1
IONIAN UNIVERSITY
DEPARTMENT OF INFORMATICS
MSc in Informatics
Subject area: Informatics and Humanistic Studies
MASTER THESIS
Computer Supported Collaborative Learning (CSCL) in
Virtual Communities (VCs):
Linguistic Analysis and Inner Speech
Stefanos Nikiforos (ΠΜ201205)
Supervisor: Katia - Lida Kermanidis
Corfu 2014
2
Supervisor: Lect. Katia - Lida Kermanidis
Comittee: Ass. Prof. Panagiotis Vlamos
Lect. Konstantinos Chorianopoulos
3
Abstract
Virtual Learning Communities constitute an aspect of particular importance for
Computer Supported Collaborative Learning offering new opportunities in education
and setting new challenges. Strong sense of community improves learning outcomes,
provides satisfaction to the students and reduces the feeling of isolation.
Furthermore, feeling of membership in a community increases the collaboration
among the students, their commitment to the common goals and motivates them to
learn.
Language has a prominent role in learning communities being a communicative
tool that serves information exchange and knowledge construction and sharing.
Collaborative learning, occurring by the active engagement of the individuals, is
strongly associated with communities. Collaborative efforts (interaction through
language) among the members of the community leads to learning.
Inner speech is an esoteric mental language, usually not outerly expressed. In communities, under certain conditions, it is externalized having a completely idiosyncratic syntax and specific structure. It consists of apparent lack of cohesion and it is also fragmented in nature and abbreviated in comparison to the formal language used in most everyday interactions. This type of language is expected to lead to misunderstanding, unless the thoughts of the individuals are in accordance and mean the same (community).
As Virtual Classes are frequently embodied in the learning procedure, a
challenge of detecting whether they were transformed into communities or not, is
arising. In this study, a detailed linguistic analysis in the discourse among the class
members is proposed in order to detect whether a Virtual Class is a community or
not. This analysis is focused on two axes: inner speech and collaborative learning as
they both are basic features of a community.
4
Contents
Introduction ......................................................................................................... 6
1. Communities
1.1 Learning communities ............................................................................... 8
1.2 Virtual communities .................................................................................. 9
2. Collaborative learning and Inner speech
2.1 Collaborative learning .............................................................................. 11
2.2 Computer Supported Collaborative Learning (CSCL) .............................. 12
2.3 Inner speech ............................................................................................ 14
3. Virtual Learning Communities Linguistic Analysis Model
3.1 Model analysis of inner speech ............................................................... 16
3.1.1. Inner Speech Analysis summary .............................................. 25
3.2 Model analysis of collaborative learning ................................................ 28
3.2.1 Collaborative Learning Analysis summary ................................ 32
4. case studies .................................................................................................... 33
5. Results
5.1 Inner speech analysis ............................................................................... 36
5.1.1 Inner speech analysis in VC 1.1 ................................................. 36
5.1.2 Inner speech analysis in VC 1.2 ................................................. 39
5.1.3 Inner speech analysis in VC 2 .................................................... 43
5.1.4 Inner speech analysis in VC 3 .................................................... 48
5.1.5 Inner speech analysis in students' essay texts (ST) ................... 51
5.1.6 Comparative results - Discussion .............................................. 56
5.2 Collaborative learning analysis ................................................................ 68
5.2.1 Collaboration analysis in VC 1.1 ............................................... 68
5
5.2.2 Collaboration analysis in VC 1.2 ............................................... 69
5.2.3 Collaboration analysis in VC 2 .................................................. 71
5.2.4 Collaboration analysis in VC 3 ................................................... 72
5.2.5 Collaboration analysis in Students' Texts (ST) ........................... 74
5.2.6 Comparative results - Discussion .............................................. 75
5.3 Statistical analysis .................................................................................... 81
6. Conclusion ...................................................................................................... 82
Appendix ............................................................................................................ 84
References ........................................................................................................ 139
6
Introduction
Virtual Learning Communities offer new opportunities in education and set new
challenges in Computer Supported Collaborative Learning. Creation of virtual classes
has become widespread. Nevertheless there are questions arising: Is every VC always
a community as well? How can we detect the existence of a VC community? What
are its idiosyncratic properties?
In this study, a detailed linguistic analysis in the discourse among the class members is proposed in order to detect whether a Virtual Class is a community or not. This analysis is focused on two axes: inner speech and collaborative learning as they both are basic features of a community.
Various works using discourse analysis have been presented in the CSCL field.
Spanger et al. (2009) analyzed a corpus of referring expressions targeting to develop algorithms for generating expressions in a situated collaboration.
Wells (2002), examined the role of dialogue in activity and interaction using analysis of discourse between two students engaged in a joint activity.
Other studies use machine learning techniques in order to build automated classifiers of affect in chat logs (Brooks, 2013).
Blau et al. (1998) analyzed the language in tutorial conversations attempting to detect collaboration and the power relationships reflected in actual conversations between tutors and clients.
Veermans and Cesareni (2005) examined four case studies from different countries through analyzing and categorizing postings in virtual environments. Their target was to examine the content of students' and teachers' work in each project.
Zhang et al. (2007) analyzed the online knowledge building discourse in a community targeting to investigate the ability of students to take responsibility for their own knowledge advancement. They focused on socio-cognitive dynamics of knowledge building reflected in the online discourse.
Maness (2008) analyzed students' chat reference conversations and created a linguistic profile of them, focusing on the formality of the language.
Rovai (2002), examined the relationship between the sense of community and cognitive learning in an online educational environment.
Kollar et al. (2005), investigated how external collaboration scripts interact with learner's internal scripts aiming at facilitating collaborative argumentation in web-based learning environments.
Innes (2007) examined the quality of discourse in problem-based learning groups focusing on the type and quality of dialogue.
7
Daniel et al. (2003) explored how the notions of social capital and trust can be extended in virtual communities.
Unlike these studies, the proposed approach takes into account the correlation between community properties and both inner speech and collaborative learning features (Bielaczyc and Collins, 1999) by applying linguistic analysis to the discourse among class members as a means for community detection. To this end, the discourse of four different types of VCs is analyzed and compared against non-conversational language use by the same students, in order to detect differences.
This approach is considered as important because since now the distinction between the Virtual Classes and the communities has not been emphasized. This distinction is essential, as the differences between them are great. In Virtual Classes, the students, having stable and predefined roles, might work individually (action) leading to competition among them or cooperatively in the best case (individual action for a common target). In contrast, in communities the students interact and collaborate each other, while their roles are dynamic and interchangeable. This interaction results in socially constructed knowledge and learning (Bielaczyc and Collins, 1999; Veermans and Cesareni ,2005; Warschauer, 1997; Stahl et al, 2006; Stahl, 2000; Dillenbourg and Fischer, 2007; Innes, 2007; Rovai, 2002; Wells, 2002; Koschmann, 1999; Knipfer et al, 2009; Daniel et al, 2003).
For this reason, two detailed linguistic analysis models focused on the features of both collaborative learning and inner speech are proposed. The combination of these models can reveal the existence of a community or not. Since now very few work has been done concerning inner speech and collaborative learning in VCs in CSCL field. When this happens, it is focused on specific features. In contrast, the proposed models for community detection are comprehensive and are based on a set of features. Furthermore, these two models are applied in non-conversational language aiming to better weighted results.
This study is organized as follows: in Section 1, communities (in general), learning communities and virtual communities are presented. In Section 2, collaborative learning and inner speech and their correlation to the communities are referred. The proposed linguistic analysis model is presented in Section 3, while in section 4 case studies are described. In section 5 the results of the two models of analysis and their statistical analysis are presented, finishing in Section 6 with the conclusion.
8
1. Communities
Community is a complex, rich and diverse term. It is usually referred as a small
social unit that shares common values. A number of conditions existing in a
community affect the identity of the participants and their degree of cohesiveness
(Mc Millan and Chavis, 1986; Daniel et al, 2003).
There are various types of communities. Three types seem to be the most
common: 1) geographic communities, where location is the basic feature, 2)
communities of culture, consisting of people sharing a common culture and 3)
community organizations, which can be either informal (family) or formal
(professional association).
The most essential elements of the sense of belonging in a community are: 1)
mutual interdependence among members, 2) connectedness, 3) trust, 4)
interactivity and 5) shared values and goals (Rovai, 2002; Daniel et al, 2003). People
communicate, interact and influence one another, within a community (Engeström,
2001). Members establish their own identity and learn to function in the community.
Interaction among members affect the community structure either negatively or
positively (Mc Millan and Chavis, 1986).
One way of interaction among community members is mutualism. It is
considered as the most beneficial interaction as it involves collaboration which has
positive effect in all members, in contrast to competition that has positive effects for
some members and negative for others (Patterson, 2005).
1.1 Learning communities
Building of a community goes through various stages targeting to the creation of
the "true" community. In this "true" community members give deep respect to their
colleagues and care for the needs of others. However, a challenge for a community is
how to incorporate individuality and differences among its members. This challenge
is crucial for learning communities (Wells, 2002; Bielaczyc and Collins, 1999).
There is a significant relationship between classroom community and learning
(Innes, 2007). The stronger is the sense of community, the greater is learning
perceived, resulting also in less isolation and greater satisfaction (Rovai, 2002; Daniel
et al, 2003). Strong feelings of community provide benefits to students by increasing
1) the commitment to group goals, 2) collaboration among them, 3) satisfaction with
their efforts and 4) motivation to learn. In addition, students having a high sense of
community "feel burned out less often at school" (Rovai, 2002).
9
1.2 Virtual Communities
A social network of individuals who interact through the internet is defined as a
virtual community. Virtual communities consist of persons having mutual interests or
goals and cross geographical, political or cultural borders (Daniel et al, 2003).
Virtual communities have been developed in the recent years, especially after
the growth and the wide spread of the internet. They resemble real life communities
and they can serve multiple purposes. Members of a virtual community are able to
interact through social networking sites, chat rooms, forums and even virtual worlds.
Sharing similar interests allow members to bond each other and possibly form
friendship (Mc Millan and Chavis, 1986). Virtual communities offer the advantage of
instant information exchange and give users a feeling of membership and belonging.
Members can also give and receive support (Warschauer, 1997).
The most common type of virtual communities is social networking services. In
recent years, virtual communities are also used for educational purposes (Virtual
Learning Communities - VLCs). A VLC should be tolerate to differences and
individualism. Virtuality offers this opportunity as the members are judged
exclusively by their effort and work, no matter of their appearance, colour, gender or
disability (Warschauer, 1997; Daniel et al, 2003).
A classroom community consists of two (2) components: 1) feelings of
connectedness among students and 2) commonality of learning expectations and
goals (Rovai, 2002).
A learning community is strong under certain circumstances: students 1) feel
connected to each other and to the teacher, 2) use immediate communication in
order to reduce any existing social or psychological distance among them, 3) share
common interests and values, 4) help and trust each other, 5) actively engage in
two-way communications and 6) pursue common learning objectives (Rovai, 2002).
If students' satisfaction is achieved virtual (online) classes can be inviting and
successful learning environments. Specifically, students standing in front of their
computer screen should feel as if they are working together with a group of peers
(Rovai, 2002; Mc Millan and Chavis, 1986; Daniel et al, 2003; Dillenbourg and Fischer,
2007; Warschauer, 1997).
Learning is the feeling that knowledge is actively constructed within the
community. The acquisition of knowledge and understanding is enhanced within the
community and the learning needs of the students are satisfied. Students in a
classroom community internalize, at least partially group's values and goals. Learning
is the goal for a learning community, it is therefore an indispensable component
(Rovai, 2002).
10
Sense of community is a powerful force in our culture. There is therefore a
challenge of building learning communities based on faith, hope and tolerance and
using sense of community as a tool for collaboration (Mc Millan and Chavis, 1986).
11
2. Collaborative Learning and Inner Speech
2.1 Collaborative Learning
In Collaborative Learning (CL) knowledge can be created within a population
where members actively interact by sharing experiences and taking asymmetry
roles. Learners in CL are engaged in a common task, they capitalize on one another's
resources and skills and each one is both depended on and accountable to each
other. CL includes both face to face conversations and computer discussions.
Conversation analysis and statistical discourse analysis are methods used for
examining CL (Daniel et al, 2003).
In CL groups of students work together to search for understanding, meaning or
solutions as well as to create artifacts or products of their learning. Collaborative
writing, joint problem solving, study teams and group projects are common activities
in CL. This kind of work and the referred activities exist also in the cooperative
learning. CL, in contrast, occurs when individuals are actively engaged in a
community in which learning takes place through collaborative efforts (Stahl et al,
2006). Learning is not just the accumulation of new knowledge to those that already
exist, but the transformation and the reorganization of the previous knowledge.
When the students are confronted with data opposed to something they already
know, they experience cognitive collision that might lead them to revision and
reorganize their own ideas (Vekyri, 2007; Daniel et al, 2003; Stahl, 2000; Knipfer et
al, 2009; Bereiter, 1994). Public discussion is a central way for a learning community
to expand its knowledge (Bielaczyc and Collins, 1999).
Students in CL learn how to be able to solve problems as a group. Groups of
three (3) to five (5) students are allocated by the instructor who assigns a problem to
be solved. Ideas and solutions of each team should be presented to entire class,
allowing groups to come together as a whole. Discussion follows, targeting to
evaluate and not to judge the students' work. The goal is to remove focus of the
instructor's authority (Bielaczyc and Collins, 1999). Collaborative scripts should
target to create roles and mediating interactions, while allowing for flexibility in
dialogue and activities (Kollar et al, 2005).
Students' motivation to learn, their interpersonal relationships and the
expectations for personal success are improved by CL methods (Stahl, 2000). It is
shown that students working in small groups achieved significantly more than
students who worked individually. Discussion-based practices improved text
comprehension and critical-thinking skill for students across ethnic and
socioeconomic backgrounds. CL can also lead to students' success by deepening the
understanding of a given topic (Stahl et al, 2006).
12
Learning through CL methods seems not to be commonly used. Teachers' role
should be differentiated (Bereiter, 1994; Bielaczyc and Collins, 1999). In CL teachers
act as guides. During the learning process, students lead discussions and work
independently with teacher oversight and help when asked, rather than explicit
direction. Participation of learners is a key element to CL as it functions as the
method by which the learning process occurs. CL occurs when children and adults in
communities switch between "knowledge performers" and "observing helpers".
