyzed in the same fashion. First an overall LMM was built for the
participants’ perceptions of translanguaging functions in the process and
product of DMC. The dependent variable was score (1-5), the fixed factors were
language (English, Mandarin, dialect), phase of DMC (process, product), and
their interaction. Two additional LMMs were built for process and product. The
dependent variable was score, the fixed factors were language, process function
(accessing information, collaborating with teammates, interacting with
community, learning) or product function (expressing identity, augmenting
meaning, negotiating meaning, learning) and their interaction.
In
all six LMMs, the random factor was intercepts of participants.
The
videos were analyzed in terms of image, subtitle, and voice-over. Specifically,
the subtitles of each video were marked as English only, Chinese only, or both
English and Chinese. The duration of the images with non-English language
(including Mandarin or Chinese dialects) in each video was calculated in second
and then converted to percentage by dividing it by the total duration of the
whole video. Likewise, the duration of non-English language in the voice-over
was calculated and converted to percentage.
4. Results
Analyses
of the questionnaire responses and videos revealed variations in the
participants’ practices and perceptions of translanguaging in the DMC project.
4.1
Translanguaging Practices in the Process and Product of DMC
The
participants’ reported translanguaging practices in the process and product of
DMC are visualized in Figure 1. The fixed effects of overall LMM (Table 2)
showed that the participants used more Mandarin (B = 1.16, SE = 0.14, p < .001) but less dialects (B = -0.57, SE = 0.07, p < .001) than English, and there was more
translanguaging in the product of DMC than in its process (B = 1.20, SE = 0.11, p < .001). In addition, there were interactions
between language and phase of DMC (Bs = -1.30, -1.81, SEs = 0.13, 0.22, ps < .001). The random intercepts of participants
were significant (B = 0.06, SE = 0.02, p < .001), suggesting participant-level
variability.
Figure
1. Comparisons of translanguaging practices: overall (top), in process
(middle), and in product (bottom)
Table 2. Fixed-effects of overall LMM for
translanguaging practices. The reference for language was English and for phase
of DMC was process (same in Tables 3-4).
|
|
B |
SE |
Z |
95% CI |
p |
|
|
|
|
Intercept |
2.16 |
0.09 |
24.69 |
1.99 |
2.33 |
<.001*** |
|
|
|
Language=Dialect |
-0.57 |
0.07 |
-8.13 |
-0.71 |
-0.43 |
<.001*** |
|
|
|
Language=Mandarin |
1.16 |
0.14 |
8.50 |
0.90 |
1.43 |
<.001*** |
|
|
|
Phase=Product |
1.20 |
0.11 |
10.63 |
0.98 |
1.42 |
<.001*** |
|
|
|
Dialect x Product |
-1.30 |
0.13 |
-10.37 |
-1.55 |
-1.05 |
<.001*** |
|
|
|
Mandarin x Product |
-1.81 |
0.22 |
-8.42 |
-2.23 |
-1.39 |
<.001*** |
|
|
|
Marginal R2 |
.48 |
|
|
|
|
|
|
|
|
Conditional R2 |
.53 |
|
|
|
|
|
|
|
The
fixed-effects of the LMM for process (Table 3a) showed
that overall the participants used less dialects (B = -0.61, SE = 0.11, p < .001) but more Mandarin (B = 1.02, SE = 0.15, p < .001) than English. They used all languages
comparably across the activities (ps > .136).
There was an interaction between language and activity (B = 0.31, SE = 0.14, p = .026). Simple main effects (Table 3b) revealed
that although in all three types of activities during the process of DMC, the
participants used less dialects and more Mandarin than English (all ps < .001), they used relatively more Mandarin in
group discussion (B
= 1.33, SE = 0.17, p < .001) than in information search and
real-life communication (Bs = 1.02, 1.14, SEs = 0.15, 0.17, ps < .001).
