doi:10.5128/ERYa12.

17

EESTI RAKENDUSLINGVISTIKA ÜHINGU AASTARAAMAT 12, 287–307

mEaSuring affEctivE languagE in
known pEEr fEEdback on l2 acadEmic
writing courSES: a novEl approach
Roger Michael Alan Yallop

Abstract. Publishing scientific articles in English is often a prerequisite

for academic success. Thus, developing effective pedagogies to support
Estonian university students develop writing skills in L2 (English) is
becoming increasingly more important. One such method is by form-
ing small writing groups where each member periodically gives written
feedback on their colleague’s writing. Here, the affective language used
in the written communication between the reviewer and writer may
strongly influence their relationship. This in turn may have a significant
impact on the writing process. This study describes the development
of a novel taxonomy to measure the cumulative effect of affective fac-
tors by accounting for the uniqueness of each individual, and how they
project their distinct personalities or ‘social presence’ to build rapport
within the group. The hypothesis is that individuals exhibiting a high
social presence are more likely to produce higher-quality feedback and
more improved subsequent texts than those with a lower social pres-
ence. The paper concludes by illustrating how this taxonomy can be
used to both test this hypothesis and gain further insight into the peer
feedback process in future studies.*
Keywords: social presence, community of inquiry, pragmatics, writ-
ing groups, peer review

1. Introduction
It is becoming common practice within L2 Academic writing courses to include peer
feedback as one, or even the main, component of the teaching process (Cho et al.
2006, Kollar, Fischer 2010, Diab 2011). This is because the process of peer review
brings cognitive, affective, social, and linguistic benefits to both the writer (Min
2006: 118–119) and the reviewer (Lundstrom, Baker 2009). It can also offer a more

287
* This research was supported by the Estonian Research Council grant PUT701.
 practical teaching solution in universities with a high demand for L2 writing instruc-
tion, but with limited resources (Su 2011). This is particularly true in this study
where the main challenge is to provide a sound pedagogical basis for conducting
writing for scientific publication courses to large cohorts of postgraduate students.
Feedback can be given anonymously or by a reviewer already acquainted with the
writer (hereinafter referred to as ‘known feedback’). Surprisingly, many researchers
(Ferris 1997, Ertmer et al. 2007, dLu, Bol 2008) advocate the use of anonymous
peer feedback over known feedback arguing that reviewers can be more critical in
their comments without fear of peer reprisal. This could be an advantage in ‘one-off’
feedback instances, but it fails to harness the full potential of collaborative learning.
This is because known feedback is more in line with constructivist learning theories
(Vygotsky 1980) and particularly so on longer courses that allow both synchronous
and asynchronous feedback stages. It allows individuals more opportunities to
‘negotiate for meaning’ (Ellis 2003) as well as allowing the groups to bond over
time, developing a greater ‘trust’ and ‘sense of community’ (Garrison et al. 2010a).
Thus, it seems likely that, in known feedback, the use of affective factors will have
a much greater impact on the uptake of peer feedback.
Randy D. Garrison, Terry Anderson and Walter Archer (1999) developed their
Community of Inquiry (CoI) Model that is commonly used on both online and
blended courses (Motteram 2009, Zhao et al. 2014) and is widely accepted amongst
scholars (see Garrison et al. 2010b). Here, they hypothesise that students with a
high social presence who project themselves both emotionally and socially within
their community of inquiry are more likely to produce higher quality written texts
and feedback than those who do not. Their model is designed for learners con-
structing knowledge collaboratively online through the medium of asynchronous
text on higher education courses. This paper describes how the author adapts their
CoI model to measure the ‘social presence’ exhibited within one small Academic
writing group as shown through their asynchronous texts.

2. Theoretical background
2.1. Affective factors within the peer feedback process

There is controversy in the current literature about how, if at all, written affective
comments between the reviewer and the writer may improve or hamper the peer
feedback process. Some researchers argue that praise increases the trust between the
writer and the reviewer and this enhances motivation (Topping 1998: 256, Cho et al.
2006, Nelson, Schunn 2009). In other words, these comments encourage the writer
to engage with the reviewers’ comments and the writing process for a longer period
of time. This results in them making a larger number of revisions (Gee 1972, Cho et
al. 2006) and, thus, being more likely to produce a more improved text. Conversely,
other studies claim that feedback comments containing affective language should
be completely discouraged, because they have a negligible influence on the writing
process (Ferris 1997). Fiona and Ken Hyland (2001: 207) argue that mitigated or
‘hedged criticism’ could even be detrimental to the process because it may confuse
the writer’s understanding of the feedback. In addition, socio-cultural factors could

288
also strongly influence the process in that ‘Estonians seem to focus more on content
than relationships in communication’ (Keevallik, Grzega 2008: 214).
What is clear, though, is that students use much affective and mitigated language
in their feedback comments (Hyland, Hyland 2001, Liu, Sadler 2003). Further-
more, affective language, or the use of motivational comments, is highly valued
and expected amongst peers in their written correspondence (Tuzi 2004, Cho et
al. 2006, Kaufman, Schunn 2011: 390). This suggests that affective language does
indeed serve a useful linguistic purpose by building rapport between the writer
and the reviewer. Here, it is speculated that the greater the rapport between the
dyad, the more receptive they may be to the uptake of each other’s comments in
their subsequent drafts.
Previous studies have measured the affective factors, ‘praise’, ‘mitigation’ and
‘criticism’ as separate and non-dependent entities within peer feedback comments,
and their effect on the writer’s revised text (Hyland, Hyland 2001, Cho et al. 2006,
Nelson, Schunn 2009). This treatment, though, may be too simplistic. Madeline E.
Ehrman, Betty L. Leaver and Rebecca L. Oxford (2003: 322) identify over a dozen
affective factors including anxiety, self-efficacy and learning attitudes that are all
inter-related, and ultimately affect the learner’s motivation. This is concurred with
many influential researchers (Gardner 1985, Williams, Burden 1996, Ellis 2003)
who stress the importance of motivation and how it greatly influences the amount
of learning both within and outside the classroom. In this context, positive affec-
tive comments (i.e., praise and mitigation) may increase the writer’s self-efficacy
leading to a higher level of motivation, whereas negative affective comments (i.e.,
criticism) may raise the writer’s anxiety leading to decreased levels of motivation.
This suggests that these ‘emotional comments’ trigger a much more intricate and
interwoven relationship that cannot be measured as independent and unrelated
variables alone, and may influence the peer feedback process considerably more
than is currently reported in the literature.
Thus, studies in peer feedback have employed both quantitative and qualitative
research methods to measure the reviewer and writer’s anxiety and self-efficacy
(Nicol, Macfarlane-Dick 2006, Kurt, Atay 2007, Choi 2013), and their attitudes to
the peer feedback process (van Zundert et al. 2010: 277, Kaufman, Schunn 2011,
Chang 2012). Although these studies stress the importance of how affective factors
influence the peer feedback process, few have attempted to measure their overall
effect on the process (Hyland, Hyland 2001: 187, Nicol, Macfarlane-Dick 2006: 212).
This study heeds their call and explains how a novel methodology is developed to
measure the ‘combined’ effect of affective language used by writing groups in known
peer feedback as expressed in their asynchronous written texts. This methodology
can then be used in tandem with other research methods in subsequent studies to
further assess the impact of affective factors on the peer feedback process.