Horizontal structure allows for flexible leadership, which is one of the key aspects of
CL (Jameson et al, 2006).
Different types of CL have been implemented:
- Collaborative Network Learning (via electronic dialogue among co-learners,
learners and experts)
- Learning Management Systems (collection of tools used by learners to assist or
to be assisted by others)
- Collaborative Learning Network (learning systems working as a network, e-
learning)
- CL in Virtual Worlds
- Computer Supported Collaborative Learning (CSCL) (use of technology in a
learning environment to help mediate and support group interactions in a CL
context. Controlling and monitoring interactions, regulating tasks, rules and roles
and mediating the acquisition of new knowledge can be achieved with the use of
technology as well) (Stahl et al, 2006; Dillenbourg and Fischer, 2007).
2.2 Computer Supported Collaborative Learning (CSCL)
Computer-supported collaborative learning (CSCL) is defined as
a pedagogical approach wherein learning takes place via social interaction using a
computer or through the Internet. It is characterized by the sharing and construction
of knowledge among participants using technology as their primary means of
communication or as a common resource (Koschmann, 1999). CSCL can be
implemented both in online and classroom learning environments and can take
place synchronously or asynchronously (Bielaczyc and Collins, 1999). It can be
applied in all levels of education. As a consequence it poses the challenge of
increasing students' access to computers and the Internet.
Virtual Learning Communities (VLCs) constitute an aspect of particular
importance for Computer Supported Collaborative Learning (CSCL). Encouraging
students to learn together in small groups has been increasingly emphasized in the
broader learning sciences. Combining these two ideas (computer support and
13
collaborative learning) is an educational challenge that CSCL is designed to address it
(Stahl et al, 2006).
CSCL emerged as a strategy rich with research implications for the growing
philosophies of constructivism and social cognitivism (Daniel et al, 2003). The rapid
development of social media technologies and the increasing need of individuals to
understand and use those technologies has brought researchers from many
disciplines to the field of CSCL. CSCL is used today in traditional and online schools
and knowledge-building communities. The field of CSCL is based on learning theories
that emphasize that knowledge is the result of learners interacting with each other,
sharing knowledge, and building knowledge as a group (Stahl et al, 2006). Since the
field focuses on collaborative activity and collaborative learning, it inherently takes
much from constructivist and social cognitivist learning theories (Daniel et al, 2003).
Roots of collaborative epistemology related to CSCL are found in Vygotsky's
Social Learning Theory. Theory's notion of internalization and the idea that
knowledge is developed by one's interaction with his or her surrounding culture and
society are very important to CSCL. Another key element is the "zone of proximal
development". A range of tasks that might be too difficult for learners to master by
themselves can be made possible with the assistance of a more skilled individual or
teacher. Individual learners have different developmental capabilities in
collaborative situations than when they are working alone (Dillenbourg and Fischer,
2007; Knipfer et al, 2009; Innes, 2007). The measure of the difference between these
two capabilities is defined as the "zone of proximal development" (Stahl et al, 2006;
Engeström, 2001; Wells, 2002). These ideas are central to CSCL as knowledge
building is achieved through interaction with others.
An essential element to the zone of proximal development is the acquisition of
language. According to Vygotsky (2008), language is fundamental to children's
cognitive growth because language provides purpose and intention so that behaviors
can be better understood. Children communicate and learn from others through
dialogue, an important tool in the zone of proximal development (Warschauer,
1997).
Learning theories related to CSCL focus on the social aspect of learning and
knowledge building, recognizing that learning and knowledge building involve inter-
personal activities including conversation, argument, and negotiation (Stahl, 2000).
Collaboration theory, suggested by Stahl (2004) as a system of analysis for CSCL,
postulates that knowledge is constructed in social interactions such as discourse. The
theory suggests that learning is not a matter of accepting fixed facts, but is the
dynamic, on-going, and evolving result of complex interactions primarily taking place
within communities of people. The goal of collaboration theory is to develop an
14
understanding of how meaning is collaboratively constructed, preserved, and re-
learned through the media of language and artifacts in group interaction (Daniel et
al, 2003; Wells, 2002; Warschauer, 1997; Koschmann, 1999) .
CSCL software environments provide various forms of pedagogical support for
collaborative learning. But in most cases their role is secondary to the interpersonal
collaboration process among the students and possibly the teacher. Human group
processes cannot be replaced by the software, which has a supportive role (Stahl et
al, 2006).
Wikis are a common approach to CSCL providing a way to encourage discussion
among learners. Learners can also experience and resolve sociocognitive conflicts
during knowledge building in wikis (Knipfer et al, 2009). Technology-mediated
discourse allows participants that may be separated by time and distance to engage
in conversations and build knowledge together. Other approaches commonly used in
CSCL are the problem-based learning and the project-based learning where need for
collaboration is essential.
In CSCL teachers still have a vital role as they facilitate learning. Teacher is
responsible for the proper design of a project and for discussion monitoring
(Dillenbourg and Fischer, 2007).
2.3 Inner Speech
Vygotsky (2008) emphasizes the inter-relationship of language development and
thought. This relationship establishes the connection between speech (inner speech
or oral language) and the development of mental concepts and cognitive awareness.
Inner speech is described by Vygotsky (2008) as being qualitatively different from
external speech. Inner speech is developed from external speech via a gradual
process of internalization (Emerson, 1983).
Language is an external tool for the child at the beginning of its life, used for
social interaction. Child's behavior is guided by using this tool in a kind of a self-talk
or "thinking out loud". Gradually, self-talk is used a tool for shelf-directed and shelf
regulating behavior (Emerson, 1983). As speaking has been appropriated and
internalized, self-talk is "evaporated in thoughts" at the age of the child's school
age. Egocentric language is not dying at the threshold of the school age, but it is
transformed in inner speech (mute language) (Ehrich, 2006). Vygotsky (1987),
suggests that self-talk "develops along a rising not declining curve; it goes through an
evolution, not an involution. In the end, it becomes inner speech". Speaking is being
developed along two lines: the line of social communication and the line of inner
speech. Inner speech mediates and regulates child's activity through its thoughts.
The thoughts are, in turn, mediated by the semiotics of inner speech. Despite
15
thinking can take place even without language, it is mediated by language and is
developed in a higher level of sophistication (Emerson, 1983).
Inner speech is a language having its own specific features. Firstly, it has a
completely specific syntax consisting of apparent lack of cohesion, fragmented
nature and the abbreviation in comparison to the outer language. Clauses are
abbreviated by keeping the predicate and its accompanying words and omitting the
subject and the parts of sentence related to it. This abbreviation does not lead in
misunderstandings if the thoughts of the individuals are in accordance and mean the
same. In a different case, misunderstanding is inevitable. The more identical the
thoughts of the individuals are, the less linguistic cues are used. In an extreme level
of common understanding, entire discourse can be served by one single word
(Vygotsky, 2008; Socolov, 1972;).
In contrast, as in written language interlocutors are absent, it is fully developed
and its syntax structure reach to its higher level. In this language, abbreviations do
not serve communication needs. As a consequence written language, in comparison
to oral language, is fully developed and has a complicated syntax structure. Oral
language can become even more abbreviated when the mental proximity of the
interlocutors creates a community perception, allowing communication with
intimation and ellipticity. This exception of the oral language is the rule for inner
speech. Word is much more meaningful in inner speech than in the outer speech
(Vygotsky, 2008).
Inner speech is an efficient cognitive resource, because the abbreviations and
short-cuts speed it up. This speed makes inner speech in some ways more useful
than outer speech (Wiley, 2006).
According to Vygotsky, higher psychological functions appear first
interpsychologically, in collaborative action, and later intrapsychologically,
internalized by the individual (Engeström, 2011).
16
3. Virtual Learning Communities Linguistic Analysis Model
Sharing of knowledge within a community is achieved through shared codes and
language (Daniel et al, 2003; Stahl, 2000; Innes, 2007). Language is not only a tool for
communication as it also serves knowledge and information exchange (Dillenbourg
and Fischer, 2007; Knipfer et al, 2009). Furthermore, it "determines how individuals
perceive information they receive from individuals within their community or
outside the community" (Daniel et al, 2003). Communication and dialogue are in a
privileged position in the learning process, due to the assumption that knowledge is
socially constructed (Innes, 2007).
As a consequence, important results might come from applying linguistic
analysis in the language expressed among the class members.
3.1 Model analysis of inner speech
Socolov (1972) suggests that, in a community, under certain conditions, the
specific features of inner speech takes characteristics and appear in outer (surface)
speech. The more inner speech appears, the stronger the indication it gives us of the
existence of a community. Taking into account Vygotsky's (2008) argument that the
stronger the specific mental action of inner speech is, the clearer the peculiarities of
its syntax structure appear and attempting to identify the existence of inner speech,
it is argued in this study that we can apply linguistic analysis on the discourse that
has been developed among the team members. Subsequently, it could be defined
whether there are any inner speech features appeared in outer speech (Wiley,
2006).
Focusing on this attempt, a linguistic analysis based on the following features is
proposed.
IS-1 Omission of Subjects
According to Vygotsky (2008), one basic characteristic of inner speech is the
absence of the subject. So, it is proposed to investigate how many verbs are used in
the discourse without having a subject.
1) Verbs without having any subject or No subject
total number of verbs Verb count
2) Verbs with one or more subjects or With subject
total number of verbs Verb count
Applying these measures, a strong indicator of inner speech could be offered to
our study.
17
Clause types
IS-2 Omission of Conjunction
Omission of the conjunctions reveals that the language is desultory, elliptical
and informal. For this reason, the count of the clauses that are not joined by a
conjunction is suggested.
number of clauses without conjunction between them or no conjunction
total number of clauses clause count
IS-3 Informal clauses
In inner speech there is a common code for communication between the
communicating parties. This common code transforms the language genre and style,
and makes it more specific (Emerson, 1983). According to Vygotsky (2008), the main
feature of inner speech is ellipticity, i.e. the clauses consist of only the predicate. The
informal clauses (IS-3) (Maness, 2008; Pérez-Sabater, 2012), the clauses having no
verb (IS-4) and the semantically abbreviated clauses being elliptical in meaning (IS-5)
are therefore counted.
number of clauses having informal features or informal
total number of clauses clause count
IS-4 omission of verbs
number of clauses having no verbs or no verb
total number of clauses clause count
IS-5 Elliptical clauses
number of clauses with elliptical meaning or elliptical
total number of clauses clause count
Another feature of ellipticity in the language is the number of words used. Wiley
(2006) suggests that in inner speech "the stock of vocabulary is much smaller than in
outer language". It is then suggested to count the total of the words used per period.
This provides us with an indicator of the predicative speech.
IS-6 Words per clause
total number of words or word count
total number of clauses clause count
18
Alternatively it could be used:
IS-7 Words per period
total number of words or word count
total number of periods period count
IS-8 Punctuation
Another qualitative feature of inner speech is the reduced use of punctuation.
As inner speech has the characteristics of informal language, punctuation is likely to
be sparse as well (Brooks et al., 2013; Mannes, 2008; Pérez-Sabater, 2012). For this
reason, focusing on the punctuation used in the discourse is proposed.
More specific, the following punctuation marks should be counted:
a) parenthesis
total number of parentheses or parentheses
total number of clauses clause count
b) commas
total number of commas or commas
total number of clauses clause count
c) question marks
total number of question marks or question marks
total number of clauses clause count
d) dots
total number of dots or dots
total number of clauses clause count
e) exclamation marks
total number of exclamation marks or exclamation marks
total number of clauses clause
count
f) full stops
total number of full stops or full stops
total number of clauses clause count
19
g) total of the punctuation marks
total number of punctuation marks or punctuation marks
total number of clauses clause count
IS-9 Word types
Investigating the word types used by the team members, we get access to
another indicator of inner speech. The counting of the following word types is
proposed:
a) adverbs
Using adverbs is essential in a discourse when the speaker/writer wants to
specify his/her meaning and make it more clear to the receiver/reader. In the case of
inner speech, use of adverbs does not seem to be so essential, due to the
common/mutual understanding among the members of a team/class (Emerson,
1983).
It should then be counted:
total number of adverbs or adverbs
total number of words word count
But focusing only at the total number of the adverbs used might be misleading,
due to the variety of the adverb types. For this reason, the counting of the adverb
types as well is proposed. There are adverb types essential for a discourse, while
some other types can be omitted. In inner speech, adverbs of place, adverbs of
manner or relative adverbs are eliminated, while viewpoint and commenting
adverbs seem to be more essential.
So, in addition to adverb counting, the counting of adverb types (IS-10) is also
proposed:
a) adverbs of place
total number of adverbs of place or adverbs of place
total number of adverbs total adverbs
b) adverbs of time
total number of adverbs of time or adverbs of time
total number of adverbs total adverbs
20
c) adverbs of manner
total number of adverbs of manner or adverbs of manner
total number of adverbs total adverbs
d) adverbs of certainty
total number of adverbs of certainty or adverbs of certainty
total number of adverbs total adverbs
e) quantitative adverbs
total number of quantitative adverbs or quantitative adverbs
total number of adverbs total adverbs
f) interrogative adverbs
total number of interrogative adverbs or interrogative adverbs
total number of adverbs total adverbs
g) relative adverbs
total number of relative adverbs or relative adverbs
total number of adverbs total adverbs
h) viewpoint and commenting adverbs
total number of viewpoint and commenting advs or viewpoint & commenting
total number of adverbs total adverbs
IS-9.b adjectives
Adjectives are used to enrich the language and make it more specific. They are
needed when one wants to define something and to reduce the ambiguity. Absence
of the adjectives makes the language elliptical, ambiguous and general. As Vygotsky
(2008) claimed, inner speech contains only the predicate, i.e. only the terms that are
related to the verb. Wiley (2006), suggests that in inner speech "adjectives and other
modifiers can usually be dispensed with". As a result, it is expected that adjectives
will not be so commonly used in inner speech. For this reason, calculating the
average of the adjectives used in the discourse is proposed.
total number of adjectives or adjectives
total number of words word count
IS-9.c greeklish words
Greeklish is a type of informal written language used by Greeks. Specifically, it is
used when someone types Greek words with Latin letters. This type of language is
21
used only in the case of an informal communication and it is commonly used in
chatting, email and sms (Short Message Service). Therefore, if greeklish words are
used in a discourse it could be argued that the expressed language is informal.
total number of greeklish words or greeklish
total number of words word count
IS-9.d informal words
Similarly to the greeklish words, the use of informal words (shortened and
simplified word forms, idioms, diminutives) indicates informal communication which
is common only in the cases of team members feeling familiarity with each other.