Table 3. LMM for translanguaging practices in process of DMC
a. Fixed-
and random effects
|
|
B |
SE |
Z |
95%CI |
p |
|
|
Intercept |
2.22 |
0.10 |
21.90 |
2.02 |
2.42 |
<.001*** |
|
Language=Dialect |
-0.61 |
0.11 |
-5.47 |
-0.83 |
-0.39 |
<.001*** |
|
Language=Mandarin |
1.02 |
0.15 |
6.89 |
0.73 |
1.31 |
<.001*** |
|
Activity=Real-life Communication |
-0.02 |
0.12 |
-0.16 |
-0.26 |
0.22 |
.873 |
|
Activity=Group Discussion |
-0.16 |
0.11 |
-1.49 |
-0.36 |
0.05 |
.136 |
|
Dialect x Real-life Communication |
0.08 |
0.14 |
0.58 |
-0.19 |
0.35 |
.563 |
|
Dialect x Group Discussion |
0.04 |
0.13 |
0.30 |
-0.22 |
0.30 |
.763 |
|
Mandarin x Real-life Communication |
0.12 |
0.17 |
0.71 |
-0.21 |
0.44 |
.478 |
|
Mandarin x Group Discussion |
0.31 |
0.14 |
2.24 |
0.04 |
0.59 |
.026* |
|
Random Intercepts of Participants |
0.05 |
0.02 |
2.55 |
0.03 |
0.12 |
.011* |
|
Marginal R2 |
.51 |
|
|
|
|
|
|
Conditional R2 |
.56 |
|
|
|
|
|
b. Simple
main effects
|
|
|
B |
SE |
Z |
95% CI |
p |
|
|
Info.
Search |
Intercept |
2.22 |
0.10 |
21.90 |
2.02 |
2.42 |
<.001*** |
|
Dialect
vs. English |
-0.61 |
0.11 |
-5.47 |
-0.83 |
-0.39 |
<.001*** |
|
|
Mandarin
vs. English |
1.02 |
0.15 |
6.89 |
0.73 |
1.31 |
<.001*** |
|
|
Group
Discus.
|
Intercept |
2.06 |
0.11 |
18.91 |
1.84 |
2.27 |
<.001*** |
|
Dialect
vs. English |
-0.57 |
0.09 |
-6.40 |
-0.74 |
-0.39 |
<.001*** |
|
|
Mandarin
vs. English |
1.33 |
0.17 |
7.84 |
1.00 |
1.67 |
<.001*** |
|
|
Real- life
Com. |
Intercept |
2.20 |
0.12 |
18.66 |
1.96 |
2.43 |
<.001*** |
|
Dialect
vs. English |
-0.53 |
0.10 |
-5.23 |
-0.73 |
-0.33 |
<.001*** |
|
|
Mandarin
vs. English |
1.14 |
0.17 |
6.83 |
0.81 |
1.47 |
<.001*** |
|
The
fixed-effects of LMM for product (Table 4a) revealed that the participants used
less Mandarin (B = -0.53, SE = 0.22, p
= .019) and dialects (B = -1.63, SE = 0.14, p < .001) than English in the product of DMC, and
they used more languages in subtitle (B = 0.26, SE = 0.09, p = .004) and voice-over (B = 0.28, SE = 0.09, p = .002) than in image. There were interactions
between language and product component (Bs = -0.55-0.20, SEs = 0.11-0.22, ps =.002-.381). Simple main effects (Table 4b) indicated
that the participants used less Mandarin and dialects than English in all three
components of their videos, but their use of Mandarin was significantly less
than English in voice-over (B = -1.08, SE = 0.21, p < .001) compared with that in image (B = -0.53, SE = 0.22, p = .019) and subtitle (B = -0.33, SE = 0.16, p = .043).