2.2. Community of inquiry model

The CoI model is based on collaborative constructivism and theoretically grounded

in the research on deep and meaningful approaches to learning (Garrison et al.
2010a: 32). The model (see Figure 1) draws on Matthew Lipman’s (2003) com-
munity of inquiry where students collaborate to build, challenge and support one
289
 another in their ideas within their community. The model assumes that learning
for the individual occurs through the dynamic interaction of three crucial elements:
cognitive presence, social presence (SP) and teaching presence. This means that
the three ‘presences’ are all interrelated in that the individual’s SP influences and is
influenced by both their cognitive presence and the teaching presence. It is supposed
that a higher SP would increase the writer or reviewer’s cognitive presence that in
turn would lead to either a more improved subsequent text or higher quality feedback
comments. Thus, the crux of the model is in creating a better learning environment
in a community of inquiry by supporting the three elements to encourage deep
learning, where deep learning is often associated with a ‘high cognitive’ presence.
Cognitive presence refers to learners constructing and confirming meaning for
reflection and discourse in a community of inquiry. Garrison et al. (2010b) draws
on John Dewey’s (1933) critical thinking model and sets out their practical inquiry
model for critical thinking. This is where the learner engages through four stages of
thinking within his or her private and/or shared world as follows: trigger, explora-
tion, integration and resolution. However, the learner does not necessarily progress
through the model in a linear order and often fails to get beyond the second or third
phases (Garrison et al. 2010b: 6).
Teaching presence relates to 1) Design and organization: ideally, the student
should have some influence on the content and approach if this is indeed a collab-
orative, constructivist framework. 2) Facilitating discourse where the onus is on the
teacher to monitor the asynchronous texts (feedback comments and cover letters)
and aid the students in their construction of personal meaning. 3) Direct instruction.

Supporting
SOCIAL Discourse COGNITIVE
PRESENCE PRESENCE
EDUCATIONAL
EXPERIENCE
Setting Selecting
Climate Content

TEACHING PRESENCE
(Structure/Process)

Communication Medium

Figure 1. Community of Inquiry Model (Garrison, Arbaugh 2007: 158)

Influential CoI researchers (Richardson, Swan 2003, Rourke et al. 2007) established
that SP serves as the ‘foundation’ for building communities of inquiry, and then they
turned their attention towards investigating how SP affects higher-order learning.
Randy D. Garrison and Terry Anderson (2003: 48) state that ‘it (social presence) is
an important antecedent to collaboration and critical discourse… It is inconceivable
to think that one could create a community without some degree of social presence.’
In other words, SP is a prerequisite for higher-order thinking to occur. This hypoth-
esis is supported through extensive research that strongly indicates that a ‘high’ SP
290
facilitates deep learning or promotes a ‘higher’ cognitive presence (Lomicka, Lord
2007, Shea, Bidjerano 2009, Shea et al. 2010). Furthermore, there are strong causal
links between the teaching, cognitive and social presences where ‘social presence
is a mediating variable between teaching presence and cognitive presence. That
is, it is a responsibility of teaching presence and a condition for creating ‘cognitive
presence’ (Garrison et al. 2010a: 32). This suggests that if the ‘teaching presence’ is
low, then the ‘SP is likely to be similarly ‘low’ which in turn results in poor learning
or a ‘low’ cognitive presence and vice-versa for a high teaching presence. In this
study, I assume a reasonably high and constant teaching presence throughout the
course negating this as a measurable variable.

2.3. Social presence

Liam Rourke, Terry Anderson, Randy D. Garrison and Walter Archer (2007: 53–54)
give a comprehensive account of the rationale behind the development of their
original taxonomy to measure social presence within the CoI framework. Here, they
categorise SP into three broad categories, 1) Affective responses that express emo-
tions, feelings and mood, 2) Interactive responses that express a willingness to build,
and sustain relationships and tacitly indicate interpersonal support, encouragement,
and acceptance of the initiator, and 3) Cohesive responses that build and sustain a
sense of group commitment. This taxonomy has since been further developed (see
Table 1) by prominent researchers in the same field.

Table 1. A comparison of social presence categories by researcher

Researcher Social presence (SP) categories

Rourke et al. (2007) Affective responses Interactive responses Cohesive responses
Garrison and Project their Communicate Identify with
Arbaugh (2007) personalities purposefully the community
Shea et al. (2010) Affective response Open communication Group cohesion
Expresses emotions, Builds and sustains Builds and sustains group
Meaning
feelings and mood relationships commitment

From this basis, Peter Shea, Suzanne Hayes, Jason Vickers, Mary Gozza-Cohen,
Sedef Uzuner, Ruchi Mehta, Anna Valchova and Prahalad Rangan (2010) used a
coding scheme (see Table 2) to measure an individual’s SP as expressed in their
asynchronous correspondence to their group members.
In the coding process, they segment the data into SP categories using a syntactic
unit of measurement of one sentence as advocated by Patrick J. Fahy (2001), and
a thematic unit (Henri 1992) when it seems more logical. Each segmented unit
is coded on a macro-level for overall meaning, and then micro-coded within this
segmentation for further meaning. Then, the number of the indicators encountered
within each message is recorded. A high frequency of indicators would denote a
warm and friendly environment (high SP) whereas a low frequency of indicators
would suggest a cold and impersonal environment (low SP). In their data analysis,
they compared the frequency of such SP indicators used collectively by the group
at each learning stage of the module.
291
 Cohen, Sedef Uzuner, Ruchi Mehta, Anna Valchova and Prahalad Rangan (2010)
used a coding scheme (see Table 2) to measure an individual’s SP as expressed in
their asynchronous correspondence to their group members.

Table2.2.Coding
Table Coding scheme
scheme forfor Social
Social Presence
Presence (Shea(Shea et al. 19–20)
et al. 2010: 2010: 19–20)
Social Presence Category = Affective (AF)
Indicators Code Definition Examples
Expressing emotions SP-AFI Conventional expressions of emotion I'm really annoyed... I'm so happy
Use of humour SP-AF2 Teasing, cajoling, irony, understatements, sarcasm The banana crop in Edmonton is
looking good this year ;-)
Self-disclosure and SP-AF3 Presents details of life outside of class, Where l work, this is what we do...
personal intent or expresses vulnerability; includes expressions of I just don't understand this question
likes, dislikes and preferences
Use of unconventional SP-AF4 Unconventional expressions of emotion; includes I just can’t stand it when…!!!;
expressions to express repetitious punctuation, conspicuous ANYBODY OUT THERE !; What does this
emotion capitalization, emoticons mean!?!?; Good idea :- )
Expressing value SP-AF5 Expressing personal values, beliefs and attitudes I think it is a necessary evil; I feel our children
have the same rights.

