General Instructions

1
2
3
4

Research Methodology
 General Instructions
Do not leave a cell blank. Use "It does not apply" if not relevant
Use the Notes field to list anything that is noteworthy: interesting, confusing, needs clarification, etc
If using quotes, include page number and use " " to clearly indicate a quote
Line returns within a cell can help the readablity of the text, keep results/themes/statements together

A line return can be inserted in a cell. Place the cursor at the end of the line of text where you want a new line to be
* on a Mac, the command is: Option+Return
* on a PC, the command is: Alt+Enter

Research Methodology
If unsure which research methodology is appropriate, refer to the MMAT diagram:
http://mixedmethodsappraisaltoolpublic.pbworks.com/w/file/fetch/127916259/MMAT_2018_criteria-manual_201
 Author(s) ((Please
Include last name and Year
first name initial))

Alonso, A. 2022
Anson, C. M. 2022
Chen, M. H., Huang, S. T.,
2015
Chang, J. S., & Liou H. C.
Dawson, P. 2020
Dinneen, C. 2021
Ducar, C, & Schocket, D.
2018
H.
Fyfe, P. 2022
Gero, K., Liu, V., &
2022
Chilton, L.
Groves, M., & K. Mundt 2022
Merine, R., Purkayastha,
2022
S.
Prentice, F. M., & Kinden,
2018
C. E.
Roe, J. & Perkins, M. 2022
Rogerson, A. M., &
2017
McCarthy, G.
 Sharples, M. 2022

Missing Full Text Studies - Interlibrary Loan Ordered, but not received/unable to fill
Exclude at Full Text with Reasons of No Full Text Avaialabe
Author(s) Year

Francke, W., &

2019
Alexander, B.

Khandelwal, K. 2021
 Citation (APA 7) Reviewer

Alonso, A. N. (2022). Online translators in online language

assessments. CALL-EJ, 23(3), 115-135.
Beatriz
http://callej.org/journal/23-3/Alonso2022.pdf
Anson, C. M. (2022). AI-Based text generation and the social
construction of “fraudulent authorship”: A revisitation.
Beatriz
Composition Studies, 50(1), 37-46.
Chen, M. H., et al. (2015). Developing a corpus-based
paraphrase tool to improve EFL learners' writing skills.
Computer Assisted Language Learning, 28(1), 22-40. Beatriz
http://dx.doi.org/10.1080/09588221.2013.783873
Dawson, P. (2020). Defending assessment security in a digital
world: Preventing e-cheating and supporting academic
integrity in higher education.
https://www.taylorfrancis.com/books/mono/10.4324/978042 Beatriz
9324178/defending-assessment-security-digital-world-phillip-
dawson
Dinneen, C. (2021). Students’ use of digital translation and
paraphrasing tools in written assignments on direct entry
english programs. English Australia Journal, 37(1). Beatriz
https://files.eric.ed.gov/fulltext/EJ1341751.pdf
 Ducar, C., & Schocket, D. H. (2018). Machine translation and
the L2 classroom: Pedagogical solutions for making peace with
Beatriz
Google Translate. Foreign Language Annals 51(4), 779-795.
Fyfe, P. (2022). How to cheat on your final paper: Assigning AI
for student writing. AI & Society.
Beatriz
https://link.springer.com/article/10.1007/s00146-022-01397-z
Gero, K. I., et al. (2022). Sparks: Inspiration for science writing
using language models. DIS '22: Designing Interactive Systems
Beatriz
Conference, Association for Computing Machinery, Inc.
Groves, M., & K. Mundt (2021). A ghostwriter in the machine?
Attitudes of academic staff towards machine translation use in
internationalised Higher Education. Journal of English for
Beatriz
Academic Purposes 50.
https://doi.org/10.1016/j.jeap.2021.100957
Merine, R., Purkayastha, S. (2022). Risks and benefits of AI-
generated text summarization for expert level content in
graduate health informatics. 2022 IEEE 10th International
Conference on Healthcare Informatics (ICHI), Institute of Beatriz
Electrical and Electronics Engineers Inc.
https://ieeexplore.ieee.org/document/9874678/
Prentice, F. M., & Kinden, C. E. (2018). Paraphrasing tools,
language translation tools and plagiarism: An exploratory
Beatriz
study. International Journal for Educational Integrity 14.
 Roe, J. & Perkins, M. (2022). What are Automated
Paraphrasing Tools and how do we address them? A review of
a growing threat to academic integrity. International Journal
Beatriz
for Educational Integrity 18(1).
Rogerson, A. M., & McCarthy, G. (2017). Using internet based
paraphrasing tools: Original work, patchwriting or facilitated
Beatriz
plagiarism? International Journal for Educational Integrity, 13.
 Sharples, M. (2022). Automated essay writing: An AIED
opinion. International Journal of Artificial Intelligence in
Beatriz
Education, 32(4), 1119-1126.

udies - Interlibrary Loan Ordered, but not received/unable to fill

at Full Text with Reasons of No Full Text Avaialabe
Citation (APA)

Francke, W., & Alexander, B. (2019). The potential influence of

artificial intelligence on plagiarism: A higher education
perspective. Proceedings of the European Conference on the
Impact of Artificial Intelligence and Robotics.

Khandelwal, K. (2021). Application of ai to education during

the global crisis. Review of International Geographical
Education Online, 11(7), 32004-3212
Country ((Please use list below))

United Kingdom
United States of America
Taiwan
Australia
Australia
United States of America
United States of America
United States of America
United Kingdom
United States of America
Australia
Singapore
Australia
United Kingdom
Multiple Countries ((If there is one than more country, please
add the names of other countries mentioned in the text))

It does not apply

It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
Vietnam
It does not apply
It does not apply
 Type of source Participants
Specific city(ies), town(s) and campus ((If applicable) ((Please use list ((Please use
below)) list below))

6=journal
It does not apply 6=multiple
article
 6=journal
It does not apply 6=multiple
article
 6=journal
National Tsing Hua University 2=students
article
It does not apply 2=book 6=multiple
 6=journal
It does not apply 2=students
article
 6=journal
It does not apply 6=multiple
article
 6=journal
NC State University 2=students
article
 6=journal
Columbia University 2=students
article
 6=journal
University of Surrey & University of Nottingham 1=Faculty
article
 4=conference
Indiana University 2=students
paper
 6=journal
It does not apply 1=Faculty
article
 6=journal
It does not apply 1=Faculty
article
 6=journal
University of Wollongong 1=Faculty
article
 6=journal
It does not apply 1=Faculty
article
 If multiple disciplines or
AI Tool Discipline
Other please list

Online
Humanities It does not apply
translator
 Text
Humanities It does not apply
generator
Paraphrasing
Humanities It does not apply
tool
AI in general Other Multidisciplinary
 Online
translator and
Humanities It does not apply
paraphrasing
tool
 Online
Humanities It does not apply
translator
 Computer
Text
Science/Engin It does not apply
generator
eering
 Computer
Text
Science/Engin It does not apply
generator
eering
 Online
Humanities It does not apply
translator
 Computer
Text
Science/Engin It does not apply
generator
eering
 Online
translator and
Other Health Sciences
paraphrasing
tool
Paraphrasing
Other Unclear
tool
Paraphrasing
Humanities It does not apply
tool
 Text
Other Educational Technology
generator
 Purpose(s) ((Please add direct quotes/page numbers))

"This study describes some of the main uses of online machine translation tools,
commenting on the pedagogical value of these, and presents the results of an online survey
that investigated language teachers’ perceptions of the impact that online language teaching
and assessment may have on the use of online translators and its connection with plagiarism
detection tools such as Turnitin and with other online language reference tools. Finally,
some reflections and recommendations will follow on good online teaching and assessment
practices and the implementation of online translators in the online language class." (p. 115)
"Here, I first revisit an article I wrote for Composition Studies in 2011, “Fraudulent Practices:
Academic Misrepresentations of Plagiarism in the Name of Good Pedagogy.” In that article, I
argued that what counts as plagiarism in some contexts occurs with impunity across a wide
range of published material. This is because definitions of plagiarism are socially constructed
and tied to context-sensitive cycles of reward for the production—and therefore the
ownership—of certain kinds of texts. Helping students to understand plagiarism means
showing them these contextually-specific constructs of text ownership, rather than assuming
that any unattributed text, published anywhere, in any form, constitutes plagiarism. I then
turn to AI-based NLP systems. Teachers who learn what these systems can do usually
respond with the same hand-wringing and defensive posture triggered by concerns about
student plagiarism." (p. 37)
"To meet the need of developing EFL learners’ paraphrasing competence through a CALL
tool, we developed a web-based paraphrase suggestion system and assessed its
effectiveness for the purposes of the study." (p. 25)
 "The book explores the current and likely future states of e-cheating, what can be done to
secure assessment in this new world, and some of the challenges we are likely to face along
the way." (p. 1)
"The purpose of this paper, however, is to look more closely at what is happening, and the
ramifications, rather than considering the why." (p. 42)
 "This article aims to advance the theoretical discussion surrounding MT in the L2 classroom
to help instructors understand what MT can and cannot do and help them equip their
students to use the technology in an educationally and interculturally respectful manner." (p.
780)
 "This paper shares results from a pedagogical experiment that assigns undergraduates to
“cheat” on a final class essay by requiring their use of text-generating AI software." (p. 1)

"In this paper, I share how students experienced those issues, connect their insights to
broader conversations in the humanities about writing and communication, and explain their
relevance for the ethical use and evaluation of language models." (p. 1)
 "In this work we study how language models can be applied to a real-world, high-impact
writing task: science writing. This introduces challenges different to those in traditional
creative writing tasks, such as writing stories and poetry, which tend to deal with common
objects and relations. Science writing support requires a system to demonstrate pro ciency
within an area of expertise." (p. 1003)

"Our work aims to study how text generated by language models might be used by writers in
a science writing task. There is relation to a natural language generation task like
summarization, because we are concerned with specific factual information (as opposed to
commonsense knowledge) but we take a human centered approach where the language
model provides suggestions, rather than a completed output." (p. 1004)
"To move this issue forward, the present study examines perceptions on the use of GT and
other freely available online translation tools by university students as reading and writing
aid." (p. 2)
 "Therefore, the purpose of this paper is to address the application of AI-generated text
summarization in academia and its risks and benefits to students. In addition, through an
experiment conducted among graduate health informatics students at Indiana University,
this paper would illustrate the quality of the text summary generated utilizing the
abstractive NLP technique." (p. 567)

"This study will look into the possibilities of using an advanced NLP technique to create texts
with expert knowledge content in the educational sector." (p. 570)
It does not apply
 "In this article, we aim to contribute to solving this problem through engaging in a detailed
literature review of a category of tool that may be used to commit academic misconduct by
aiding text-based plagiarism, that of Automated Paraphrasing Tools (APTs). We begin by
describing the origins of APTs and their use in academic work. We then explore the
relationship between language proficiency and APT use, and how APTs may or may not be
used for in an academically dishonest way, referring to case studies from Dinneen (2021)
and Prentice and Kinden (2018). Finally, we propose solutions and relevant limitations to
tackling the problem of APTs in academia, as well as areas for future research." (p. 2)
 "Having had our attention drawn to the existence and use of paraphrasing tools it was
decided to investigate the phenomenon. What became apparent was that the ease of access
to and use of such tools was greater than first thought. Consequently it is important to bring
the use and operation of paraphrasing tools to a wider audience to encourage discussion
about developing individual writing skills and improve the detection of these emerging
practices, thereby raising awareness for students, teachers and institutions." (p. 2)
"How do such Transformer AI systems work? How can we detect whether an aca- demic
essay has been written by an AI? Could any good come from widespread use of such
technology? What does this mean for the AIED community?" (p. 1120)
 Research question(s) ((Please add direct quotes/page numbers))

"This pedagogical research reports on results from an online survey directed to mostly higher
education and secondary education teachers. The main aims of this research are the following:
1. How do students and teachers use online translators in language class?
2. What are language teachers’ opinions on the impact of online translators in online summative
assessment types and the various levels of learning, on online plagiarism detection tools, and on the
use of online translators in combination with other online
language reference tools?
3. What can be done to improve online assessment practice, preserve academic integrity, and
integrate online translators as valuable pedagogical tools in the language class?" (p. 120)
It does not apply
"1) Which reference tool best benefits EFL learners’ writing performance: PREFER, Longman English
Dictionary Online or Thesaurus.com?
(2) To what extent does PREFER benefit different EFL learners’ paraphrasing performance?
(3) What are the EFL learners’ perceptions of paraphrasing and the effectiveness of PREFER?" (p. 25)
It does not apply
It does not apply
"What challenges and opportunities does this technology present to L2 students and instructors? Is it
possible to deter students from overusing this technology? How can teaching students about the
strengths and weaknesses of this technology transform some of the challenges into learning
opportunities?" (p. 779)
"What would that look like? While many writers are familiar with various assistive technologies, there
are practically no examples for training writers with text- generating AI and language models like GPT-
2. Thus, the research questions from my experiment also explore students’ strategies for AI-assisted
composition. How might we write with these tools? What skills might writing with AI newly require?
How does this challenge our assumptions about textual communication? And what potential risks and
harms must we navigate? “Cheating” offers a starting point, but it opens onto much more complex and
consequential topics, from human and nonhuman agency to the threats of disinformation and
algorithmic bias. Thus, the assignment moves students from a familiar ethical context (is this
plagiarism?) to new ethical questions (whose writing is this? what effects does it have?) and to new
forms of writing practice (can I collaborate with AI?)." (p. 2)
 "We structure our work around the following research question:
RQ: How can language model outputs support writers in a creative but constrained writing task?" (p.
1003)

"In particular, we pose the following research questions:

RQ1: In what ways do writers make use of language model outputs? RQ2: What attributes of language
model outputs, if any, correlate with writer usage and satisfaction?" (p. 1010)
 "overarching research questions:
1. To what degree do academic staff see the use of MT by students for their academic work as
acceptable?
2. Is use of MT as an aid for academic reading and writing considered academic misconduct; and, if so,
under what
circumstances?
3. Is there currently a university policy on student use of MT? Should there be?" (p. 9)
It does not apply
 "This led us to consider the following questions:
1. Were students using online paraphrasing tools to manipulate work which was written in English and
which had not been authored by them?
2. Were students who had English as an Additional Language (EAL) composing work in their first
language and then translating this through online language translation tools?
3. Are there indicators which can identify the use of on-line paraphrasing tools?" (p. 1)
It does not apply
It does not apply
It does not apply
Study Design ((Please use list below))

Quantitative Descriptive
Theoretical
Quantitative Randomized
Theoretical
Theoretical
Theoretical
Qualitative
Mixed Methods
Qualitative
Quantitative Descriptive
Qualitative
Theoretical
Qualitative
Theoretical
 Participants Information (Added on March 10, 2023)

"the participants (n=32) were teaching in Higher Education

(90%), although some of them were also teaching in secondary
schools and most of them have experience teaching various
languages such as Spanish, French, English, German, Italian,
Chinese, Arabic, Russian, Korean and Danish, and at various
levels of the CEF from A1 to C2 (see Figure 1 below)." (p. 120)
It does not apply
 "An assessment study of the effectiveness of PREFER was
conducted with 55 Chinese- speaking EFL college freshmen in
an Asian country." (p. 23)

"A total of 64 Chinese-speaking EFL college freshmen in a

public university of an Asian country were recruited for
participation in this study. These students came from the De-
partment of Mathematics (n = 34), Material Science and
Engineering (14), Chemical En- gineering (12), and Industrial
Engineering and Engineering Management (4). The students
had at least six years of formal instruction from junior to senior
high schools and were estimated to be at the intermediate
level regarding their overall English competence." (p. 28)
It does not apply
It does not apply
It does not apply
"The class enrolled 20 students from first-years to seniors with
a variety of majors and con- centrations." (p. 4)
Study 1: "For biology, we had 3 senior undergraduate students
majoring in biology; for environ- mental science, we had 2
senior undergraduate students majoring in environmental
science; for computer science, we had 2 PhD students from
the computer science department." (p. 1008)

Study 2: "We recruited 13 STEM graduate students" (p. 1011)

"Participants were recruited at two UK universities, resulting in
data being collected from eleven members of staff (!ve from
one university, six from the other). The universities are
members of the Russell Group, a coalition of high-ranking UK
universities. After ethical approval was granted by the ethics
committee at both universities, the researchers used their
professional networks to identify potential participants.
Obtaining informed consent from the participants involved a
written introduction of the aims and processes of the research,
along with the assurance of anonymity of the participants and
the option to withdraw from the study at any time.
Staff from language departments and EAP units were
deliberately excluded, since it was anticipated that their take
on student use of MT would be affected by their speci!c
interest in language acquisition. Their opinions would warrant
a separate study taking into account their speci!c interests and
teaching and learning aims." (p. 4)
"58 health informatics graduate students" (p. 570)
It does not apply
It does not apply
It does not apply
It does not apply
Intervention if applicable (Added on March 14, 2023)

It does not apply

It does not apply
 "In this study, we developed PREFER, an online corpus-based
paraphrasing assistance system. Allowing multi-word input and
returning promptly with a list of paraphrases in En- glish and
Chinese, along with usage patterns and example sentences,
PREFER pro- vides substantial support for EFL learners to vary
their expressions during writing." (p. 22)
It does not apply
It does not apply
It does not apply
It does not apply
 "We present a system for generating “sparks”, sentences
related to a scientific concept intended to inspire writers." (p.
1002)

"We present a system that aims to inspire domain experts

when writing tweetorials on a topic of their expertise. This
system provides what we call “sparks”: sentences intended to
spark ideas in the writer. Our system generates sparks using a
mid-sized language model (GPT-2 [43]) and a custom decoding
method to encourage specific and diverse outputs." (p. 1003)
It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
 Data collection (if relevant) ((Please add direct quotes and their pa

"The method of investigation used to find an answer to the above research questions was an online surve
year online assessments and exam boards were over. Most of the participants (n=32) were teaching in Hi
also teaching in secondary schools and most of them have experience teaching various languages such as
Arabic, Russian, Korean and Danish, and at various levels of the CEF from A1 to C2 (
It does not apply
 "Design of the experiment
For the purpose of assessing whether PREFER benefits EFL learners’ writing skills, we used a single-group
writing performance and developed a questionnaire to explore participants’ perceptio

"The format of the test items is a short paragraph with its corresponding translation, which avoided obsc
with the help of translation, “apart from” could be clearly defined as “besides” in the sentence “Apart fr
protecting our environment.” Moreover, a total of 22 phrases were embedded in items of both pretest a
quantity of students’ paraphrasing performance." (p. 28

"Experimental procedures
To evaluate the effectiveness of PREFER, we drew a comparison between PREFER and the two most popu
and Thesaurus.com among students in this country. A complete administration of the test and the questio
the pretest in 25 minutes with no tool support, all students were introduced to and familiarized with PRE
English Dictionary Online, and Thesaurus.com for 40 minutes. Then, in another 35 minutes, they were ran
paraphrasing task using different tools: PREFER (Group P), Longman English Dictionary Online (Group L), a
their assigned tool." (p. 28)

"A program was written to record the frequency that students consulted PREFER to track their online beh
compare students’ motivation and paraphrasing performance
It does not apply
It does not apply
 "this article examines current research on the use of MT, highlights the strengths and limitations of th
solutions designed to harness the capabilities of both MT and alternative technologies, and suggests ve
learners’ academic growth in line with ACTFL's Can-Do Statements for Intercultur
"For the course’s final module on AI, students had to write an essay that integrated the output from a text
but without reveal- ing which was which. Styled as the “Professor Fyfe Turing Test,” students were encour
text came from the AI or from them- selves. I imposed no quotas for the amount of AI-generated text nor
final section, students composed exclusively in their own voices (i.e. without AI), reflecting on the followi
easy or not was it to write this way? What worked or what didn’t? How did the AI-generated content rel
have thought about or written? Do you feel like you “cheated”? To what degree is this paper “your” writ- in
versus your own? Would you use this tool again, and in what circumstances? And, ultimately, what ideas
test or change? Finally, all papers included an Appendix in which students provided a “revealed” version o
The assignment yielded 20 student essays of 1500 + words each. All students gave their permission to h
reports about the experiment." (p. 4)

"Many of the assignment’s questions were open ended, leading students into certain considerations wh
While evaluating these papers for our class, I captured responses to those guiding questions to includ
 "Study 1: Spark Quality" (p. 1007)