Counting the informal words could offer an indication of inner speech.
total number of informal words or informal
total number of words word count
IS-9.e Emoticons
Another language feature found mainly in computing is the use of emoticons. It
offers an indication of informal communication (Brooks et al., 2013; Mannes, 2008;
Pérez-Sabater, 2012), a basic feature of inner speech.
total number of Emoticons or Emoticons
total number of words word count
IS-11 Syntax
Another aspect of the ellipticity of inner speech is expected to be the reduced
use of subordination and prepositional phrases. Use of subordinate clauses and
prepositional phrases enriches language and makes it more understandable.
According to Wiley (2006), inner speech is "simpler in both semantics and syntax,
using fewer words and fewer parts of speech".
It is argued that inner speech can be detected by counting these syntax terms.
total number of subordinate clauses or subordinate clauses
total number of clauses clause count
Special attention must be given to subordinate clauses that are used as an
object of the verb. In this case, subordinate clauses should not be taken into
account, because they constitute a predicate and are therefore essential for the
meaning of the sentence, and their existence is a basic feature of inner speech.
22
Similar to the subordinate clauses is the use of prepositional phrases. Elliptical
language contains as little linguistic terms as possible. Therefore, prepositional
phrases are not likely to be commonly used in inner speech.
total number of prepositional phrases or prepositional phrases
total number of clauses clause count
Additional terms
It is considered that using additional terms in a sentence makes the language
more explanatory and precise. This seems to be essential for achieving
communication. But does the same hold in the case of inner speech? Its elliptical
form and the mutual understanding should be kept in mind. So, it is expected that
these additional terms will not be commonly used in inner speech. Such terms as
apposition (IS-13), epexegesis (IS-14) and terms in the possessive (IS-15) and in the
accusative case (IS-16) should be included in the model of inner speech analysis.
Terms in the possessive or accusative case may reveal a lot of things (owner,
subject, property, time, value, object, content or relation). These terms should be
counted as a whole, despite the big variety, because counting them by category is
not likely to add something to the analysis.
IS-13 total number of appositions or appositions
total number of clauses clause count
IS-14 total number of epexegesis or epexegesis
total number of clauses clause count
IS-15 total number of terms in possessive case or terms in possessive case
total number of words word count
IS-16 total number of terms in accusative case or terms in accusative case
total number of words word count
It is expected that the less these terms are used, the greater is the indication of
the existence of inner speech.
IS-12 Article types
The two types of articles (definite/indefinite) give different meaning to a
discourse. Indefinite articles are used when someone wants to specify and describe
the meaning of his message. In inner speech, where common/mutual understanding
exists, the message is definite and clear to the receiver (Emerson, 1983; Mairesse et
al, 2007). For this reason, it is expected that the use of indefinite articles will be
limited in inner speech. Additionally, definite articles are likely to constitute the
majority of the articles in inner speech. The following metrics are then proposed:
23
a) total number of definite articles or definite articles
total number of articles total articles
b) total number of indefinite articles or indefinite articles
total number of articles total articles
In addition to these metrics, counting of the average number of articles to the
total words and to the total periods used, is also suggested, because it is likely to
offer an indication of the ellipticity of language.
c) total number of articles or articles
total number of words word count
d) total number of articles or articles
total number of periods period count
IS-17 Abbreviations
Abbreviations is a core feature of inner speech (Vygotsky, 2008; Socolov, 1972;
Wiley, 1996). Using an abbreviation in a communication means that one knows that
the receiver of the message is able to understand it. This requires a common
understanding between the sender and the receiver of the message which leads to
the result that the communication has inner speech features. Counting the
abbreviations might give an indicator of the existence of inner speech.
The use of the following metric is proposed:
total number of abbreviations or abbreviation
total number of words word count
IS-18 Metaphors
Similar to the use of abbreviations, it is considered that someone uses a
metaphor only in the case that the receiver is expected to understand it.
Additionally, metaphors are powerful for creating and exchanging rich sets of
meaning (Daniel et al, 2003). Accepting this, it is argued that metaphors are an index
mark of mutual understanding and therefore an index mark of inner speech.
total number of metaphors or metaphors
total number of clauses clause count
IS-19 Similes
In contrast to metaphors, similes are used in order to give more
details/explanations. Subsequently, their presence indicates a necessity for
24
additional information for achieving the common understanding. So, the absence of
similes can be an indicator for inner speech.
total number of similes or similes
total number of clauses clause count
IS-20 Word variety
According to Vygotsky (2008), the number of distinct words used in the
discourse within a community is usually restricted, due to the mutual understanding.
As a result, it is expected that, in the case of communities, the variety of the words
used is likely to be poor. "Vocabulary can be quite small and the same words can be
used over and over again" (Wiley, 2006). In contrast, when someone is addressing a
general audience, he/she is expected to use a richer variety of words (Mairesse et al,
2007). Targeting to examine whether a discourse indicates the existence of a
community or not, it is proposed to use the metric of word variety as follows:
variety of words or variety
total number of words word count
Calculating the word variety, we count how many different words are used in a
discourse. Different forms of words (i.e. verbs in present or past tense, nouns in
single or plural) are counted as one single word. Articles are excluded in the
aforementioned counts due to their frequency.
Within a community this metric is likely to be small, indicating a robust and
condensed communication.
25
3.1.1. Inner Speech Analysis summary
IS-1 Omission of Subjects No subject
Verb count
With subject
Verb count
IS-2 Omission of Conjunction No conjunction
clause count
IS-3 Informal clauses Informal clauses
clause count
IS-4 Omission of verbs No verb
clause count
IS-5 Elliptical clauses Elliptical clauses
clause count
IS-6 Words per clause word count
clause count
IS-7 Words per period word count
period count
IS-8 Punctuation marks a) parenthesis
clause count
b) commas
clause count
c) question marks
clause count
d) dots
clause count
e) exclamation marks
clause count
f) full stops
clause count
g) punctuation (total)
clause count
IS-9 Word types a) adverbs
word count
b) adjectives
word count
c) greeklish
word count
d) informal
word count
26
e) emoticons
word count
IS-10 Adverb types a) Adverbs of place
Total adverbs
b) Adverbs of time
Total adverbs
c) Adverbs of manner
Total adverbs
d) Adverbs of certainty
Total adverbs
e) quantitative adverbs
Total adverbs
f) interrogative adverbs
Total adverbs
g) relative adverbs
Total adverbs
h) Viewpoint and commenting advs
Total adverbs
IS-11 Syntax a) Subordinate clause
clause count
b) prepositional phrases
clause count
IS-12 Article types a) Definite
Total articles
b) indefinite
Total articles
c) articles
word count
d) articles
periods
IS-13 Apposition Apposition
clause count
IS-14 Epexegesis Epexegesis
Clause count
IS-15 Additional terms in
possessive case
terms in possessive case
word count
IS-16 Additional terms in
accusative case
terms in accusative case
word count
27
IS-17 Abbreviations Abbreviations
word count
IS-18 Metaphors Metaphors
clause count
IS-19 Similes Similes
clause count
IS-20 Word variety Variety
word count
28
3.2 Model analysis of collaborative learning
Collaboration is considered to be the most important shared characteristic in
Virtual Learning Communities (Daniel et al, 2003). Focusing on the discourse among
the members of a team, a group, a class or a community, it is suggested that we
might be able to investigate index marks of collaborative learning. Linguistic analysis
could then be applied on the discourse, focusing on specific characteristics.
CA-1 Verbs in first plural person
Use of verbs in the first person plural form (we) can constitute an indicator of
team action, work or knowledge that has been produced collaboratively. According
to Mc Millan and Chavis (1986) "personal investment is an important contributor to
a person's feeling of group membership and to his or her sense of community". This
could be applied by calculating the average of the verbs in the first plural person to
the total number of the verbs used.
total number of verbs in 1st person plural or Verbs (we)
total number of verbs Verb count
Clause types
The types of clauses used in the discourse could be another indicator of
collaboration. Specifically, the clauses expressing emotion and the ones that express
rewarding could be searched.
CA-2 Emotional clauses
As it has already been mentioned (Mc Millan and Chavis, 1986), emotion is a
unique characteristic of the discourse within a community. Additionally, emotion is
directly related to inner speech as well. Wiley (2006) suggests that "inner language is
so pervaded with emotion". As a result, it is suggested to focus on the clauses
expressing emotions in order to investigate whether there were developed
emotional relationships among the members of a group (Brooks et al., 2013). These
clauses (emotional) can be counted and calculate their average to the total number
of the clauses used.
total number of clauses showing emotion or emotional clauses
total number of clauses clause count
CA-3 Rewarding clauses
Members of a group that have developed strong relationships and strong ties
feel the need to reward their partners for their effort. According to Mc Millan and
Chavis (1986), "it is obvious that for any group to maintain a positive sense of
29
togetherness, the individual-group association must be rewarding for its members".
As a result, it is probable for rewarding clauses to be used in the discourse within
the group members. Focusing on the rewarding clauses it can be examined whether
there were strong ties in a class (Bielaczyc and Collins, 1999; Mairesse et al, 2007). If
one reaches at this result, it can then be argued that the class members have
collaborated enough.
total clauses expressing reward or rewarding clauses
total number of clauses clause count
Word types
In addition to the clause types metrics, it is considered that the type of the
words used must also be examined. Subsequently, the use of the following metrics is
proposed.
CA-4 Clauses of negation
Clauses of negation can offer an index mark of the attitude of the
team/community members (Mairesse et al, 2007). This kind of clauses is likely to be
less as the collaboration increases. Community members tend to offer constructive
proposals/suggestions to each other, rather than disagreements. That is why it is
argued that we should count the clauses of negation of a discourse. Clauses
containing negative words ("no", "not", "don't") are considered as negative. It is
proposed to count them in percentage to the total clauses.
It is argued that a large number of clauses of negation might indicate lack of
collaboration.
total number of clauses of negation or clauses of negation
total number of clauses clause count
CA-5 Clauses of reason
Clauses of reason are used in a discourse in order to argue, express an opinion
and explain proposals. When this happens, it shows that a member of a
team/community respects his team members and that he is proposing something,
without giving orders. Using clauses of reason might offer us an index of
collaboration. Subsequently, they include in the model of analysis.
total number of clauses of reason or clauses of reason
total number of clauses clause count
30
CA-6 Familiarity words
Use of familiarity words indicates the familiarity intimacy that has been
developed among the members of a team which has been transformed into a
community (Mairesse et al, 2007). Especially in the cases of a virtual class that the
students did not know each other before the creation of the class, this metric can be
considered as a strong indication of the existence of a community.
total number of words showing familiarity or familiarity words
total number of words word count
CA-7 Inclusive words
The existence of inclusive words in the discourse might offer an index of a
feeling of membership. Words such as "together", "team", "company", "community"
etc. can be considered as inclusive words (Mairesse et al, 2007).
total number of inclusive words or inclusive words
total number of words word count
CA-8 Social words
Social words (like "friend", "colleague", "mate") can provide an index mark for
the existence of a community (Mairesse et al, 2007). The more social words are used
in the discourse, the more likely it is for strong social relationships among the
members of a class to exist.
total number of social words or social words
total number of words word count
CA-9 Emotional words
In addition to the evaluation of the emotional sentences used in the discourse,
the evaluation of the emotional words is considered as useful as well (Brooks et al.,
2013).
total number of emotional words or emotional words
total number of words word count
This metric should be enriched by another one in order to investigate whether
the emotional words express positive (CA-10) or negative emotion (CA-11). So, the
use of the following is proposed:
total number of positive emotion words or positive emotion
total number of emotional words emotional words
31
total number of negative emotion words or negative emotion
total number of emotional words emotional words
It is expected that in the case of collaborative learning, the majority of the
emotional words will express positive emotion, as there is strong correlation
between the members' positive experience and the community bond (Mc Millan and
Chavis, 1986; Brooks et al., 2013; Mairesse et al, 2007).
CA-12, CA-13 Use of first person plural pronouns
Using possessive or personal pronouns in the first person plural form (we, ours)
indicates the sense of belonging in a team (Mc Millan and Chavis, 1986). It also
indicates the sense of the common/mutual belonging, the co-construction of
knowledge and the feeling of sharing items with others. Pronouns are also directly
related to inner speech. Wiley (2006) refers that " this pronominal scheme I will treat
as a niche or circuitry or set of channels within which inner speech goes on. The
pronouns are the saddle, and if you want to ride through this linguistic land you have
to get on that saddle and inhabit the pronouns".
CA-12 total number of pronouns in the 1st plural person or (we, ours)
total number of pronouns pronouns
For a more precise specification the following could be used:
CA-13 total number of pronouns in the 1st plural person or (we, ours)_
total number of possessive + personal pronouns possessive + personal
32
3.2.1 Collaborative Learning Analysis summary
CA-1 Verbs in first plural person Verbs (we)
Verb count
CA-2 Emotional clauses emotional clauses
clause count
CA-3 Rewarding clauses rewarding clauses
clause count
CA-4 Clauses of negation Clauses of negation
clause count
CA-5 Clauses of reason Clauses of reason
clause count
CA-6 Familiarity words familiarity words
word count
CA-7 Inclusive words inclusive words
word count
CA-8 Social words social words
word count
CA-9 Emotional words emotional words
word count
CA-10 Positive emotion positive emotion words
emotional words
CA-11 Negative emotion negative emotion words
emotional words
CA-12 Use of first person plural
pronouns
Us, ours
Pronouns
CA-13 Use of first person plural
pronouns
Us, ours
Possessive + personal pronouns
33
4. Case studies
The five different learning communities used as case studies in this work are
described in this section.
Virtual class 1 (VC 1)
Virtual Class 1 (VC 1) was created during the spring semester of the school year
between two (2) schools, an elementary school (ES) and a high school (HS). These
schools were located in two different towns in Greece. The elementary school
students (20) who participated (ages 11-12) were attending the 6th grade of the
elementary education. The high school students (20) (ages 12-13) were attending
the 1st grade class of the secondary education. The main target of that project was
the collaboration between the two classes in order for the students to create a wiki
and exchange information about the town/location they live in.