Social Presence Category = Open Communication (OC)

Indicators Code Definition Examples
Continuing a thread SP-OC1 Using reply feature of software, rather than Software dependent, e.g. Subject: Re- or Branch
starting a new thread from
Quoting from others' SP-OC2 Using software features to quote others' entire Software dependent_ e.g. “Martha writes;” or text
messages message or cut and passing selection of others' prefaced by less than symbol <
messages
Referring explicitly to SP-OC3 Direct references to contents of others' posts In your message you talked about Moore’s
others' messages. distinction between...
Asking questions SP-OC4 Students ask questions of other students or the Anyone else had experience with BlackBoard ?
moderator
Complimenting. SP-OC5 Complimenting others or contents of others' I really like your interpretation of the reading.
expressing messages
Expressing agreement SP-OC6 Expressing agreement with others or contents of I was thinking the same thing.
others messages You really hit the nail on the head.
Expressing SP-OC7 Expresses disagreement with other or contents of I don't think... I think it is different...
disagreement others messages
Personal advice SP-OC8 Offering specific advice to classmates The CEC web site might have some references

Social Presence Category = Cohesion (CH)

Indicators Code Definition Examples
Vocatives SP-CHI Addressing or referring to the participants by I think John made a good point.
name John, what do you think?
Addresses or refers to SP-CH2 Addresses the group as we, us, our, group Our textbook refers to...; I think we veered off
the group using track...
inclusive pronouns
Phatics, salutations and SP-CH3 Communication that serves a purely Hi all; Hi John; That's it for now; We're having the
greetings social function ; greetings or closures most beautiful weather here.
Social sharing SP-CH4 Sharing information unrelated to the course Happy Birthday!! To both of you!!
Course reflection SP-CH5 Reflection on the course itself A good example was the CD-ROM we read about

3.
5 The study

This study follows one discipline-specific writing group containing four L2 post-
graduate students in humanities (see Table 3 for the participants’ demographic data
along with their fictitious names for the purpose of this study). They are participating
in a three-month writing for scientific publication course at an Estonian university,
and they are all under 30 years old.

Table 3. Demographic data on the four participants

Name Gender Nationality Mother Tongue Discipline

Ann Female Estonian Estonian Literature
Bob Male Latvian Russian Russian Literature
Carol Female Russian Russian Semiotics (Folklore)
Dawn Female Latvian Russian Anthropology (Cultural)
292
Prior to the writing process, they are given peer feedback training and this is fol-
lowed by face-to-face (f2f) and online genre-based instruction (Swales 1995) peri-
odically throughout the course. The peer feedback process involves the following
cyclic process on each draft, 1) the students submit their original drafts and their
cover letters electronically in MS Word, 2) all group members give asynchronous
feedback on their colleagues’ drafts and submit these online using track changes
and/or comment boxes in MS Word, 3) all members attend a writing group meeting
to allow the reviewers and writers to discuss (synchronous feedback given f2f), 4)
the writers decide whether to implement or not implement their peer’s feedback
comments, and 5) the writers submit their revised draft electronically in MS Word.
The writers typically follow the IMRaD (Introduction, Methods, Research and
Discussion) structure common in their discipline, where the writer’s original and
subsequent draft, their cover letter, and their reviewers’ feedback are collected for
each stage. To allow for the course to settle, there are seven stages in total; three
stages to write one introduction section, followed by one stage each for the methods,
the results, the discussion and the abstract sections. Aside from the f2f lectures,
there is no teacher intervention throughout the course. Thus, the course adopts
a blended learning approach (asynchronous text and f2f meeting) in a strongly
constructivist paradigm. Even though f2f meetings may affect the group’s sense
of community, only asynchronous text discussions are considered in this analysis.

4. Methodology
4.1. Devising novel taxonomy

The taxonomy is devised through the analysis of two data sets, 1) non-revision-
oriented comments (NRCs) contained within the reviewer’s feedback letters and,
2) the writer’s cover letters. The reviewer’s feedback comments at each stage
(IMRaD structure) of the writing process are segmented according to their nature
into revision-oriented comments (RC) and NRCs according to Jun Liu and Ran-
dall W. Sadler’s (2004) taxonomy. RCs are those that request the writer to make a
direct change to their text (i.e., ‘the title is too long’), whereas NRCs more reflect
the writer’s personality and contain ‘affective comments’ (i.e., ‘this is an excellent
piece of writing’). As affective comments are the subject of this research, only NRCs
are analysed. Cover letters are the means by which the writer communicates to the
group how their writing should be assessed. They contain much affective language
and are also analysed for the purpose of formulating this taxonomy. Next, the
feedback letters (i.e., Ann’s review of Bob) and the cover letters are segmented into
thematic units (Henri 2002) and categorised at the macro-level only.
Firstly, deductive reasoning is used on the segmented NRCs and cover letters to
translate Shea et al.’s (2010) taxonomy from their context to a revised taxonomy to
measure social presence (SP) within the peer feedback process in asynchronous text
environments. Table 4 shows the analysis of data collected from Ann, in the form
of a cover letter and her feedback comments to Bob, and Bob’s feedback comments
to Ann, at stage 1 (introduction) of the writing process.

293
 Table 4. Segmenting feedback comments and cover letters
Stage 1: Introductions Coding System
Adapted Shea et al. (2010)
Writer Draft Cover Letter
Thematic units (Henri, 1992)
Ann 1 Hello all! Greeting
The introduction you are about to read is
unfortunately not a part of already written
Ann 1 Self-disclosure
article that I'm currently perfecting, but instead
something I wrote for this assignment.
I'm still struggling with 'global issues' such as
Ann 1 planning and structure of the article, Self-doubt
argumentation, reasoning and focus.
Thus I would like you to focus on those 'global
Ann 1 Advice
problems' and anything connected to them
and maybe consider 'local problems' such as
Ann 1 Advice
wording and grammar not a priority.
Ann 1 Looking forward to our meeting Future contact
Ann 1 and lots of feedback! Expressing emotion
Ann 1 Ann Name (no closure)
Stage 1 (feedback comments Ann to Bob) Coding System
Liu and Sadler (2003)
Draft Author Text
NRCs coded with Shea et al. (2010)
You have referred to 8 studies in your
introduction. Even if you use all of them
1 Bob Revision-oriented (Global comment)
extensively in the rest of your article (and
especially if you don't), this is too much.
The paragraph thus far has covered the second
move „establishing a niche“ as in bringing the
1 Bob Non-revision-oriented (Praise)
song lyrics (territory from the first move) into
the context of cognitive studies.
If you write the highlighted part of the
1 Bob paragraph into a longer third paragraph, you'll Revision-oriented (Global comment)
have all three moves.
Stage 1 (feedback comments Bob to Ann) Coding System
Liu and Sadler (2003)
Draft Author Text
NRCs coded with Shea et al. (2010)
a lack of references is very noticable here (and
1 Ann in the next part), as you speak of tradition and Revision-oriented (Global comment)
establish your niche.
1 Ann But it's understandable in the draft, of course. Non-revision-oriented (Empathy)

Ann’s cover letter is segmented according to theme and then categorised using
Shea et al.’s (2010) taxonomy as the basis. Ann’s feedback comments to Bob, and
vice-versa, are segmented into RCs and NRCs using Liu and Sadler (2004). Then,
the NRCs are sub-categorised in a similar way the cover letters. Sub-categories are
developed within the three categories: affective (i.e., expressing emotions), open
communication (i.e., advice) and cohesion (i.e., phatics), and they adhere to the
principles of the original taxonomy (Rourke et al. 2007) and its further develop-
ment (Shea et al. 2010). Finally, this developed taxonomy (Table 5) is tested on data
consensually obtained on other similar Academic writing courses.