"we recruited domain experts to annotate outputs for coherence on a 0 - 4 scale, in line with knowledge
(“Not true”), 2 (“Opinion/Don’t know”), 3 (“Sometimes true”), and 4 (“Generally true”).13 For biology, w
biology; for environmental science, we had 2 senior undergraduate students majoring in environmental sc
from the computer science department.14 Each discipline had 900 sentences to annotate (300 human gen
the custom decoding).250 randomly selected outputs from each discipline were annotated b

"Study 2: User Evaluation" (p. 1010)

"The study was run remotely via video chat and screen sharing. Participants were first asked to read an
tweetorials are and walked through an example tweetorial. They were then introduced to the system a
system’s features and showed an example use case of the system. Participants could ask clarifying question
minutes. At this point the participant was asked to pick a topic to write about, as well as provide a ‘contex
the system to correctly interpret their topic. Then they were given 20 minutes to interact with the system
presses while the participant interacted with the system were collected, as well as a

"After this, the participant filed out a short survey, which included the Creativity Support Index [13], an
facilitator. During the interviews, participants were asked questions about the usefulness of the system a
writing process. They were encouraged to review what they had written / the sparks they had seen to g
questions can be found in the appendix." (p. 1011)

"Finally, all sparks seen by participants were collected and annotated for common computer- generated t
‘In-coherent’ [17]. These annotations were done by graduate stude

"Participant interviews were transcribed and the authors performed a thematic analysis [6] on the trans
helpful or unhelpful, how writing with the system compared to their normal writing process, and owners
Relevant quotes were selected from the transcripts and collated in a shared document, where the authors
emergent themes. Finally, all sparks seen by participants were collected and annotated for common com
‘Redundant’, and ‘In- coherent’ [17]. These annotations were done by gradua
 "This study adopted a qualitative approach, using semi-structured interviews. While a purely qualitative
generalisability of the findings and sample size, it allows for a more detailed exploration of individual opin
data and its analysis that facilitates ‘collecting stories, analysing their contents, !nding patterns and sha
discussion that is in its infancy, it was felt suitable to start with a small-scale study and take the opport
stimulate further debate in HE and be used as a basis for future larger-scale research. Further, qualitative
they emerge during the interviews, which proved to be a fruitful activity du

"The research was conducted by means of semi-structured interviews (Dörnyei, 2007). All interviews w
analysis." p. 4)
"We conducted a case study among graduate health informatics students at Indiana University using an ab
Representations from Transformers (BERT) language model to evaluate the current state-of-the-art AI-g
create expert knowledge content. As part of the assignments, students were required to prepare annotate
topic every week. 58 health informatics graduate students wrote annotated bibliography for 25 articles e
summaries. We then asked the students to each grade 20 annotated bibliography, of which 10 were stud
the author information. We ensured that the students did not receive their own annotations. Additiona
students to classify if the generated text was AI-generated or not. For our analysis, we used the Kruskal
differences in the peer grades between student-generated and AI-generated
"In order to investigate the extent to which paraphrasing tools substituted recognised and expected med
essays which we had identified as particularly unusual. We did not know the provenance of these essays
might have arisen from a single seed document which was an essay submitted by one st
It does not apply
"In order to test the quality of output generated by some free Internet based paraphrasing tools and how
the following experiment was conducted. A paragraph from an existing publication by this article’s author
Educations Integrity (IJEI) was selected to be the original source material (McCarthy & Rogerson, 2009, p.
in-text citation, one in-text citation was included (Thatcher, 2008). A set of three bibliographic entries fr
selected to test how references are interpreted." (p. 4

"The next step was to compare the outputs from the original journal article material to the outputs of Tool
observed, tagged and highlighted in grey. Matches between the two paraphrasing outputs that did not m
the relevant text in a box. Contractions and unusual matches were highlighted by double underlining the t
an in-text citation) the following summary characteristics were calculated: total word counts, total word
paragraph." (p. 5)

"In order to identify how Turnitin® interpreted the paragraph and bibliographic out- puts from the parap
paraphrasing outputs were uploaded to Turnitin® to check whether the journal public
It does not apply
"The results show that only 36.7% of the respondents/teachers had used online translators in their languag

(From table 1) "Some examples of how teachers used online translators in the language class [...] MT outp
and their prope

"The vast majority of teachers agreed that students use online translators on their phones

"As shown in Table 2 below, most of the summative online assessments set up at the various institutions
involving various skills such as reading, grammar, and writin

"Looking at the results in table 3, it is interesting how level B1 reported the majority of assessment stud
tutors. However, students’ responses

"Looking at the results per assessment type in table 4, we notice that reading and listening assessments,

"Only 39% of respondents reported that the average ma

"The vast majority of teachers (75%)

"A third of the participants (n=10) reported that it was easy to spot students’ plagiarised work if you know
outside their remit. Finding incon

"Almost a third of the participants (n=8) thought that spotting plagiarism was n

"Table 5 summarises the features that made tutors suspect that students might have used online translat
levels) do

"Other features mentioned were: the choice of vocabulary, highly accurate language, and too complex
references. For this reason, it is good

"The main reason reported was “because students want to obtain a higher mark” (78%), followed by the st
respondents also mentioned that some students worry that other studen

"It was rather shocking to find out that 75% of the respondents did not report any case of online translatio
with instructors on these matters because they felt they were not directly responsible fo

"Amongst those who reported it, these were the reasons behind: because in language teaching

"Most of the p

"However, there were also 40% of respondents thought that copying the output fro

"What is clear is that there does not seem to be a consensus on this and every HE institution seems to hav
reference tools in the exam instru

"Table 6 summarises responses to the question ‘What forms of online language reference do you think
search engines (n=16) which usually involve a rather more com

"The vast majority of respondents (84.4%) indicated that their institutions do have a s

"Having said this, the jury is out on this issue since one can argue that by making use of va

"One respondent highlighted the need to “come to terms with the fact that online translators are a

"Re-designing assessments fi

"Quite a few respondents were in

 proficiency, moti- vation

"To answer the first research question, the paraphrasing scores of the three student groups were analyzed
of Group L and Group T decreased by 31.6% and 6.2%, respectively. A one-way ANOVA was conducted and
and the results showed significant differences between Group P’s and Group L’s paraphrasing scores (t = 4
is, the

"All students achieved expected gains with the help of reference tools except for Group T’s perfor

"Interestingly, the accuracy rates of Group T decreased from 68.6% to 65.5%, whereas Group P showed a 7
other hand, the Longman English Dictionary Online offered more relevant info

"Looking at the usefulness of the three tools in helping the students to accomplish the task, the Longma
have enough information for paraphrasable or corresponding items. PREF

"Regarding the second research question, we targeted the issue of the extent to which PREFER affecte
indicator of students’ proficiency levels), frequency of consultation (using the log data as an ind

"Overall, PREFER evidently helped students show better paraphrasing performance, es

"The third research question explores students’ perceptions

"About three quarters (75.9%) of the participants agreed that sometimes they could barely generate any p
of various classifications of student groups reveal that more than 90% of the less proficient, highly motiv

"Overall, almost

"We also explored students’ opinions about whether PREFER facilitates their language learning. Approxim
 translation, tra

"GT also excels at spelling. Not on

"GT has also shown that it can translate high-frequency idioms rather than simply reproducing literal trans

"the shield cannot always be trusted; for example, in French, GT's verified translation of mettre son grai
addition to similar errors in meaning, error

"many of the previous er

"GT has also improve

"Similarly, since the im

"In addition, GT

"GT continues to fail in instances where ag

"issues of for

"Telltale signs of MT use by students include producing verb tenses that have not yet been studied, usin

"while GT's ability to conjugate verbs and produce coherent sentences has vastl

"In sum, after probing GT to evaluate the advances that have been made and acknowledging that it is inc
"Students rightly presented a lot of critique, perceiving their difference from GPT-2 in terms of viewpoints
even worked nicely: “I was genuinely surprised with how well some of the content flowed with my perso
predicted text,” a student shared that “I felt like I was reading potential sentences that I would have w
controlled by the author, but seemed almost fractal, as if multiplying the author a

"A few students were startled that GPT-2 gave them

"In a related way, other studen

"At

"In a sense, these students are “patchwriting,” or borrowing and manipulating the structures of expert
writing or being outsiders. These students adapted AI outputs in

"As students accommodated AI outputs into their essays, whether gratefully or begrudgingly, they a
acknowledged that results were somehow not the

"Other students took the opposite

"Ironically, when encouraged to cheat, these students instead e

"The eight students who concluded that

"But other students were less sure they had really cheated, accepting GPT-2’s contributions less as “plag
One student insightfully explained how this requires conceiving AI less antagonistically: “AI assistants w

"Thus, this student did not mistake GPT-2 for a “peer” at all, but approached it instead as a tool for her o
things. First, stu- dents can perceive creative and critical opportunities when allowed to experiment beyon
participation in new ways and to writers of diverse backgrounds. Second, conventional definitions of plagi

"Similarly, another student noted the strange- ness of not recognizing the sources of their own essay aft
Turing test. Rather, this student implies how t

"Ten students overall answered that they had not cheated. Some of them hedged that while they didn’t
firmly defended their work as “not cheating” even within the con- ventional framework of plagiarism.
grammatical text, bu

"While the student did not pursue the analogy to Franken- stein, we might consider the provocative relatio
hard questions about the relation of creature and creator. The student had played mad scientist in a text

"When writing with AI becomes more like t

"When new text tech- nologies appear, those terms have to be renegotiated, eventu- ally becoming norm

"This assignment asks students to reneg

"Thus, not only does this assignment aim at d

"My students realized that we were nowhere close to push-button solutions to writing essays automa
 "First, v

"These participa

"Second, six of the participants reported using sparks to help them with translation by providing detailed

"Participants discussed the difficulty of writing technical de nitions or including technical details, and rem

"Third, three of the participant

"Interestingly, participants discussed sparks that were factually wron

"The ‘translation’ use case corresponds to much back and forth between

"The ‘translation’ participants requested more sparks and used a higher variety of prompts to do so tha

"We also see that ‘translation’ users moved back and forth between requ

"Looking at how sparks were incorporate

"We find that s

"we find that starred sparks have significantly hig

"The aggregate measure nicely matches our

"We calculate the Pearson correlation coeficient and p-value to look for a linear relations

"These participants responded to the same error (misinterpreted topic) in d

"Participants tended to agree th

"In the interviews participants were able to be more precise about how they perceived the di erences. So

"Most participants had no ownership concerns about incorporating sparks. Several reported that beca

"While most participants had no ownership concerns, all participants expressed concerns about plagiarism

We found that there was no correlation

"It might be that people’s attitudes towards in uence and inspiration has a large impact on their attitude t
ge

"Overall, we believe that participant attitudes are a major unknown factor when studying human-AI collab

Most participants were worried about providence and plagiarism, bringing up these issues independent

"Presumably the assumption of commercial writing support systems is that

The bias of large language models is well-documented and a serious concern for anyone making use of th
However,
"The theme of the use of MT in terms of academic integrity was one of the key points to emerge from this
try and identify what migh

"Responses, somewhat unsurprisingly, were varied. Also, there was an evident difference between how

"Generally, the data indicates that the par

"In sum, there was no indication in the data that any participant considered the use of GT as a rea

"Participants were also asked whether they would regard it as acceptable in terms of academic integrity if
the !

"The main tentative direction emerging from the data above was that if students had written their papers

"In essence, the data shows that there is no agreement amongst university staff regarding whether using G
tool, with clearly diverging opinions and c

"Among the participants, there was a clear awareness that the degree that the students earn at UK unive

"Views expressed with regard to the job market tended to fall into three main areas. The !rst was

"In addition to this, there was concern (n = 7) about protecting the reputation, or brand, of the university
 "We did not find a statistically significant differenc

"Thus, the Sensitivity=0.6000, Specificity 0.4517, F1=0.5586 with an Accuracy=0.5259. This

"Interestingly, in the 348 summaries that were correctly identified by the stude
 "The scenario (Appendix) was put through a number of paraphrasing tools, and in e

"The scenario document was subject to iterative language translation (Day et al. 201
refreshed Google TranslateTM page for translation back into English. The target lang
principal first languages of the EAL students enrolled in this subject. The translat
translations were Chinese (Simplified and Traditional) and Vietnamese, and the highes

"The original scenario was then put through six paraphrasing tools selected as the top
the assumption that students would u

"The results from the output texts were analysed for synonym substitution of recognis
tech- nique was used for convenience purposes as the intention was to gain

"Although it is not within the scope of this brief exploratory study to state that there
 "There are obvious differences in how the online paraphr

"Both tools

"The output from Tool 1 has used 77 words or 50% of the words in the original paragraph but these were
Chinese, but did capitalise seven random words mid-sentence (Audit, Numerous, Conc

"The Turnitin® results for both the paragraph and reference list uploads identified the original source as 1
iden

"the machine-based originality similarity checking software continues to ha

"It is interesting that the change to lower case for authors’ initials appeared to impact on Turnitin’s® capac

"A further examination of both sets of output

Author Identified Limitations ((Please add direct quotes and their page numbers))

It does not apply

It does not apply
 "Some phrases’ paraphrases were not available mainly because of the size of the corpus. The corpus size s
by integrating other bilingual corpora such as the Sinorama bilingual corpus (SPC) (http://www.aclclp.org.tw
other cases, the phrases might not be able to be paraphrased with one lexical substitution, even though th
paraphrases. For example, “on the whole” and “to sum up” share the same Chinese translations “!!"%”
bilingual corpus. However, “to sum up”, compared with “on the whole”, seldom appears in the beginning
usages and positions in sentences should be included, which are supposed to facilitate learners’ phrase u
pragmatic competence). Meanwhile, the current techniques are still unable to generate the inconsecutiv
(e.g., “take something into consideration”)." (p. 35)

"Last, the student experiments should be pilot tested on a small number of subjects who have identical or
as the group participating in this study." (p. 35)
It does not apply
It does not apply
It does not apply
It does not apply
 "And while we picked our task to be representative of constrained and creative writing tasks, it di ers gr
writing tasks people might be in- terested in like writing stories, academic papers, newspaper articles, or m
1017)

"Because we wanted our user study to closely mimic a realis- tic writing scenario, we had participants sele
However, this introduced a large confounding factor, as di erent topics are more or less di cult to explain a
and di erent topics may elicit di erent levels of spark quality from the system, as seen in Study 1.

"future work could bene t from comparative studies, either large-scale ones where participants can still p
but the size of the study minimizes topic as a factor, or smaller-scale ones where participants are assigne

"The small sample size of our study may have limited our ability to nd signi cant correlations. Perhaps in
would nd that the quality of system outputs does correlate with perceived usefulness." (p.
It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
It does not apply
 Conclusions/ Closing ((Please add direct quotes an

"We have learned that students do use online translators in class and that it is in the lower levels where u
to take place." (p. 132)

"In terms of assessment methods, we learned that reading, listening, oral, and written production tests
translation tests." (p. 132)

"Firstly, tools such as Turnitin alone cannot be used to detect online language learnin

"Equally important is the need to provide clear guidelines that contemplate how online translators sho
(especially in summative assessments) will also be contemplated in marking schemes an

"Secondly, we learned that it is not all “black or white” when it comes to labelling (or not) online machine
agreed that it constitutes an illicit form of language in that the students did not put in the intellectual effo
translation, content is not compromised, and students can be taught how to use these tools effectively to
tools such as dictionaries or concordanc

"online machine translation tools have an important part to play in the academic community and beyon
lifelong learning to name some exampl

"Outside the academic context, the efficient use of these tools can constitute a valuable asset for futu
language, but also in contracting individuals with the capacity to analyze content in various languages to
reality. This is particularly needed in minority languages with limited (and m

"The proliferation of online translator tools and neural models online can help break down language b
Language learning will not be disadvantaged but essential to operate and train th
It does not apply
"The results revealed that PREFER, as expected, achieved considerable success in helping English learners p
based on students’ proficiency (paraphrasing score in pretest), motivation (frequency of consultation), an
the less proficient, highly motivated, and conservative students showed significant progress in the para
students who need more support. Moreover, students’ attitudes toward
It does not apply
 "Banning the use of digital paraphrase and translation tools would be a thwarted attempt to manage aca
relevant, transferable skills." (p

"The paper has demonstrated a number of illicit uses of translation and paraphrase tools identified by high
to reveal a systematic failure to appropriately define it in light of

"This silence on the appropriate use of digital tools in academic integrity policy needs to be addressed
institutions, leaving learners less equipped to deal with the r
"The omnipresence of MT in the L2 classroom presents a dilemma for instructors, raising concerns about b
improving features that render it so us

"Today's L2 educators can help students recognize that reasoned, nuanced, and culturally sensitive use of
technology can certainly inform L2 learning, it cannot —at least at present— fu

"The issue then is not whether instructors can prevent learners from consulting such technologies, but r
proficiency and ethical use of technologies are critical 2
"Collectively, my class did not conclude one way or another that writing with AI was tantamount to plagiar
about their votes. But this assignment was never meant to settle a debate. Instead, it immersed students
other questions we had not yet collectively

"The “hands-on” experience let students explore the arguments for themselves, as well as test out new c
AI." (p. 9)

"The assignment’s framework of AI-powered plagiarism led students to assume that they would somehow
or expanding students’ definition of what that work

"Whether or not writing with AI becomes cheating is a much more complicated question than software
humanities scholars can play an important role in help- ing explain

"How do institutionalized definitions of plagiarism block the possibilities of writing with computational as
plagiarism policies impose or preclude, especially compared to what we really value for students? How ar
to impact (like writing essays)? How might we modify these activities or emphasize different outcomes?
students? Are there new modes of creativity, critical reasoning, rhetoric, assemblage, and expression tha
exercises align with frameworks for teaching AI literacy? What are the risks and harms of these technol
perspectives on these questions to bear upon the development

"Embedding students within these debates helps show them not only the important issues at stake, but in
9)
 "We found that our sparks were higher quality than a baseline system, and app

"we found that participants used the sparks in three main ways: as inspiration by providing ideas, to help w
that helped them understand their read

"We also found that while participants preferred higher quality sparks, across participants average

"In the discussion we propose that participant attitudes towards writing may be in uencing how they pe
inherent biases in large language models to develop mor
"There are two factors which have brought the issue of MT in HE to the fore. The first is the undoubted im
widely available and virtually effort free. The second is the fact that this has become a part of an internati
they do not share expertise in the language of th

"In response to the research questions, among the academic staff interviewed for this paper there was s
students, but this is far from universal. The use of MT as a writing aid seems to be viewed as more contro
view the use of MT as academic misconduct. However, our findings suggest that there is a clear and u
guidelines that are robust, realistic and in line with the wider v

"It should also be noted that it would be a mistake to consider this issue only relevant to Anglosphere univ
other similar institutions with EMI programmes outside of

"MT would provide a solution to many of the issues that international students face in their learning and a
to focus their energy on their academic endeavours, not on improving the
 "Through our paper, we have seen that state-of-the-art language models have a capabilities to generate
expert-level understanding and knowled

"While these models are simply mimicking the kind of text they are supplied, graduate-level students i
student-generated summaries." (p

"We see that educators will have a challenge in the near future to assess learning that depends on summa
order tasks that might be better ways to assess learning at the graduate level. However, with larger langu
might be these higher-order tasks."
 "The availability of free online paraphrasing tools may appear to them as a realistic solution to these cha
students who write original work in their first language and then use online translation tools to convert t
that the submitted work is a result of their own intellec- tual endeavours. Unfortunately, students who us
matching software, have committed an overt act of ac

"In academic writing in the health science discipline, there is an expectation that standard medical termin
and investigated the outputs of both paraphrasing tools and Google TranslateTM. We noted that paraphra
accepted medical nomenclature, whereas Google TranslateTM large

"When paraphrasing tools have been applied to text the output is frequently of such poor quality as to ren
generated will be notable for the use of unidiomatic words and phrases; expected vocabulary such as s
synonyms; word matching software, such as Turnitin® (n.d.), may not recognise the re-engineered tex
indicative of the actual level of plagiari

"When using online translation tools, such as Google TranslateTM, to convert text from a language o
nomenclature, such as standard medical terminology, will be changed to t

"This study demonstrates that there are a number of distinct features which can be identified in the text g
process of detecting plagiarism. While the emphasis should be on sup- porting students to develop the sk
provide evidence of the use of paraphrasing tools will be of benefit in the overall m
"As technology continues to accelerate, the rate of development in advanced tools which manipulate langu
and illicitly, will continue to grow. The role of academics is to decipher their use, understand why and ho
unacceptable usage." (p. 8)

"This article has sought to remedy this through the review of current literature pertaining to APTs and offe
with this growing threat among both native English speaking and EFL students. We have also identified
development of technical solutions is promising but may continue