Table 4. LMM for translanguaging practices in product of DMC
a. Fixed-and
random effects
|
|
B |
SE |
Z |
95%
CI |
p |
|
|
Intercept |
3.18 |
0.11 |
28.03 |
2.95 |
3.40 |
<.001*** |
|
Language=Dialect |
-1.63 |
0.14 |
-11.52 |
-1.91 |
-1.35 |
<.001*** |
|
Language=Mandarin |
-0.53 |
0.22 |
-2.36 |
-0.97 |
-0.09 |
.019* |
|
Component=Voice-over |
0.28 |
0.09 |
3.12 |
0.10 |
0.45 |
.002** |
|
Component=Subtitle |
0.26 |
0.09 |
2.93 |
0.08 |
0.43 |
.004** |
|
Dialect x Voice-over |
-0.35 |
0.11 |
-3.10 |
-0.58 |
-0.13 |
.002** |
|
Dialect x Subtitle |
-0.37 |
0.12 |
-3.09 |
-0.61 |
-0.14 |
.002** |
|
Mandarin x Voice-over |
-0.55 |
0.22 |
-2.45 |
-0.99 |
-0.11 |
.015* |
|
Mandarin x Subtitle |
0.20 |
0.22 |
0.88 |
-0.24 |
0.64 |
.381 |
|
Random Intercepts of Participants |
0.05 |
0.02 |
2.04 |
0.02 |
0.13 |
.041* |
|
Marginal R2 |
.49 |
|
|
|
|
|
|
Conditional R2 |
.52 |
|
|
|
|
|
b. Simple main effects
|
|
|
B |
SE |
Z |
95% CI |
p |
|
|
Image |
Intercept |
3.18 |
0.11 |
28.03 |
2.95 |
3.40 |
<.001*** |
|
Dialect
vs. English |
-1.63 |
0.14 |
-11.52 |
-1.91 |
-1.35 |
<.001*** |
|
|
Mandarin
vs. English |
-0.53 |
0.22 |
-2.36 |
-0.97 |
-0.09 |
.019* |
|
|
Sub- title
|
Intercept |
3.43 |
0.10 |
35.35 |
3.24 |
3.62 |
<.001*** |
|
Dialect
vs. English |
-2.00 |
0.14 |
-14.15 |
-2.28 |
-1.72 |
<.001*** |
|
|
Mandarin
vs. English |
-0.33 |
0.16 |
-2.04 |
-0.65 |
-0.01 |
.043* |
|
|
Voice- over |
Intercept |
3.45 |
0.10 |
35.47 |
3.26 |
3.64 |
<.001*** |
|
Dialect
vs. English |
-1.98 |
0.15 |
-13.15 |
-2.28 |
-1.68 |
<.001*** |
|
|
Mandarin
vs. English |
-1.08 |
0.21 |
-5.07 |
-1.50 |
-0.66 |
<.001*** |
|
This
pattern of language use is supported by the analysis of the videos (Table 5).
Of the ten videos, six used both English and Chinese in their subtitles, all
had very limited Chinese in images (ranging from 1.09% to 5.26%), and only one
contained one second of Mandarin (0.30%) in the background of the English
narration.
Table 5. Language use in images, subtitles, and voice-overs
|
Video No. |
Length |
Non-English Image in Second (%) |
Language(s) in Subtitle |
Non-English Voice-over in Second (%) |
|
1 |
5’34” |
11(3.29%) |
English |
1(0.30%) |
|
2 |
4’35” |
3(1.09%) |
English
& Mandarin |
- |
|
3 |
5’04” |
16(5.26%) |
English
& Mandarin |
- |
|
4 |
5’04” |
5(1.65%) |
English
& Mandarin |
- |
|
5 |
5’22” |
13(4.04%) |
English
& Mandarin |
- |
|
6 |
7’01” |
6(1.43%) |
English
& Mandarin |
- |
|
7 |
4’49” |
4(1.38%) |
English |
- |
|
8 |
6’07” |
6(1.64%) |
English |
- |
|
9 |
3’16” |
10(5.10%) |
English
& Mandarin |
- |
|
10 |
6’52” |
10(2.43%) |
English |
- |
In
summary, the participants demonstrated different translanguaging practices in
the process and product of the DMC project. In the process, they relied heavily
on their L1 Mandarin, especially for group discussion, whereas in the product,
they mostly used English, with Chinese in subtitles and very small fractions of
images.
4.2
Perceptions of Translanguaging in the Process and Product of DMC
The
participants’ perceptions of the functions of translanguaging in the process
and product of DMC are visualized in Figure 2. The fixed-effects
of the overall LMM showed that the participants were more affirmative about the
functions of translanguaging in the product than in the process, B = 0.14, SE = 0.47, p = .003. The intercepts of participants were
statistically significant, B
= 0.23, SE = 0.05, p < .001. Marginal R² was .01, showing that phase of DMC alone explained
only 1% of the variance. Conditional R² was .63, reflecting that adding the random effect
increased the explained variance to 63%.
Figure
2. Comparisons of perceptions of translanguaging functions: overall (top), in
process (middle), and in product (bottom)
The
fixed-effects of the LMM for perceptions of
translanguaging in the process of DMC showed variations by function (p < .001). The intercepts of participants were
statistically significant, B
= 0.22, SE = 0.05, p < .001. Marginal R² was .01, meaning that function alone explained
merely 1% of the variance. Conditional R² was .62, indicating that the fixed and random
effects together explained 62% of the variance. Post-hoc pairwise comparisons
(Table 6) revealed that the participants were more affirmative about
translanguaging for accessing information (B = 0.16, SE = 0.06, p = .004) and learning (B = 0.19, SE = 0.07, p = .008) than collaborating with peers.