294
Table 5. Coding scheme for measuring social presence within known peer feedback
(adapted from Shea et al. 2010: 19–20)

295
 4.2. Measuring social presence

The following taxonomy is then used to measure the number of SP indicators

exhibited by the group members both individually and then collectively as expressed
through their cover letters, and feedback letters. For the purpose of the initial
analysis and in order to make comparisons from incomplete data sets, all data
collected from the three introduction stages are averaged out and are assumed to
be one stage only. This resulted in data from the five writing stages (Introduction,
Methods, Results, Discussion and Abstract) being available for analysis. The final
analysis concerning the feedback letters examines the data over all the seven stages
in order to establish whether all the proposed sub-categories within the proposed
taxonomy are used at the macro-level of coding.

4.2.1. Cover letters

Apart from the abstract stage in which only Ann wrote a cover letter, all participants
wrote one cover letter for each stage. Thus, the abstract stage was discarded and this
analysis was conducted on the cover letters written for the other four stages only.
The number of words and the quantity of SP indicators, both overall and within
the SP sub-categories, in each student’s cover letter were measured along with
their respective mean values. From this, the group’s mean number of SP indicators
exhibited per cover letter in each writing stage was calculated. The group results are
reported graphically and individual results are reported descriptively.

4.2.2. Feedback letters

Collectively, there were a total of 44 out of a possible 60 feedback letters (73%)

written between the students over the five writing stages (Table 6). All the feedback
letters were completed for the introduction and method stages, 9 out of 12 for the
results, half for the abstract and 5 out of 12 for the discussion. Individually, Ann
and Dawn both wrote 12, Carol 11, and Bob 9 letters. Bob wrote no feedback letters
for the results and abstract stage, and Carol wrote none for the discussion stage.

Table 6. Feedback letters written by individual with averaged introduction stages

296
 For the analysis of the group as a whole, the mean number of words in the non-
revision-oriented comments (NRCs) and the mean length of its respective feedback
letter were calculated per writing stage and student. On a similar basis, the mean
quantity of SP indicators (overall and within the SP sub-categories) contained within
these letters were determined and are presented graphically.
Data from the feedback letters was analysed from all the seven stages (Table 7)
in order to assess the frequency of SP indicators in the sub-categories and identify
possible redundant indicators within the proposed taxonomy.

Table 7. Feedback letters written by individual over the seven stages

Collectively, there were a total of 65 out of a possible 84 feedback letters (77%)

completed between the students with the vast majority of the introduction letters
being completed. From this data, the total number of SP indicators used by the
whole group according to category was calculated and are reported graphically.
Individual results are reported descriptively.

5. Results
5.1. Cover letters

Each participant wrote one cover letter for each writing stage. Overall, the group on
average wrote 77 words per cover letter (Figure 2). Bob wrote the most and Dawn
the least at an average of 94 and 57 words per stage respectively. Collectively, most
words were written in the introduction stage (100), followed by results (76) and
methods (71), and least in the discussion stage (60).

Group
Figure 2. mean
Group meannumber
numberofofwords
wordsin
in cover letters
lettersper
perstage
stage

Individually, the amount of SP indicators per cover letter varies from 4 (Bob
in methods and discussion; Carol in methods) to 7.5 (Ann in introduction). Ann has 297
the highest mean SP (6.3 indicators per letter) and Bob the lowest SP (4.7 indicators
per letter). Despite Carol’s low SP in her methods letter, she displays a similar SP to
Dawn overall. Furthermore, writing stages containing a higher mean word count (see
Figure 2) have a larger number of SP indicators (Figure 3). Figure 3 shows that
collectively the number of SP indicators is highest in the category, affective, followed
by group cohesion and then open communication in all the stages. The total number of
 Group mean number of words in cover letters per stage

Individually, theamount
Individually, the amount ofof
SPSP indicators
indicators perper cover
cover letter
letter varies
varies fromfrom
4 (Bob4 (Bob in
methodsand
in methods and discussion;
discussion; Carol
Carol in methods)
in methods) to 7.5to(Ann
7.5 (Ann in introduction).
in introduction). Ann hasAnn has
thethe
highest mean SP (6.3 indicators per letter) and Bob the lowest SP (4.7 indicators
highest mean SP (6.3 indicators per letter) and Bob the lowest SP (4.7 indicators
perper
letter). Despite
letter). Carol’s
Despite low SP
Carol’s lowinSPherinmethods letter, she
her methods displays
letter, a similaraSP
she displays to
similar SP
Dawn overall. Furthermore, writing stages containing a higher mean word
to Dawn overall. Furthermore, writing stages containing a higher mean word count count (see
Figure
(see2)Figure
have a2)larger
havenumber of number
a larger SP indicators
of SP(Figure 3). Figure
indicators 3 shows
(Figure that 3 shows
3). Figure
collectively the number
that collectively theofnumber
SP indicators
of SP is highest inisthe
indicators category,
highest affective,
in the followed
category, affective,
by group cohesion and then open communication in all the stages. The total number of
followed by group cohesion and then open communication in all the stages. The
SP indicators varies according to category from 6.7 (introduction) to 4.8 per reviewer
total number of SP indicators varies according to category from 6.7 (introduction)
(discussion).
to 4.8 per reviewer (discussion).

Figure 3. Group
Group meannumber
mean numberofofSP
SP indicators
indicators per
per cover
coverletter
letterand
andwriting
writingstage
stage
Group mean number of SP indicators per cover letter and writing stage

Figure 4 shows the cumulative effect of the individual’s SP expressed in their four
Figure
Figure44shows thecumulative
shows the cumulative effect
effect of individual’s
of the the individual’s SP expressed
SP expressed in their in their
cover
four cover
four letters
coverletters according
lettersaccording to sub-category
according totosub-category
sub-category andandand category
category
category (see Table
(see Table
(see Table 2 for2 explanation
for explanation
2 for explanation
of
of codes).
codes).
of codes).

12

Figure 4. Number of SP indicators in all cover letters by sub-category and category

Number of SP indicators in all cover letters by sub-category and category
298

All the students show a high affective presence in self-disclosure (AF3) and in
self-doubt and mitigation (AF6). There are a few instances of expressing emotion
(AF1), unconventional expressions of emotion (AF4) and expressing value (AF5)
with a complete absence of humour (AF2). Advice (OC7) and apologising (OC8) are