"We therefore advocate for training as the most important tool in both reducing the use of APTs by stud

"Finally, as recommended by Rogerson (2020) additional inves- tigations should aim to

"This study has demonstrated that students can use online paraphrasing tools or article spinners in ways t

"the intent of this paper is to ensure that those involved in teaching and learning are aware of the practice
the perils of using such tools." (p

"The proliferation of fee-based and free Internet-based tools designed to re-engineer text is a concern. O
cannot necessarily be used at this time to identify where writi

"Further work is needed to identify linguistic markers indicating use of online paraphrasing tools such as
may be strongly in favour of upholding academic standards, they may also be reluctant to undert

"Research is also needed in exploring the most effective techniques or combination of educational, deterr
online paraphrasing tools and article spinners and other form
It does not apply
 Source's information on eth

"A particularly pertinent issue in this regard is academic dishonesty. Some argue that the use of online t

"In the language learning context, the use of online translators such as Google Translate at lower levels wh
students for language learning purposes concluded that, although its use helps them to experiment with t
the use of online t

"Another important factor to combat academic misconduct is the creation of proper guidance for both ed

"I believe academic integrity policies often need to

"Having said this, the jury is out on this issue since one can argue that by making use of va

"One respondent highlighted the need to “come to term

"there is a need to train language educators in the various forms of online machine

"we learned that it is not all “black or white” when it comes to labelling (or not) online machine translation
others argue that in the process of generating the automatic translation, content is not compromised, and
 "This is because definitions of plagiarism are socially constructed and tied to context-sensitive cycles of re
text ownership, rather t

"In this system of rewards, text ownership and attribution are sacrosanct, and “theft” is forbidden. But if li
syllabi, assignments, and other teaching-related documents w

"When we consider the possibility that students could generate entire authenti

"Recent advances in the production of natural language—text that mirrors the characteristics of writing
process akin to the crea

"To consider how GPT-3 (or GPT-4, or Google’s system) might subvert stude

"Although imperfect, these examples show that even at this stage of development—with sweeping impr

"Like the acceptable use of unattributed text, this range of writing processes widens our conception of wh

"Some applications of NLP systems may s

"In this case, I provided the in

"Many researchers in STEM fields might find auto-written methods sections acceptable, just as many STE
Anson, Hall, Pemberton, and Moskovitz; Pemberton, Hall, Moskovitz, and Anson; and Anson, Moskovitz,
used in interpreting the theme o

"In a college biology course, an instructor might accept an NLP-generated lab report as

"Solutions include having students write at computers that block access to the Internet, write by hand in cl
writing activities (see Anson, “Cops”; “Defining”; Vie) But because automated writing systems are here to

"the process of exploring GPT technology is artfully demonstrated in an assignment developed by Paul Fy
incorporate the material into their final essay. However, the students were required to highlight whi
perspectives, and how the material might or might not be considered plagiarism

"In addition, students need to learn the sinister side of NLP systems. Because they generate text base

"In other words, we need to help students understand that the act of writing is always situated within a
years to come. As such, it is im
 "Whatever we do, we need to engage in dialogue with students about what the allowa

"There is nothing obviously ‘robotic’ about it. In coming years, translation software may become so adept

"It is even harder for tec

"We, as an education com

"Cheating is ultimately about ethics, and while that is something we may be able to outsource to machines
"This incident highlights student behaviours and perceptions about the use of digital tools that are not exp

"In reality, the illicit use of digital translation and paraphrasing tools does not fit neatly

"While our policies on academic integrity have evolved, to some extent, to encompass the digital

"Students’ use of digital tools for language enhancement needs to be a consideration in the design of DEE
"Machine translation (MT) has become inescapable—from the Google Chrome Web browser that automati

"These new realities force educators to ask tough questions. Does students’ reliance on MT represent an

"Love it (students) or hate it (instructors), the capabi

"In other words, the greater threat would be to deny the extent to which MT is a rad

"Taken together, the emerging body of work has shown that MT can no longer be viewed as the perhaps
While instructors may disagree about the extent to which the use of MT violates an institution

"If indeed MT has radically altered what it means to communicate in another languag

"However, when evaluating the quality of GT's actual translations and their impact on L2 teaching and lea
whose expre

"Christopher Manning, professor of machine learning at Stanford, has argued that “there is no reason w

"In sum, after probing GT to evaluate the advances that have been made and acknowledging that it is inc
take into account the less p

"The preceding exposé of GT's improvements, new strengths, and remaining weaknesses and failures m

"Today's digital natives, who grew up in a “google, copy, and paste” environment, may not intuitively und

"The fundamental ethical problem is that learners must agree to abide by stated codes of ethics and inte
their call for creating a space for honest and respectful work. Regardless of whether students see their
include an exhaustive l
 "Interestingly, artificial intelligence has upped the ante, disabling the usual strategies of detecting plagiar
able by current plagiarism detection software. For example, claiming in a tagline to be “Empowered by
giarism concerns.” Marketing itself as a “bot” desig

"In response, learning management systems and plagiarism detection software are no

"These days, computer- and AI-assisted writing is already deeply embedded into practices that students
autocorrect, autocomplete, grammar suggestions, smart compose, and others? Asking students to “write
thresholds? At what point, in

"Additionally, in this paper, I want to suggest how di

"When instructors strat

"Throwing AI into the mix updates these questions in interesti

'Discussions of posthumanism are mo

"Writing with AI may force us into a heightened awareness of these posthuman dependencies, not only as
of t

"Whether or not writing with AI becomes cheating is a much more complicated question than softwa

""How do institutionalized definitions of plagiarism block the possibilities of writing with computational a
these values supported (or not) by the learning activities that AI seems likely to impact (like writing essay
creativity, critical reasoning, rhetoric, assemblage, and expression that computational assistance helps us
we bring different discip
"Research on language models is attuned to if the model is copying from the data it is trained on [42], beca
may still need to

"Dehouche argues that GPT3-generated text raises basic questions about authorship, because the autho
access to the mode

"Presumably the assumption of commercial writing support systems is that the writer is also the author of

"Bias in language models and the value of a biased perspective. The bias of large language models is wel
sexism, and other kinds of prejudices broug
 "However, as the technology developed, voices more supportive of GT in the academic space began to em

"There is an understandable tension when new technologies threaten to undermine or disrupt standa

"More nuanced thinking in the area of copy-paste plagiarism has led to a better understanding of how pa
not all unoriginality, or intertextuality, is in fact transgressive (Chandr

"This implies that it is not solely the student’s responsibility to develop their communicative competence f
in the curriculum. Should that not be the case, studen

"Translation has played an important role in the dissemination of knowledge for millennia. It can be use
approaches the is
 "Apart from using AI-GTS for online learning and online di

"Several research has shown that AI can improve teaching and learning outcomes in education. One su
example, instead of relying on a single textbook, students can use interactive multimedia versions of co
widely employed to acquire insights into students’ behavior dur

"On the other hand, it’s just as important to understand the potential risks that AI could pose in educati
students or lecturers are erroneous. In this case, who or what is to be held responsible? At some poin
algorithms

"However, it is critic

"Impair cognitive skills: The summarizing technique allows readers to identify critical elements (main id
connecting subsidiary ideas to the core idea, removing extraneous information, and arranging essential an
t

"Lack comprehension skills : AI-GTS could be a worry in students life as it doesn’t require pupils to utilize t
it helps to improve memory and understanding of topics by utilizing mental skills effectively [18]. The sum
the use of AI-GTS, the c

"Lack explanatory skills: In addition to improving cognitive and memory skills, summarization enhances
summarization, summarizing a topic into brief and understandable paragraphs necessitates high lang

"Damage writing skills: Writing a summary requires a lot of effort, and writing and editing skills are imp

"Impair cooperative learning: Task of summarizing facilitate collaborative learning. It provide students

"Over-dependence and plagiarism: Automatically available summaries can sometimes exploit students and
as their reliance on machine-driven activities increases. As a result, students are hesitan

"Inaccurate information: The inaccuracy of these systems causes a separate set of concerns. These tools s
year, Google search — which uses AI to summ
"The literature is replete with the lamentations of academics who feel that pursuing academic misconduct
within the cultural norms of academia (Brimble and Steve
 "Access to online resources, accelerating Internet connection speeds, and global interconnectedness c
technologically advanced metho

"The lines between acceptable and unacceptable academic behavior are not univ

"Our position on this is that despite not meeting the technical definition of academic mis- conduct based o
counteract these kinds of cases, the case study hi

"While then, there are many areas of debate surrounding APT use, the fact remains that they are a serious
to help identify potential cases of plagiarism, thus weakening one of the most effective current diagnostic
researchers who may wish to expand their output through publishing paraphrased versions of the same
happen

"Under the arms-race scenario

"In terms of the arms-race metaphor however, it may not be long before proficient speakers start to find
"Developments in the treatment of translating natural language as a machine learning problem (known a
free tools available via the Internet lack constant updates and improvements as the code is controlled by
Consequently there are issues with the qualit

"students using an online paraphrasing system fail to demonstrate their understanding of the assessment
guilty of academic miscon
 "Impressive, yet it gave a similar response when presented with the introduction to a highly-cited resea
somewhat stilted, awkward style of writing”. Rather than attempt further embarrassing comparisons, we n

"Humans fare no better than machines at detecting AI-generated essays. In a small study by EduRef.net6,
the machine-written essay was given a grade of C, while the human essays were graded B and D. For US Hi
was failed, while human-written essays were

"Transformers are models of language not experiential knowledge. They are not designed to be scholarly

"This could be a pivotal time for education, as students equip themselves with powerful new AI tools that
al., 2021). An education system th

"Every new educational technology arrives with affordances and limitations. AI Transformer technology is a
 Source's information on artificial intelligence used for c
((Use direct quotes and their page n

"The use of online translators in online formative or summative assessments can be regarded as ‘plagiaris
thoughts) from a source other than the students’ and in presenting this output as their own without prope
work together in an unauthorized way to develop a submission for an online assessment when such inpu
online assessments as “cheating” in that students are receiving unauthorized assistance from online trans

"We have learned that students do use online translators in class and that it is in the lower levels where u
to take place." (p. 132)
"Let’s imagine that a student generates the content of an essay using a “robo-writer” like GPT-3 and subm
this would not technically be a case of plagia- rism—until or unless computers are considered authors—bu
Clarke), a violation of student codes of conduct similar to submitting a customized paper written by some
NLP-written texts may not look like ethical viola- tions, in much the way that vast amounts of re

"Auto-generated interview questions may be acceptable to some teachers, but what about a brief biograp
scholar) asks GPT-3 to write the methods sec- tion of an articl
It does not apply
"However, educators are often unaware of the full range of possible e-cheating approaches, which creates
use to rewrite text and fool text-matching tools like Turnitin (Rogerson & McCarthy, 2017): while few ed
probably regard them as cheating.

"Essay spinning, auto-paraphrase and back-translation tools [...] These tools rewrite text for students, effe
these tools to avoid having to manually paraphrase (at best) or to reuse someone else’s work without getti
paraphrase, students have also used translation tools for paraphrasing. This approach is known as back-tr
language (e.g. English) to another (e.g. Spanish) and then converting the tran

"As translation tools have become more sophisticated, they have become

"Translation tools [...] Where the specific outcomes being assessed include the ability to produce work in
amount to e-cheating [...] in the future, it may be that all but the most fluent second-language speakers m
language, rather than attempt- ing to write in their s

"Bespoke essay writing tools [...] These tools conduct online research, find sources, and write and structur
not good) grades from using these too

"BYPASSING TEXT-MATCHING SOFTWARE [...] The auto-paraphrase tools discussed earlier successfully defe
database (M. Jones & Sheridan, 2015; Lancaster & Clarke, 2009; R

"If I take a paragraph from a journal article, paraphrase it, and use it in a new article without citing the orig
of semantic content-matching tool. In the long term, with increasing sophistication and adoption of para
approaches, otherwise they will only catch the most blatant

"Text- based methods like bespoke essay writing, translation, and auto-paraphrase tools appear to have
databases of source content." (p

"In Chapter 1, I identified several e-cheating affordances enabled by AI that students are already using to
basic foreign language translation, paraphrasing and algebra can all be done for free by computer over the
the sort of HLMI proposed by Müller and Bostrom (2016), in that they are much more narrowly scoped
dealing with a hypothetical." (p.

"The use of free auto-paraphrase tools allows a student to copy-paste existing work and c

"current content-matching tools cannot detect paraphrased work; this means students can resubmit thei
use an auto-paraphrase tool." (p.

"But the shift to e-cheating has been accompanied by the rapid proliferation of new ways to cheat, to wh
tools, automatic content generation and homework-completing apps are just a few of the unsolved proble
once been devel- oped in isolated pockets by expert cheaters, now cheating can be bought and sold anony
and coded into an app so it’s easy for anyone to do.Things are only going to get worse as we progress tow
experts believe has a 90% chance of happening within our
"Anti-plagiarism software has gone some way toward reducing the incidence of plagiarism, but new techn
challenges, with particular regard to monitoring the origina

"A growing body of literature is emerging around the practice of cross-language plagiarism (Eshan & Shake
work from Language 1 through a translation tool to produce a translated piece of work in Langu

"Cross-language plagiarism has emerged and proliferated in line with the advancement of digital transl
numbers of writers claim other people’s work in an

"Students undertaking back translation identify a sentence, paragraph or even a lengthy passage from a
translate it into a foreign language. They then copy and paste the translated text back into the translation
undetected through plagiarism softwa

"Prentice and Kinden (2018) were alerted to the use of paraphrasing tools by undergraduate students on
flowery expressions that lacked coherence. On investigation, they discovered the prolific use of paraphr
produced. In particular, the tools did not recognise or preserve medical terminology, but instead sought to
terms substantially." (p. 43)
 "Multiple studies have confirmed that students consult online translators, even when instructors explicit
Kashani, 2014; Luton, 2003; Niño, 2009; Williams, 2006). The differing perceptions betw

"Niño credited students’ dependency on MT to its wide availability, immediacy, and multilingualism, as w
(Niño, 2009, p. 245). In fact, Niño's study of advanced Spanish students revealed that 75% reported turnin
MT in the future (Niño, 2009, p. 249).

"Inputting data into GT and reproducing those results patently violates the code of academic conduct, and
788)
 "Interestingly, artificial intelligence has upped the ante, disabling the usual strategies of detecting plagiar
Turnitin). Some online resources now advertise AI to students to generate usable, unique text that is unt
tagline to be “Empowered by Artificial Intel- ligence,” EssayBot.com promises itself as “your personal AI w
paraphrases for you to erase plagiarism concerns.” Marketing itself as a “bot” designed to outmaneuver pl
erasure of human effort. In response, learning management systems and plagiarism detection software a
defined antagonists: systems to cheat artificially versus syste

"What skills might writing with AI newly require? How does this challenge our assumptions about textu
“Cheating” offers a starting point, but it opens onto much more complex and consequential topics, from h
bias. Thus, the assignment moves students from a familiar ethical context (is this plagia- rism?) to new et
forms of writing practice (can I collaborate w

"Fabricated quotes like this shifted students’ attention from plagiarism to the threats of “synthetic disinfo
think that the false information could be classified as cheating a lot m

"The eight students who concluded that they had indeed “cheated” on the assignment were co

"Using GPT-2 only deepened their desire for control, as another student echoed: “Ironically, I feel like ‘che
liked.”" (p. 7)

"But about a quarter of the class had no need of such nuances. Intriguingly, they remained confident in t
cheating” even within the conventional framework of plagiarism. They claimed to have done original, int
effort more in terms of assembly and editing. The AI could generate grammatical text, but until that text w
(p. 7)

"In this light, while one student confessed they were “barely writing,” they similarly defended their wor
voices.” Ultimately, as another student concluded, “that involvement in my thought process turns the use
makes me feel like I did more writing rather than less.” Here, the word “writing” represents a far more c
ironically required them to do more o

"Whether or not writing with AI becomes cheating is a much more complicated question than software
humanities scholars can play an important role in help- ing explain

"Embedding students within these debates helps show them not only the important issues at stake, but in
9)
 "Most participants had no ownership concerns about incorporating sparks. Several reported that becau
ownership over anything they wrote. Others said that since they are writing about public know

"While most participants had no ownership concerns, all participants expressed concerns about plagiaris
sparks were coming from, and they wanted to make sure that anything they took from the sparks was

"Providence and plagiarism as major writi

Most participants were worried about providence and plagiarism, bringing up these issues independent of
sparks came from” and were worried that copying too much would
 purposes." (p. 6)

"It can be seen here that concerns about MT as writing tool are not only related to legal aspects (i.e. po
learning outcomes of UK degrees and the development of students to function

"More troubling to the participants, however, is the idea that students would be able to graduate with a
demonstrated the ability to communicate effectively in the language of instruction. This is clearly a concer
that, just as academia is having to adjust itself to the advent of MT, so the wo

"There was less consensus on the acceptability of the students’ use of the technology to write whole te
accepting of this, generally indicating that this is not optimal, but at the same time not

"There was, on the other hand, a clear feeling among our participants that students should leave their deg
was most commonly articulated as part of the university brand. However, there was little discus

"The general idea is that students could use MT at the start of a degree, but by the end of the degree, sho
an inherent contradiction, as pointed out by one of the participants. This is that, unlike a dictionary, for
language skills, which requires sustained and fo

"While the technology could be seen as a levelling mechanism during a student’s degree, which could po
challenging degrees in an additional language, it also would mean that these students will not have gai
communication skills, at least not as traditional

"There is also a clear need for policy within this area, and we strongly believe that this should be institutio
inconsistent and contradictory. This is necessary because, even in the small sample of academic staff
acceptability of the use of the technology. In addition, given the concerns over the dilution of the univers
ability in graduate attributes, and whether or not this is assumed to be una

"There is also a clear need for policy within this area, and we strongly believe that this should be institutio
inconsistent and contradictory. This is necessary because, even in the small sample of academic staff
acceptability of the use of the technology. In addition, given the concerns over the dilution of the univers
ability in graduate attributes, and whether or not this is assumed to be una
It does not apply
 rapidly escalated to the identification of a cluster of essays which bore remarkable similarity in the use
standard recognised terminology within the health sciences dis- cipline. Further to this, there was significa
text citations, were provided in an identical sequence. In some cases, the Turnitin® (n.d.) similarity index
resulted in an index of 0%. It became clear that para- phrasing tools were probably being used and

"The breakthrough came when an essay was so alarmingly absurd that we were able to trace the origin to
nario regarding a young Indigenous man’s experiences in the Au

"In some cases the original source was taken from the internet, notably Wikipedia, but in one instance the
student did not provide an in-text citation, however the ori- ginal source was identified by the student i
referred to as illicit paraphrasing (Curtis and Vardanega 2016), and actions such as this may

"We were also able to recognise patchwriting in text that had been appropriated from multiple sources, an
a ‘rainbow’ of colours in the similarity report demonstrating different sources. However, in the essays und
writing which was largely unintelligib

"Whereas in patchwriting synonyms are manually substituted by the student, online paraphrasing tools
posited by Rogerson and McCarthy (2017), as to whether the use of online paraphrasing tools transcen
Pecorari 2003, p.9)." (p. 5)

"One of the most obvious issues we encountered in the essays was the use of synonyms for standard med
throughout health care to avoid ambiguity in documentation and communication. This provides the inter
allied health information regarding patient care, and is an essential element of safety and

"The lack of standard medical terminology and the inclusion of unusual synonyms for this terminology wa
difficulties with English expression, or were manually substituting syno- nyms as seen in patchwriting, it w
us to suspect, and subsequently investigate, online p

"Spinning is a technique used to produce a new document, or documents, from an original text source by r
avoiding machine-based text matching tools used to identify plagiarism. Machine based paraphrasing t
rankings in Google search results and are part of a suite of search engine optimisation (SEO) techniques r
McCarthy 2017; Zhang et al. 2014)

"Cursory exam- ination revealed that many are duplicate sites with the same tool offered under different n
links to essay purchasing services. Anticipating the vulnerability of the student, some sites offer a

"In this study we assumed that the students were accessing the fee free version of online paraphrasing too
more words treated as mutables and thus less discretionar

"When determining whether there is a potential breach in academic integrity, it is important to distinguish
device, and the generation of text through a para

"If a student has used paraphrasing tools to alter a text to evade detection of plagiarism, then

"The method of detection we suggest, identifying the absence of expected nomenclature such as disciplin
expected immutables of recognised medical terms are substituted with synonyms, and thus treated by the
or external text, that is, the text containing the medical terminology wh

"In this exploratory study we identified linguistic features of spun text which indicated the use of parap
messengers to point towards collusio
 method, software, or system is developed for engaging in breaches of edu- cational integrity, a technologi
can be seen through the work of Foltynek, Meuschke, & Gipp (2019), who found that between 2013 and
identify complex forms of academic plagiarism. Some of these show great promise, with one tool develope
machine-translated paraphrased text docu

"From this beginning in website development, APTs have found a second user-base in academia, allowing w
2020) and bypass plagiarism detection services which use text-matching algorithms. The underlying

"Rogerson (2020) also argues that professional scholars and researchers may equally make use of these
Sixsmith (2021) described a single case of an article published in a journal which was later found to be like
identify 817 unique differences between the suspected source text (another journal article) and the publis
by a machine, suggesting the use of an

"The lack of consensus on what constitutes appropriate paraphrasing may be one factor that affects stud
use of APTs and to what extent they constitute academic misconduct. By reviewing the types of APT and h
AD can be formulated." (p. 4

"There are several different varieties of APTs, and all are not created equal. Prentice and Kinden (2018) hig
results from a search engine query for paraphrasing tools, the number of results had reached 3 milli
approximately 4.5 million; highlighting not only the growing number of APTs available, but also the increas
(p. 4)

"As an example, Prentice and Kinden (2018) found that in the discipline of health sciences, the use of
incomprehensible words that lacked meaning. This can be one of the clear

"Users first locate texts which are relevant to the subject at hand, and then copy material verbatim from
enter it into the tool. Students may also engage in ‘back translation’ (Jones, 2009; Dinneen, 2021) in whi
(again using a MT tool such as Google Translate) and then translate it back to English, resulting in a paraph
a 3-step process. By doing so, the writer may believe they are able to bypass plagiarism detection softw
paraphrase, or may simply feel that they have successfully engaged in paraphrasing, thus not committing a
text is incoherent, writers may attempt to proof- read and edit the paraphrased text to increase readabili
our view paraphrasing plagiarism.