Table 6. Pairwise comparisons between the
functions of translanguaging in process
|
Comparisons |
B |
SE |
Z |
95% CI |
p |
|
|
Accessing Info. vs. Collaborating w. Peer |
0.16 |
0.06 |
2.88 |
0.05 |
0.26 |
.004** |
|
Accessing Info. vs. Interacting w. Com. |
0.06 |
0.08 |
0.71 |
-0.10 |
0.22 |
.478 |
|
Accessing Info. vs. Learning |
-0.03 |
0.08 |
-0.35 |
-0.19 |
0.14 |
.726 |
|
Collaborating w. Peer vs. Interacting w. Com. |
-0.10 |
0.08 |
-1.20 |
-0.26 |
0.06 |
.234 |
|
Collaborating w. Peer vs. Learning |
-0.19 |
0.07 |
-2.69 |
-0.32 |
-0.05 |
.008** |
|
Interacting w. Com. vs. Learning |
-0.09 |
0.06 |
-1.40 |
-0.21 |
0.04 |
.163 |
The
fixed-effects of the LMM for perceptions of
translanguaging in the product of DMC showed a marginal main effect of function
(p = .078). The intercepts of participants were
statistically significant, B
= 0.26, SE = 0.05, p < .001. Marginal R² was .002, indicating that function alone
explained only 0.2% of the variance. Conditional R² was .88, revealing that the
fixed and random effects together explained 88% of the variance. Pairwise
comparisons (Table 7) revealed less affirmative
perceptions for enhancing meaning (B = -0.05, SE = 0.02, p = .020) and marginally less affirmative perceptions
for engaging audiences (B
= -0.07, SE = 0.04, p = .082) than learning.
Table 7. Pairwise comparisons between the
functions of translanguaging in product
|
Comparisons |
B |
SE |
Z |
95% CI |
p |
|
|
Constructing Identity vs. Enhancing Meaning |
0.02 |
0.04 |
0.47 |
-0.06 |
0.10 |
.637 |
|
Constructing Identity vs. Engaging Audience |
0.04 |
0.04 |
0.95 |
-0.04 |
0.12 |
.343 |
|
Constructing Identity vs. Learning |
-0.03 |
0.04 |
-0.73 |
-0.11 |
0.05 |
.465 |
|
Enhancing Meaning vs. Engaging Audience |
0.02 |
0.04 |
0.50 |
-0.06 |
0.10 |
.617 |
|
Enhancing Meaning vs. Learning |
-0.05 |
0.02 |
-2.35 |
-0.09 |
-0.01 |
.020* |
|
Engaging Audience vs. Learning |
-0.07 |
0.04 |
-1.75 |
-0.15 |
0.01 |
.082 |
In
brief, the participants were more affirmative about translanguaging in the
product of DMC than in its process. For process, they valued the functions of
accessing information and learning more than collaborating with peers. For
product, they valued the function of learning more than enhancing meaning and
engaging audiences.
5. Discussion
This
study investigated emergent Chinese-English
bilinguals’ practices and perceptions of translanguaging in the process and
product of a DMC project, revealing some patterns of “shifts” and “stance” of
translanguaging in DMC.
5.1
Practices of Translanguaging in Process and Product of DMC
The
first research question asked about the participants’ translanguaging practices
in the process and product of DMC. The results showed that the participants
reported more translanguaging practices in the product of DMC (i.e., the video)
than in its process. Moreover, their translanguaging practices differed in the
process and product of DMC. In the process, they reported primarily using
Mandarin and translanguaging to English and Chinese dialects. In the product,
they reported primarily using English and translanguaging to Mandarin and
Chinese dialects, which was supported by the qualitative analysis of their
videos. These findings offer insights into how the emergent bilinguals
strategically deploy their linguistic repertoires across the different phases
of DMC, aligning with and extending existing research findings of
translanguaging in L2 DMC contexts.
First,
the finding that the participants reported more frequent translanguaging
practices in the DMC product (i.e., the video) than in the DMC process
highlights that translanguaging serves a critical performative function in the
final multimodal artifact, which supports Ho and Tai’s (2024) framework
distinguishing interactional and performative translanguaging spaces in DMC. In
the product phase, the participants skillfully used translanguaging as a
deliberate design choice to enhance meaning-making and self-expression, which
are goals inherent to the public nature of DMC products compared to the more
private or collaborative nature of the composing process. The qualitative
analysis of the videos further corroborates this with tangible evidence that
translanguaging in the product was not accidental but a purposeful act of
multimodal meaning-making.