12
 All the students show a high affective presence in self-disclosure (AF3) and in
self-doubt and mitigation (AF6). There are a few instances of expressing emotion
(AF1), unconventional expressions of emotion (AF4) and expressing value (AF5)
with a complete absence of humour (AF2). Advice (OC7) and apologising (OC8) are
the highest and almost the only open communication indicators used by students
Ann, Bob and Dawn. Carol is the only student to refer to past messages from her
reviewer (OC1), to never apologise (OC8) and to appreciate the advice given (OC3)
thein
highest and almost
one instance. the only
Asking open communication
questions of other group indicators
members used by students
(OC2), Ann,
encouragement
Bob and Dawn.
(OC4), Carol agreement
expressing is the only student
(OC5),to refer to (OC6)
criticism past messages from clarification
and seeking her reviewer (OC9)
(OC1), to never apologise (OC8) and to appreciate the advice given
are never used. In the category, group cohesion, everyone uses inclusive (OC3) in onepronouns
instance. Asking questions of other group members (OC2),
(CH2), most use both closures and vocatives (CH3a) with Ann alone encouragement (OC4),
preferring to
expressing agreement (OC5), criticism (OC6) and seeking clarification (OC9) are
close with the use of her name only (CH3c). However she is the only one to refer to
never used. In the category, group cohesion, everyone uses inclusive pronouns (CH2),
future group meetings (CH6). The other cohesive indicators, conventional saluta-
most use both closures and vocatives (CH3a) with Ann alone preferring to close with
thetions
use ofonly (CH1b),
her name onlyclosures
(CH3c). only (CH3b),
However she issocial
the onlysharing
one to(CH4),
refer to course reflection
future group
(CH5) (CH6).
meetings and phatics (CH7),
The other are not
cohesive used. conventional salutations only (CH1b),
indicators,
closuresAnn
onlyhas the highest
(CH3b), socialSP in both
sharing opencourse
(CH4), communication and group
reflection (CH5) cohesion. Carol
and phatics
has the highest
(CH7), are not used. affective SP, but the lowest in open communication. Overall, Ann
has the
Annhighest
has the number
highest SP ofinSPboth
indicators in the group.and
open communication The other
group members have a
cohesion.
Carol has the
similar highestofaffective
amount SP. SP, but the lowest in open communication. Overall,
Ann has the highest number of SP indicators in the group. The other members have a
similar amount of SP.
5.2. Feedback comments

Figure 5 shows the group’s mean number of words in NRCs and RCs within each
Figure 5 shows the group’s mean number of words in NRCs and RCs within each
feedback letter and writing stage as calculated from 44 feedback letters (see Table 6).
feedback letter and writing stage as calculated from 44 feedback letters (see Table 6).

Figure 5. Group’smean
Group’s meanword
word count
count per
per feedback
feedback letter
letterand
andstage
stage
Feedback
Feedback letters
letters are are on average
on average longest
longest forfor
thethe discussion(99
discussion (99words)
words)and
and introduc-
introduction (94 words)
tion (94 words) stages,
stages, andand shortest
shortest forforthe
theabstract
abstract (49
(49 words)
words)with
withaamean
meanofof 77
77 words per stage. The mean number of words used for NR comments overall is 19
words per stage. The mean number of words used for NR comments overall is 19
words, ranging from 22 to 25 words for the introduction, methods and discussion
words, ranging from 22 to 25 words for the introduction, methods and discussion
letters, and 12 to 14 words for the abstract and results respectively.
letters,
Theand 12 toof14SPwords
number for the
indicators perabstract
feedbackand results
letter respectively.
(Figure 6) is highest in the
categories open communication or cohesion, and lowest in the category6)affective.
The number of SP indicators per feedback letter (Figure is highest in the
categories open communication or cohesion, and lowest in the category affective.

299
 letters, and 12 to 14 words for the abstract and results respectively.
The number of SP indicators per feedback letter (Figure 6) is highest in the
categories open communication or cohesion, and lowest in the category affective.

Figure 6. Group’s
Group’smean social
mean presence
social indicators
presence per feedback
indicators letterletter
per feedback and stage
and stage
The combined effect of SP per reviewer is lowest in the abstract (1.7
13 The combined effect of SP per reviewer is lowest in the abstract (1.7 indicators) and
indicators) and highest in methods (2.8 indicators). As feedback letters are longer on
highest
average in methods
in the (2.8 indicators).
introduction As feedback
and method stages letters
(see Figure 5), are longer
longer ondoaverage
letters not in
the introduction
necessarily result and method
in a higher stagesof(see
number SP Figure 5), longer letters do not necessarily
indicators.
result inThe
a higher number of SP indicators.
amount of SP indicators contained within all the feedback letters over the
sevenThe amount
stages is 147ofSP
SPindicators
indicators contained
(Figure 7). within all the feedback letters over the
seven stages is 147 SP indicators (Figure 7).

Figure 7. Number of SP indicators in all feedback letters (65) by sub-category and category
Number of SP indicators in all feedback letters (65) by sub-category and category

The totalThe total number

number of words
of words written
written in these
in these 65 feedback
65 feedback exchanges
exchanges is is 4 085words
4 085
words
of whichof 1which 1 702 words
702 words (32.5%)(32.5%)
are NRarecomments.
NR comments.Open Open communication
communication is the
is the larg-
largest category with 45 instances of praise (OC3) and 12 instances of encouragement
est category with 45 instances of praise (OC3) and 12 instances of encouragement
(OC4). There is also a small amount of agreement (OC5), advice (OC7), asking for
(OC4). There is also a small amount of agreement (OC5), advice (OC7), asking for
clarification (OC9), and single instances of both criticism (OC6) and apologising
clarification (OC9), and single instances of both criticism (OC6) and apologising
(OC9). There are no instances of the students referring to past messages (OC1) or
(OC9). There areabout
asking questions no instances of theofstudents
other members the groupreferring
(OC2). to past messages (OC1) or
asking questions
Cohesion hasabout other members
36 instances of the the
of addressing group (OC2). by name with a
respondent
conventional salutation (CH1a) and 16 instances of closing the feedback letter with
300 both the reviewer’s name and a closing remark (CH3a). There are also a few instances
of using a closure only (CH3b), name only (CH3c) or opening the letter by name only
(CH1c). There are no other instances of cohesion of social sharing (CH4), course
reflection (CH5), references to future meetings (CH6) or the use of phatics (CH7).
Affective is the smallest category with four instances of using humour (AF2)
and mitigation (AF6), a couple of instances of unconventional expressions to express
emotion (AF4), and no instances of conventional expressions of expressing emotion
 Cohesion has 36 instances of addressing the respondent by name with a con-
ventional salutation (CH1a) and 16 instances of closing the feedback letter with both
the reviewer’s name and a closing remark (CH3a). There are also a few instances of
using a closure only (CH3b), name only (CH3c) or opening the letter by name only
(CH1c). There are no other instances of cohesion of social sharing (CH4), course
reflection (CH5), references to future meetings (CH6) or the use of phatics (CH7).
Affective is the smallest category with four instances of using humour (AF2) and
mitigation (AF6), a couple of instances of unconventional expressions to express
emotion (AF4), and no instances of conventional expressions of expressing emotion
(AF1), self-disclosure (AF3) and expressing value (AF5).

6. Interpretation of results
This section discusses whether the proposed taxonomy can be used to understand
more fully how the group and its individual members use affective language in
both their cover and feedback letters. It analyses the type and frequency of SP
indicators used in these sources on a category-by-category basis and coded at the
macro-level only. It accounts for the length of the cover letters, and the amount of
NRCs as compared to RCs in feedback letters. Then, it identifies variations in the
way individuals use affective language towards the group in their cover letters and
towards each other in their feedback letters that can be further analysed through
qualitative analysis. Finally, and through the interpretation of these findings, it
concludes by suggesting ways in which the proposed taxonomy can be used in future
research to shed further light on how affective language is used in the process of peer
feedback.