"One example of such seemingly unintentional use of an APT to commit paraphrasing plagiarism is given b
APT to paraphrase text from file-sharing sites, while providing the origi- nal source in a reference list. Altho
the student did not intend to deceive, this can under most definiti

"A further case that may create debate is a report from Dinneen (2021), who describes a student who had
that as they had used in-text citations, and changed the wording of the authors’ original text (through us
Based on the interpretation of the institution’s plagiarism policy, it was found that there was no indication
that in essence the student was correct (Dinn
 "One particular submission from a previous subject instance had phrasing that included “constructive em
employee performance reviews. The student was interviewed at the time about why they had submitte
plausible explanation was provided

"Nor is the issue of paraphrasing or article spinning tool us

"Roig (2016) highlights that some forms of text recycling are normal in academic life such as converting c
between editions of books, as long as there is appropriate acknowledgement of the original source. Ho
previous work as with technological advances it will not be long before all forms of academic written wo
inconceivable at the present time” (Roig, 201

"The case of a student submitting work generated by an online tool without appropriate acknowledgemen
to reframe texts for alternate publications could be considered as a form of self- plagiarism. Both scenari
seeks to use some form of easily accessible Internet-based source to prepare or supplement submission
Stamatatos, 2011)." (p. 4)

"If an unintelligible string of words was submitted as part of an assessment task it may be a reason to hav
writing, and to determine if paraphrasing tools or article spinners have contributed. Where a citation is pr
writing conventions. Where there is no citation or any reference to the original source the situation m
procedures." (p. 9)

"A further investigation of the results from the Google search on ‘paraphrasing tools’ identified that many
that direct users back to the same paraphrasing machine. The purpose behind the existence of the sites is n
faces may be related to a way to generate income. Alarm- ingly the sites examined in this study showed ad
by users as tacit approval for the sites and the

"Other sites highlight that rudimentary paraphrasing tools are highly inaccurate but promote their paid
another form of contracted plagiarism (Clarke & La

"Using text-matching as a basis for detection instead of semantic matching means that uses of online par
detect at this time. Therefore for the foreseeable future the onus of detection of unoriginal material r
"Now, an unintended consequence of generative “Transformer” AI systems such as GPT-3 is that they dem
of around 50 US cents" (p. 112

" Equally worrying, would-be academic researchers can call on these systems to generate

"Gaining access to GPT-3 is straightforward. Anyone with internet can sign up to the OpenAI website3, ga
essay, set the maximum length of output (up to 4000 language “tokens”, or approximately 3000 words) a
and formatted text. Some companies are already promoting AI-b
 Source's information on artificial intelligence for ethical support in w
((Use direct quotes and their page numbers

"From the literature, we learned that online translators have some undeniable pedagogical value, i.e. can
language written production and translation, with proper guidance, can be used for independent learni
written production. In its very basic text-to-speech mode, some beginners have also acknowledged their u
speech online translators and oral corpora are growing in no time, we will be exploring their use/value for
mind that online translators can process various languages in ever-increasing language pair combinati
multilingual interaction is needed." (p. 118

"To illustrate some of the uses of online translators suggested by those teachers who used them in cla
examples ranging from the use of post-editing techniques (e.g. spotting and correcting/improving online tr
or for the practice of translation techniques, without forgetting the training and demonstration of all the o
output, or the contrastive translation evaluation of various online translators and getting the students
particular communicative purpose taking into consideration linguistic and cultural parameters such as
phraseology or the presence any sociocultural items to nam

"It is essential that language tutors of the various languages to and from which online translators transla
them, can contrast online translators’ output with students’ production, and, in the process, learn to r
condemn students’ use of it (this is unethical and very complicated to evidence) but to inform their teachin
how they can benefit from using these tools and how and under which conditions th

"It is also advisable that educators are aware of how pedagogic translation and translanguaging can be use
on MT will also help educators design more appropriate, authentic, and realistic formative and summative
so that the barrier that determines when their use if licit or not i

"Equally important is the need to provide clear guidelines that contemplate how online translators shoul
illicit use (especially in summative assessments) will also be contemplated in marking schemes and U

"online machine translation tools have an important part to play in the academic community and beyond
and aiding lifelong learning to name some examples
"At the least, these auto-generated examples might spark some new human- generated ideas—or perhaps
serve as invention prompts, or as text generators whose outputs are either
 "Since the development of paraphrasing ability presents a daunting challenge for many EFL learners, bo
practical solutions for enhancing learners’ paraphrasing com

"we developed PREFER (PREFabricated Expression Recognizer) (Figure 1), a web-based and corpus-based
expand their lexical knowledge (McInnis, 2009) as well as providing instant assistance to learners in their w
phrasal paraphrases as well as lexical paraphrasing (i.e., synony

"Owing to the fact that current paraphrase instructions and materials fail to systemati- cally and substan
support is a pressing need. However, few computer-assisted paraphrase-learning systems

"Similar to these concordances, PREFER generates paraphrases based on a statistical ap- proach. The
pedagogically to facilitate general language learners’ acquisition and expan

"we developed a paraphrase reference tool for educational purpose. We utilized the pivot-based app
paraphrases if they share the same foreign translatio

"PREFER targets at providing substantial paraphrases together with their corresponding individual usage
skill." (p. 27)

"We also explored students’ opinions about whether PREFER facilitates their language learning. Approx
PREFER in achieving better language proficiency. Notably the highly motivated, conservative students
competence." (p. 34)

"In this study, we focus on assisting EFL learners in paraphrasing short paragraphs with a reference tool, P
by applying natural language processing techniques, or more precisely, statistical machine translation, t
parallel cor- pus." (p. 34)
It does not apply
It does not apply
 "From a more practical perspective, the research synthesis and examples presented here suggest topics o
the strengths and limitations of MT. Learners need to realize that even when words seem to map directly
practices, beliefs, and values that are conveyed are not necessar

"In addition, educators might implement some of the translation activities suggested in previous resear
Crespo, 2017) and think carefully about the role of translation as a pedagogical tool that can heighten lea
that are involved in precisely conveying one's intended me

"Pedagogical activities, such as translating a popular song from English into the target language and then
that translations are rarely verbatim reproductions of the original text and that the act of translating does
in another. Focusing on more basic issues like syllabification, rhyming, and alliteration or on more comple
cognitive) can further illustrate to students both the value and the weakne

"In the age of GT, the very choice of a word processor sets the stage for the language learning that follows
moment they set foot in their first L2 classroom instructors must help them learn to use the other, more r
using Google Docs may be tempted by the “Translate document” option listed under Tools, if they instead
then the word processor will help them catch spelling and grammatical issues; thus students will garner
independence and integrity." (p. 790)

"In ACTFL's Statement on the Role of Technology in Language Learning, ACTFL (2017a) calls for techno
instructor-designed, learner-centered, and aimed at developing proficiency in the target language throu
experiences” (n.p.)." (p. 791)
 "Noting the “similarity between my writing and the predicted text,” a student shared that “I felt like I w
another timeline, like I was glancing into multiple futures.” In this case, AI unsettled the writer’s orientati
thread controlled by the author, but seemed almost fractal, as if multiplying the author and the possible
unsettling the familiar dimen- sions of their writing and how the

"But other students were less sure they had really cheated, accepting GPT-2’s contributions less as “plagia
we have yet to define. In general, these students were more open to different configurations of authors
requires conceiving AI less antagonistically: “AI assistants were not meant to replace or impersonate hum
hybrid.” (p. 7)

"Leah Henrickson has noted that “we do not know where computer-generated texts fit within our curren
they don’t fit at all, scrambling the ways we recognize and define how texts communicate. These ambigu
one wrote of their hybridized essay, “it doesn’t feel like something I’d write but it also doesn’t not feel
appropriate, as the demarcations of agency (what “I’d write”) are

"Several scholars have tried to articulate what this frame- work might be. Instead of using a “simple auth
Henrick- son proposes that we place them along “a continuum from authorship to generatorship” (2019
“algorithmic author- ship” (Henrickson 2019; 2021). David Rieder sees this less in terms of authorsh
configurations of digital media con- tribute to the rhetor’s goals of persuasion.14 Other scholars have p
writing with AI and the hybridity of computer-gen- erated texts (Manjavacas et al. 2017). Notably, these ar
new media art, and electronic literature rather than exposi-

"Ten students overall answered that they had not cheated. Some of them hedged that while the
It does not apply
"it would seem that academic staff are generally, but guardedly, not opposed to the use of MT by students
be of major concern, since there is a clear view that if students are demonstrating their understanding a
"awed language skills to present this, if there is a clear and effective alternative. However, to go some way
to declare the level of technological intervention that they have employed in

"No participant opposed the use of MT as a substitute for a dictionary, or to expand conventional dictiona
academic terminology." (p. 8)
 "AI-GTS play a significant role in the domain of education to improve the effectiveness and e

"Online learning platform: Implementing AI-generated text summarization in an online learning platform h
online discussion forums, learning content, and assess- ments [10]. AI-GTS aided in exploring ideas, ins
summarization techniques. Similarly, thread summariza- tion in an online discussion forum helped learner
research utilized summarization methods to identify qual- ity posts in the online discussion forum

"Learning disability: Students with learning disabilities find it challenging to comprehend and organize info
readers in focusing on the primary ideas and other critical skill concepts while ignoring the less import
individuals with learning disabilities to overcome the challenges and be m

"Summarize lecture content: Another study showed an exciting application of AI-GTS in education. They u
research and memory refresher tool that assisted university students by summarizing the instructor’s lec-
to students in the form of a brief document [14]."

"Language barriers: Text summarizing techniques are also used to construct summaries from one langu
linguistic barriers they confront in today’s globalized world. Texts in Russian, German, French, Spanish, E
translated into any other language [15]." (p. 5

"Increases productivity level: AI-GTS allows students to swiftly scan the contents of a paper for the e
unnecessary information. AI-GTS decreases the text size by approximately 20%, b

"Peer assessments : AI-GTS made peer review and evalu- ations easier by summarizing the feedback an
capturing the work’s strengths and weaknesses. When a group of peers ignores some comments becaus
immediate, actionable feedback, allowing them to assess their work qua
It does not apply
 "One other variety of APTs are those which are used for pedagogical purposes and do not constitute a vi
these can be indispensable tools for teaching paraphrasing as a skill. Chen et al. (2015) for example, dem
paraphrases using a parallel Chinese-English corpus, and found that 90% of the sample (N = 55) preferred t
the tool benefited their writing. This demonstrates that for students who are practicing English writing as
resource for learning." (p. 4)

"However, if student training is to be used as an initial proactive approach to deal- ing with APTs, then a
students understand the difference between the use of such tools pedagogically in the English as a Foreign
use individually to produce assessed work in their discipline
It does not apply
 "An imaginative way to incorporate AI-generated text into teaching could be for the teacher to employ Tr
then ask students to critique these and write their own better versions. Or set a complex question then as
the student to evaluate these responses in relation to the mark

"Transformer AI can be a tool for creative writing. For example, the student writes a first paragraph, the A
partner helps maintain a flow of words and also takes the story in unexpected direc- tions, to which the st
to a story may help a student writer see creative writing not as a linear progression, but an

"AI assisted writing exercises could focus on skills of critical reading, accuracy, argumentation and structu
style, expression, voice and personal reflection." (

"Additionally, teachers could explore with students the ethics and limits of genera- tive AI. How does it
morals and no experience of the world? Is writing with AI tantamount to p
Source's information on equity, diversity, and inclusion elements in artificial intelligence for teaching, le
assessment in higher education
((Use direct quotes and their page numbers))

"In this respect, there is scope for online machine translation technology to be turned into inclusive aut
systems which can cater to different learning styles and add a more robust and rounded dimension to a
course." (p. 132)

"instead, exploring alternative forms of inclusive assessment designs to obtain a more trustworthy perf
students against the learning objectives." (p. 132)

"Outside the academic context, the efficient use of these tools can constitute a valuable asset for future em
only interested in the accurate use of the target language, but also in contracting individuals with the c
content in various languages to extract or summarise specific information concerning a particular langu
particularly needed in minority languages with limited (and mostly monolingual) online resource

"The proliferation of online translator tools and neural models online can help break down language barri
sharing of knowledge and information worldwide. Language learning will not be disadvantaged but esse
train these constantly evolving forms of technology." (p. 133)
It does not apply
It does not apply
It does not apply
"In my work with DEEPs, I am often called upon to adjudicate on alleged acts of plagiarism, collusion and c
teachers have identified in their students’ work. In one case, a campus manager brought forward a stude
contained writing that was well above the student’s own standard of in-class writing. Suspecting a case of
cheating, we interviewed the student who denied receiving help from anyone. She was unable to recount
within her essay, however, and explained that while she understood the content in her own language, she
English. Eager to demonstrate that she had not contracted a ghost writer, the student displayed her inte
revealing the various translation and paraphrase tools she had used in preparing her essay. By piecing tog
language from her search engine history, we were able to ascertain that although the text had passed thr
detection software without incident, less than 25% of the submitted work had actually been produced b
majority of the writing comprised of chunks of text that had been copied directly from the paraphrasing a
(p. 41)

"What became a more perplexing matter, however, was the student’s adamant stance that she had not
integrity norms. She had, in accordance with our plagiarism policy, used in-text citations throughout her w
the names of the original authors and the years of publication, and she had changed the wording of the au
this student’s opinion, the fact that the writing she submitted was not the result of her own cognition, bu
an algorithm, did not constitute an academic integrity breach. On review of the language centre’s own p
found that she actually had a point." (p. 42)

"For many EAL students and L1 English speakers, the rigours of academic writing are challenging and the m
and synthesis of other writers’ words and ideas takes time. Students may not view their use of digital t
academic integrity norms, but rather as a necessary scaffold, or in Dawson’s words a ‘cognitive offloadin
development of mastery (2020, p.40)." (p. 42)
 "What is more, research on second language acquisition (SLA) theories has repeatedly shown that stude
engaged in their learning and that language acquisition is tied to input, output, and frequency effects (for
theories, see Mitchell, Myles, & Marsden, 2013). If our goal is language acquisition, then the use of tran
without understanding said technology is counterproductive. Thus, instructors have to help learners want
autonomously seek to further their proficiency rather than simply complete an assigned task.

"By engaging students in project- and community-based learning that emphasize communication but are r
that is tied to students’ interests and needs rather than “covering the book,” students’ tendency to con
decrease." (p. 789)

"Since students often turn to GT when an assignment seems overwhelmingly daunting, instructors must fa
with what learners typically can and cannot do at each level of proficiency, align assignments with the ACTF
(ACTFL, 2017b) that target the appropriate level of production, remind students that first language (L1) a
shows that it takes young children many months of what is usually intensive exposure before they begin
word phrases together (Clark, 2016), and point out that mistakes are part of the language acquisition p
common L1 overgeneralizations that are actually indicative of language growth, see Clark, 2016; for an
accessibility hierarchy in the L2 context, see Mitchell et al., 2013)." (p. 792)

"In sum, to ensure that classroom pedagogies help students gain proficiency rather than push them towa
instructors must keep in mind several key points. They must (1) evaluate their own knowledge of the ava
tools, (2) directly teach learners how to use appropriate technology responsibly, (3) review their beliefs ab
supportive technologies, (4) familiarize themselves with their institution's policies on academic honesty, an
intend to act and react when such policies are violated, all while offering engaging and motivating instructi
(p. 793)
It does not apply
It does not apply
"It should also be noted that it would be a mistake to consider this issue only relevant to Anglosphere univ
more pressing and may have greater impacts in HEIs and other similar institutions with EMI programmes o
L1 countries" (p. 10)

"MT would provide a solution to many of the issues that international students face in their learning and
some degree levelling the playing !eld and allowing students to focus their energy on their academic en
improving their expertise in a challenging language. From another point of view, this also puts the pri- mac
at risk, if it is to be believed that the English language is a selling point for prospective students and also p
employers. Given the advantages that universities have when they are based in the English- speaking wo
justness of this primacy is open to question." (p. 10)
It does not apply
 "When the student is from an English as an Additional Language (EAL) background, poor expression in wr
attributed to lack of facility with the language, clumsy patchwriting, or the use of an online translation t
TranslateTM (n.d.) (https://translate.google.com.au)." (p. 2)

"As such, it may not be a deliberate or intentional breach of academic conduct. In students with EAL, th
linguistic facility to represent the meaning of a text without resorting to reproducing the author’s actual w
than the few months that our students have been studying at an English-speaking University

"In the essays considered in this exploratory study, we encountered examples of English expression which
student was struggling to develop fluency" (p. 4)
 "The relationship between language proficiency and plagiarism may lead to the conclusion that APT users
students who are not native English speakers, but are instead using English as a Foreign Language (EFL) (R
2017)." (p. 3)

"Non-native English writers were found by Keck (2006) to use more ‘near copies’ of phrases than native En
and the relationship between language proficiency and ability to paraphrase has also been shown as relat
dents’ text comprehension (Erhel & Jamet, 2006). Insufficient knowledge may also lead to students being
way to restate an idea (Rogerson & McCarthy, 2017). Therefore, a lower level of ability in English may
comprehension, resulting in poorer paraphrasing. Several studies have equally found a negative associa- ti
proficiency and engaging in plagiarism, such as Bretag (2007), Li (2015), Pennycook (1996), Marshall and G
Gezgin and Roe (2018), and Chen and Ku (2007). However, Keck (2014) also found that novice writers hav
rely more heavily on copying from source material, so experience may also play a role in the ability to p

"One further complicating factor when understanding APT use and its role as an academically dishonest
clarity as to what constitutes appropriate and inappropriate paraphrasing." (p. 4)

"If an EFL student is introduced to an APT by a teacher, for example in a university English class environme
may find it confusing if it is deemed unacceptable for use in an assess- ment and results in them subseque
plagiarism." (p. 4)

"On the other hand, an EFL student writing in their first language, and then translating it to English, foll
through an APT, may be considered poor academic practice, or a disingenuous representation of their own
definition, plagiarism, This is a debatable example, given that the answer to whether the text is in the stu
not clear cut. Some may argue that the student’s ideas were initially created by the student, and only the p
medium has been changed, where others may state that the student has not met the requirements of w
language and has attempted to deceive the assessor that they have done so, constituting Academic
paraphrasing plagiarism." (p. 5)
 "Research in this area [paraphrasing] appears to concentrate more specifically on second language (L2) s
students per se (For a review see Cumming et al. 2016) although many native English writers may also lack
disseminate academic discourse in their own voice (Bailey & Challen, 2015)." (p. 2)

"Where a student is considered to lack the necessary linguistic skills, the errors or inaccuracies may be inte
as a student having a poor understanding of academic writing conventions rather than recognising that a s
written the work themselves. Where an academic is working in an additional language, they may find the d
or inaccuracies more difficult to identify." (p. 3)
It does not apply
 Research Team Comments (Added on March 10, 2023)

Beatriz Moya's comments (March 14, 2023):

This is a survey study. Participants are teachers who have had various levels of exposure to online translato
methodology would best apply to this case. In my opinion, the methodology that is more closely connecte
be analytical cross-sectional study because it is a study implemented at one point in time which provides i
and outcomes. It is both non-experimental, and it uses a survey for data collection purpo

Beatriz Moya's comments (May 12, 2023):

This study qualifies as a quantitative descriptive study. MMAT has tools to critically appraise this k
 Beatriz Moya's comments (March 15, 2023):
Even when this is not an empirical study, it falls well into the text and opinion catego
opinion is clear; this paper is also of interest to faculty members. The opinions of t
established in logical ways and are built from extant literature, and the sources are
defended. Chris M. Anson is a well-established scholar.
 Sarah Eaton's comments (March 5, 2023):
I agree it is quantitative randomized. I note that on in the abstract and in the conclusion (p. 34) the researc
but on p. 28 they say that N = 64. I could find no indication of why there was a discrepancy between these
note this on the spreadsheet. There can be an additional column for our notes.