Students of each school were divided into working groups, each one consisting
of two or three students. Seventeen (17) groups were created. Each group took over
a topic, collected the relevant material and uploaded it. Both the topic and the
material were selected by the members of the team. The teachers who were
involved in the project had a supporting and inspirational role and tried to minimize
their involvement.
Wikispaces was the collaborative platform used in this project.
The students uploaded artifacts (links, photos, text) associated with their own
town on topics of their choice. Consequently, students of the partner school used
these artifacts to present that town. So, the students of ES used the material posted
by students of HS to present their (HS's) town. HS students worked reversely.
Firstly, a web page was created to be used by the students for introducing each
other. Besides that, students of the two schools introduced themselves through a
video conference that took place at the beginning of the project. During the
implementation of the project students were exchanging communication messages
in a special web page (discussion posts). So they had the ability to communicate, to
post their comments, to express their proposals or arguments. Each group created
its own page and every member (not only the members of the team) was able to
post his comments. The discourse among the members of the groups was recorded
at this web page. Furthermore, the log files of the actions (insertions, corrections,
deletions) could be accessed through the revision history provided by the selected
tool (wikispaces).
34
Subgroups:
VC 1.1
Discourse in VC 1 has been divided into two subgroups for the needs of the
analysis. The first subgroup (VC 1.1) contains the discourse among the team
members after having completed their task. Students expressed their impressions,
opinions and feelings for the already completed project. In this case, there was no
problem to be solved and the students chatted in a more free frame. This
communication took place on a special web page, which functioned as a forum for all
the students of both schools.
VC 1.2
The second subgroup (VC 1.2) contains the discourse among the team members
during the project, i.e. during the effort to solve the given problem (construction of a
wiki that presents each other's place). In this case, communication is focused on the
task and serves students' needs of each team for collaboration.
Virtual class 2 (VC 2)
Virtual Class 2 (VC 2) was created during the winter semester of the school year
between two (2) elementary schools (ES1 and ES2) located in different towns in
Greece. Each class had twenty (20) students (ages 11-12).
The elementary students (20) from ES1 who participated in this VC were the
same ones described in VC 1.
The design of this project’s teams was the same as the previous one described in
VC 1. The groups created in this VC were eighteen (18).
The two main differences that have to be mentioned are
i) the difference between the educative level of the students in VC 1, which does
not exist in this VC, and
ii) the previous experience for the elementary students gained through their
participation in VC 1.
Virtual class 3 (VC 3)
In this case study a real class was transformed into a virtual one through running
an one month project using online collaborative tools. The target of the project was
the creation of presentations for a national holiday.
The students who participated in the Virtual Class 3 (VC 3) were the same
students of the elementary school ES1 that joined the two aforementioned virtual
35
classes (VC 1 and VC 2). During this project the students were divided into nine (9)
groups. Each group consisted of two (2) or three (3) students. Creation of the groups
was made according to the students' choice.
Each group took the responsibility of creating a presentation related to the
general topic of the project. Topics were selected by the members of the group in
collaboration with the other group members. Nine (9) presentations were therefore
produced at the end of the project.
Teachers had an active instructive role. The selected environment for the
project was Google Drive due to its capability to allow for collaborative work from a
distance.
Students were guided to create two (2) files in order to create a collaborative
platform: one (1) presentation file and one (1) document file for the necessities of
the communication among the group members. In these files both the members of
each group and the teachers were defined as common users.
Log files of the activities were kept through the functionality provided by the
tool used (historical revision). The communication messages among the participants
(students and teachers) were also recorded at the aforementioned document file.
Student's essay texts (ST)
Having in mind that chat communication is generally informal (Brooks, 2013;
Bielaczyc and Collins, 1999) it had to be examined whether the results of the analysis
appeared exclusively under these conditions (case studies) or whether they were
constant attributes of the students' language use. An index mark of the students'
formal language was also required. For this reason, students' essay texts (ST) were
also used in the analysis in comparison to the language used in VCs.
These essay texts were written by the same elementary students of ES1 that
took part in the case studies (VC 1, VC 2 and VC 3). They were written throughout
the same school year when the case studies took place. Students wrote them within
the linguistics course in their school.
These texts are narrative essays written by 7 different students (4 boys and 3
girls) out of a total of 20 in the class. They contain a total of 3.577 words and 666
clauses, while VC 1.1 had 210 and 52, VC 1.2 had 453 and 106, VC 2 had 471 and 102
and VC 3 had 704 and 147 respectively.
36
5. Results
5.1 Inner speech analysis
Having completed the model of inner speech analysis, it was applied in the case
studies in order to evaluate its performance.
5.1.1 Inner speech analysis in VC 1.1
Omission of subjects
In VC 1.1 88% of the verbs used had not any subject. Only 12% of the verbs had
a subject.
IS-1 Omission of Subjects No subject
Verb count
44/50 0,88
With subject
Verb count
6/50 0,12
Clause types
Performing the model for clause types, 12 informal clauses were found out of a
total of 52. Two (2) clauses had no verb and six (6) were elliptical.
The total words used were 210, when the total of the clauses were 52 and the
total of periods were 17. So, students used 4,04 words per clause or 12,35 words per
period.
IS-2 Omission of Conjunction No conjunction
clause count
0/52 0
IS-3 Informal clauses Informal clauses
clause count
12/52 0,23
IS-4 Omission of verbs No verb
clause count
2/52 0,04
IS-5 Elliptical clauses Elliptical clauses
clause count
6/52 0,12
IS-6 Words per clause word count
clause count
210/52 4,04
IS-7 Words per period word count
period count
210/17 12,35
37
Punctuation
The total number of punctuation marks used in this discourse was 37 (27
exclamation marks, 7 full stops, 1 comma, 1 question mark and 1 dots).
It is remarkable that exclamation marks constitute the majority, while full stops
are only 7. Furthermore, total of the "ending" marks for a clause are 37, while the
clauses are 52.
IS-8 Punctuation
marks
a) parentheses
clause count
0/52 0
b) commas
clause count
1/52 0,02
c) question marks
clause count
1/52 0,02
d) dots
clause count
1/52 0,02
e) exclamation marks
clause count
27/52 0,52
f) full stops
clause count
7/52 0,13
g) punctuation (total)
clause count
37/52 0,71
Word types
Adverbs were the most common type of the words used in the discourse. They
were 25, adjectives were 3, informal words 4 and 1 emoticon, when total words
were 210.
IS-9 Word types a) adverbs
word count
25/210 0,12
b) adjectives
word count
3/210 0,01
c) greeklish
word count
0/210 0
d) informal
word count
4/210 0,02
e) emoticons
word count
1/210 0
38
Adverb types
The adverbs used in this discourse were counted according to their types. So, 4
adverbs of place, 7 adverbs of manner and 13 quantitative adverbs were counted.
Quantitative adverbs constitute the majority. No other kind of adverbs was used.
Syntax types
In this discourse there were 15 subordinate clauses and 13 prepositional
phrases, when the total clauses were 52.
IS-11 Syntax a) Subordinate clauses
clause count
15/52 0,29
b) prepositional phrases
clause count
13/52 0,25
Article types
The definite articles were 19 while indefinite articles were not used at all.
IS-10 Adverb types a) Adverbs of place
Total adverbs
4/25 0,16
b) Adverbs of time
Total adverbs
0/25 0
c) Adverbs of manner
Total adverbs
7/25 0,28
d) Adverbs of certainty
Total adverbs
0/25 0
e) quantitative adverbs
Total adverbs
14/25 0,56
f) interrogative adverbs
Total adverbs
0/25 0
g) relative adverbs
Total adverbs
0/25 0
h) Viewpoint and commenting advs
Total adverbs
0/25 0
39
IS-12 Article types a) Definite
Total articles
19/19 1,00
b) indefinite
Total articles
0/19 0
c) articles
word count
19/210 0,09
d) articles
periods
19/17 1,12
Additional terms
Additional terms were 2 (1 epexegesis and 1 term in possessive case).
IS-13 Apposition Apposition
Clause count
0/52 0
IS-14 Epexegesis Epexegesis
Clause count
1/52 0,02
IS-15 Terms in possessive case possessive case
word count
1/210 0
IS-16 Terms in accusative case accusative case
word count
0/210 0
Abbreviations-Metaphors-Similes
No abbreviations were used. Students also did not use any metaphors or similes.
Word variety
The variety of the words used in the discourse was examined. There were 65
different words, when the total words were 191.
IS-20 Word variety Variety
word count
65/191 0,34
5.1.2 Inner speech analysis in VC 1.2
The discourse in VC 1.2 took place during the effort of the students to solve the
problem of creating their project.
Omission of subjects
Nearly all the verbs used in this discourse had no subject. In a great percentage,
77 verbs out of a total of 84, had no subject.
40
IS-1 Omission of Subjects No subject
Verb count
77/84 0,92
With subject
Verb count
7/84 0,08
Clause types
The clauses used were at mainly informal (80) or elliptical (60). Also, 22 clauses
had no verb and 2 had no conjunction.
IS-2 Omission of Conjunction No conjunction
clause count
2/106 0,02
IS-3 Informal clauses Informal clauses
clause count
80/106 0,75
IS-4 Omission of verbs No verb
clause count
22/106 0,21
IS-5 Elliptical clauses Elliptical clauses
clause count
60/106 0,57
IS-6 Words per clause word count
clause count
453/106 4,27
IS-7 Words per period word count
period count
453/64 7,08
Punctuation
Students in this VC did not seem to be very typical in use of punctuation. In a
total of 106 clauses, they used 28 full stops, 30 exclamation marks, 8 commas, 3
question marks and 3 dots. It should be noticed that the majority of them are the
exclamation marks which indicates an informal type of communication (Brooks et al.,
2013; Pérez-Sabater, 2012).
41
IS-8 Punctuation
marks
a) parentheses
clause count
0/106 0
b) commas
clause count
8/106 0,08
c) question marks
clause count
3/106 0,03
d) dots
clause count
3/106 0,03
e) exclamation marks
clause count
30/106 0,28
f) full stops
clause count
28/106 0,26
g) punctuation (total)
clause count
72/106 0,68
Word types
Students used 453 words. 33 of them were adverbs, 30 adjectives, 29 informal
and 2 greeklish.
IS-9 Word types a) adverbs
word count
33/453 0,07
b) adjectives
word count
30/453 0,07
c) greeklish
word count
2/453 0
d) informal
word count
29/453 0,06
e) emoticons
word count
0/453 0
Adverb types
The total number of adverbs were 33. The majority of them (13) were
quantitative. 8 were adverbs of manner, 6 adverbs of place, 4 adverbs of time, 1
relative and 1 viewpoint and commenting adverb.
42
IS-10 Adverb types a) Adverbs of place
Total adverbs
6/33 0,18
b) Adverbs of time
Total adverbs
4/33 0,12
c) Adverbs of manner
Total adverbs
8/33 0,24
d) Adverbs of certainty
Total adverbs
0/33 0
e) quantitative adverbs
Total adverbs
13/33 0,39
f) interrogative adverbs
Total adverbs
0/33 0
g) relative adverbs
Total adverbs
1/33 0,03
h) Viewpoint and commenting advs
Total adverbs
1/33 0,03
Syntax types
Prepositional phrases were 22 and the subordinate clauses 13 out of a total of
106 clauses.
IS-11 Syntax a) Subordinate clauses
clause count
13/106 0,12
b) prepositional phrases
clause count
22/106 0,21
Article types
In a total of 453 words, there were 53 definite articles, while the indefinites
were just 1.
IS-12 Article types a) Definite
Total articles
53/54 0,98
b) indefinite
Total articles
1/54 0,02
c) articles
word count
54/453 0,12
d) articles
periods
54/64 0,84
43
Additional terms
In a total of 106 clauses neither epexegesis nor apposition were used. Attributes
in possessive or accusative case were nearly null (3 and 1 respectively) out of a total
of 453 words.
IS-13 Apposition Apposition
Clause count
0/106 0
IS-14 Epexegesis epexegesis
Clause count
0/106 0
IS-15 Terms in possessive case possessive case
word count
3/453 0,01
IS-16 Terms in accusative case accusative case
word count
1/453 0
Abbreviations-Metaphors-Similes
Similar to the above mentioned terms, abbreviations were just a few (2) and
metaphors and similes were null.
IS-17 Abbreviations Abbreviations
word count
2/453 0
IS-18 Metaphors Metaphors
clause count
0/453 0
IS-19 Similes Similes
clause count
0/453 0
Word variety
Students used 157 different words out of a total of 399.
IS-20 Word variety Variety
word count
157/399 0,39
5.1.3 Inner speech analysis in VC 2
Omission of subjects
In VC 2 85% of the verbs had no subject.
44
IS-1 Omission of Subjects No subject
Verb count
66/78 0,85
With subject
Verb count
12/78 0,15
Clause types
In this VC, informal clauses were the majority (59). A significant number (35)
were elliptical, while in 26 clauses the verb was omitted and 12 clauses had no
conjunction.
Counting of the words per clause gave the average of 4,62, while counting them
per period the average was 7,03.
IS-2 Omission of Conjunction No conjunction
clause count
12/102 0,12
IS-3 Informal clauses Informal clauses
clause count
59/102 0,58
IS-4 Omission of verbs No verb
clause count
26/102 0,25
IS-5 Elliptical clauses Elliptical clauses
clause count
35/102 0,34
IS-6 Words per clause word count
clause count
471/102 4,62
IS-7 Words per period word count
period count
471/67 7,03
Punctuation
In VC 2 the majority of the punctuation marks were the full stops (24), while a
significant number of question marks (20) and exclamation marks (15) appeared.
Total punctuation marks were 66 in 102 clauses.
45
IS-8 Punctuation marks a) parentheses
clause count
3/102 0,03
b) commas
clause count
2/102 0,02
c) question marks
clause count
20/102 0,20
d) dots
clause count
2/102 0,02
e) exclamation marks
clause count
15/102 0,15
f) full stops
clause count
24/102 0,24
g) punctuation (total)
clause count
66/102 0,65
Word types
Extended use of greeklish words (61) appeared in this discourse. Informal words
were 35, adverbs 27 and adjectives 21 out of a total of 471 words.
IS-9 Word types a) adverbs
word count
27/471 0,06
b) adjectives
word count
21/471 0,04
c) greeklish
word count
61/471 0,13
d) informal
word count
35/471 0,07
e) emoticons
word count
2/471 0
Adverb types
Four types of adverbs were used in this discourse. Adverbs of place, quantitative
and viewpoint and commenting adverbs were in almost equal frequency.