6.1. Affective

There is a much higher level of affective SP indicators exhibited in the cover let-
ters (48% of all SP indicators) as compared to those in the feedback letters (7%).
This is unsurprising, because cover letters are inherently affective in content. This
is where, unlike in feedback letters, the author is mainly writing about themselves
and their ‘feelings’ about how their text should be reviewed. The vast majority of
these SP indicators show the authors expressing mitigation or self-doubt (AF6)
in what they have written such as ‘I am still struggling with global issues… (Ann)’
or disclosing information (AF3) ‘I wrote something, that I refer to myself as the
“last part” of the paper… (Bob).’ The few instances of expressing emotion either
conventionally (AF1) or unconventionally (AF4) could be because the group either
prefers to adopt a more content-oriented approach as suggested by Leelo Keevallik
and Joachim Grzega (2008) and/or these indicators would be more frequent in the
micro-coding of such comments that has been disregarded in this analysis. The very
occasional use of expressing value (AF5), i.e., ‘I believe the following part is the part
that proves the validity of my research’ (Ann) is unexpected and its low use may
indicate tentativeness in the assertions of both the reviewers and writers. Humour
(AF4) is a personal issue and perhaps is more frequent within certain dyads where

301
 they become more comfortable with each other over time as in Dawn’s ‘me and my
speller.’ The absence of humour in the group letters may be due to differences in
pragmatic norms or personalities within the group.

6.2. Open communication

Conversely to the affective category, there is a much higher level of Open Commu-
nication SP indicators exhibited in the feedback letters (50% of all SP indicators)
as compared to those in the cover letters (7%). This is because the aim of feedback
letters is for reviewers to give advice on how to improve their colleague’s writ-
ing. Unlike in the feedback letters, most of these SP indicators (57 instances) are
motivational comments with the majority of such containing praise (OC3), i.e., ‘I
like everything! (Carol to Dawn).’ There are many remarks of encouragement and
empathy (OC4) such as Dawn’s comment to Ann of ‘Good luck with your paper!’ and
to a lesser extent of agreement (OC5) where Dawn again expresses agreement with
Ann with her cover letter (OC5) of ‘I am feeling totally fine with three last words
of your title (viz.).’ The high levels of Open Communication SP indicators suggest
that the reviewers do indeed deem motivational comments as important and this
concurs with Min’s (2006) findings. Purely critical remarks (OC6) are extremely
rare which tends to suggest that the reviewers do not wish to be overtly critical with
their feedback and so they soften such comments with positive comments. There are
sometimes instances of the reviewer seeking further clarification (OC9) in aspects
of the text that are unclear, and this signals issues that need to be addressed in the
synchronous f2f meeting. Reviewers offering, and writers seeking advice (OC7) are
reasonably common, and this is what one would expect in such exchanges. There
are instances of both the reviewers and writers apologising (OC8) for the lateness
of their submissions. This gives a good indication of the punctuality of the reviews,
cover letters and drafts and whether the group members feel obliged to apologise
if they submit them late. The few instances of referring to past texts (OC1) could
be due to these references being mentioned in the f2f group meetings, and the f2f
lectures may account for why there are no comments asking questions concerning
the instructor or of other students (OC2).

6.3. Cohesion

The amount of cohesion (CH) indicators is reasonably high for both the reviewers
(43% of all SP indicators) and writers (34%). Most reviewers opened and closed
their feedback letters with a friendly salutation with the name of the recipient such
as ‘Dear Bob’ (CH1a) and similarly their cover letters by addressing the whole group
in, ‘Dear All’ (CH2). Both cover letters and feedback letters ended with a conven-
tional closure with the author’s name as in ‘Sincerely yours, Bob’ (CH3a). This
adheres to the more formal social norms of opening and closing letters. Frequently,
reviewers and writers end their letter informally using only a closure without their
name (CH3b) or with their name only (CH3c), or even abruptly with no closure.
The writers may have decided to use these more informal structures because they

302
feel they are bonding closer to the group. Alternatively, this could be an individual
style or an over-sight caused by time constraints.
However, this lack of formality in the letters could indirectly affect whether the
more pragmatically minded reader might lose trust in their reviewer. This may be
because they view this feedback as overtly informal or even rude. Conversely, if
the dyads are both content-driven individuals, this informality will enable them to
build more trust in each other. Whatever these permutations are within the dyad,
the perceived ‘pragmatic’ formality intended by the writer and correspondingly
interpreted by the reader may have profound effects on whether the feedback is
indeed engaged with and subsequently implemented. Future studies could examine
whether the perceived understanding of the use of salutations and closures has
any effect on the peer feedback process. Group cohesion is further built by refer-
ences on a few occasions to future meetings (CH6) or more specifically to the f2f
group meetings (‘Looking forward to our meeting’ Ann). Surprisingly, there is no
use of phatics or language that serves a purely social function (CH7) and sharing
information unrelated to the course (CH5). This may also be because the group is
more ‘content-driven’. The purpose of the course is to improve their writing skills
for publication only. Thus, the use of unnecessary language such as phatics or social
language is deemed a waste of time. There is also no reference to course reflection
(CH5), but this may have been covered during the synchronous exchanges at the
group meetings or f2f lectures.

6.4. Word count

Generally, the results seem to show that the number of SP indicators within the
cover letters is normally proportional to its length. Thus, the longer the letter, the
more likely it is to contain a higher amount of SP indicators, explaining why there
are more SP indicators in the introduction than in the other cover letters. This is
logical as the vast majority of content in cover letters contains affective language.
However, this pattern is not necessarily exhibited at an individual level. This may
be because the segmentation of SP indicators is based on the unit of meaning rather
than the linguistic unit. In other words, SP indicators with the sub-categories of
group cohesion can be as short as one or two words, i.e., ‘sincerely Ann’ (CH3a),
as compared to other SP sub-categories such as praise (OC3) that can contain over
thirty words. Thus, differences in the word count within different sub-categories
(word length per SP indicator) as well as using other ways to analyse the word count
such as taking the density of SP indicators (i.e., by dividing the number of SP indica-
tors used by the number of words written) to give meaningful information should
be included in subsequent studies. However, determining the type of SP indicators
used as compared to the total written amount of words per cover letter seems to
give a good indication of the amount of affective language used.
The methods cover letter contains the highest SP and mean percentage of NRCs
(28.6%), but is only the third largest word count. As such, it seems that in addition
to word length and word length per SP indicator, the number of SP indicators per
feedback letter is also dependent on the percentage of NRCs. In other words, a
lengthy letter containing a high percentage of NRCs is more likely to contain more

303
 SP indicators than a long letter with a low percentage of NRCs or a short letter with
a high percentage of NRCs. Thus, calculating the density of SP indicators (i.e., SP
indicators/ 100 words) in follow-up studies may also give a good indication of the
amount of SP exhibited by the group and individually in each individual writing
stage.