The small sample size is a limitation not mentioned by the researchers in their paper, but we can menti
appraisal.

Beatriz Moya's comments (March 15, 2023):

This study is also organized into three parts "Grounded in the statistical analyses, three main re- search
addressed: (1) whether PREFER benefits learners' writing per- formance, compared with the other two t
tools; (2) the relation between the effects of PREFER and learner differences (i.e., language proficiency
involvement); (c) students' perceptions of paraphrasing and of the effectiveness of PREFER." The first part
test data analysis of students' performance. The students were randomly assigned to groups. The second p
as it seems they only analyzed data from the group that used PREFER. The last part involved a question
unclear how data was coded. Considering that the authors consider this study an experiment, I believe it w
analyze this as a randomized controlled study.

If this were a randomized controlled study, I could also say that the information provided in the article all
that there was true randomization and that the allocation was concealed. Participants were blind to treatm
also the outcomes were assessed in ways that ensured anonymity. It also seemed that the treatment gr
identically and that the participants' results were analyzed in the right groups. My only concerns were the
similarity of groups at the baseline and whether those who delivered the treatment were blind to the tre
 Beatriz Moya's comments (March 15, 2023):
Even when this is not an empirical study, it falls well into the text and opinion categor
source of opinion is clear; this paper is also of interest to faculty members. The opinio
are established in logical ways and are built from extant literature. Additionally, the
logically defended. Phillip Dawson is a well-established scholar in the academic in
 Beatriz Moya's comments (March 15, 2023):
Even when this is not an empirical study, it falls well into the text and opinion catego
and 7, the source of opinion is clear; this paper is also of interest to faculty members
the author are established in logical ways and are built from extant literature. Also, th
logically defended. Cara Dinneen is the Associate Director, Learning and Teaching
University College.
 Beatriz Moya's comments (March 15, 2023):
Even when this is not an empirical study, it falls well into the text and opinion categor
and 8, the source of opinion is clear; this paper is also of interest to faculty members
the author are established in logical ways and are built from extant literature. The s
logically defended.
 Beatriz Moya's comments (March 15, 2023):
The study is well articulated and shows a clear congruity between research methodol
questions, research methodology and data collection methods, and research metho
interpretation of the results. In my opinion, the students' voices are represented pr
conclusions clearly flow from data. Some elements that were a bit unclear from my p
the philosophical perspective of the researcher and his positionality. A last element
been improved is the fact that the study involved the author's students, and I wish I
information regarding the strategies taken to ensure that the research was carried
 Beatriz Moya's comments (March 14, 2023):
This is a two-part study. The first one focuses on comparing Sparks outputs with human-generated text
another language model: "In the first study, we compare the outputs from our custom decoding metho
baseline as well as to a human-written gold standard, reporting on the diversity and coherence of all outp
three hypotheses behind the first study are the following "We have three hypotheses

• H1: The custom decoding produces more coherent and diverse outputs than a baseline system, but less
outputs than a human-written gold standard.
• H2:The custom decoding performs consistently across many different topics.
• H3: There is significant variance across output quality in topic+prompt combinations." (p.

Data collection involved experts annotating outputs on a 0 - 4 scale focused on coherence. On this scale, 0
any sense, one was Not true, 2 was Opinion/Don't know, three was Sometimes true, and 4 was Generally
worked on 250 randomly assigned outputs.

The second study addressed the perceptions of STEM graduate students as Sparks' users: "In our second
graduate students from STEM disciplines write tweetorials with our system and report on how they thou
use of the sparks." (p. 1003).

The research questions of the second study were the following: "RQ1: In what ways do writers make use
outputs? RQ2: What attributes of language model outputs, if any, correlate with writer usage and satis

Data collection included a survey and a semi-structured interview to discuss the usefulness of the system
with other kinds of systems experienced by the participants.

Beatriz Moya's comments (May 12, 2023):

This study qualifies as a mixed methods study, and MMAT has tools to critically appraise the quality of t
 Beatriz Moya's comments (March 15, 2023):
The study is well articulated and shows a clear congruity between research methodol
questions, research methodology and data collection methods, and research metho
interpretation of the results. In my opinion, the students' voices are represented pr
conclusions clearly flow from data. Some elements that were a bit unclear from my p
the researcher's positionality. A last element that could have been improved is the fa
did not show whether it was approved by a Research Ethics Board.
 Beatriz Moya's comments (March 16, 2023):
Students were asked to produce summaries and grade a group of summaries developed by AI and other
were also asked to classify the summaries into two categories: AI and human generate

Beatriz Moya's comments (May 12, 2023):

This study qualifies as a quantitative descriptive study. MMAT has tools to critically appraise this k
 Beatriz Moya's comments (March 16, 2023):
The study is focused on the capabilities of paraphrasing and language translation tools. The authors explor
some examples. When describing the method of analysis, the authors say: "In order to investigate the
paraphrasing tools substituted recognized and expected medical terms for unusual synonyms, we selecte
we had identified as particularly unusual. We did not know the provenance of these essays, although th
evidence that they might have arisen from a single seed document which was an essay submitted by one s
cohort" (p. 8). To me, this is a novel kind of research, but the authors clearly developed a deep analysis o
suggesting this is qualitative research.
 Beatriz Moya's comments (March 16, 2023):
The authors suggest this is a "detailed literature review" (p. 2), I did not see a description of the literature
 Beatriz Moya's comments (March 16, 2023):
The study is focused on comparing the outputs of paraphrasing tools. I recognize some parallels with Pre
study (row 16). When the authors explain the methodology they said: "In order to test the quality of outpu
free Internet based paraphrasing tools and how the originality of the output is assessed by Turnitin®, the f
was conducted. A paragraph from an existing publication by this article’s authors from a prior edition of the
of Educations Integrity (IJEI) was selected to be the original source material (McCarthy & Rogerson, 2009, p
paraphrasing tool processes an in-text citation, one in-text citation was included (Thatcher, 2008). A set o
entries from the reference list of the same article were also selected to test how references are interpreted
a novel kind of research, but the authors clearly developed a deep analysis of the tools' outputs suggesti
research.
 Beatriz Moya's comments (March 16, 2023):
Even when this is not an empirical study, it falls well into the text and opinion categor
8, and 9, the source of opinion is clear; this paper is also of interest to faculty member
the author are established in logical ways and are built from extant literature. The s
logically defended. Michael Sharples is a renowned scholar and currently an Emeritus
Open University.
AI used for ethical support of teaching AI used for academic dishonesty (cheating,
and learning? (Y/N) plagiarism, etc.)? (Y/N)

Y Y
Y Y
Y N
N Y
Y Y
Y Y
Y Y
N Y
Y Y
N N
N Y
Y Y
N Y
Y Y
 Methodology Include? (Y/N)

Quantitative descriptive (MMAT) Y

Text and Opinion (JBI) Y
Randomized controlled trials (JBI) Y
Text and Opinion (JBI) Y
Text and Opinion (JBI) Y
Text and Opinion (JBI) Y
Qualitative research (JBI) Y
Mixed methods (MMAT) Y
Qualitative research (JBI) Y
Quantitative descriptive (MMAT) Y
Qualitative research (JBI) Y
Text and Opinion (JBI) Y
Qualitative research (JBI) Y
Text and Opinion (JBI) Y
 QUALITATIVE: Is there congruity
between the stated philosophical
Assessment perspective and the research
methodology? (JBI) (Please write 1
for YES and 0 for NO)

Medium
High
Medium
High
High
High
High 1
High
High 1
High
Medium 1
High
High 1
High
 CRITICAL APPRAISAL

QUALITATIVE: Is there congruity QUALITATIVE: Is there congruity

between the research methodology between the research methodology
and the research question or and the methods used to collect
objectives? (JBI) (Please write 1 for data? (JBI) (Please write 1 for YES
YES and 0 for NO) and 0 for NO)
1 1
1 1
1 1
1 1
 CRITICAL APPRAISAL

QUALITATIVE: Is there congruity QUALITATIVE: Is there congruity

between the research methodology between the research methodology
and the representation and analysis and the interpretation of results?
of data? (JBI) (Please write 1 for YES (JBI) (Please write 1 for YES and 0 for
and 0 for NO) NO)
1 1
1 1
1 1
1 1
 QUALITATIVE: Is there a statement QUALITATIVE: Is the influence of the
locating the researcher culturally or researcher on the research, and vice-
theoretically? (JBI) (Please write 1 for versa, addressed? (JBI) (Please write
YES and 0 for NO) 1 for YES and 0 for NO)
0 0
0 1
0 0
0 1
 QUALITATIVE: Is the research ethical
QUALITATIVE: Are participants, and according to current criteria or, for
their voices, adequately recent studies, and is there evidence
represented? (JBI) (Please write 1 for of ethical approval by an
YES and 0 for NO) appropriate body? (JBI) (Please
write 1 for YES and 0 for NO)
1 1
1 0
0 1
0 1
 QUALITATIVE: Do the conclusions
drawn in the research report flow
QUALITATIVE: Critical appraisal
from the analysis, or interpretation,
points
of the data? (JBI) (Please write 1 for
YES and 0 for NO)

0
0
0
0
0
0
1 8
0
1 8
0
1 7
0
1 8
0
 RANDOMIZED CONTROLLED TRIALS:
RANDOMIZED CONTROLLED TRIALS:
Was true randomization used for
Was allocation to treatment groups
assignment of participants to
concealed? (JBI) (Please write 1 for
treatment groups? (JBI) (Please write
YES and 0 for NO)
1 for YES and 0 for NO)
1 1
RANDOMIZED CONTROLLED TRIALS: RANDOMIZED CONTROLLED TRIALS:
Were treatment groups similar at Were participants blind to
the baseline? (JBI) (Please write 1 for treatment assignment? (JBI) (Please
YES and 0 for NO) write 1 for YES and 0 for NO)
0 1
RANDOMIZED CONTROLLED TRIALS: RANDOMIZED CONTROLLED TRIALS:
Were those delivering treatment Were outcomes assessors blind to
blind to treatment assignment? (JBI) treatment assignment? (JBI) (Please
(Please write 1 for YES and 0 for NO) write 1 for YES and 0 for NO)
0 1
 RANDOMIZED CONTROLLED TRIALS:
RANDOMIZED CONTROLLED TRIALS:
Was follow up complete and if not,
Were treatment groups treated
were differences between groups in
identically other than the
terms of their follow up adequately
intervention of interest? (JBI)
described and analyzed? (JBI)
(Please write 1 for YES and 0 for
(Please write 1 for YES and 0 for
NO)
NO)
1 0
RANDOMIZED CONTROLLED TRIALS: RANDOMIZED CONTROLLED TRIALS:
Were participants analyzed in the Were outcomes measured in the
groups to which they were same way for treatment groups?
randomized? (JBI) (Please write 1 (JBI) (Please write 1 for YES and 0 for
for YES and 0 for NO) NO)
1 1
RANDOMIZED CONTROLLED TRIALS: RANDOMIZED CONTROLLED TRIALS:
Were outcomes measured in a Was appropriate statistical analysis
reliable way? (JBI) (Please write 1 used? (JBI) (Please write 1 for YES
for YES and 0 for NO) and 0 for NO)
1 1
RANDOMIZED CONTROLLED TRIALS:
Was the trial design appropriate,
and any deviations from the
standard RCT design (individual RANDOMIZED CONTROLLED TRIALS
randomization, parallel groups) Critical appraisal points
accounted for in the conduct and
analysis of the trial? (JBI) (Please
write 1 for YES and 0 for NO)

0
0
0 9
0
0
0
0
0
0
0
0
0
0
0
 TEXT AND OPINION: Does the
TEXT AND OPINION: Is the source of
source of opinion have standing in
the opinion clearly identified? (JBI)
the field of expertise? (JBI) (Please
(Please write 1 for YES and 0 for NO)
write 1 for YES and 0 for NO)
1 1
1 1
1 1
1 1
1 1
1 1
 TEXT AND OPINION: Are the TEXT AND OPINION: Is the stated
interests of the relevant population position the result of an analytical
the central focus of the opinion? process, and is there logic in the
(JBI) (Please write 1 for YES and 0 for opinion expressed? (JBI) (Please
NO) write 1 for YES and 0 for NO)
1 1
1 1
1 1
1 1
1 1
1 1
 TEXT AND OPINION: Is any
TEXT AND OPINION: Is there
incongruence with the
reference to the extant literature?
literature/sources logically
(JBI) (Please write 1 for YES and 0 for
defended? (JBI) (Please write 1 for
NO)
YES and 0 for NO)
1 1
1 1
1 1
1 1
1 1
1 1
 QUANTITATIVE DESCRIPTIVE: Is the
TEXT AND OPINION: Critical
sampling strategy relevant to
appraisal score
address the research question?

0 0
6
0
6
6
6
0
0
0
0 1
0
6
0
6
QUANTITATIVE DESCRIPTIVE: Is the
QUANTITATIVE DESCRIPTIVE: Are
sample representative of the target
the measurements appropriate?
population?

0 1
0 1
 QUANTITATIVE DESCRIPTIVE: Is the
QUANTITATIVE DESCRIPTIVE: Is the
statistical analysis appropriate to
risk of nonresponse bias low?
answer the research question?

1 1
1 1
 MIXED METHODS: Is there an
QUANTITATIVE DESCRIPTIVE: Critical adequate rationale for using a
appraisal score mixed methods design to address
the research question?

3
0
0
0
0
0
0
0 1
0
4
0
0
0
0
MIXED METHODS: Are the different MIXED METHODS: Are the outputs
components of the study effectively of the integration of qualitative and
integrated to answer the research quantitative components
question? adequately interpreted?
1 1
MIXED METHODS: Are divergences MIXED METHODS: Do the different
and inconsistencies between components of the study adhere to
quantitative and qualitative results the quality criteria of each tradition
adequately addressed? of the methods involved?
0 1
MIXED METHODS: Critical appraisal
score

0
0
0
0
0
0
0
4
0
0
0
0
0
0
 Multiple
Countries
((If there is
one than
Author(s) Specific
more
((Please Country city(ies),
country,
Include last Citation ((Please town(s)
Year Reviewer please add
name and (APA 7) use list and
the names
first name below)) campus ((If
of other
initial)) applicable)
countries
mentioned
in the
text))