46
IS-10 Adverb types a) Adverbs of place
Total adverbs
9/27 0,33
b) Adverbs of time
Total adverbs
0/27 0
c) Adverbs of manner
Total adverbs
2/27 0,07
d) Adverbs of certainty
Total adverbs
0/27 0
e) quantitative adverbs
Total adverbs
8/27 0,30
f) interrogative adverbs
Total adverbs
0/27 0
g) relative adverbs
Total adverbs
0/27 0
h) Viewpoint and commenting advs
Total adverbs
8/27 0,30
Syntax types
Total clauses were 102. 14 of them were subordinate clauses. The prepositional
phrases were 26.
IS-11 Syntax a) Subordinate clauses
clause count
14/102 0,14
b) prepositional phrases
clause count
26/102 0,25
Article types
Similar to the previous analyzed case studies (VC 1.1, VC 1.2), VC 2 also used,
almost exclusively, definite articles.
47
Additional terms
Four additional terms (1 epexegesis, 2 possessive case and 1 accusative case
attributes) were counted.
IS-13 Apposition Apposition
Clause count
0/102 0
IS-14 Epexegesis epexegesis
Clause count
1/102 0,01
IS-15 Terms in possessive case possessive case
word count
2/471 0
IS-16 Terms in accusative case accusative case
word count
1/471 0
Abbreviations-Metaphors-Similes
This VC did use abbreviations (8) and metaphors (4). No similes were used.
IS-17 Abbreviations Abbreviations
word count
8/471 0,02
IS-18 Metaphors Metaphors
clause count
4/102 0,04
IS-19 Similes Similes
clause count
0/102 0
Word variety
Students used 147 different words. Total words were 413, giving a score of 0,36.
IS-20 Word variety Variety
word count
147/413 0,36
IS-12 Article types a) Definite
Total articles
56/58 0,97
b) indefinite
Total articles
2/58 0,03
c) articles
word count
58/471 0,12
d) articles
periods
58/67 0,87
48
5.1.4 Inner speech analysis in VC 3
Omission of subjects
In VC 3 the two categories, "with subject" and "no subject", were almost equal.
Verbs without having any subject were slightly more.
IS-1 Omission of Subjects No subject
Verb count
64/125 0,51
With subject
Verb count
61/125 0,49
Clause types
Informal (101) and elliptical (89) clauses outmatched in this discourse. There
were also 21 clauses without having a verb and 20 without having conjunctions. It is
remarkable that students used 704 words in this discourse. Words per clause were
4,79 and words per period 8,09.
IS-2 Omission of Conjunction No conjunction
clause count
20/147 0,14
IS-3 Informal clauses Informal clauses
clause count
101/147 0,69
IS-4 Omission of verbs No verb
clause count
21/147 0,14
IS-5 Elliptical clauses Elliptical clauses
clause count
89/147 0,61
IS-6 Words per clause word count
clause count
704/147 4,79
IS-7 Words per period word count
period count
704/87 8,09
Punctuation
The majority of the punctuation marks in VC 3 were full stops (37) and commas
(20). Question marks were 10 and exclamation marks were 9.
49
IS-8 Punctuation marks a) parentheses
clause count
1/147 0,01
b) commas
clause count
20/147 0,14
c) question marks
clause count
10/147 0,07
d) dots
clause count
3/147 0,02
e) exclamation marks
clause count
9/147 0,06
f) full stops
clause count
37/147 0,25
g) punctuation (total)
clause count
80/147 0,54
Word types
Adjectives (55) and informal words (53) were almost equal, while adverbs were
43.
IS-9 Word types a) adverbs
word count
43/704 0,06
b) adjectives
word count
55/704 0,08
c) greeklish
word count
5/704 0,01
d) informal
word count
53/704 0,08
e) emoticons
word count
0/704 0
Adverb types
Quantitative adverbs, adverbs of manner and adverbs of place were the most
common types of the adverbs used in this discourse.
50
IS-10 Adverb types a) Adverbs of place
Total adverbs
11/43 0,26
b) Adverbs of time
Total adverbs
4/43 0,09
c) Adverbs of manner
Total adverbs
11/43 0,26
d) Adverbs of certainty
Total adverbs
0/43 0
e) quantitative adverbs
Total adverbs
15/43 0,33
f) interrogative adverbs
Total adverbs
0/43 0
g) relative adverbs
Total adverbs
0/43 0
h) Viewpoint and commenting advs
Total adverbs
2/43 0,05
Syntax types
Prepositional phrases were the majority (42). Subordinate clauses were just 11.
IS-11 Syntax a) Subordinate clauses
clause count
11/147 0,07
b) prepositional phrases
clause count
42/147 0,29
Article types
It is remarkable that, also in VC 3, articles were almost exclusively definite.
IS-12 Article types a) Definite
Total articles
103/110 0,94
b) indefinite
Total articles
7/110 0,06
c) articles
word count
110/704 0,16
d) articles
periods
110/87 1,26
51
Additional terms
Additional terms were nearly null, with the exception of attributes in possessive
case (11).
IS-13 Apposition Apposition
Clause count
0/147 0
IS-14 Epexegesis Epexegesis
Clause count
1/147 0,01
IS-15 Terms in possessive case possessive case
word count
11/704 0,02
IS-16 Terms in accusative case accusative case
word count
0/704 0
Abbreviations-Metaphors-Similes
Abbreviations were 30 in a total of 704 words. No similes were used in the
discourse. Metaphors were 18.
IS-17 Abbreviations Abbreviations
word count
30/704 0,04
IS-18 Metaphors Metaphors
clause count
18/147 0,12
IS-19 Similes Similes
clause count
0/147 0
Word variety
223 different words were found in this discourse, giving a score of 0,38.
IS-20 Word variety Variety
word count
223/594 0,38
5.1.5 Inner speech analysis in students' essay texts (ST)
Omission of subjects
The majority (67%) of the verbs used in the students' texts did not have any
subject.
52
IS-1 Omission of Subjects No subject
Verb count
447/663 0,67
With subject
Verb count
216/663 0,33
Clause types
Informal and no verb clauses were nearly null. Neither omission of conjunction
nor elliptical clauses were commonly used.
It is remarkable that the words per period were nearly up to 13.
IS-2 Omission of Conjunction No conjunction
clause count
17/666 0,03
IS-3 Informal clauses Informal clauses
clause count
3/666 0
IS-4 Omission of verbs No verb
clause count
6/666 0,01
IS-5 Elliptical clauses Elliptical clauses
clause count
26/666 0,04
IS-6 Words per clause word count
clause count
3577/666 5,37
IS-7 Words per period word count
period count
3577/279 12,82
Punctuation
Use of commas and full stops is increased and indicates a rather formal language
(Brooks et al., 2013; Mannes, 2008). Question marks were null. Exclamation marks
were just 13 in 666 clauses.
53
IS-8 Punctuation marks a) parentheses
clause count
0/666 0
b) commas
clause count
120/666 0,18
c) question marks
clause count
0/666 0
d) dots
clause count
3/666 0
e) exclamation marks
clause count
13/666 0,02
f) full stops
clause count
264/666 0,40
g) punctuation (total)
clause count
403/666 0,61
Word types
Adverbs and adjectives were the most common word types used in students'
texts. There was only 1 informal word, which strongly indicates formal language.
IS-9 Word types a) adverbs
word count
205/3577 0,06
b) adjectives
word count
315/3577 0,09
c) greeklish
word count
3/3577 0
d) informal
word count
1/3577 0
e) emoticons
word count
0/3577 0
Adverb types
Adverbs of time were the most common type (69). Quantitative adverbs were
55, adverbs of manner 46 and adverbs of place 32.
54
IS-10 Adverb types a) Adverbs of place
Total adverbs
32/205 0,16
b) Adverbs of time
Total adverbs
69/205 0,34
c) Adverbs of manner
Total adverbs
46/205 0,23
d) Adverbs of certainty
Total adverbs
1/205 0
e) quantitative adverbs
Total adverbs
55/205 0,27
f) interrogative adverbs
Total adverbs
0/205 0
g) relative adverbs
Total adverbs
2/205 0,01
h) Viewpoint and commenting advs
Total adverbs
0/205 0
Syntax types
Prepositional phrases (238) and subordinate clauses (153) seem to be very
common syntax types in the students' texts.
IS-11 Syntax a) Subordinate clauses
clause count
153/666 0,23
b) prepositional phrases
clause count
238/666 0,36
Article types
Definite articles were the majority (467), but use of indefinites is increased. The
total use of articles seems increased as well.
55
IS-12 Article types a) Definite
Total articles
467/531 0,88
b) indefinite
Total articles
64/531 0,12
c) articles
word count
531/3577 0,15
d) articles
periods
531/279 1,90
Additional terms
Additional terms were not commonly used by students.
IS-13 Apposition Apposition
Clause count
0/666 0
IS-14 Epexegesis Epexegesis
Clause count
5/666 0,01
IS-15 Terms in possessive case possessive case
word count
108/3577 0,03
IS-16 Terms in accusative case accusative case
word count
33/3577 0,01
Abbreviations-Metaphors-Similes
Abbreviations were very few in students' texts (16 in a total of 3577 words).
Metaphors were also few (24) in a total of 666 clauses. Similes were just 10.
IS-17 Abbreviations Abbreviations
word count
6/3577 0
IS-18 Metaphors Metaphors
clause count
24/666 0,04
IS-19 Similes Similes
clause count
10/666 0,02
Word variety
Word variety is calculated for each one student individually, in order to identify
any possible existing differences between them.
Student 1 got a score of 0,47 by using 323 different words in a total of 688.
Student 2 scored 0,50 (250/503). The lowest score was for student 3: 0,41 (271/660).
56
Student 4 wrote short texts (232 words) with a variety score of 0,47. Student 5 got
0,55 (184/335). Student 6 used 387 words and scored 0,48. Student 6 had the
highest score of 0,65, having written the shortest texts (241 words). In conclusion,
students achieved an average score of 0,50.
IS-20 Word
variety
Student
1
Student
2
Student
3
Student
4
Student
5
Student
6
Student
7
Variety/
word count
323/688 250/503 271/660 109/232 184/335 187/387 156/241
Score 0,47 0,50 0,41 0,47 0,55 0,48 0,65
Average
score
0,50
5.1.6 Comparative results - Discussion
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
IS-1 Omission
of
Subjects
No subject
Verb count
0,88 0,92 0,85 0,51 0,67
With subject
Verb count
0,12 0,08 0,15 0,49 0,33
In VC 1.1, VC 1.2 and VC 2 omission of subjects is very frequent providing a
strong indicator of inner speech existence (Vygotsky, 2008). Especially in VC 1.2 this
indication is the strongest one.
In contrast, this feature is rather sparse in VC 3. This is possibly the case because
of the active instructive role of the teachers, which affects language and makes it
more formal (Bielaczyc and Collins, 1999; Maness, 2008). It is expected that language
in VC 3 is influenced by the discrimination of roles between the teachers and the
students. As they do not have the same role (teachers are the instructors), members
of the VC have to specify their identity i.e. use of subjects (Pérez-Sabater, 2012).
Despite the fact that in Students' texts omission seems to be commonly used, it
is significantly lower than in VC 1.1, VC 1.2 and VC 2 (tables IS-1/1-IS-1/3, page 84).
This feature (omission of subjects) is a basic feature of inner speech (Vygotsky,
2008) providing a strong indicator of inner speech existence.
57
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
IS-2 Omission of
Conjunction
No conjunction
clause count
0 0,02 0,12 0,14 0,03
IS-3 Informal
clauses
Informal clauses
clause count
0,23 0,75 0,58 0,69 0
IS-4 Omission of
verbs
No verb
clause count
0,04 0,21 0,25 0,14 0,01
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,12 0,57 0,34 0,61 0,04
IS-6 Words per
clause
word count
clause count
4,04 4,27 4,62 4,79 5,37
IS-7 Words per
period
word count
period count
12,35 7,08 7,03 8,09 12,82
In IS-2 feature (omission of conjunction) significant differences are recorded
between both VC 2 and VC 3 and the other cases (tables IS-2/1-IS-2/10, pages 86-
87). VC 1.1 has zero clauses that are not linked together, being thus in accordance
with VC 2, where these clauses are rare (0,02).
Feature IS-3 strongly indicates that the language in all VCs was informal.
Informal clauses might be expected in VC 3, which had already been a community,
but they are also greatly used in all other VCs. This indicates the familiarity among
the team members and therefore the existence of the community.
Two thirds of the total clauses in VC 1.2 were informal. This is undoubtedly an
impressive feature of this discourse. Also in VC 2 and VC 3 informal clauses were very
common, indicating an informal language. This feature gives a very strong indication
of inner speech, as in ST informal clauses are null. Even in VC 1.1, where the
proportion of informal clauses reaches 0,23, there is a great difference with ST (0)
(table IS-3/1, page 88).
IS-4 and IS-5 are much higher in VCs than in ST, confirming the existence of
informal language (Santaholma, 2005). Specifically in IS-4 one fourth of the total
clauses in VC 2 had no verb. It was approximately the same in VC 1.2. In all VCs this
feature is higher than in ST. Even in VC 1.1 which has the lowest score, it is four
times greater than in ST. Omission of verbs appears rarely in ST, confirming that this
a feature of informal language (Vygotsky, 2008; Santaholma, 2005).
58
IS-5 (elliptical clauses) have the lowest percentage in ST (0,04). In VC 1.2 and in
VC 3 ellipticity reaches up to 60%. But also in VC 2 a considerable percentage (34%)
of the total clauses are elliptical. VC 1.1 has the lowest score (0,12), but still remains
three times higher than in ST. As ellipticity is a basic feature of inner speech
(Vygotsky, 2008), this feature provides us with a strong indication of inner speech
existence. VC 3, which presented the highest percentage, was a priori a community
confirming what Vygotsky (2008) has argued. IS-5 shows that VC 1.2 is using inner
speech at the same level with VC 3, which is considered as an index mark for the
ellipticity, revealing the similarity between these two VCs. We should therefore
accept that VC 1.2 is also a community.
It is also obvious from IS-6 and mainly from IS-7, that the number of the words
used in VCs is restricted, which indicates a strong correlation with inner speech.
Exception to the restricted use of words is VC 1.1, possibly affected of the absence of
a problem-based project. It should be kept in mind that in this VC students had
completed their task and, in that discourse, they expressed their impressions,
thoughts and feelings in a free frame.