7. Limitations and future research

This paper shows the development of a novel taxonomy that can be used to measure
affective factors within Academic writing groups. It measures the cumulative effect
of these by accounting for the uniqueness of each individual, and how they project
their distinct personalities or ‘social presence’ (SP) over time to express their emo-
tions, and build and sustain group commitment. The hypothesis is that reviewers
and writers exhibiting a high SP are more likely to produce higher-quality reviews
and more improved subsequent texts respectively than if they exhibit a ‘low SP’.
On a practical level, this taxonomy shows instances when teacher intervention may
be helpful. In research, it highlights interesting phenomena in the data that can
be further explored using qualitative methods. Although not ideal, as mentioned
in the methodology, a number of assumptions are made in the analysis in order to
draw meaningful comparisons between the participants. This is necessary because
not every participant completed all the cover and feedback letters on the course.
Furthermore, the impact of the synchronous f2f writing group meetings on the
participants’ SP should be included in future research designs. As such, this research
needs to be repeated on much larger and complete data sets before drawing more
conclusive deductions. This, though, should not distract from the strong potential
and versatility of using this taxonomy in tandem with other research questions
to gain a much clearer understanding of how affective factors can influence the
peer feedback process. One such way this could be explored is by comparing these
findings with the impact of revision-oriented comments on the quality of the texts
produced.

Abbreviations
AF affective
CH cohesion
CoI community of inquiry model
Disc. discussion
f2f face to face
GC group cohesion
Intro. introduction
IMRaD Introduction, Methods, Research and Discussion
NR non-revison
NRC non-revision-oriented comment
OC open communication
RC revision-oriented comment
SP social presence

304
References
Chang, Ching-Fen 2012. Peer review via three modes in an EFL writing course. – Computers
and Composition, 29 (1), 63–78. http://dx.doi.org/10.1016/j.compcom.2012.01.001
Cho, Kwangsu; Schunn, Christian, D.; Charney, Davida 2006. Commenting on writing typol-
ogy and perceived helpfulness of comments from novice peer reviewers and subject
matter experts. – Written Communication, 23 (3), 260–294.
Choi, Jaeho 2013. Does peer feedback affect L2 writers’ L2 learning, composition skills, meta-
cognitive knowledge, and L2 writing anxiety? – English Teaching, 68 (3), 187–212.
http://dx.doi.org/10.15858/engtea.68.3.201309.187
Dewey, John 1933. How We Think: A Restatement of the Relation of Reflective Thinking to
the Educational Process. Lexington, MA: Heath.
Diab, Nuwar, M. 2011. Assessing the relationship between different types of student feedback
and the quality of revised writing. – Assessing Writing, 16 (4), 274–292. http://dx.doi.
org/10.1016/j.asw.2011.08.001
dLu, Ruiling; Bol, Linda 2007. A comparison of anonymous versus identifiable e-peer review
on college student writing performance and the extent of critical feedback. – Journal
of Interactive Online Learning, 6 (2), 100–115.
Ehrman, Madeline, E.; Leaver, Betty, L.; Oxford, Rebecca, L. 2003. A brief overview of indi-
vidual differences in second language learning. – System, 31 (3), 313–330. http://
dx.doi.org/10.1016/S0346-251X(03)00045-9
Ellis, Rod 2003. Task-Based Language Learning and Teaching. Oxford: Oxford University
Press.
Ertmer, Peggy A.; Richardson Jennifer C.; Belland, Brian; Camin, Denise; Connolly,
Patrick; Coulthard, Glen; Lei, Kimfong; Mong, Christopher 2007. Using peer
feedback to enhance the quality of student online postings: An exploratory study. –
Journal of Computer-Mediated Communication, 12 (2), 412–433. http://dx.doi.
org/10.1111/j.1083-6101.2007.00331.x
Fahy, Patrick J. 2001. Addressing some common problems in transcript analysis. – The
International Review of Research in Open and Distributed Learning, 1 (2), 1–6.
Ferris, Dana R. 1997. The influence of teacher commentary on student revision. – Tesol
Quarterly, 31 (2) 315–39. http://dx.doi.org/10.2307/3588049
Gardner, R. 1985. Social Psychology and Second Language Learning. London: Edward Arnold.
Garrison, Randy, D.; Anderson, Terry; Archer, Walter 1999. Critical inquiry in a text-based
environment: Computer conferencing in higher education. – The Internet and Higher
Education, 2 (2), 87–105. http://dx.doi.org/10.1016/S1096-7516(00)00016-6
Garrison, Randy, D.; Anderson, Terry 2003. E-learning in the 21st Century: A Framework
for Research and Practice. Abingdon: Routledge Farmer.
Garrison, Randy, D; Arbaugh, Ben J. 2007. Researching the community of inquiry framework:
Review, issues, and future directions. – The Internet and Higher Education, 10 (3),
157–172. http://dx.doi.org/10.1016/j.iheduc.2007.04.001
Garrison, Randy, D; Cleveland-Innes, Martha; Fung Tak, S. 2010a. Exploring causal rela-
tionships among teaching, cognitive and social presence: Student perceptions of the
community of inquiry framework. – The Internet and Higher Education, 13 (1), 31–36.
http://dx.doi.org/10.1016/j.iheduc.2009.10.002
Garrison, Randy D.; Anderson, Terry; Archer, Walter 2010b. The first decade of the com-
munity of inquiry framework: A retrospective. – The Internet and Higher Education,
13 (1), 5–9. http://dx.doi.org/10.1016/j.iheduc.2009.10.003
Gee, Thomas C. 1972. Students’ responses to teacher comments. – Research in the Teaching
of English, 212–221.
Henri, France 1992. Computer Conferencing and Content Analysis. Berlin: Springer. http://
dx.doi.org/10.1007/978-3-642-77684-7_8