Alonso, A.
N. (2022).
Online
translators
in online
language
assessment
United It does not It does not
Alonso, A. 2022 s. CALL-EJ, Beatriz
Kingdom apply apply
23(3), 115-
135.
http://callej
.org/journa
l/23-3/Alon
so2022.pdf
 Anson, C.
M. (2022).
AI-Based
text
generation
and the
social
constructio
n of United
Anson, C. It does not It does not
2022 “fraudulent Beatriz States of
M. apply apply
authorship” America
:A
revisitation
.
Compositio
n Studies,
50(1), 37-
46.
 Chen, M.
H., et al.
(2015).
Developing
a corpus-
based
paraphrase
tool to
improve
EFL
Chen, M. learners'
H., Huang, writing National
It does not
S. T., 2015 skills. Beatriz Taiwan Tsing Hua
apply
Chang, J. S., Computer University
& Liou H. C. Assisted
Language
Learning,
28(1), 22-
40.
http://dx.d
oi.org/10.1
080/09588
221.2013.7
83873
 Dawson, P.
(2020).
Defending
assessment
security in
a digital
world:
Preventing
e-cheating
and
supporting
academic
integrity in
higher
education. It does not It does not
Dawson, P. 2020 Beatriz Australia
https://ww apply apply
w.taylorfra
ncis.com/b
ooks/mono
/10.4324/9
780429324
178/defend
ing-
assessment
-security-
digital-
world-
phillip-
dawson
 Dinneen, C.
(2021).
Students’
use of
digital
translation
and
paraphrasi
ng tools in
written
assignment
s on direct
It does not It does not
Dinneen, C. 2021 entry Beatriz Australia
apply apply
english
programs.
English
Australia
Journal,
37(1).
https://files
.eric.ed.gov
/fulltext/EJ
1341751.p
df
 Ducar, C., &
Schocket,
D. H.
(2018).
Machine
translation
and the L2
classroom:
Pedagogica
Ducar, C, & United
l solutions It does not It does not
Schocket, 2018 Beatriz States of
for making apply apply
D. H. America
peace with
Google
Translate.
Foreign
Language
Annals
51(4), 779-
795.
 Fyfe, P.
(2022).
How to
cheat on
your final
paper:
Assigning
AI for
United
student It does not NC State
Fyfe, P. 2022 Beatriz States of
writing. AI apply University
America
& Society.
https://link.
springer.co
m/article/1
0.1007/s00
146-022-
01397-z
 Gero, K. I.,
et al.
(2022).
Sparks:
Inspiration
for science
writing
using
language
Gero, K., models. DIS United
It does not Columbia
Liu, V., & 2022 '22: Beatriz States of
apply University
Chilton, L. Designing America
Interactive
Systems
Conference,
Association
for
Computing
Machinery,
Inc.
 Groves, M.,
& K. Mundt
(2021). A
ghostwriter
in the
machine?
Attitudes of
academic
staff
towards
machine
translation University
Groves, M., use in of Surrey &
United It does not
& K. 2022 internation Beatriz University
Kingdom apply
Mundt alised of
Higher Nottingham
Education.
Journal of
English for
Academic
Purposes
50.
https://doi.
org/10.101
6/j.jeap.20
21.100957
 a, S. (2022).
Risks and
benefits of
AI-
generated
text
summarizat
ion for
expert level
content in
graduate
health
informatics
. 2022 IEEE
10th
Merine, R., United
Internation It does not Indiana
Purkayasth 2022 Beatriz States of
al apply University
a, S. America
Conference
on
Healthcare
Informatics
(ICHI),
Institute of
Electrical
and
Electronics
Engineers
Inc.
https://iee
explore.iee
e.org/docu
ment/9874
 Prentice, F.
M., &
Kinden, C.
E. (2018).
Paraphrasi
ng tools,
language
translation
Prentice, F. tools and
M., & plagiarism: It does not It does not
2018 Beatriz Australia
Kinden, C. An apply apply
E. exploratory
study.
Internation
al Journal
for
Educational
Integrity
14.
 Roe, J. &
Perkins, M.
(2022).
What are
Automated
Paraphrasi
ng Tools
and how do
we address
them? A
review of a
Roe, J. & It does not
2022 growing Beatriz Singapore Vietnam
Perkins, M. apply
threat to
academic
integrity.
Internation
al Journal
for
Educational
Integrity
18(1).
 Rogerson,
A. M., &
McCarthy,
G. (2017).
Using
internet
based
paraphrasi
ng tools:
Rogerson, Original University
A. M., & work, It does not of
2017 Beatriz Australia
McCarthy, patchwritin apply Wollongon
G. g or g
facilitated
plagiarism?
Internation
al Journal
for
Educational
Integrity,
13.
 Sharples,
M. (2022).
Automated
essay
writing: An
AIED
opinion.
Sharples, Internation United It does not It does not
2022 Beatriz
M. al Journal Kingdom apply apply
of Artificial
Intelligence
in
Education,
32(4),
1119-1126.
 "This study
online
describes
survey
some of
directed to
the main
mostly
uses of
DATAonline
EXTRACTION higher
CRITERIA
education
machine
and
translation
secondary
tools,
education
Research
commentin
Purpose(s) teachers.
Type of on the question(s) Study
Participant If multiple g((Please The main
source ((Please Design
s ((Please disciplines pedagogica
add direct aims of this
((Please Discipline l value of add direct ((Please
use list or Other quotes/pag research
use list these, quotes/pag use list
below)) please list e and are the
below)) presents e below))
numbers)) following:
the results numbers))
1. How do
of an online
students
survey that
and
investigate
teachers
d language
use online
teachers’
translators
perceptions
in language
of the
class?
impact that
2. What are
online
language
language
teachers’
teaching
opinions on
and
the impact
assessment
of online
may have
translators
on the use
in online
of online
summative
translators
assessment Quantitativ
6=journal It does not and its
6=multiple Humanities types and e
article apply connection
the various Descriptive
with
levels of
plagiarism
learning, on
detection
online
tools such
plagiarism
as Turnitin
detection
and with
tools, and
other
on the use
online
of online
language
translators
reference
in
tools.
combinatio
Finally,
n with
some
other
reflections
online
and
language
recommen
reference
dations will
tools?
follow on
3. What
good online
can be
teaching
done to
and
improve
assessment
online
practices
assessment
and the
practice,
implement
preserve
ation of
 published
material.
This is
because
definitions
of
plagiarism
are socially
constructe
d and tied
to context-
sensitive
cycles of
reward for
the
production
—and
therefore
the
6=journal It does not ownership It does not
6=multiple Humanities Theoretical
article apply —of certain apply
kinds of
texts.
Helping
students to
understand
plagiarism
means
showing
them these
contextuall
y-specific
constructs
of text
ownership,
rather than
assuming
that any
unattribute
d text,
published
anywhere,
in any
form,
constitutes
plagiarism.
I then turn
to AI-based
NLP
systems.
Teachers
who learn
what these
systems
can do
usually
respond
with the
same hand-
wringing
and
 tool best
benefits
EFL
learners’
writing
"To meet performanc
the need of e: PREFER,
developing Longman
EFL English
learners’ Dictionary
paraphrasi Online or
ng Thesaurus.
competenc com?
e through a (2) To what
CALL tool, extent does
Quantitativ
we PREFER
6=journal It does not e
2=students Humanities developed benefit
article apply Randomize
a web- different
d
based EFL
paraphrase learners’
suggestion paraphrasi
system and ng
assessed its performanc
effectivene e?
ss for the (3) What
purposes of are the EFL
the study." learners’
(p. 25) perceptions
of
paraphrasi
ng and the
effectivene
ss of
 "The book
explores
the current
and likely
future
states of e-
cheating,
what can
be done to
secure
Multidiscipl It does not
2=book 6=multiple Other assessment Theoretical
inary apply
in this new
world, and
some of
the
challenges
we are
likely to
face along
the way."
(p. 1)
 "The
purpose of
this paper,
however, is
to look
more
closely at
6=journal It does not what is It does not
2=students Humanities Theoretical
article apply happening, apply
and the
ramificatio
ns, rather
than
considering
the why."
(p. 42)
 and
"This article opportuniti
aims to es does this
advance technology
the present to
theoretical L2 students
discussion and
surroundin instructors
g MT in the ? Is it
L2 possible to
classroom deter
to help students
instructors from
understand overusing
what MT this
6=journal It does not can and technology
6=multiple Humanities Theoretical
article apply cannot do ? How can
and help teaching
them equip students
their about the
students to strengths
use the and
technology weaknesse
in an s of this
educational technology
ly and transform
intercultura some of
lly the
respectful challenges
manner." into
(p. 780) learning
opportuniti
 compositio
l
n. How
experiment
might we
that assigns
write with
undergradu
these
ates to
tools?
“cheat” on
What skills
a final class
might
essay by
writing
requiring
with AI
their use of
newly
text-
require?
generating
How does
AI
this
software."
challenge
(p. 1)
our
assumption
"In this
s about
paper, I
textual
Computer share how
6=journal It does not communica
2=students Science/En students Qualitative
article apply tion? And
gineering experience
what
d those
potential
issues,
risks and
connect
harms must
their
we
insights to
navigate?
broader
“Cheating”
conversatio
offers a
ns in the
starting
humanities
point, but it
about
opens onto
writing and
much more
communica
complex
tion, and
and
explain
consequent
their
ial topics,
relevance
from
for the
human and
ethical use
nonhuman
and
agency to
evaluation
the threats
of language
of
models."
disinformat
(p. 1)
ion and
algorithmic
bias. Thus,
the
assignment
moves
students
from a
familiar
ethical
context (is
this
plagiarism?
) to new
ethical
questions
(whose
 which tend around the
to deal following
with research
common question:
objects and RQ: How
relations. can
Science language
writing model
support outputs
requires a support
system to writers in a
demonstrat creative
e pro but
ciency constrained
within an writing
area of task?" (p.
expertise." 1003)
(p. 1003)
"In
Computer
6=journal It does not "Our work particular, Mixed
2=students Science/En
article apply aims to we pose Methods
gineering
study how the
text following
generated research
by questions:
language RQ1: In
models what ways
might be do writers
used by make use
writers in a of language
science model
writing outputs?
task. There RQ2: What
is relation attributes
to a natural of language
language model
generation outputs, if
task like any,
summarizat correlate
ion, with writer
because we usage and
are satisfaction
concerned ?" (p. 1010)
with
specific
factual
information
(as
opposed to
commonse
nse
knowledge)
but we take
a human
centered
approach
where the
language
model
provides
suggestions
 questions:
1. To what
degree do
academic
staff see
the use of
MT by
"To move students
this issue for their
forward, academic
the present work as
study acceptable
examines ?
perceptions 2. Is use of
on the use MT as an
of GT and aid for
6=journal It does not
1=Faculty Humanities other freely academic Qualitative
article apply
available reading and
online writing
translation considered
tools by academic
university misconduct
students as ; and, if so,
reading and under what
writing circumstan
aid." (p. 2) ces?
3. Is there
currently a
university
policy on
student use
of MT?
Should
 and its risks
and
benefits to
students. In
addition,
through an
experiment
conducted
among
graduate
health
informatics
students at
Indiana
University,
Computer Quantitativ
4=conferen It does not this paper It does not
2=students Science/En e
ce paper apply would apply
gineering Descriptive
illustrate
the quality
of the text
summary
generated
utilizing the
abstractive
NLP
technique."
(p. 567)

"This study
will look
into the
possibilities
 1. Were
students
using
online
paraphrasi
ng tools to
manipulate
work which
was written
in English
and which
had not
been
authored
by them?
2. Were
students
who had
English as
6=journal Health It does not an
6=multiple Other Qualitative
article Sciences apply Additional
Language
(EAL)
composing
work in
their first
language
and then
translating
this
through
online
language
translation
tools?
3. Are
there
indicators
which can
identify the
use of on-
line
paraphrasi
ng tools?"
(p. 1)
 academic
misconduct
by aiding
text-based
plagiarism,
that of
Automated
Paraphrasi
ng Tools
(APTs). We
begin by
describing
the origins
of APTs and
their use in
academic
work. We
then
explore the
6=journal relationshi It does not
1=Faculty Other Unclear Theoretical
article p between apply
language
proficiency
and APT
use, and
how APTs
may or may
not be used
for in an
academicall
y dishonest
way,
referring to
case
studies
from
Dinneen
(2021) and
Prentice
and Kinden
(2018).
Finally, we
propose
solutions
and
relevant
limitations
to tackling
the
problem of
APTs in
academia,
as well as
areas for
future
research."
(p. 2)
 phenomen
on. What
became
apparent
was that
the ease of
access to
and use of
such tools
was greater
than first
thought.
Consequen
tly it is
important
6=journal It does not to bring the It does not
1=Faculty Humanities Qualitative
article apply use and apply
operation
of
paraphrasi
ng tools to
a wider
audience to
encourage
discussion
about
developing
individual
writing
skills and
improve
the
 "How do
such
Transforme
r AI
systems
work? How
can we
detect
whether an
aca- demic
essay has
been
6=journal Educational written by It does not
1=Faculty Other Theoretical
article Technology an AI? apply
Could any
good come
from
widespread
use of such
technology
? What
does this
mean for
the AIED
community
?" (p. 1120)
 “because when it and
students comes to students on
want to labelling (or plagiarism
"The
obtain a not) online detection,
method of
higher machine designing
investigatio
mark” translation appropriate
ethical
n used to
(78%), tools as a implication
online
find an
Data to followed by source of assessment
s of
answer Summary Author plagiarism.
collection the Conclusion s,
artificial
Participant Interventio the above students’ of Identified Whereas encouragin
(if
research s/ Closing intelligence
s n if Findings
lack of / Limitations many of ingteaching,
proper
relevant) ((Please
Informatio applicable questions language Results ((Please
((Please addthe direct reference
learning,
n (Added (Added on was an ((Use knowledgedirect add direct respondent of work
add direct
online quotes and quotes and quotes and research,
on March March 14,
quotes and (59.4%), s agreed
their page submitted,
and
10,"the
2023) 2023) survey their page their page
laziness that it assessment
and writing
participants their page
carried out numbers)) numbers)) numbers))
numbers)) (46.9%), constitutes instructions
in higher
(n=32) in June
lack of time an illicit education
on what it
were 2020 right
(40.6%), form of ((Useis anddirect
it is
teaching in after all the
checking language in not
Higher end-of-year
purposes that the considered
Education online
(37.5%) students licit before
(90%), assessment
and lack of did not put attempting
although s and exam
academic in the an online
some of boards
malpractice intellectual assessment
them were were over.
awareness effort to ." (p. 119)
also Most of the
(34.4%). In produce it,
teaching in participants
a smaller others "I believe
secondary (n=32)
percentage argue that academic
schools and were
, in the integrity
most of teaching in
respondent process of policies
them have Higher
s also generating often need
experience Education
mentioned the to be
teaching (90%),
that some automatic customized
various It does not although It does not
students translation, for modern
languages apply some of apply
worry that content is foreign
such as them were
other not language
Spanish, also
students compromis courses
French, teaching in
use online ed, and where
English, secondary
translators students online
German, schools and
and obtain can be translators
Italian, most of
higher taught how are part of
Chinese, them have
marks to use the widely
Arabic, experience
leaving these tools available
Russian, teaching
them effectively online
Korean and various
behind to enhance tools." (p.
Danish, and languages
(3.1%), or their 129)
at various such as
that they language
levels of Spanish,
feel they learning "Having
the CEF French,
lack experience said this,
from A1 to English,
confidence in the jury is
C2 (see German,
in their combinatio out on this
Figure 1 Italian,
language n with issue since
below)." (p. Chinese,
skills other one can
120) Arabic,
(3.1%)." (p. online tools argue that
Russian,
126) such as by making
Korean and
dictionaries use of
Danish, and
"It was or various
at various
rather concordanc language
levels of
shocking to ers." (p. referencing
the CEF
find out 132) online tools
from A1 to
 context.
But where
does
acceptabilit
y end?" (p.
43)

"In this
case, I
provided
the
information
, but who
lays claim
to the
expression
of that
information
in a
It does not It does not It does not It does not It does not It does not coherent
apply apply apply apply apply apply string of
sentences?
" (p. 43)

"Many
researchers
in STEM
fields might
find auto-
written
methods
sections
acceptable,
just as
many STEM
editors and
researchers
find the
verbatim
replication
of methods
sections in
different
published
articles
acceptable
in a process
my co-
researchers
and I have
called “text
recycling”
(see Anson,
Hall,
Pemberton,
and
Moskovitz;
Pemberton,
Hall,
Moskovitz,
and Anson;
EFL college PREFER, an assist the
paraphrasi example, n, based on
freshmen online students’
ng “on the students’
in a public corpus- paraphrasi
performanc whole” and proficiency
university based ng process
e." (p. 28) “to sum (paraphrasi
of an Asian paraphrasi in looking
up” share ng score in
country ng for a
"Experimen the same pretest),
were assistance proper
tal Chinese motivation
recruited system. phrase.
procedures translations (frequency
for Allowing Theasarus.c
To evaluate “!!"%” and of
participatio multi-word om
the “!!! !” in consultatio
n in this input and sometimes
effectivene our n), and
study. returning did not
ss of bilingual involvemen
These promptly have
PREFER, we corpus. t (the
students with a list enough
drew a However, number of
came from of information
comparison “to sum paraphrase
the De- paraphrase for
between up”, d phrases It does not
partment s in En- paraphrasa
PREFER and compared in pretest), apply
of glish and ble or
the two with “on revealed
Mathemati Chinese, correspond
most the whole”, that the
cs (n = 34), along with ing items.
popular seldom less
Material usage PREFER
dictionaries appears in proficient,
Science and patterns does not
, Longman the highly
Engineering and have either
English beginning motivated,
(14), example of these
Dictionary of an and
Chemical sentences, limitations,
Online and article. conservativ
En- PREFER and ranked
Thesaurus. Phrase e students
gineering pro- vides highest in
com among usages and showed
(12), and substantial the group
students in positions in significant
Industrial support for as a viable
this sentences progress in
Engineering EFL and
country. A should be the
and learners to effective
complete included, paraphrasi
Engineering vary their productive
administrat which are ng task.
Manageme expressions language
 experts will
not be able
to spot
computer
translations
.What can
we do
when
assessing
language
translations
– or any
other
offloadable
tasks – if
we cannot
spot the AI
anymore?"
(p. 88)
It does not It does not It does not It does not It does not It does not
apply apply apply apply apply apply
"It is even
harder for
technology
tools to
navigate
the murky
ethical
territory
between
cheating
and
legitimate
assistance."
(p. 89)

"We, as an
education
community
, need to
engage in
sophisticat
ed debate
about the
boundaries
between
help and
cheating."
(p. 90)