VC 1.1 used 4,04 words per clause, when the words used in ST are 5,37. If this
feature is examined in comparison with IS-7, we can see that IS-7 gives a more clear
indication of ellipticity. ST, constituting the formal language of students, contain
12,82 words per period, while in VC this number reached merely 7,03. We can notice
that in VC 1.1 students' language for a non-problem based task contains almost the
same amount of words as in ST. In contrast, in all other VCs, where the discourse is
about a problem-based task, the words per period are significantly lower. In VC 3,
this feature is increased, but it might be influenced by the teachers' language who
had an instructive role in the procedure (Bielaczyc and Collins, 1999).
59
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
IS-8 Punctuation
marks
a) parentheses
clause count
0 0 0,03 0,01 0
b) commas
clause count
0,02 0,08 0,02 0,14 0,18
c) question marks
clause count
0,02 0,03 0,20 0,07 0
d) dots
clause count
0,02 0,03 0,02 0,02 0
e) exclamation marks
clause count
0,52 0,28 0,15 0,06 0,02
f) full stops
clause count
0,13 0,26 0,24 0,25 0,40
g) punctuation (total)
clause count
0,71 0,68 0,65 0,54 0,61
Parenthesis
Parentheses are rarely used. In VC 1.1 and VC 1.2 they are null. But in VC 2 and
VC 3 they are 0,03 and 0,01 respectively. In ST parentheses are null, giving an index
point of students' formal language (Brooks et al., 2013).
Commas
Use of commas is clearly higher in ST than in VCs. Taking this into account we
note that use of commas in VCs is restricted, indicating informal language (Brooks et
al., 2013).
Question marks
Students did not use question marks in their essays. In contrast, students in VC
2 used them a lot (0,20). In VC 1.1 lower use was expected as students expressed
their impressions, i.e. they made statements. Low use is observed also in VC 1.2. It
seems that the students in this VC did not have the same need to ask a lot of
questions. The score in VC 3 was 0,07.
Once again, the increased use of question marks in VCs indicates informal
language (Brooks et al., 2013).
60
Dots
Similarly to the use of question marks, dots were not used in ST. In VCs, dots had
a score of 0,02 to 0,03.
Exclamation marks
In formal language exclamation marks are used in a low level basis (Brooks et al.,
2013; Pérez-Sabater, 2012). Their use can therefore give an indication of informal
language.
According to the results, score in ST was 0,02 when the lowest score in VCs was
three times higher (VC 3: 0,06). In VC 2 the score was increased (0,15) and in VC 1.2
reached up to 0,28. In VC 1.1, where the students were asked to express their
impressions of the project that they had already completed, it was expected that the
use of the exclamation marks would be increased. Nevertheless, the score was
extremely higher (0,52).
Full stops
Full stops are a basic feature of formal language (Mannes, 2008). It is essential
to use at least one full stop per sentence. Alternatively, someone could use a
question mark or an exclamation mark or dots.
Use of full stops seems to be very common in ST (0,40), while in VCs this feature
varies from 0,24 to 0,26. VC 1.1 is an exception because the use of full stops is much
lower (0,13). This may be attributed to the increased use of exclamation marks. Once
again, low use of full stops indicates that VCs used informal language (Brooks et al.,
2013).
Punctuation (total)
Examining the total use of the punctuation marks might offer a general view of
the students' language. But because there is a big variety of punctuation marks and
each one has a different use, i.e. a different meaning, examining only the total use of
punctuation marks cannot analyze the language in depth. For this reason the total
use of punctuation marks should be examined in combination with the use of each
type of mark.
VC 1.1 used the punctuation marks at the highest rate (0,71). Having in mind
that they needed to express their impressions, it seems that punctuation marks are
used as an additional tool of expression. This is confirmed by the increased use of
the exclamation marks.
61
VC 1.2, which presented also a high percentage in this feature (0,68), appears to
show a more balanced use of the different punctuation marks. In VC 3 the
percentage was the lowest one.
In ST the percentage is 0,61, with use of only three types of punctuation marks.
One of them is used rarely (exclamation marks: 0,02). So the percentage in ST (0,61)
results, formed by the use of only two types of punctuation marks (full stops and
commas), indicates formal language (Brooks et al., 2013).
Overall, in VCs the use of punctuation marks is more complicated as it results
from various types of punctuation marks, bringing out the need to use richer
expressive means.
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
IS-9 Word
types
a) adverbs
word count
0,12 0,07 0,06 0,06 0,06
b) adjectives
word count
0,01 0,07 0,04 0,08 0,09
c) greeklish
word count
0 0 0,13 0,01 0
d) informal
word count
0,02 0,06 0,07 0,08 0
e) emoticons
word count
0 0 0 0 0
Adverbs
The highest percentage is observed in VC 1.1. Use of adverbs is approximately
the same in all other cases. It is then necessary to analyze their use by the various
adverb types (which will follow below).
Adjectives
Adjectives are commonly used in ST (0,09), in VC 3 (0,08) and in VC 1.2 (0,07).
But in VC 2 the result is 0,04 and in VC 1.1 it is 0,01 indicating a more precise and
restricted language, as adjectives offer additional information to the meaning, i.e.
modification (Wiley, 2006; Vygotsky, 2008).
Greeklish
Use of greeklish in VC 2 (0,13) is quite extensive, which indicates strongly an
informal language. In all the other cases greeklish is not substantially used.
62
Informal words
Informal words in ST are null. In VC 1.2, VC 2 and VC 3 they are 0,06, 0,07 and
0,08 respectively, adding one more indication of the informal language used by the
students in the VCs.
Emoticons
Emoticons cannot offer anything in the analysis as they were not used in any
case.
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
IS-10 Adverb
types
a) Adverbs of place
Total adverbs
0,16 0,18 0,33 0,26 0,16
b) Adverbs of time
Total adverbs
0 0,12 0 0,09 0,34
c) Adverbs of manner
Total adverbs
0,28 0,24 0,07 0,26 0,23
d) Adverbs of certainty
Total adverbs
0 0 0 0 0
e) quantitative adverbs
Total adverbs
0,56 0,39 0,30 0,33 0,27
f) interrogative adverbs
Total adverbs
0 0 0 0 0
g) relative adverbs
Total adverbs
0 0,03 0 0 0,01
h) Viewpoint and commenting
Total adverbs
0 0,03 0,30 0,05 0
Adverb types
Average use of adverbs of place in ST is approximately equal to VC 1.1 and VC
1.2. In VC 2 and VC 3 it is higher than in the other cases, i.e. 0,33 and 0,26
respectively. It is observed that in these two VCs (VC 2 and VC 3), use of adverbs of
place is higher than in the other cases.
Adverbs of time seem to play an important role in ST (0,34), because of their
narrative nature. This is not observed in VCs, where the highest score is 0,12 (VC
1.2). In VC 3 the score is 0,09, while in VC 1.1 and VC 2 use of adverbs of time is null.
Specifically in VC 1.1, time was expected to have a minor importance because the
students had already completed their project.
63
In VC 2 adverbs of manner were used very infrequently (0,07), in contrast to the
other VCs which used them in an approximately equal percentage (VC 1.2: 0,24 - VC
3: 0,26 - VC 1.1: 0,28). In ST the score is lower: 0,23.
Adverbs of certainty are so rarely used that they cannot offer any index.
Quantitative adverbs are commonly used in VC 1.1, where students needed to
describe their impressions. In this case, quantitative adverbs helped students to
express the degree of their impressions, feelings, emotions. It is therefore expected
that in VC 1.1, where the use of emotional clauses was increased, the use of
quantitative adverbs would be increased as well. In VC 1.2 use of quantitative
adverbs had the score of 0,39, in VC 2 0,30 and in VC 3 0,33. The lowest percentage
is observed in ST (0,27).
Interrogative adverbs are not used in any case, neither in VCs nor in ST.
A rather low use of relative adverbs was also observed. In ST the score is 0,01
(lowest) and in VC 1.2 it is 0,03 (highest). In all other cases (VC 1.1, VC 2 and VC 3)
they are null.
Viewpoint and commenting adverbs are not used in ST. The same holds in VC
1.1. In VC 1.2 and in VC 3 the use of this type of adverbs is increased (0,03 and 0,05
respectively). The most impressive point is the high percentage in VC 2 (0,30).
It is therefore clear that students of VC 2 used this type of adverbs in order to
express their comments and views about their project. This happens, even to a
smaller extent, in all VCs where a problem-based project existed. In the non
problem-based project (VC 1.1) and also in ST, this type of adverbs is not used.
Case study Percentage of adverbs Number of adverb types Most frequent
VC1.1 0,12 3 quantitative
VC1.2 0,07 6 quantitative
VC2 0,06 4 Adverbs of place
VC3 0,06 5 quantitative
ST 0,06 5 Adverbs of time
In VC 1.1 adverbs had the most frequent use (0,12). This increased use results
from only three types of adverbs (adverbs of place, adverbs of manner and
quantitative adverbs). Quantitative adverbs were the most frequent (0,52), adverbs
64
of manner were used in a percentage of 0,28 and adverbs of place in a percentage of
0,16.
Far lower, VC 1.2 used adverbs in a percentage of 0,07. Despite this fact, the
types of the adverbs used were six. Students in this VC did not use adverbs of
certainty and interrogative adverbs (these types are not also used in any other case).
Once again, the most frequent type are the quantitative adverbs.
In VC 2, use of adverbs is approximately equal to this of VC 1.2. But in this case
the result comes by using four types. In this VC quantitative and viewpoint and
commenting adverbs are equally used (0,30). Adverbs of place reached
approximately the same percentage (0,27), while the proportion of adverbs of
manner was 0,07 (the lowest use of this type in all cases).
In VC 3, use of adverbs is equal to VC 2 (0,06). The students in this case used five
types of adverbs. It is clear that in one more case, quantitative adverbs were the
majority (0,35). Adverbs of manner and adverbs of place were also commonly used
(0,26 for both types).
Despite the fact that in ST use of adverbs is approximately the same as in the
most VCs (with the exception of VC 1.1), it is observed that the types were five. The
majority of them, in this case, are not the quantitative adverbs, but the adverbs of
time (0,34). It seems that time was not so important in VCs as it was in ST. Adverbs of
time play an important role in ST, because of their narrative nature. The difference in
this type is significantly high (0,34 in ST, 0,12 in VC 1.2, 0,09 in VC 3 - tables IS-10/1-
IS-10/4, pages 105-106). Additionally, it should be noted that viewpoint and
commenting adverbs are not used in ST, while in VCs the percentage vary from 0,03
to 0,30.
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
IS-11 Syntax a) Subordinate clauses
clause count
0,29 0,12 0,14 0,07 0,23
b) prepositional phrases
clause count
0,25 0,21 0,25 0,29 0,36
Use of subordinate clauses makes the language more precise, providing details
and additional information to the communication. In the case that these clauses are
missed, the language becomes elliptical, harsh and difficult to understand. Having in
mind that the members of a community develop a common code for mutual
understanding (Emerson, 1983), this additional information could be omitted.
65
It is clear that in both VC 1.1 and ST, subordinate clauses are used frequently (VC
1.1: 0,29 and ST: 0,23). It seems that VC 1.1 is influenced by the absence of the
"problem" in their task and the students felt more free to express their feelings and
impressions. The students, prompted by intention to express their emotions, used a
lot of subordinate clauses. High use of subordinate clauses is also observed in ST.
This might be considered as expected because the students in their essays are
addressing a general readership. Therefore their language has to be more formal and
more detailed in order to avoid misunderstandings (Mairesse et al, 2007). In
contrast, use of subordinate clauses in VCs is quite lower (VC 1.2: 0,12 - VC 2: 0,14).
The lowest use is in VC 3 (0,07) confirming the argument for the ellipticity of
language in communities (VC 3 was a priori a community). As VC 1.2 and VC 2 tend
to have similar low use of subordinate clauses, one more index of the community
existence in these VCs is added.
In addition to use of subordinate clauses, investigating the prepositional phrases
used shows clearly that there is lower use in VCs than in ST. In ST the score was 0,36,
while in VCs it reached 0,25 in VC 1.1, 0,21 in VC 1.2, 0,25 in VC 2 and 0,29 in VC 3. It
is then obvious that language in ST contains more details and it is richer in additional
terms, i.e. additional information, showing the ellipticity of the language in VCs.
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
IS-12 Article
types
a) Definite
Total articles
1,00 0,98 0,97 0,94 0,88
b) indefinite
Total articles
0 0,02 0,03 0,06 0,12
c) articles
word count
0,09 0,12 0,12 0,16 0,15
d) articles
periods
1,12 0,84 0,87 1,26 1,90
Definite articles constitute the majority in every case. Applying a more detailed
analysis, a difference between ST and VCs is revealed. The lowest score in VCs is
0,94, while the score in ST is 0,88. The difference increases if we compare ST to VC 2
(0,97), VC 1.2 (0,98) or VC 1.1 (1,00).
In VC 1.1 students were at the end of their project. The community was already
constructed then, while in the other VCs the discourse contains communication also
from the beginning of the project, when it would be impossible for a community to
exist. It is then expected that in VC 1.1 inner speech was developed at a higher
degree. Definite articles are referred to something already known indicating that the
66
language was more accurate and served the ability of the developed mutual
understanding among the class members (Emerson, 1983).
Counting the total articles per period is showing once more that there is a
difference between ST and VCs (ST: 1,90 - VC 3: 1,26 - VC 1.1: 1,12 - VC 2: 0,87 - VC
1.2: 0,84 - tables IS-12/1-IS-12/4, pages 110-111).
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
IS-13 Apposition Apposition
Clause count
0 0 0 0 0
IS-14 Epexegesis Epexegesis
Clause count
0,02 0 0,01 0,01 0,01
IS-15 Terms in
possessive case
possessive case
word count
0 0,01 0 0,02 0,03
IS-16 Terms in
accusative case
accusative case
word count
0 0 0 0 0,01
Apposition is rarely used and cannot add anything in the analysis. The same
happens with IS-14 (epexegesis).
Examining the additional terms in the possessive or the accusative case shows
that they are used more in ST, despite the fact that the difference with VCs is small.
Absence of these additional terms, in addition to the absence of apposition and
epexegesis, is confirming the elliptical language of the students in VCs (tables IS-
15/1-IS-15/4, page 113) and tables IS-16/1-IS-16/4, pages 113-114).