Second,
the distinct patterns of translanguaging across the DMC process and product
reflect the dynamic alignment of linguistic choices with the goals and
constraints of each phase. During the DMC process, the participants likely
prioritized communicative efficiency when searching for information or
discussing in groups, leading them to rely on Mandarin, their shared dominant
language, while using English and dialects to clarify specific concepts, share
technical terminology, or draw on cultural-specific expressions. They also
translated information from Chinese to English, as revealed by some
participants during their consultation with the course teacher. This supports
the previous finding that L1 is often used as a supportive language by emergent
bilinguals during communication (Trinh, 2025; Wang, 2016) and reveals the role
of translanguaging as a scaffolding tool for collaborative problem-solving in
interactive contexts. In contrast, the primary use of English in the final
video product suggests that the participants were attuned to the requirements
of the DMC task, which positions English as the default language for their
videos, while still leveraging Mandarin and dialects to enrich content (e.g.,
including Chinese menus and interviews), convey cultural nuances (e.g., showing
Chinese characters at entrances of restaurants), or ensure clarity (e.g.,
English-Chinese subtitles). This pattern suggests that emergent bilinguals are
not passive language users but active agents who negotiate between task requirements,
audience expectations, and their own linguistic repertoires.
In
conclusion, the different patterns of translanguaging in the process and
product of DMC demonstrate the participants’ “shifts” in their translanguaging
practices, revealing that these learners can actively adjust their language use
in response to different purposes, communicative needs, and contextual demands
when engaging in the DMC project.
5.2
Perceptions of Translanguaging in Process and Product of DMC
The
second research question was about the participants’ perceptions of
translanguaging in the process and product of DMC. Overall, these emergent
bilinguals were positive about translanguaging in DMC, with more affirmative
attitudes to translanguaging in the product than in the process. Moreover, they
were more affirmative about accessing information and learning than
collaborating with their group members in the process, and more affirmative
about learning than enhancing meaning and engaging audiences in the product.
These findings further enrich our understanding of how emergent bilinguals view
translanguaging as a meaning-making resource in relation to task phases and
communicative goals.
The
overall positive attitudes toward translanguaging in DMC align with studies
highlighting emergent bilinguals’ recognition of translanguaging for
facilitating L2 learning (e.g., Trinh, 2025; Wang & Shen, 2023; Zhou &
Mann, 2021). More notably, the stronger affirmative attitudes toward
translanguaging in the product than in the process echo the earlier finding of
more frequent translanguaging practices in the DMC product, suggesting a
consistency between the participants’ perceptions and their actual linguistic
behaviors. This discrepancy in attitudes across the DMC phases may stem from
the distinct functions of translanguaging in each context. In the final
product, translanguaging is a deliberate, creative choice that enhances the
quality of the DMC product, which likely brings a more positive evaluation. In
contrast, translanguaging during the process may be perceived as a supportive
tool rather than a value-adding practice, thus evoking a slightly less
affirmative stance.
Moreover,
the differences in attitudes toward translanguaging for specific purposes
reveal that these emergent bilinguals’ perceptions are context-dependent.
During the DMC process, translanguaging for accessing information and
facilitating learning directly addresses the participants’ individual needs,
leading to their positive attitudes. In contrast, translanguaging for
collaboration may involve more negotiation and potential misalignment of
linguistic preferences, which could weaken the participants’ positive
perceptions. In the DMC product, the participants’ stronger affirmation of
translanguaging for learning reflects their inherent awareness of DMC as a course
assignment, where learning gains are a core expectation. The relatively weaker
attitudes toward translanguaging for meaning enhancement and audience
engagement may stem from the participants’ belief that English use should be
prioritized for these functions in L2 tasks, or their limited experience in
designing multimodal products for specific audiences in real-life contexts.
This suggests that while emergent bilinguals recognize translanguaging’s value
for learning, they may need more guidance to fully leverage its potential for
effective multimodal communication.
Taken
together, the participants’ perceptions reflect their “stance” toward
translanguaging, which is generally favorable yet predominantly
learning-oriented, prioritizing translanguaging’s function for learning over
its function for promoting interpersonal interactions.