"Cheating is
ultimately
about
ethics, and
while that
is
something
we may be
able to
outsource
to
 ng tools
"The paper
does not fit
has
neatly into
demonstrat
current
ed a
definitions
number of
of
illicit uses
academic
of
integrity,
translation
highlighting
and
the need
paraphrase
for a broad
tools
review of
identified
policies and
by higher
considerati
education
on of
professiona
behaviours
ls and has
they may
reviewed
be failing to
definitions
identify."
It does not It does not It does not It does not It does not of
(p. 44)
apply apply apply apply apply academic
integrity to
"While our
reveal a
policies on
systematic
academic
failure to
integrity
appropriate
have
ly define it
evolved, to
in light of
some
technologic
extent, to
al
encompass
developme
the digital
nts." (p. 47)
world, the
abovement
"This
ioned
silence on
articles on
the
cross-
appropriate
language
use of
plagiarism,
digital tools
backtransla
in
tion and
academic
paraphrase
integrity
fraud are
policy
evidence
needs to be
that
addressed.
practice
Its absence
and misuse
impacts
is evolving
learning
at a greater
and
pace." (p.
teaching in
44)
ELICOS and
higher
"Students’
education
use of
institutions,
digital tools
leaving
for
learners
language
less
enhanceme
equipped
nt needs to
to deal
be a
with the
considerati
 the recognize across the
"GT has
strengths that K–12 and
also
and reasoned, postsecond
improved
limitations nuanced, ary
its ability to
of this and spectrum
detect and
technology, culturally work
accurately
explores sensitive primarily
translate
21st- use of with
so-called
century language is beginning-
archaic or
pedagogica a uniquely and
less-
l solutions human intermediat
commonly-
designed to ability — e-level
used
harness the and indeed language
words." (p.
capabilities a learners
783)
of both MT marketable whose
It does not It does not and It does not skill. expression
"Similarly,
apply apply alternative apply Although in the L2 is
since the
technologie technology basic
implement
s, and can enough
ation of the
suggests certainly that GT can
new AI
venues for inform L2 generally
algorithm,
future learning, it handle
GT has
research cannot —at their
improved
with the least at queries
its ability to
goal of present— with
avoid literal
ensuring fully success."
translations
learners’ replicate (p. 782)
." (p. 783)
academic natural
growth in language "Christoph
"In
line with production. er
addition,
ACTFL's " (p. 793) Manning,
GT also
Can-Do professor
demonstrat
 less about institutiona
didn’t? plagiarism,
the l plagiarism
How did authorship,
content." policies
the AI- and writing
(p. 6) would have
generated pedagogy."
us believe.
content (p. 2)
"Other And
relate to
students humanities
your own? "When
took the scholars
How did it instructors
opposite can play an
affect what strategicall
approach: important
you might y
chang- ing role in
have decriminali
their own help- ing
thought ze
writing explain and
about or plagiarism,
style to navigate
written? Do they also
"The class better those
you feel open
enrolled 20 match GPT- complexitie
like you opportuniti
students 2’s outputs s." (p. 9)
“cheated”? es for
from first- and “to
To what creative
years to create a "How do
degree is and critical
seniors It does not smoother It does not institutiona
this paper exploration
with a apply flow.”" (p. apply lized
“your” ." (p. 3)
variety of 6) definitions
writ- ing?
majors and of
Do you "Throwing
con- "Ironically, plagiarism
expect a AI into the
centrations when block the
reader mix
." (p. 4) encouraged possibilities
would updates
to cheat, of writing
notice GPT- these
these with
2’s text questions
students computatio
versus your in
instead nal
own? interesting
ended up assistance
Would you ways—and
reinforcing or even
use this helps to
their own partnership
tool again, move
capacities ? What
and in what discussions
as writers, kinds of
circumstan about
the creative or
ces? And, language
distinctiven critical
ultimately, models
ess of their framework
what ideas beyond
voices, and s do
about simplistic
the plagiarism
writing, AI, contrasts of
tradeoffs if policies
or human
not impose or
humanness originality
outright preclude,
did the vs. machine
sacrifices of especially
experiment imitation."
using AI." compared
test or (p. 3)
(p. 6) to what we
change?
really value
Finally, all 'Discussion
"The eight for
papers s of
students students?
included an posthuman
who How are
Appendix in ism are
concluded these
which more open
that they values
students to the
had indeed supported
provided a alternatives
“cheated” (or not) by
“revealed” , and help
on the the
version of signal how
assignment learning
their essay to
were activities
with the AI- approach
committed that AI
generated writing
to their seems
text with AI
own voice likely to
highlighted. beyond the
and impact (like
 Study 1: scientific with
area’ that select their help with model, the
"For concept translation
would give own topics. translation computer
biology, we intended to can occur
context to However, by scientists
had 3 inspire more fre-
their topic this providing who
senior writers." (p. quently
and aid the introduced detailed developed
undergradu 1002) throughout
system to a large sentences, and trained
ate this setting
correctly confoundin and by the model,
students "We of writing,
interpret g factor, as providing the
majoring in present a or perhaps
their topic. di erent perspective company o
biology; for system that that the
Then they topics are s that ering
environ- aims to translation
were given more or helped access to
mental inspire use case
20 minutes less di cult them the model,
science, we domain requires
to interact to explain understand or the
had 2 experts more
with the and make their various
senior when sparks as
system and interesting, reader." (p. anonymous
undergradu writing part of its
complete and di 1017) authors
ate tweetorials process."
the writing erent whose text
students on a topic (p. 1013)
task. topics may "We also makes up
majoring in of their
Mouse elicit di found that the training
environme expertise. "We also
clicks and erent levels while data of the
ntal This system see that
key presses of spark participants model." (p.
science; for provides ‘translation
while the quality preferred 1016)
computer what we ’ users
participant from the higher
science, we call moved
interacted system, as quality "Presumabl
had 2 PhD “sparks”: back and
with the seen in sparks, y the
students sentences forth
system Study 1." across assumption
from the intended to between
were (p. 1017) participants of
computer spark ideas requesting
collected, average commercial
science in the sparks and
as well as "future spark writing
departmen writer. Our writing
all sparks work could quality did support
t." (p. system more often
generated. bene t from not systems is
1008) generates than
" (p. 1011) comparativ correlated that the
sparks others;
e studies, with writer is
Study 2: using a ‘inspiration’
"After this, either perceived also the
"We mid-sized and
the large-scale usefulness. author of
recruited language ‘perspectiv
participant ones where " (p. 1017) the
13 STEM model e’ users
filed out a participants generated
graduate (GPT-2 tended to
short can still "In the text, thus
students" [43]) and a write for
survey, pick their discussion removing
(p. 1011) custom longer
which own topic we propose any
decoding periods of
included but the size that concern of
method to time
the of the participant plagiarism,
encourage uninterrupt
Creativity study attitudes but we saw
specific and ed." (p.
Support minimizes towards that was
diverse 1013)
Index [13], topic as a writing may not the
outputs."
and factor, or be in assumption
(p. 1003) "Looking at
partook in smaller- uencing in our
how sparks
a semi- scale ones how they study.
were
structured where perceive More work
incorporate
interview participants system is needed
d,
with the are outputs, as to
‘translation
facilitator. assigned well as investigate
’ users
During the topics." (p. discuss this
tended to
interviews, 1017) how important
copy longer
participants designers question."
portions of
were asked "The small might work (p. 1016)
sparks
questions sample size with the
directly
about the of our inherent "Bias in
into their
usefulness study may biases in language
writing
of the have large models and
than
 their 2003). In
, the had written seems to
contents, ! addition,
researchers their be viewed
nding there is
used their papers as more
patterns now an
professiona themselves controversi
and sharing understand
l networks and then al than as a
what is ing that not
to identify used GT to reading aid.
learned’ all
potential change the Overall,
(Mears, unoriginalit
participants language, there
2012, p. y, or
. Obtaining this was seemed to
175). For a intertextual
informed not be the
discussion ity, is in
consent considered tendency
that is in its fact
from the to to not view
infancy, it transgressi
participants constitute the use of
was felt ve
involved a academic MT as
suitable to (Chandraso
written misconduct academic
start with a ma et al.,
introductio It does not . However, It does not misconduct
small-scale 2004), and
n of the apply this view apply . However,
study and students
aims and was by no our findings
take the are often
processes means suggest
opportunit encouraged
of the universal that there
y to have to
research, among the is a clear
professiona incorporate
along with participants and urgent
l generic
the ." (p. 6) need for
discussions academic
assurance university
that might language
of "In policy
stimulate into their
anonymity essence, makers to
further own output
of the the data engage
debate in (Davis &
participants shows that with this
HE and be Morley,
and the there is no issue and
used as a 2018)." (p.
option to agreement offer
basis for 3)
withdraw amongst guidelines
future
from the university that are
larger-scale "This
 papers
"Thus, the text they understand
relevant to
Sensitivity= are ing and
their topic
0.6000, supplied, rememberi
every
Specificity graduate- ng are vital
week. 58
0.4517, level for
health
F1=0.5586 students in successful
informatics
with an Health learning,
graduate
Accuracy=0 Informatics which is
students
.5259. This were why
wrote
poor unable to summarizat
annotated
identificati distinguish ion is
bibliograph
on of AI- the AI- employed
y for 25
generated generated as a
articles
summaries text from learning
"58 health each, to
by expert- student- approach in
informatics which we
It does not level It does not generated education
graduate added the
apply students apply summaries. since it
students" 25 AI-
demonstrat " (p. 573) helps to
(p. 570) generated
es that the improve
article
AI- "We see memory
summaries.
generated that and
We then
model educators understand
asked the
generated will have a ing of
students to
fairly challenge topics by
each grade
human-like in the near utilizing
20
article future to mental
annotated
summaries. assess skills
bibliograph
" (p. 572) learning effectively
y, of which
that [18]. The
10 were
"Interesting depends on summarizin
student-
ly, in the summarizin g process,
written and
348 g or according
10 AI-
 the
paraphrasi preserved pursuing
ng tools highest these academic
substituted number terms misconduct
recognised of errors intact." (p. forces
and occurred 13) them in to
expected in Arabic, the role of
medical Hindi and "When detective.
terms for Punjabi. paraphrasi Collecting
unusual ng tools evidence,
synonyms, In the have been analysing
we latter applied to scenarios,
selected language text the motives
three s there output is and prior
essays were frequently offences
which we more of such and
had poor operating
identified substituti quality as in a quasi-
as ons for to render judicial, if
particularly standardi the text not
It does not It does not It does not
unusual. sed unintelligibl criminologi
apply apply apply
We did not health e. We also cal
know the terms noted the paradigm,
provenance following does not sit
of these (Table features: well within
essays, 2)." (p. 8) the the cultural
although language norms of
there was "The generated academia
structural original will be (Brimble
evidence scenario notable for and
that they the use of Stevenson-
might have was then unidiomatic Clarke
arisen from put words and 2006;
a single through phrases; Burke and
seed six expected Sanney
document paraphra vocabulary 2018;
which was sing tools such as Coren
an essay standard 2011;
selected
submitted medical Keith-
by one as the terminolog Spiegel et
student in top y will al. 1998;
the current entries usually be Sutherland-
cohort." (p. generate substituted Smith
8) d by a with 2005:
Google inappropria Thomas
te and De
search synonyms; Bruin
using the word 2012). (p.
term matching 2)
‘paraphr software,
asing such as
tools’. Turnitin®
(n.d.), may
This not
techniqu recognise
e follows the re-
that used engineered
by text from
Rogerson the source
and thus
and provide a
 this becomes
through more
the review widespread
of current ." (p. 6)
literature
pertaining "While
to APTs and then, there
offer are many
insight into areas of
issues debate
which surroundin
institutions g APT use,
and faculty the fact
might face remains
when con- that they
fronted are a
with this serious and
growing current
threat threat to
It does not It does not It does not It does not It does not
among academic
apply apply apply apply apply
both native integrity,
English which can
speaking hide plagia-
and EFL rism and
students. help to
We have facilitate
also collusion
identified (Wahle et
that the al., 2021).
current APTs can
approach serve to
of reduce the
combating ability of
the illicit text-
use of APTs matching
through software
the used to
developme help
nt of identify
technical potential
solutions is cases of
promising plagiarism,
but may thus
continue to weakening
form an one of the
arms-race most
scenario." effective
(p. 8) current
diagnostic
"We tools for
therefore academic
advocate misconduct
for training and
as the most plagiarism
important (Wahle et
tool in both al., 2021).
reducing These tools
the use of not only
APTs by represent a
students, risk for
 have been Inkpen,
test how concern. Of
reprocesse 2012). This
references greater
d to means
are concern is
‘camwood’. advances in
interpreted that tools
" (p. 6) methods
." (p. 4) contracted
and
to identify
"The algorithms
"The next original
Turnitin® are not
step was to source
results for always
compare materials
both the available to
the outputs cannot
paragraph individuals
from the necessarily
and relying on
original be used at
reference free
journal this time to
list uploads Internet
article identify
identified based
material to where
the original tools.
It does not It does not the outputs It does not writing has
source as Consequen
apply apply of Tool 1 apply been
100% tly there
and Tool 2. repurposed
match to are issues
Exact ." (p. 13)
the online with the
matches to
location of quality of
the original "Further
the journal MT which
text were work is
supporting may
observed, needed to
Turnitin’s® require a
tagged and identify
claim in level of
highlighted linguistic
relation to post-
in grey. markers
identifying editing to
Matches indicating
legitimate correct the
between use of
academic raw output
the two online
resources. so that it is
paraphrasi paraphrasi
What is of fit for
ng outputs ng tools
concern is purpose
that did not such as
Turnitin’s® (Inaba et al.
 essays
ranged
from A- to
F. For one
topic,
Creative
Writing,
the
machine
essay was
failed,
while
human-
written
essays
It does not It does not It does not It does not It does not It does not were
apply apply apply apply apply apply graded
from A- to
D+. The
professors
gave
similar
written
feedback to
the
machine
production
s as to
human
writers." (p.
1122)
 thoughts) of various assessment This is a
from a online designs to survey
source translators obtain a study.
other than and getting more Participants
the the trustworth are
students’ Source'sto diversity,
students y teachers
Source's
and in informatio
conclude performanc and who have
informatio
presenting n on
the einclusion
from the had various
n on
this artificial
output comparabl students elements levels of
artificial AI used for
as their intelligence
e in artificial
against the Research
exposure AI used for
intelligence ethical
used
for ethical intelligence
ownfor translation learning toTeam
online support of
academic
support in for Comments
withoutin quality of objectives." translators. dishonesty Methodolo
cheating teaching
proper writing in teaching, (Added on (cheating, gy
higher these for a (p. 132) I am unsure and
acknowled higher learning, March which10, plagiarism,
education particular learning?
gment. education research, 2023) etc.)? (Y/N)
((Use direct communica "Outside methodolo (Y/N)
quotesOtherand ((Use direct
tive and
the gy would
people quotes
purpose and assessment
academic best apply
their page
label this their page
taking into in higher
context, to this
numbers))
numbers)) education
use as considerati the case. In my
“collusion” on efficient opinion,
in that linguistic use of the
students and these tools methodolo
work cultural can gy that is
together in parameters constitute a more
an such as valuable closely
unauthoriz register, asset for connected
ed way to context, future to this
develop a the employers study
submission regional who are would be
for an variety, not only analytical
online jargon, interested cross-
assessment terminolog in the sectional
Quantitativ
when such y, accurate study
e
input is not phraseolog use of the because it Y Y
descriptive
permitted. y or the target is a study
(MMAT)
Finally, presence language, implement
some any but also in ed at one
people sociocultur contracting point in
regard the al items to individuals time which
use of name a with the provides
online few." (p. capacity to insight into
translators 121) analyze exposures
in online content in and
assessment "It is various outcomes.
s as essential languages It is both
“cheating” that to extract non-
in that language or experiment
students tutors of summarise al, and it
are the various specific uses a
receiving languages information survey for
unauthoriz to and from concerning data
ed which a particular collection
assistance online language purposes.
from online translators reality. This
translators translate is Beatriz
(not from nowadays particularly Moya's
another have a first- needed in comments
person) to hand minority (May 12,
complete experience languages 2023):
an experiment with This study
 this is
(Curtis and
Clare; not an
Lancaster empirical
and study, it
Clarke), a "At the falls well
violation of least, these into the
student auto- text and
codes of generated opinion
conduct examples
similar to might spark category.
submitting some new The
a human- source of
customized generated opinion is
paper ideas—or clear;
written by perhaps this
someone one of
else. them paper is
However, would be also of
when we sufficient. interest Text and
It does not
consider In this to faculty Y Y Opinion
apply
contexts sense, NLP members (JBI)
beyond systems . The
teaching can serve
and as opinions
learning, invention of the
NLP- prompts, or author
written as text are
texts may generators establish
not look whose ed in
like ethical outputs are
viola- tions, either logical
in much revised or ways and
the way used are built
that vast verbatim." from
amounts of (p. 42) extant
replicated literature
and
, and the
unattribute
d text do sources
not are also
constitute logically
plagiarism. defended
" (p. 41) . Chris M.
Anson is
"Auto-
generated a well-
interview establish
questions ed
may be scholar.
acceptable
to some
teachers,
but what
about a
brief
biography
spliced into
the start of
the
interview
 a statistical
t); (c)
ap- proach.
students'
The
perceptions
consequent
of
abundant
paraphrasi
paraphrase
ng and of
information
the
can be
effectivene
used
ss of
pedagogica
PREFER."
lly to
The first
facilitate
part
general
involves
language
pre and
learners’
post-test Randomize
acquisition
It does not It does not data d
and Y N
apply apply analysis of controlled
expansion
students' trials (JBI)
of lexical
performanc
knowledge.
e. The
" (p. 25)
students
were
"we
randomly
developed
assigned to
a
groups. The
paraphrase
second part
reference
is less clear
tool for
to me as it
educational
seems they
purpose.
only
We utilized
analyzed
the pivot-
data from
based
 falls well
grades
from using into the
these text and
tools." (p. opinion
13) category.
As in row
"BYPASSIN 5, the
G TEXT- source of
MATCHING
SOFTWARE opinion is
[...] The clear;
auto- this
paraphrase paper is
tools also of
discussed interest
earlier
successfully to faculty
defeat text- members
matching . The Text and
It does not It does not
tools, as opinions N Y Opinion
apply apply
they of the (JBI)
produce author
original
text that are
cannot be establish
found in a ed in
database logical
(M. Jones & ways and
Sheridan, are built
2015;
Lancaster & from
Clarke, extant
2009; literature
Rogerson & .
McCarthy, Additiona
2017)." (p. lly, the
15)
sources
"If I take a are also
paragraph logically
from a defended
journal . Phillip
article, Dawson
paraphrase is a well-
it, and use
it in a new establish
article ed
without scholar
citing the in the
original academi
article, this c
would
elude text- integrity
matching, field.
but could
be caught
by some
sort of
semantic
content-
 tools and, software text and
undetected without opinion
, is poised incident, category.
to become less than As in
big 25% of the
business as submitted rows 5
increasing work had and 7,
numbers of actually the
writers been source of
claim other produced opinion is
people’s by the clear;
work in student. this
another The
language." majority of paper is
(p. 41) the writing also of
comprised interest
"Students of chunks to faculty
undertakin of text that members
g back had been . The Text and
translation It does not copied
opinions Y Y Opinion
identify a apply directly
of the (JBI)
sentence, from the
paragraph paraphrasi author
or even a ng and are
lengthy translation establish
passage tools." (p. ed in
from an 41)
English text logical
that they "What ways and
want to use became a are built
and pass it more from
through a perplexing extant
translation matter, literature
tool to however,
translate it was the . Also,
into a student’s the
foreign adamant sources
language. stance that are also
They then she had not logically
copy and breached defended
paste the academic . Cara
translated integrity
text back norms. She Dinneen
into the had, in is the
translation accordance Associate
tool for with our Director,
reformulati plagiarism Learning
on into policy, used and
English. in-text
The citations Teaching
resultant throughout at
back- her work Macquari
translated which e
text passes included Universit
undetected the names y
through of the
plagiarism original College.
software." authors
(p. 43) and the
years of
 simply
(p. 781) with what
involve
learners study, it
substituting
"Niño typically falls well
words in
credited can and into the
one
students’ cannot do text and
language
dependenc at each opinion
for those in
y on MT to level of category.
another.
its wide proficiency, As in
Focusing
availability, align
on more
immediacy, assignment rows 5, 7
basic issues
and s with the and 8,
like the
multilingual ACTFL Can-
syllabificati
ism, as well Do- source of
on,
as its Statements opinion is
rhyming,
success in (ACTFL, clear;
and Text and
producing 2017b) that
alliteration this Y Y Opinion
short target the
or on more (JBI)
lexical units appropriate paper is
complex also of
and simply level of
issues such
structured production, interest
as
texts (Niño, remind to faculty
metaphors
2009, p. students members
and
245). In that first
metonymie . The
fact, Niño's language
s (both opinions
study of (L1)
literary and
advanced acquisition of the
cognitive)
Spanish research author
can further
students shows that are
illustrate to
revealed it takes establish
students
that 75% young
both the ed in
reported children
value and logical
turning to many
the
 conception interpret
reflected, “I
s of ation of
think that
authorship the
the false
and results.
information
reading”
could be In my
(2021). In
classified as opinion,
many ways,
cheating a the
they don’t
lot more
fit at all, students'
than the
scrambling voices
actual
the ways are
process
we represen
itself.”" (p.
recognize
5) ted
and define
how texts properly,
"The eight and the
communica
students
te. These conclusio
who
ambiguities ns clearly
concluded
tended to flow from
that they Qualitative
interest It does not
had indeed data. Y Y research
students in apply
“cheated” Some (JBI)
the “not
on the elements
cheating”
assignment
camp. As that were
were
one wrote a bit
committed
of their unclear
to their
hybridized
own voice from my
essay, “it
and perspecti
doesn’t feel
pedagogica ve were
like
l value of
something the
writing." (p.
I’d write philosop
5)
but it also hical
doesn’t not
"Using GPT- perspecti
feel like
2 only ve of the
something
deepened research
I’d write.”
their desire
The er and
for control,
vagueness his
as another
of this positiona
student
statement
echoed: lity. A
is
“Ironically, last
appropriate
I feel like element
, as the
‘cheating’ that
demarcatio
actually
ns of could
sabotaged
agency have
my ability
(what “I’d been
to write
write”) are
this lab improved
no longer
report how is the
clear." (p.
I would fact that
7)
have
liked.”" (p. the study
"Several involved
7)
scholars the
have tried
"But about author's
to
a quarter of students,
articulate
the class and I
what this
had no
frame- wish I
need of
work might
 "While
pt
most
combinatio
participants
ns." (p.
had no
1007).
ownership
concerns,
Data
all
collection
participants
involved
expressed
experts
concerns
annotating
about
outputs on
plagiarism.
a 0 - 4 scale
Several
focused on
participants
coherence.
brought up
On this
that they
scale, 0
were
meant
unsure
Doesn't
exactly
make any Mixed
where the It does not It does not
sense, one N Y methods
sparks apply apply
was Not (MMAT)
were
true, 2 was
coming
Opinion/Do
from, and
n't know,
they
three was
wanted to
Sometimes
make sure
true, and 4
that
was
anything
Generally
they took
true. Two
from the
annotators
sparks was
worked on
adequately
250
changed, to
randomly
alleviate
assigned
any
outputs.
concerns
about
The second
plagiarism.
study
" (p. 1015)
addressed
the
"Providenc
perceptions
e and
of STEM
plagiarism
graduate
as major
students as
writing
Sparks'
concerns.
users: "In
Most
our second
participants
study, we
were
have 13
worried
graduate
about
students
providence
from STEM
and
disciplines
plagiarism,
write
bringing up
tweetorials
these
with our
issues
system and
independe
report on
nt of any
how they
prompting
 ing and and
example, that
argumentat
heavy use internation research
ion, there is
of MT does al students methodol
little reason
not face in ogy and
to insist
necessarily their
that they the
lead to the learning
use "awed interpret
developme and
language
nt of assessment ation of
skills to
language , by to the
present
skills, which some results.
this, if
requires degree In my
there is a
sustained levelling
clear and opinion,
and the
effective the
focused playing !eld
alternative. students'
effort." (p. and
However, Qualitative
9) allowing voices
to go some Y Y research
students to are
way to (JBI)
"While the focus their represen
remove
technology energy on
doubt, ted
could be their
there is a
seen as a academic properly,
case that
levelling endeavours and the
students
mechanism , not on conclusio
should be
during a improving ns clearly
asked to
student’s their flow from
declare the
degree, expertise in data.
level of
which a
technologic
could challenging Some
al
potentially language. elements
interventio that were
go some From
n that they
way to another a bit
have
remove the point of unclear
employed
 Students
[13]." (p. were asked
568) to produce
summaries
"Summariz and grade a
e lecture group of
content: summaries
Another developed
study by AI and
showed an other
exciting students.
application Students
of AI-GTS in were also
education. asked to
They used classify the
natural summaries Quantitativ
It does not language It does not into two e
N N
apply processing apply categories: descriptive
models to AI and (MMAT)
create a human
com- generated.
pelling
research Beatriz
and Moya's
memory comments
refresher (May 12,
tool that 2023):
assisted This study
university qualifies as
students by a
summarizin quantitativ
g the e
instructor’s descriptive
 TranslateT
unintelligibl the authors
M (n.d.)
e." (p. 4) say: "In
(https://tra
order to
nslate.goog
"Whereas investigate
le.com.au).
in the extent
" (p. 2)
patchwritin to which
g synonyms paraphrasi
"As such, it
are ng tools
may not be
manually substituted
a
substituted recognized
deliberate
by the and
or
student, expected
intentional
online medical
breach of
paraphrasi terms for
academic
ng tools unusual
conduct. In
achieve this synonyms,
students
through an we
with EAL,
automatic selected
the
function, three Qualitative
It does not acquisition
and thus essays N Y research
apply of the
the which we (JBI)
linguistic
question had
facility to
arises, as identified
represent
posited by as
the
Rogerson particularly
meaning of
and unusual.
a text
McCarthy We did not
without
(2017), as know the
resorting to
to whether provenance
reproducin
the use of of these
g the
online essays,
author’s
paraphrasi although
actual
ng tools there was
words may
transcends structural
take more
patchwritin evidence
than the
g to that they
few
become might have
months
what arisen from
that our
Walker a single
students
describes seed
have been
as illicit document
studying at
paraphrasi which was
an English-
ng (in an essay
speaking
Pecorari submitted
University."
2003, p.9)." by one
(p. 3)
(p. 5) student in
the current
"In the
"One of the cohort" (p.
essays
most 8). To me,
considered
obvious this is a
in this
issues we novel kind
exploratory
encountere of research,
study, we
d in the but the
encountere
essays was authors
d examples
the use of clearly
of English
synonyms developed
expression
for a deep
which
standard analysis of
indicated
medical the tools'
that the
terminolog outputs
EAL
 2018. A found that (2007).
search for 90% of the However,
this term in sample (N = Keck (2014)
November 55) also found
2021 preferred that novice
obtained to write writers
results of using their have also
approximat assistive been
ely 4.5 paraphrasi shown to
million; ng tool, rely more
highlighting and 75% heavily on Beatriz
not only felt that copying Moya's
the the tool from comments
growing benefited source (March 16,
number of their material, so 2023):
APTs writing. experience The
available, This may also authors
but also the demonstrat play a role suggest this
increased es that for in the is a
Text and
interest in students ability to "detailed
Y Y Opinion
this field who are paraphrase literature
(JBI)
shown by practicing ." (p. 3) review" (p.
both English 2), I did not
scholars writing as "One see a
and the English as a further description
gen- eral Foreign complicatin of the
public Language g factor literature
alike." (p. (EFL), such when review
4) APTs can understand methodolo
be a ing APT use gy.
"As an valuable and its role
example, resource as an
Prentice for academicall
and Kinden learning." y dishonest
(2018) (p. 4) behavior is
found that a lack of
in the "However, clarity as to
discipline if student what
of health training is constitutes
sciences, to be used appropriate
the use of as an initial and
paraphrasi proactive inappropria
ng tools approach te
resulted in to deal- ing paraphrasi
medical with APTs, ng." (p. 4)
terminolog then a clear
y being communica "If an EFL
substituted tion student is
for strategy introduced
incompreh should be to an APT
ensible devised to by a
words that ensure that teacher, for
lacked students example in
meaning. understand a university
This can be the English
one of the difference class
clear between environme
indications the use of nt, it
that an APT such tools follows
has been pedagogica that they
used." (p. lly in the may find it
 by others disseminat the
(Granitz, e academic following
2007; discourse experiment
Scanlon & in their was
Neumann, own voice conducted.
2002; (Bailey & A
Stamatatos Challen, paragraph
, 2011)." (p. 2015)." (p. from an
4) 2) existing
publication
"If an "Where a by this
unintelligibl student is article’s
e string of considered authors
words was to lack the from a
submitted necessary prior
as part of linguistic edition of Qualitative
It does not
an skills, the the N Y research
apply
assessment errors or Internation (JBI)
task it may inaccuracie al Journal
be a reason s may be of
to have a interpreted Educations
conversatio by Integrity
n with a assessors (IJEI) was
student to as a selected to
understand student be the
how they having a original
are going poor source
about their understand material
writing, ing of (McCarthy
and to academic &
determine writing Rogerson,
if convention 2009,
paraphrasi s rather p.49). To
 second As in
these
paragraph,
systems to rows 5,
and so on.
generate 7, 8, and
The AI
articles for 9, the
writing
submission
partner source of
to
helps opinion is
conference
maintain a clear;
s and
flow of
journals." this
words and
(p. 1120) paper is
also takes
the story in also of
"Gaining
unexpected interest
access to
direc- tions, to faculty
GPT-3 is
to which members
straightfor
the student Text and
ward. It does not . The
must Y Y Opinion
Anyone apply opinions
respond. (JBI)
with of the
Generating
internet
a few author
can sign up
alternative are
to the
continuatio establish
OpenAI
ns to a
website3, ed in
story may
gain an logical
help a
account, ways and
student
click the are built
writer see
“Playgroun
creative from
d” tab, type
writing not extant
a prompt
as a linear literature
such as the
progression
title of an . The
, but an
essay, set sources
exploration
 CRITICAL
VE: Is there VE: Is there VE: Is there VE: IsAPPRAISAL
there VE: Is there
congruity congruity congruity congruity congruity
between between between between between
the stated the the the the
philosophic research research research research
al methodolo methodolo methodolo methodolo
Include? Assessmen perspective gy and the gy and the gy and the gy and the
(Y/N) t and the research methods representat interpretati
research question or used to ion and on of
methodolo objectives? collect analysis of results?
gy? (JBI) (JBI) data? (JBI) data? (JBI) (JBI)
(Please (Please (Please (Please (Please
write 1 for write 1 for write 1 for write 1 for write 1 for
YES and 0 YES and 0 YES and 0 YES and 0 YES and 0