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
IS-17 Abbreviations Abbreviations
word count
0 0 0,02 0,04 0
IS-18 Metaphors Metaphors
clause count
0 0 0,04 0,12 0,04
IS-19 Similes Similes
clause count
0 0 0 0 0,02
Abbreviations indicate informal language (Mannes, 2008) and inner speech
(Vygotsky, 2008). It was therefore expected that they were used in VCs (VC 2 and VC
3) and they were not used in ST.
67
Metaphors could be considered as an indication of a common code for mutual
understanding (Daniel et al, 2003), as they do not always give a clear message (the
receiver has to "decode" it). This is confirmed by the score in VC 3 (0,12). Students of
this VC, which were a priori a community, do not hesitate to use metaphors, in
contrast to the other cases.
Similes, unlike metaphors, are used in order to add details and make the
meaning of the language more clear. Null use of similes in VCs indicates the mutual
understanding among the team members, i.e. it indicates lack of additional
information that could be considered as ellipticity by the non class members.
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 Average
IS-20 Word
variety
Variety
word count
0,34 0,39 0,36 0,38 0,37
Feature Metric ST 1 ST 2 ST 3 ST 4 ST 5 ST 6 ST 7 Average
IS-20 Word
variety
Variety
word count
0,47 0,50 0,41 0,47 0,55 0,48 0,65 0,50
Word variety in VCs varies from 0,34 to 0,39, with an average of 0,37. The
lowest score was in VC 1.1, and can be explained by the restricted frame of the
discourse (expressing their impressions). The highest score is in VC 1.2 (0,39), not
much higher than VC 3 (0,38) and VC 2 (0,36).
Examining the word variety in ST gives scores between 0,41 and 0,65, with an
average of 0,50.
Comparison of the two average scores confirms that the variety of the words
used in VCs is significantly lower than in ST (table IS-20/1, page 116). It is therefore
clear that the students' language in VCs appears to be restricted (using a narrower
vocabulary).
68
5.2 Collaborative learning analysis
5.2.1 Collaboration analysis in VC 1.1
Verbs in first person plural form
Students used fifty (50) verbs in this discourse. Thirty eight (38) of them (76%)
were in the first person plural form (we), giving a strong indication that the students
felt as members of a team.
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
38/50 0,76
Clause types
The total number of clauses used in this discourse were fifty two (52). Seventeen
(17) of them were expressing emotions and one (1) was giving reward. No clauses of
negation existed in VC 1.1. In contrast, a considerable number of clauses of reason
appeared (15).
CA-2 Emotional clauses Emotional clauses
clause count
17/52 0,33
CA-3 Rewarding clauses Rewarding clauses
clause count
1/52 0,02
CA-4 Clauses of negation Clauses of negation
clause count
0/52 0
CA-5 Clauses of reason Clauses of reason
clause count
15/52 0,29
Word types
There were 210 words used in VC 1.1. The majority of them were familiarity
words (70), while the emotional words were thirty seven (37). Twelve (12) social
words were also counted. It should be stressed that all emotional words were
expressing positive emotion.
69
CA-6 Familiarity words Familiarity words
word count
70/210 0,33
CA-7 Inclusive words Inclusive words
word count
2/210 0,01
CA-8 Social words Social words
word count
12/210 0,06
CA-9 Emotional words Emotional words
word count
37/210 0,18
CA-10 Positive emotion Positive emotion words
emotional words
37/37 1,00
CA-11 Negative emotion Negative emotion words
emotional words
0/37 0
Pronouns in first person plural
The total number of pronouns used in this discourse were thirty (30). Eleven (11)
of them were possessive and eighteen (18) were personal pronouns. Ten (10) of
these twenty nine (29) (personal + possessive) pronouns were in the first person
plural form.
CA-12 Use of first plural
person pronouns
Us, ours
pronouns
10/30 0,33
CA-13 Us, ours
Possessive + personal pronouns
10/29 0,34
5.2.2 Collaboration analysis in VC 1.2
Verbs in first person plural form
In VC 1.2 a significant percentage (43%) of the verbs used were in the first
person plural form (we).
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
36/84 0,43
Clause types
A remarkable number of emotional clauses was used in VC 1.2. Rewarding
sentences were at a rather high percentage as well. Clauses of negation were three
70
(3) out of a total of 105 clauses, giving an index of positive attitude. Clauses of
reason were nearly null (2/105).
CA-2 Emotional clauses Emotional clauses
clause count
22/105 0,21
CA-3 Rewarding clauses Rewarding clauses
clause count
17/105 0,16
CA-4 Clauses of negation Clauses of negation
clause count
3/105 0,03
CA-5 Clauses of reason Clauses of reason
clause count
2/105 0,02
Word types
Emotional words (47) constitute the majority of the word types used in this
discourse. Familiarity words were 24. Despite the fact that emotional words were
divided approximately in half (positive - negative), positive emotional words were
the majority (55%).
CA-6 Familiarity words Familiarity words
word count
24/453 0,05
CA-7 Inclusive words Inclusive words
word count
0/453 0
CA-8 Social words Social words
word count
1/453 0
CA-9 Emotional words Emotional words
word count
47/453 0,10
CA-10 Positive emotion Positive emotion words
emotional words
26/47 0,55
CA-11 Negative emotion Negative emotion words
emotional words
21/47 0,45
Pronouns in first plural person
The total number of pronouns used were sixteen (16). Eleven (11) of them were
possessive and four (4) were personal pronouns. Seven (7) out of the latter fifteen
(15) (personal and possessive) were in the first person plural, indicating the feeling
of belonging to a team.
71
CA-12 Use of first plural
person pronouns
Us, ours
pronouns
7/16 0,44
CA-13 Us, ours
Possessive + personal pronouns
7/15 0,47
5.2.3 Collaboration analysis in VC 2
Verbs in first plural person form
In VC 2, almost one third (1/3) of the total verbs were used in the first person
plural form.
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
24/78 0,31
Clause types
Emotional and rewarding clauses were the most commonly used types of
clauses in VC 2. Specifically, emotional clauses constituted almost one fourth of the
total clauses.
CA-2 Emotional clauses Emotional clauses
clause count
22/101 0,22
CA-3 Rewarding clauses Rewarding clauses
clause count
16/101 0,16
CA-4 Clauses of negation Clauses of negation
clause count
7/101 0,07
CA-5 Clauses of reason Clauses of reason
clause count
1/101 0,01
Word types
Emotional words were the most frequent (34). Almost all of these words (32/34)
expressed positive emotions.
72
CA-6 Familiarity words Familiarity words
word count
10/466 0,02
CA-7 Inclusive words Inclusive words
word count
0/466 0
CA-8 Social words Social words
word count
3/466 0,01
CA-9 Emotional words Emotional words
word count
34/466 0,07
CA-10 Positive emotion Positive emotion words
emotional words
32/34 0,94
CA-11 Negative emotion Negative emotion words
emotional words
2/34 0,06
Pronouns in first plural person
It is undeniable that the majority of the pronouns used were in the first person
plural form.
CA-12 Use of first plural
person pronouns
Us, ours
pronouns
28/41 0,68
CA-13 Us, ours
Possessive + personal pronouns
28/40 0,70
5.2.4 Collaboration analysis in VC 3
Verbs in first plural person
It is impressive that in VC 3 the use of verbs in the first plural person is very low
(3%).
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
4/125 0,03
Clause types
In a total of 147 clauses, emotional clauses were fourteen (14). Clauses of
negation and rewarding clauses were equally used (9), while there was one (1)
clause of reason.
73
CA-2 Emotional clauses Emotional clauses
clause count
14/147 0,10
CA-3 Rewarding clauses Rewarding clauses
clause count
9/147 0,06
CA-4 Clauses of negation Clauses of negation
clause count
9/147 0,06
CA-5 Clauses of reason Clauses of reason
clause count
1/147 0,01
Word types
Familiarity (29) and emotional words (28) were the most common word types. It
is obvious that, in VC 3 the positive emotional words constitute the majority of the
emotional words, which indicates a positive attitude to the team/class (Mairesse et
al, 2007; Mc Millan and Chavis, 1986).
CA-6 Familiarity words Familiarity words
word count
29/704 0,04
CA-7 Inclusive words Inclusive words
word count
1/704 0
CA-8 Social words Social words
word count
15/704 0,02
CA-9 Emotional words Emotional words
word count
28/704 0,04
CA-10 Positive emotion Positive emotion words
emotional words
19/28 0,68
CA-11 Negative emotion Negative emotion words
emotional words
9/25 0,32
Pronouns in first plural person
No pronouns in the first person plural were used. As we can note, this feature
appears to correlate with CA-1 (Verbs in first person plural).
CA-12 Use of first plural
person pronouns
Us, ours
pronouns
0/32 0
CA-13 Us, ours
Possessive + personal pronouns
0/12 0
74
5.2.5 Collaboration analysis in Students' Texts (ST)
Verbs in first plural person form
Use of verbs in the first person plural form was rather low in ST (only 103 verbs
in a total of 663 used).
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
103/663 0,16
Clause types
Emotional clauses in ST were very few (33 in a total of 666 clauses). Rewarding
clauses were null. Clauses of negation and clauses of reason were minimally used.
CA-2 Emotional clauses Emotional clauses
clause count
33/666 0,05
CA-3 Rewarding clauses Rewarding clauses
clause count
0/666 0
CA-4 Clauses of negation Clauses of negation
clause count
36/666 0,05
CA-5 Clauses of reason Clauses of reason
clause count
11/666 0,02
Word types
Familiarity, inclusive and social words were null. Emotional words were rarely
used. It is clear that the positive emotional words outweigh the negative.
CA-6 Familiarity words Familiarity words
word count
0/3577 0
CA-7 Inclusive words Inclusive words
word count
5/3577 0
CA-8 Social words Social words
word count
12/3577 0
CA-9 Emotional words Emotional words
word count
82/3577 0,02
CA-10 Positive emotion Positive emotion words
emotional words
53/82 0,65
CA-11 Negative emotion Negative emotion words
emotional words
29/82 0,35
75
Pronouns in first plural person
Use of pronouns in the first person plural was low. There were just eighteen (18)
out of a total of 179 (personal and possessive pronouns).
CA-12 Use of first plural
person pronouns
Us, ours
pronouns
18/237 0,08
CA-13 Us, ours
Possessive + personal pronouns
18/179 0,10
5.2.6 Comparative results - Discussion
Verbs in first person plural
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
CA-1 Verbs in
first plural
person
Verbs (we)
Verb count
0,76 0,43 0,31 0,03 0,16
There is no doubt that VC 1.1 presents the strongest indication of team working.
This can be explained if one takes into account that students were at the end of their
project, when the ties among them were stronger than in the beginning when they
did not even know each other.
This feature is considerably higher in VC 1.2 than in VC 2, possibly due to the
different level of education of the two collaborated subgroups. According to
Vygotsky (1978), zone of the proximal development offers students a chance of
learning. In VC 1.2, where one of the two collaborated subgroups was composed of
elementary school students and the second one of high school students, it would be
expected that there was a challenge for the elementary students for increased
participation and collaboration, i.e. a challenge for learning.
But even in the case of subgroups being in the same level of education (VC 2),
feature CA-1 is almost the double compared to the ST.
An impressive point is the very low index in VC 3. Two points should be
reminded:
a) In this case a physical class was transformed into a virtual one and students
collaborated distantly through the internet and
b) teaching was directed by the teachers who had an active instructive role.
76
Τhe active role of the teachers might influence students and prevent them from
using the first person plural (we), due to the existing distance between them and the
teachers. Another explanation could be offered if we consider that when the
teachers were giving instructions, they did it usually in the second person plural
(you), as they did not participate in the completion of the problem solving task.
These instructions given by the teachers were very common, according to their
instructive role.
Nevertheless, it is undoubted that, if we take into account the results in VC 1.1
and VC 1.2, which substantially constitute one class (VC 1), students appear to
behave as a community.
Clause types
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
CA-2 Emotional clauses Emotional clauses
clause count
0,33 0,21 0,22 0,10 0,05
CA-3 Rewarding clauses Rewarding clauses
clause count
0,02 0,16 0,16 0,06 0
CA-4 Clauses of negation Clauses of negation
clause count
0 0,03 0,07 0,06 0,05
CA-5 Clauses of reason Clauses of reason
clause count
0,29 0,02 0,01 0,01 0,02
Emotional clauses were much higher in VC 1.1 than in the other schemata. This
is not strange, because, in this case students are requested to express their feelings
about the already completed project. But, as we can see, emotional clauses reach a
high percentage in VC 1.2 and VC 2 as well, showing that the students felt
comfortable to express their feelings even during their effort to "solve" the problem
based task. Considering that the subgroups were unfamiliar with each other in the
beginning of the project and it was therefore difficult for them to express their
emotions, this result is more significant than it seems.
It is also remarkable that VC 1.1 and VC 1.2 have almost the same result (0,21-
0,22).
Attempting to explain the low percentage in VC 3, one should focus on the
formality of the procedure given by the active role of the teachers (Bielaczyc and
Collins, 1999; Maness, 2008). Another factor might be the existing close relations
among the students, which had already been working as a team for seven years
(from kindergarten till the 6th grade). VC 3 was also influenced by the existing daily
communication among the students in their physical class. It is then expected that a
77
student under these circumstances, does not greatly feel the need of expressing his
emotions. But whenever he feels that need, he is able to express them in person.
This was impossible in the other VCs where face-to-face communication was
impossible.
In each case, emotional clauses are clearly higher in VCs than in ST. Even in VC 3,
they are double in number.
Rewarding clauses reached the same percentage in VC 1.2 and VC 2 indicating
that the students felt the need to reward their partners for their effort.
Subsequently, this fact reveals the existence of strong relations among them, i.e. the
existence of a community.
In VC 1.1 rewarding clauses were expected to be low, as the discourse was
developed after the completion of the project. But even in this case, rewarding
clauses were also used.
In VC 3 rewarding clauses were low, possibly due to the possibility of students to
express their rewards during their daily interaction in the physical classroom.
Nevertheless, all VCs used rewarding clauses at an obvious higher level than in ST
(tables CA-3/1-CA-3/4, pages 120-121).
Clauses of negation had a rather similar use in VC 2 and VC 3. In ST they were
approximately the same in number, while in VC 1.2 they were about half as many.