5.2
Pedagogical Implications
These
findings carry important implications for L2 teachers implementing
translanguaging pedagogies in CLIL contexts. Specifically, the distinct
patterns and functions of translanguaging in the DMC process and product
highlight the need for pedagogical designs that explicitly guide learners to
recognize and deploy translanguaging across both phases. For example, teachers
could incorporate pre-composing activities that guide learners to plan how they
will use translanguaging in their final product and reflective activities that
help learners articulate the rationale behind their linguistic choices in both
process and product.
Moreover,
the unbalanced perceptions of translanguaging’s functions in the process and
product of DMC necessitate teachers’ explicit guidance to raise learners’
awareness of the various benefits and functions of translanguaging. To this
end, teachers need to highlight how translanguaging can support collaboration,
enhance content quality, engage audience, and foster identity expression. This
can be achieved by discussing how translanguaging contributes to different
goals and analyzing exemplary DMC products that effectively use translanguaging
for these goals.
6. Conclusion
In
conclusion, this study’s findings on emergent bilinguals’ translanguaging
practices and perceptions in DMC reveal two core insights. First, emergent
bilinguals’ translanguaging practices differ across the process and product of
DMC, reflecting strategic alignment with communicative needs. Second, their
overall stance toward translanguaging is favorable but predominantly
learning-oriented, prioritizing its utility for information access and
individual learning over interpersonal functions. These results indicate that
DMC affords rich translanguaging spaces where emergent bilinguals actively
adjust their language use, while also highlighting the need for a
translanguaging-informed DMC pedagogy that validates both interactional and
performative translanguaging, and guides learners to leverage translanguaging’s
multifaceted value beyond individual learning. Ultimately, this study
contributes to a nuanced understanding of translanguaging in DMC, laying a
foundation for more inclusive and effective CLIL instruction.
This
study has its limitation. Data on DMC process and perceptions relied on
self-reports, which may suffer from inaccuracy or social desirability bias.
Additionally, more qualitative data on situational language use is needed to
capture emergent bilinguals’ linguistic repertoire deployment in DMC and their
perceptions. To address this, future research should collect fine-grained data
through interviews and audio/video recordings of in-process translanguaging.
Despite this limitation, the findings advance our understanding of
translanguaging in DMC and highlight the need to consider task phases and
purposes when designing translanguaging-informed DMC pedagogy.
Acknowledgements:
This work was supported by Xiangtan University’s Annual Teaching Reform Project
(2025) “Reform and Practice of the Integrated English Course for English Majors
under the ‘Double First-Class’ Initiative”.
References
Ainsworth, S., Griffiths, D., Macrory, G., & Pahl,
K. (Eds.). (2023). Multimodality and multilingualism: Towards an
integrative approach. Channel View Publications.
Alrayes, M. (2024). Correlation factors of translanguaging
practices in English as a foreign language contexts:
The perception of teachers. Education and Information
Technologies, 29, 7659-7669. https://doi.org/10.1007/s10639-023-12137-1
García, O., Johnson, S. I., & Seltzer, K. (2017). The translanguaging classroom: Leveraging student
bilingualism for learning. Caslon.
Ho, W. Y. J. (2022). The construction of translanguaging space
through digital multimodal composing: A case study of students’ creation of
instructional videos. Journal of English for
Academic Purposes, 58, 101134. https://doi.org/10.1016/j.jeap.2022.101134
Ho, W. Y. J., & Tai, K. W. H. (2024).
Translanguaging in digital learning: the making of translanguaging spaces in
online English teaching videos. International
Journal of Bilingual Education and Bilingualism, 27(9), 1212-1233. https://doi.org/10.1080/13670050.2021.2001427
Jiang, L. (2018). Digital multimodal composing and
investment change in learners’ writing in English as a foreign language. Journal of Second Language Writing, 40, 60-72. https://doi.org/10.1016/j.jslw.2018.03.002
Jiang, L., & Hafner, C. (2025). Digital multimodal
composing in L2 classrooms: A research agenda. Language Teaching, 58(4), 528-546. https://doi.org/10.1017/S0261444824000107
Kessler, M. (2024). Digital multimodal composing: Connecting theory,
research and practice in second language acquisition. Multilingual Matters. https://doi.org/10.2307/jj.22679723
Li, D., & Wang, Y. (2024). Translanguaging as a
mediating and regulating tool in the second language writing feedback
process. Chinese Journal of Applied
Linguistics, 47(1), 121-136, 159-160.