305
 Hyland, Fiona; Hyland, Ken 2001. Sugaring the pill: Praise and criticism in written feedback. –
Journal of Second Language Writing, 10 (3), 185–212. http://dx.doi.org/10.1016/
S1060-3743(01)00038-8
Kaufman, Julia H.; Schunn, Christian D. 2011. Students’ perceptions about peer assessment
for writing: Their origin and impact on revision work. – Instructional Science, 39 (3),
387–406. http://dx.doi.org/10.1007/s11251-010-9133-6
Keevallik, Leelo; Grzega Joachim 2008. A few notes on conversational patterns in Estonian. –
Journal for EuroLinguistiX, 5, 80–87.
Kollar, Ingo; Fischer Frank 2010. Peer assessment as collaborative learning: A cognitive per-
spective. – Learning and Instruction, Unravelling Peer Assessment, 20 (4), 344–348.
Kurt, Gokce; Atay Derin 2007. The effects of peer feedback on the writing anxiety of prospec-
tive Turkish teachers of EFL. – Online Submission, 3 (1), 12–23.
Lipman, Matthew 2003. Thinking in Education. Cambridge: Cambridge University Press.
http://dx.doi.org/10.1017/CBO9780511840272
Liu, Jun; Sadler, Randall W. 2003. The effect and affect of peer review in electronic versus
traditional modes on L2 writing. – Journal of English for Academic Purposes, 2 (3),
193–227. http://dx.doi.org/10.1016/S1475-1585(03)00025-0
Lomicka, Lara; Lord, Gillian 2007. Social presence in virtual communities of foreign
language (FL) teachers. – System, 35 (2), 208–28. http://dx.doi.org/10.1016/j.
system.2006.11.002
Lundstrom, Kristi; Baker, Wendy 2009. To give is better than to receive: The benefits of peer
review to the reviewer’s own writing. – Journal of Second Language Writing, 18 (1),
30–43. http://dx.doi.org/10.1016/j.jslw.2008.06.002
Min, Hui-Tzu 2006. The effects of trained peer review on EFL students’ revision types and
writing quality. – Journal of Second Language Writing, 15 (2), 118–41. http://dx.doi.
org/10.1016/j.jslw.2006.01.003
Motteram, Gary 2009. Social computing and teacher education: An agenda for course devel-
opment. – International Journal of Innovation in Language Learning and Teaching,
3 (1), 83–97. http://dx.doi.org/10.1080/17501220802655508
Nelson, Melissa M.; Schunn, Christian D. 2009. The nature of feedback: How different types
of peer feedback affect writing performance. – Instructional Science, 37 (4), 375–401.
http://dx.doi.org/10.1007/s11251-008-9053-x
Nicol, David J.; Macfarlane-Dick, Debra 2006. Formative assessment and self-regulated
learning: A model and seven principles of good feedback practice. – Studies in Higher
Education, 31 (2), 199–218. http://dx.doi.org/10.1080/03075070600572090
Richardson, Jennifer C.; Swan, Karen 2003. Examining social presence in online courses in
relation to students’ perceived learning and satisfaction. – Journal of Asynchronous
Learning Networks, 7 (1), 68–88.
Rourke, Liam; Anderson, Terry; Garrison, Randy D.; Archer, Walter 2007. Assessing social
presence in asynchronous text-based computer conferencing. – International Journal
of E-Learning & Distance Education, 14 (2), 50–71.
Shea, Peter; Bidjerano, Temi 2009. Community of inquiry as a theoretical framework to foster
‘epistemic engagement’ and ‘cognitive presence’ in online education. – Computers
& Education, 52 (3), 543–553. http://dx.doi.org/10.1016/j.compedu.2008.10.007
Shea, Peter; Hayes, Suzanne; Vickers, Jason; Gozza-Cohen, Mary; Uzuner, Sedef; Mehta,
Ruchi; Valchova, Anna; Rangan, Prahalad 2010. A re-examination of the community
of inquiry framework: Social network and content analysis. – The Internet and Higher
Education, 13 (1), 10–21. http://dx.doi.org/10.1016/j.iheduc.2009.11.002
Su, Zhang 2011. Peer feedback: A new approach to English writing instruction in a Chinese
college setting. – Sino-US English Teaching, 8 (6), 364–68.
Swales, John M. 1995. The role of the textbook in EAP writing research. – English for Specific
Purposes, 14 (1), 3–18. http://dx.doi.org/10.1016/0889-4906(94)00028-C
306
Topping, Keith 1998. Peer assessment between students in colleges and uni-
versities. – Review of Educational Research, 68 (3), 249–276. http://dx.doi.
org/10.3102/00346543068003249
Tuzi, Frank 2004. The impact of e-feedback on the revisions of L2 writers in an academic
writing course. – Computers and Composition, 21 (2), 217–235. http://dx.doi.
org/10.1016/j.compcom.2004.02.003
Van Zundert, Marjo; Sluijsmans, Dominique; Van Merriënboer, Jeroen 2010. Effective
Peer Assessment Processes: Research findings and future directions. – Learning and
Instruction, 20 (4), 270–279. http://dx.doi.org/10.1016/j.learninstruc.2009.08.004
Vygotsky, Lev S. 1980. Mind in Society: The Development of Higher Psychological Processes.
Harvard: Harvard University Press.
Williams, Marion; Burden, Robert 1996. Psychology in Language Teaching: A Social
Constructivist Approach. Cambridge: Cambridge University Press.
Zhao, Huahui; Sullivan, Kirk P.; Mellenius, Ingmarie 2014. Participation, interaction and
social presence: An exploratory study of collaboration in online peer review groups. –
British Journal of Educational Technology, 45 (5), 807–819.

Roger Michael Alan Yallop is a PhD student researching academic L2 writing

at the University of Tartu.
Jakobi 2, 50090 Tartu
[email protected]

307
akadEEmiliStE tEkStidE tagaSiSidES ESinEva
afEktiivSE kEElEkaSutuSE mõõtminE uudSE
lähEnEmiSviiSiga

Roger Michael Alan Yallop

Tartu Ülikool

Ingliskeelsete teadusartiklite avaldamine on sageli üks akadeemilise edasijõudmise

eeltingimusi. Seepärast muutub üha olulisemaks selliste toimivate õppemeetodite
väljatöötamine, mis aitaksid Eesti ülikoolides õppivatel üliõpilastel arendada aka-
deemiliste tekstide kirjutamise oskust võõrkeelena (L2) õpitud inglise keeles. Üks
võimalikke meetodeid on moodustada väikesed kirjutamisrühmad, kus kõik liikmed
annavad regulaarselt tagasisidet teiste rühmakaaslaste kirjutatu kohta. Niisugusel
puhul võib tagasiside kommentaarides esinev afektiivne keelekasutus tugevasti
mõjutada teksti autori ja hindaja vahelisi suhteid. See omakorda võib määrata, kas
tekstikirjutaja võtab tehtud märkusi oma töö edaspidisel redigeerimisel arvesse
või mitte. Artiklis on vaadeldud, kas on võimalik kasutada uudset lähenemisviisi,
et mõõta ühes magistriõppe üliõpilaste L2 kirjutamisrühmas afektiivseid tegureid
tagasisidestamise protsessis.
Esitatav lähenemisviis on üks sellise metoodika aspekt, mille kaudu uuritakse,
kuidas afektiivsed tegurid võivad mõjutada vastastikuse tagasiside protsessi;
kõnealune uuring on ühes Eesti ülikoolis läbiviidava pikiuuringu osa. Uudne
lähenemisviis mõõdab afektiivsete tegurite kumulatiivset mõju, võttes arvesse iga
isiku eripära ja seda, kuidas aja jooksul konkreetne isiksus avaldub ehk milline on
tema sotsiaalne kohalolu emotsioonide väljendamisel ning grupikuuluvuse loomi-
sel ja hoidmisel. Kõnealust lähenemisviisi kasutatakse nii rühma kui ka iga isiku
sotsiaalse kohalolu mõõtmiseks.
Püstitatud hüpoteesi kohaselt on tõenäoline, et suure kohalolufaktoriga
tagasisidestajad ja autorid annavad vastavalt kvaliteetsemat tagasisidet ja koos-
tavad lõpuks parema teksti kui need, kelle kohalolufaktor on väike. Lähenemis-
viisi aluseks on sotsiaalkultuurilisel teoorial põhinev mudel, mis on kinnistunud
sotsiaalkonstruktivistlikus paradigmas. Töö järeldustes on osutatud, kuidas saab
seda lähenemisviisi edaspidi kasutada nii kvantitatiivsetes kui ka kvalitatiivsetes
uurimismeetodites.
Võtmesõnad: sotsiaalne kohalolu, koostöörühm, pragmaatika, kirjutamisrühm,
vastastikune tagasiside

Trükitud 100 eksemplari