Y Medium
Y High
Y Medium
Y High
Y High
Y High
Y High 1 1 1 1 1
Y High
Y High 1 1 1 1 1
Y High
Y Medium 1 1 1 1 1
Y High
Y High 1 1 1 1 1
Y High
AL according VE: Do the
VE: Is there VE: Is the ED TRIALS:
QUALITATI to current conclusions
a influence Was true
VE: Are criteria or, drawn in
statement of the randomizat
participant for recent the
locating researcher ion used
s, and their studies, research
the on the for
voices, and is report flow QUALITATI
researcher research, assignment
adequately there from the VE: Critical
culturally and vice- of
represente evidence of analysis, or appraisal
or versa, participant
d? (JBI) ethical interpretati points
theoreticall addressed? s to
(Please approval on, of the
y? (JBI) (JBI) treatment
write 1 for by an data? (JBI)
(Please (Please groups?
YES and 0 appropriat (Please
write 1 for write 1 for (JBI)
for NO) e body? write 1 for
YES and 0 YES and 0 (Please
(JBI) YES and 0

0
0
0 1
0
0
0
0 0 1 1 1 8
0
0 1 1 0 1 8
0
0 0 0 1 1 7
0
0 1 0 1 1 8
0
 RANDOMIZ up
ED ED ED ED ED TRIALS:
ED complete
CONTROLL CONTROLL CONTROLL CONTROLL Were
CONTROLL and if not,
ED TRIALS: ED TRIALS: ED TRIALS: ED TRIALS: treatment
ED TRIALS: were
Was Were Were those Were groups
Were differences
allocation treatment delivering outcomes treated
participant between
to groups treatment assessors identically
s blind to groups in
treatment similar at blind to blind to other than
treatment terms of
groups the treatment treatment the
assignment their
concealed? baseline? assignment assignment interventio
? (JBI) follow up
(JBI) (JBI) ? (JBI) ? (JBI) n of
(Please adequately
(Please (Please (Please (Please interest?
write 1 for described
write 1 for write 1 for write 1 for write 1 for (JBI)
YES and 0 and
YES and 0 YES and 0 YES and 0 YES and 0 (Please
for NO) analyzed?
1 0 1 0 1 1 0
 ED RANDOMIZ e, and any
CONTROLL RANDOMIZ
CONTROLL ED deviations TEXT AND
ED TRIALS: ED
ED TRIALS: CONTROLL from the OPINION:
Were CONTROLL
Were ED TRIALS: standard Is the
participant ED TRIALS: RANDOMIZ
outcomes Were RCT design source of
s analyzed Was ED
measured outcomes (individual the opinion
in the appropriat CONTROLL
in the same measured randomizat clearly
groups to e statistical ED TRIALS
way for in a ion, identified?
which they analysis Critical
treatment reliable parallel (JBI)
were used? (JBI) appraisal
groups? way? (JBI) groups) (Please
randomize (Please points
(JBI) (Please accounted write 1 for
d? (JBI) write 1 for
(Please write 1 for for in the YES and 0
(Please YES and 0
write 1 for YES and 0 conduct for NO)
write 1 for for NO)
YES and 0 for NO) and

0
0 1
1 1 1 1 0 9
0 1
0 1
0 1
0
0
0
0
0
0 1
0
0 1
 TEXT AND OPINION: TEXT AND
stated
OPINION: Are the TEXT AND OPINION:
position QUANTITA
Does the interests of OPINION: Is any
the result TIVE
source of the Is there incongruen
of an DESCRIPTI
opinion relevant reference ce with the
analytical TEXT AND VE: Is the
have population to the literature/s
process, OPINION: sampling
standing in the central extant ources
and is Critical strategy
the field of focus of literature? logically
there logic appraisal relevant to
expertise? the (JBI) defended?
in the score address
(JBI) opinion? (Please (JBI)
opinion the
(Please (JBI) write 1 for (Please
expressed? research
write 1 for (Please YES and 0 write 1 for
(JBI) question?
YES and 0 write 1 for for NO) YES and 0
(Please
for NO) YES and 0 for NO)

0 0
1 1 1 1 1 6
0
1 1 1 1 1 6
1 1 1 1 1 6
1 1 1 1 1 6
0
0
0
0 1
0
1 1 1 1 1 6
0
1 1 1 1 1 6
 MIXED MIXED
QUANTITA METHODS: METHODS:
QUANTITA
TIVE Is there an Are the
TIVE QUANTITA QUANTITA
DESCRIPTI adequate different
DESCRIPTI TIVE TIVE QUANTITA
VE: Is the rationale component
VE: Is the DESCRIPTI DESCRIPTI TIVE
statistical for using a s of the
sample VE: Are the VE: Is the DESCRIPTI
analysis mixed study
representat measurem risk of VE: Critical
appropriat methods effectively
ive of the ents nonrespon appraisal
e to design to integrated
target appropriat se bias score
answer the address to answer
population e? low?
research the the
?
question? research research
question? question?

0 1 1 1 3
0
0
0
0
0
0
0 1 1
0
0 1 1 1 4
0
0
0
0
METHODS: MIXED
METHODS:
Are the METHODS:
Do the
outputs of Are
different
the divergence
component
integration s and
s of the MIXED
of inconsisten
study METHODS:
qualitative cies
adhere to Critical
and between
the quality appraisal
quantitativ quantitativ
criteria of score
e e and
each
component qualitative
tradition of
s results
the
adequately adequately
methods
interpreted addressed?

0
0
0
0
0
0
0
1 0 1 4
0
0
0
0
0
0
Include? (Y/N) Y

Count of Assessment Assessment

Methodology High Medium Total Result
Mixed methods (MMAT) 1 1
Qualitative research (JBI) 3 1 4
Quantitative descriptive (MMAT) 1 1 2
Randomized controlled trials (JBI) 1 1
Text and Opinion (JBI) 6 6
Total Result 11 3 14
Beatriz
Helen
Sarah
Unknown / Not Stated
Not Relevant (i.e. Opinion Article)
Afghanistan
Albania
Algeria
Andorra
Angola
Antigua and Barbuda
Argentina
Armenia
Australia
Austria
Azerbaijan
The Bahamas
Bahrain
Bangladesh
Barbados
Belarus
Belgium
Belize
Benin
Bhutan
Bolivia
Bosnia and Herzegovina
Botswana
Brazil
Brunei
Bulgaria
Burkina Faso
Burundi
Cambodia
Cameroon
Canada
Cape Verde
Central African Republic
Chad
Chile
China
Colombia
Comoros
Congo, Republic of the
Congo, Democratic Republic of the
Costa Rica
Cote d'Ivoire
Croatia
Cuba
Cyprus
Czech Republic
Denmark
Djibouti
Dominica
Dominican Republic
East Timor (Timor-Leste)
Ecuador
Egypt
El Salvador
Equatorial Guinea
Eritrea
Estonia
Ethiopia
Fiji
Finland
France
Gabon
The Gambia
Georgia
Germany
Ghana
Greece
Grenada
Guatemala
Guinea
Guinea-Bissau
Guyana
Haiti
Honduras
Hungary
Iceland
India
Indonesia
Iran
Iraq
Ireland
Israel
Italy
Jamaica
Japan
Jordan
Kazakhstan
Kenya
Kiribati
Korea, North
Korea, South
Kosovo
Kuwait
Kyrgyzstan
Laos
Latvia
Lebanon
Lesotho
Liberia
Libya
Liechtenstein
Lithuania
Luxembourg
Macedonia
Madagascar
Malawi
Malaysia
Maldives
Mali
Malta
Marshall Islands
Mauritania
Mauritius
Mexico
Micronesia, Federated States of
Moldova
Monaco
Mongolia
Montenegro
Morocco
Mozambique
Myanmar (Burma)
Namibia
Nauru
Nepal
Netherlands
New Zealand
Nicaragua
Niger
Nigeria
Norway
Oman
Pakistan
Palau
Panama
Papua New Guinea
Paraguay
Peru
Philippines
Poland
Portugal
Qatar
Romania
Russia
Rwanda
Saint Kitts and Nevis
Saint Lucia
Saint Vincent and the Grenadines
Samoa
San Marino
Sao Tome and Principe
Saudi Arabia
Senegal
Serbia
Seychelles
Sierra Leone
Singapore
Slovakia
Slovenia
Solomon Islands
Somalia
South Africa
South Sudan
Spain
Sri Lanka
Sudan
Suriname
Swaziland
Sweden
Switzerland
Syria
Taiwan
Tajikistan
Tanzania
Thailand
Togo
Tonga
Trinidad and Tobago
Tunisia
Turkey
Turkmenistan
Tuvalu
Uganda
Ukraine
United Arab Emirates
United Kingdom
United States of America
Uruguay
Uzbekistan
Vanuatu
Vatican City (Holy See)
Venezuela
Vietnam
Yemen
Zambia
Zimbabwe
1=blog
2=book
3=book section
4=conference paper
5=conference proceedings
6=journal article
7=magazine article
8=newspaper article
9=thesis
10=webpage
1=Faculty
2=students
3=teaching assistants
4=academic support staff
5=educational developers
6=multiple
Qualitative
Quantitative Randomized
Quantitative Non-Randomized
Quantitative Descriptive
Mixed Methods
Program Description
Opinion
Theoretical
Other
Unable to determine
Type Category

JBI Analytical cross-sectional studies (JBI

JBI Case control studies (JBI)

JBI Case reports (JBI)

JBI Case series (JBI)

JBI Cohort studies (JBI)

JBI Qualitative research (JBI)

JBI Quasi-experimental studies (JBI)

JBI Randomized controlled trials (JBI)

JBI Systematic Reviews (JBI)

MMAT Quantitative descriptive (MMAT)

MMAT Mixed methods (MMAT)

JBI Text and Opinion (JBI)

Unable to Determine
Definition
Source 1: An analytical cross-sectional study is a type of quantitative, non-experimental res
purpose is to measure the association between an exposure and a disease, con
participants (Schmidt & Brown, 2019, pp. 206-207).

Source 2: "In analytical cross-sectional studies, investigators collect data for both ex
unexposed subjects. The exposures and outcomes are measured simultaneously; therefore

Source 1: "A study that compares patients who have a disease or outcome of i
the exposure to a risk factor is present in each group to determine the relationship be
Case control studies are observational because no intervention is attempted and no attempt i
the two groups of individuals: cases and controls. These studies are designed t
Case control studies are also known as "retrospective studies" and "case-referent studies." (The Him

Source 1: "A case report is a detailed report of the symptoms, signs, diagnosis,
of the cornerstones of medical progress and provide many new ideas in medicine. Some reports co
to conduct large scale research." (Heart Views, 2017, p. 1)

Source 2: "An article that describes and interprets an individual case, often w
- Unique cases that cannot be explained by known diseases or syndromes
- Cases that show an important variation of a disease or condition
- Cases that show unexpected events that may yield new or useful information
- Cases in which one patient has two or more unexpected diseases or disorders
Case reports are considered the lowest level of evidence, but they are also the first line of evidence
importance of the observation being reported.
If multiple case reports show something similar, the next step might be a case-control study to dete

Source 1: Observations are made on a series of individuals, usually all receiving the s
Source 1: A cohort study is a type of quantitative research. They are often described as "n
their exposure status. Researchers then investigate whether something individuals are ex

Source 2: "Cohort studies are a type of longitudinal study—an approach that follows rese
characteristic, such as a particular occupation or demo- graphic similarity. During the period o
of time, it is then possible to explore the impact of this vari- able (eg, ide
understanding of what factors increase or decrease the likelihood of developing disease." (Barret &

"Usually, cohort studies should adopt a purely observational approach. However, some resear
practitioner (NP)-led critical care in a large university hospital in the USA. They collected data on all
(primarily 90-day mortality) were monitored. By comparing the groups, the researchers established

Source 1: Qualitative research, in contrast to quantitative research, analyzes words and te

and phenomena among study participants. Qualitative data is generated from pa
2019, pp. 221-224).

Source 2: “Qualitative research is an approach for exploring and understanding the m

Source 1: Quasi-experimental studies are a type of quantitative research. Similar to exp

commonly, they do not randomly assign participants to study groups. Also,

Source 2: "Quasi-experimental studies encompass a broad range of non- randomized inter

"Quasi-experiments are studies that aim to evaluate interventions but that do no

outcome. Quasi-experimental studies can use both preintervention and postinte

Source 1: A randomized controlled trial (RCT) is a type of quantitative, experimental

RCTs investigate the effects of an intervention or treatment on study participants. Participants are
untreated subjects" (Schmidt & Brown, 2019, p. 172).

Source 2: A clinical study in which individual participants are allocated to interve

Centre for Evidence-based Medicine (2016); Porta et al. (2014)

Source 1: Systematic reviews are "rigorous and systematic syntheses of research

Essentially, a systematic review is a study of other studies on the same topic.

Source 1: Quantitative descriptive studies are “concerned with and designed only to
2014, p. 72). They are used to monitoring the population, planning, and generating hypothesis (Gri
Source 1: Mixed methods (MM) research involves combining qualitative (QUAL) and quantitative (Q
method are combined; (b) each method is used rigorously in accordance to the generally accepted
emerging) and the integration of the QUAL and QUAN phases, results, and data.
Definition source(s)

Source 1: https://libguides.utoledo.edu/nursingappraisal/crosssectionalSource 2: https://reader.els

Source 1: https://himmelfarb.gwu.edu/tutorials/studydesign101/casecontrols.cfm

Source 1: https://libguides.utoledo.edu/nursingappraisal/cohort

Source 2: https://himmelfarb.gwu.edu/tutorials/studydesign101/casereports.cfm

Source 1: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5513097/
Source 1: https://libguides.utoledo.edu/nursingappraisal/cohort Source 2: https://click.endnote.co

Source 1: https://libguides.utoledo.edu/nursingappraisal/qualitativeSource 2: http://mixedmethod

Source 1: https://libguides.utoledo.edu/nursingappraisal/quasiSource 2: ncbi.nlm.nih.gov/pmc/arti

Source 1: https://libguides.utoledo.edu/nursingappraisal/rctSource 2: http://mixedmethodsapprais

Source 1: https://libguides.utoledo.edu/nursingappraisal/sysrev
Source 1: http://mixedmethodsappraisaltoolpublic.pbworks.com/w/file/fetch/127916259/MMAT_

Source 1: http://mixedmethodsappraisaltoolpublic.pbworks.com/w/file/fetch/127916259/MMAT_
Discipline
Social Sciences
Humanities
Computer Science/Engineering
Natural Sciences
Fine Arts
Other
1st round
2nd round
3rd round
4th round
Data extraction without Critical appraisal
Include other categories in critical appraisal that could apply to your list
Review documents and make the critical appraisal
Include the articles that were in the list but could not be found
START FINISH
Table 2
Article Categorization Scores

Methodology High Medium Low

Mixed methods
5 to 6 points 3 to 4 points 0 to 3 points
(MMAT)

Qualitative
8 to 10 points 5 to 7 points 0 to 4 points
research (JBI)

Quantitative
descriptive 5 to 6 points 3 to 4 points 0 to 3 points
(MMAT)

Randomized
controlled trials 11 to 13 points 7 to 10 points 0 to 7 points
(JBI)

Text and Opinion

5 to 6 points 3 to 4 points 0 to 3 points
(JBI)
Table 4
Critical appraisal

Methodology (Tool) High Medium Total

Mixed methods (MMAT) 1 1
Qualitative research (JBI) 3 1 4
Quantitative descriptive (MMAT) 1 1 2
Randomized controlled trials (JBI) 1 1
Text and Opinion (JBI) 6 6
Grand Total 11 3 14
Year Count of Year
2015 1
2017 1
2018 2
2020 1
2021 1
2022 8
(empty)
Total Result 14
Table 5
Study Characteristics: Country of Study

Country Number Percentage

Australia 4 29%
Singapore 1 7%
Taiwan 1 7%
United Kingdom 3 21%
United States of America 5 36%
Type of source ((PleaseCount of Type of source ((Please use list below))
2=book 1
4=conference paper 1
6=journal article 12
(empty)
Total Result 14
Include? (Y/N) Y

Participants ((Pl Count of Participants ((Please use list below))3

1=Faculty 5
2=students 5
6=multiple 4
Total Result 14

1=Faculty 5 36
2=students 5 36
6=multiple 4 29
Include? (Y/N) Y

Discipline Count of Discipline

Computer Science/Engineering 3
Humanities 7
Other 4
Total Result 14
Study Design ((Please use lis Count of Study Design ((Please use list below))
Mixed Methods 1
Qualitative 4
Quantitative Descriptive 2
Quantitative Randomized 1
Theoretical 6
(empty)
Total Result 14

Theoretical 6 43
Qualitative 4 29
Quantitative Descriptive 2 14
Mixed Methods 1 7
Quantitative Randomized 1 7
AI Tool Count of AI Tool
AI in general 1
Online translator 3
Online translator and paraphrasing tool 2
Paraphrasing tool 3
Text generator 5
(empty)
Total Result 14

Text generator 5 36
Online translator 3 21
Paraphrasing tool 3 21
Online translator and paraphrasing tool 2 14
AI in general 1 7