Li, W. (2018). Translanguaging as a practical theory of
language. Applied Linguistics, 39(1), 9-30. https://doi.org/10.1093/applin/amx039
Liu, D., Deng, Y., & Wimpenny, K. (2024). Students’
perceptions and experiences of translanguaging pedagogy in teaching English for
academic purposes in China. Teaching
in Higher Education, 29(5),
1234-1252. https://doi.org/10.1080/13562517.2022.2129961
Pacheco, M. B., Smith, B. E., Combs, E., & Amgott, N. A. (2022). Translanguaging within multimodal
composition products and processes: A systematic review. Pedagogies: An International Journal, 17(4), 389-407. https://doi.org/10.1080/1554480X.2022.2139256
Talley, J. (2025). Community college faculty perceptions
and use of translanguaging in academic ESL. International Journal of Bilingual Education and
Bilingualism, 1-20. https://doi.org/10.1080/13670050.2025.2531246
Trinh, N. K. (2025). EFL learners’ perceptions of
translanguaging in English classes at an HCM English Centre. International Journal of Language
Instruction, 4(1). https://doi.org/10.54855/ijli.25414
Wang, D. (2016). Translanguaging in Chinese foreign language
classrooms: students and teachers’ attitudes and practices. International Journal of Bilingual Education and
Bilingualism, 22(2), 138-149. https://doi.org/10.1080/13670050.2016.1231773
Wang, D. (2019). Multilingualism and translanguaging in Chinese language
classrooms. Palgrave Macmillan. https://doi.org/10.1007/978-3-030-02529-8_3
Wang, D. (2024). Digital multimodal composing as a
translanguaging space: Understanding students’ initial experiences and
challenges. Chinese Language and
Discourse, 15(2), 164-187. https://doi.org/10.1075/cld.24016.wan
Wang, Z., & Shen, B. (2023). “I think I am
bilingual, but…”: Teachers’ practices of and students’ attitudes toward
translanguaging in a Chinese intercultural communication class. International Journal of Applied
Linguistics, 34(1), 242-260. https://doi.org/10.1111/ijal.12491
Wang, Z., Shen, B., Chen, Y., & Fang, F. (2025).
Unpacking teachers’ beliefs about translanguaging in language learning: A
qualitative meta-synthesis. International
Journal of Multilingualism, 22(3), 1774-1793. https://doi.org/10.1080/14790718.2025.2463539
Wu, J. G., Yang, Z., Lee, S., &
Teng, M. F. (2025). Multimodal composition
as a catalyst for technology self-efficacy. International Journal of Applied Linguistics. https://doi.org/10.1111/ijal.12860
Zhang, M., Akoto, M., & Li, M. (2023). Digital
multimodal composing in post-secondary L2 settings: A review of the empirical
landscape. Computer Assisted
Language Learning, 36(4), 694-721. https://doi.org/10.1080/09588221.2021.1942068
Zhou, D., & Pan, L. (2025). How Chinese senior high
English teachers utilize and perceive translanguaging practices in classrooms:
A mixed-methods study. New Perspectives on Languages, 3,
283500818. https://doi.org/10.56395/n41p3331
Zhou, X., & Mann, S. (2021). Translanguaging in a
Chinese university CLIL classroom: Teacher strategies and student attitudes. Studies in Second Language Learning and
Teaching, 11(2), 265-289. https://doi.org/10.14746/ssllt.2021.11.2.5
Author
bio
Conchao Hua is an Assistant Professor
at City University of Macau. She obtained her PhD degree in linguistics from
City University of Hong Kong. Her research interests mainly include second
language pronunciation teaching and learning, second language prosody
acquisition, and learner autonomy in second language pronunciation learning.
Her research has appeared in journals such as Language Teaching Research,
Journal of Multilingual and Multicultural Development, International Review of
Applied Linguistics, and Asia Pacific Journal of Education.
Email Address: cchua@cityu.edu.mo
ORCID: https://orcid.org/0000-0002-7200-6773
Yongli Liu is an Associate
Professor at Xiangtan University, specializing in English language teaching and
translation studies. Her academic focus lies in language acquisition and the
theoretical and practical aspects of translation. Her work aims to enhance the
quality and effectiveness of translation education, offering valuable insights
and practical strategies for both the academic community and translation
practitioners.
Email Address: liuyongli@xtu.edu.cn
ORCID: https://orcid.org/0009-0004-7217-0030