It is impressive that in VC 1.1 Clauses of negation were not used, clearly
indicating positive emotion, a basic feature within a community.
Clauses of reason were extremely high in VC 1.1, where the students felt the
need to give reasons for their impressions and feelings expressed in the discourse. In
all other cases (VCs and ST), this type of clauses were almost equally used.
78
Word types
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
CA-6 Familiarity
words
Familiarity words
word count
0,33 0,05 0,02 0,04 0
CA-7 Inclusive
words
Inclusive words
word count
0,01 0 0 0 0
CA-8 Social
words
Social words
word count
0,06 0 0,01 0,02 0
CA-9 Emotional
words
Emotional words
word count
0,18 0,10 0,07 0,04 0,02
CA-10 Positive
emotion
Positive emotion words
emotional words
1,00 0,55 0,94 0,68 0,65
CA-11 Negative
emotion
Negative emotion words
emotional words
0 0,45 0,06 0,32 0,35
There was a great difference in the use of familiarity words. One third of the
total clauses in VC 1.1 were familiarity ones, giving a strong indication for community
existence. This indication is more significant because it appears at the end of the
project.
In VC 1.2 the percentage of familiarity words was about the same as in VC 3. But
one should keep in mind that VC 3 was already a community, as students have been
collaborating for seven years (from kindergarten till the 6th grade). Taking this into
account, index in VC 1.2 seems greater. VC 2 gives a lower index in this feature.
In every case, analyzing ST shows that use of familiarity words should not be
considered as a typical feature of students' language.
It is difficult to consider inclusive words as a criterion for collaboration, because
of their sparse use.
Social words are not commonly used. Despite this, VC 1.1 once again uses them
in the highest percentage, empowering the indication of the community existence.
Even in VC 1.2 and VC 3, where social words are rarely used, their average is higher
than in ST.
Use of emotional words emphasizes the results given by the previous features
and confirms the evidence of the existence of a community. Almost one in every five
79
words used by the students in VC 1.1 was expressing emotions. This could be
considered as an expected feature of their language, as they were asked to express
their impressions on the completed project. But we should notice that the second
most frequent use of emotional words is counted in VC 1.2 (the other subgroup of
VC 1). That means that these students had the need to express their emotions both
during their effort to solve the problem based project, and also after its completion .
Use of emotional words was rather high also in VC 2. In VC 3, these words were
more rarely used. But one should keep in mind that the students in VC 3 had the
chance to express their emotions in person during their everyday school life. But
even in this case, emotional words were double compared to ST.
Examining whether the emotions that appeared in the discourse were positive
or negative, we can see that 100% of the emotions in VC 1.1 were positive, while in
VC 1.2 they were almost equal (nevertheless positive were the majority). In all VCs
and also in ST positive emotions are always the majority. An impressive point is the
percentage of positive emotion in VC 3 (94%), confirming the existence of the
community.
Pronouns in first person plural
Feature Metric VC 1.1 VC 1.2 VC 2 VC 3 ST
CA-12 Use of first
person
plural
pronouns
Us, ours
pronouns
0,33 0,44 0,68 0 0,08
CA-13 Us, ours
Possessive + personal
pronouns
0,34 0,47 0,70 0 0,10
In all VCs the pronouns in the first person plural form are widely used. In VC 2
there is a remarkable percentage of 70%. Also in VC 1.2 these pronouns are about
half of the total pronouns used, consisting a significant percentage, while in VC 1.1
one third of the pronouns are in the first person plural form. As we can see, there is
a great difference between VCs and ST, where these pronouns, that indicate the
feeling of belonging in a team, reach only 10% of the total number of pronouns
(tables CA-12/1-CA-12/3, pages 132-133).
An exception to these results is VC 3, where these pronouns are null. Having in
mind that VC 3 was a priori a community, it was expected that these students would
be focused on their task, concentrated on their teachers' directions, without being
80
concerned to build relationships among them, as these already existed (table CA-
12/4, page 133).
81
5.3 Statistical analysis
Attempting to investigate whether there were statistically significant differences between the results of the analysis, a two-tailed independent samples t-test was applied. This test is considered to be appropriate for this study, as the samples in every case study were not the same and the population of them was unequal. The choice of a two-tailed test was made, because no attempt was made for any prediction about the result of the comparison (Roussos and Tsaousis, 2006).
This test was applied for each feature of the analysis, except of those that
presented no difference between the case studies. In every case it was checked whether there were significant differences both between the VCs and the ST and also between the VCs. The null hypothesis was the same in every comparison:
Ho: the metrics of the feature are not different H1: the metrics of the feature are different
Whenever the result of the t-test is greater than the critical value the null
hypothesis is rejected and the H1 hypothesis is accepted. If the result is less than the
critical value, the null hypothesis is accepted (Roussos and Tsaousis, 2006).
82
6. Conclusion
Results of the analysis show that the VCs examined in this study were transformed into communities, providing students with the benefits of the community membership. In VC3, which was a priori a community, community existence was confirmed.
Comparison between VCs and ST revealed statistical significant differences in
the language used. Conversations in VCs was mainly informal, elliptical in meaning
and abbreviated, i.e. it had the basic features of inner speech.
Collaborative learning analysis revealed as well that students of these VCs
collaborated enough and had the membership feeling.
Active instructive role of the teachers affects the language and makes it more
formal.
There are differences in the language use between problem-based and non-
problem based projects.
The existence of a common code and the mutual understanding in communities
was confirmed.
Existence of emotion among community members and their positive attitude
was confirmed as well.
Combining the results of the two models of analysis lead to the result that the
VCs, being analyzed in this study, were transformed into communities.
It is therefore suggested that the proposed model for linguistic analysis in the
discourse among the members of a VC can provide useful results. These results can
be used in both ways:
1) evaluate a VC by detecting whether it was transformed into a community or
not, and
2) improve the design of VCs
Each VC, during its transformation into a community, will have to "pass" through two axes: inner speech and collaborative learning. These two axes are not always
distinct from each other, but they are complementary and have common features, i.e. there is a common ground between them. As a result, a clear distinction of these axes is not always possible. Nevertheless, a VC has to immerse within inner speech and collaborative learning in order to transform into a community (Figure 1).
83
Figure 1. Transformation of a VC into a community
Applying linguistic analysis in the discourse among the members of a VC can provide us with useful results. Combining the result of the two categories (inner speech and collaboration) we can get strong indications for the community existence. Furthermore, results of the analysis can help us to improve the design of the VCs.
However there is a field for future research, e.g. applying this model and evaluating it on a larger corpus and different case studies.
Virtual Class
Community
Collaborative
learning Inner Speech
84
Appendix
Statistical analysis (Inner Speech)
IS-1 Omission of subjects
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-1 Omission of
Subjects
No subject
Verb count
0,88 0,67
t(18)=2,35, p<,05
Table IS-1/1
Feature Metric VC 1.2 ST
IS-1 Omission of
Subjects
No subject
Verb count
0,92 0,67
t(16)=4,92, p<,01
Table IS-1/2
Feature Metric VC 2 ST
IS-1 Omission of
Subjects
No subject
Verb count
0,85 0,67
t(20)=2,42, p<,05
Table IS-1/3
Feature Metric VC 3 ST
IS-1 Omission of
Subjects
No subject
Verb count
0,51 0,67
t(10)=-0,96, n.s.
Table IS-1/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-1 Omission of
Subjects
No subject
Verb count
0,88 0,92
t(19)=-0,49, n.s.
Table IS-1/5
85
Feature Metric VC 1.1 VC 2
IS-1 Omission of
Subjects
No subject
Verb count
0,88 0,85
t(28)=0,12, n.s.
Table IS-1/6
Feature Metric VC 1.1 VC 3
IS-1 Omission of
Subjects
No subject
Verb count
0,88 0,51
t(17)=2,41, p<,05
Table IS-1/7
Feature Metric VC 1.2 VC 2
IS-1 Omission of
Subjects
No subject
Verb count
0,92 0,85
t(22)=0,70, n.s.
Table IS-1/8
Feature Metric VC 1.2 VC 3
IS-1 Omission of
Subjects
No subject
Verb count
0,92 0,51
t(10)=3,31, p<,01
Table IS-1/9
Feature Metric VC 2 VC 3
IS-1 Omission of
Subjects
No subject
Verb count
0,85 0,51
t(16)=2,40, p<,05
Table IS-1/10
86
IS-2 Omission of Conjunction
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-2 Omission of
Conjunction
No conjunction
clause count
0
0,03
t(6)=-2,69, p<,05
Table IS-2/1
Feature Metric VC 1.2 ST
IS-2 Omission of
Conjunction
No conjunction
clause count
0,02
0,03
t(15)=-1,03, n.s.
Table IS-2/2
Feature Metric VC 2 ST
IS-2 Omission of
Conjunction
No conjunction
clause count
0,12
0,03
t(19)=2,06, n.s.
Table IS-2/3
Feature Metric VC 3 ST
IS-2 Omission of
Conjunction
No conjunction
clause count
0,14 0,03
t(11)=1,57, n.s.
Table IS-2/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-2 Omission of
Conjunction
No conjunction
clause count
0 0,02
t(13)=-1,47, n.s.
Table IS-2/5
87
Feature Metric VC 1.1 VC 2
IS-2 Omission of
Conjunction
No conjunction
clause count
0 0,12
t(16)=-3,08, p<,01
Table IS-2/6
Feature Metric VC 1.1 VC 3
IS-2 Omission of
Conjunction
No conjunction
clause count
0 0,14
t(8)=-2,97, p<,02
Table IS-2/7
Feature Metric VC 1.2 VC 2
IS-2 Omission of
Conjunction
No conjunction
clause count
0,02
0,12
t(19)=-2,53, p<,05
Table IS-2/8
Feature Metric VC 1.2 VC 3
IS-2 Omission of
Conjunction
No conjunction
clause count
0,02
0,14
t(11)=-2,19, n.s.
Table IS-2/9
Feature Metric VC 2 VC 3
IS-2 Omission of
Conjunction
No conjunction
clause count
0,12
0,14
t(24)=0,75, n.s.
Table IS-2/10
88
IS-3 Informal clauses
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-3 Informal
clauses
Informal clauses
clause count
0,23 0
t(14)=-2,78, p<,02
Table IS-3/1
Feature Metric VC 1.2 ST
IS-3 Informal
clauses
Informal clauses
clause count
0,75 0
t(13)=7,26, p<,01
Table IS-3/2
Feature Metric VC 2 ST
IS-3 Informal
clauses
Informal clauses
clause count
0,58 0
t(16)=5,24, p<,01
Table IS-3/3
Feature Metric VC 3 ST
IS-3 Informal
clauses
Informal clauses
clause count
0,69 0
t(8)=7,16, p<,01
Table IS-3/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-3 Informal
clauses
Informal clauses
clause count
0,23 0,75
t(27)=-2,55, p<,02
Table IS-3/5
89
Feature Metric VC 1.1 VC 2
IS-3 Informal
clauses
Informal clauses
clause count
0,23 0,58
t(27)=-1,05, n.s.
Table IS-3/6
Feature Metric VC 1.1 VC 3
IS-3 Informal
clauses
Informal clauses
clause count
0,23 0,69
t(22)=-2,25, p<,05
Table IS-3/7
Feature Metric VC 1.2 VC 2
IS-3 Informal
clauses
Informal clauses
clause count
0,75 0,58
t(28)=1,74, n.s.
Table IS-3/8
Feature Metric VC 1.2 VC 3
IS-3 Informal
clauses
Informal clauses
clause count
0,75 0,69
t(20)=0,41, n.s.
Table IS-3/9
Feature Metric VC 2 VC 3
IS-3 Informal
clauses
Informal clauses
clause count
0,58 0,69
t(21)=-1,38, n.s.
Table IS-3/10
90
IS-4 Omission of verbs
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-4 Omission of
verbs
No verb
clause count
0,04 0,01
t(16)=0,96, n.s.
Table IS-4/1
Feature Metric VC 1.2 ST
IS-4 Omission of
verbs
No verb
clause count
0,21 0,01
t(14)=4,01, p<,01
Table IS-4/2
Feature Metric VC 2 ST
IS-4 Omission of
verbs
No verb
clause count
0,25 0,01
t(17)=3,39, p<,051
Table IS-4/3
Feature Metric VC 3 ST
IS-4 Omission of
verbs
No verb
clause count
0,14 0,01
t(8)=2,60, p<,05
Table IS-4/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-4 Omission of
verbs
No verb
clause count
0,04 0,21
t(24)=-2,71, p<,02
Table IS-4/5
91
Feature Metric VC 1.1 VC 2
IS-4 Omission of
verbs
No verb
clause count
0,04 0,25
t(24)=-2,56, p<,02
Table IS-4/6
Feature Metric VC 1.1 VC 3
IS-4 Omission of
verbs
No verb
clause count
0,04 0,14
t(14)=-1,70, n.s.
Table IS-4/7
Feature Metric VC 1.2 VC 2
IS-4 Omission of
verbs
No verb
clause count
0,21 0,25
t(28)=-0,38, n.s.
Table IS-4/8
Feature Metric VC 1.2 VC 3
IS-4 Omission of
verbs
No verb
clause count
0,21 0,14
t(18)=0,68, n.s.
Table IS-4/9
Feature Metric VC 2 VC 3
IS-4 Omission of
verbs
No verb
clause count
0,25 0,14
t(23)=0,93, n.s.
Table IS-4/10
92
IS-5 Elliptical clauses
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,12 0,04
t(15)=1,25, n.s.
Table IS-5/1
Feature Metric VC 1.2 ST
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,57 0,04
t(14)=6,77, p<,01
Table IS-5/2
Feature Metric VC 2 ST
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,34 0,04
t(17)=3,46, p<,01
Table IS-5/3
Feature Metric VC 3 ST
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,61 0,04
t(9)=10,70, p<,01
Table IS-5/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,12 0,57
t(27)=-3,86, p<,01
Table IS-5/5
93
Feature Metric VC 1.1 VC 2
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,12 0,34
t(30)=-1,66, n.s.
Table IS-5/6
Feature Metric VC 1.1 VC 3
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,12 0,61
t(22)=-5,08, p<,01
Table IS-5/7
Feature Metric VC 1.2 VC 2
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,57 0,34
t(29)=2,08, p<,05
Table IS-5/8
Feature Metric VC 1.2 VC 3
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,57 0,61
t(21)=-0,63,n.s.
Table IS-5/9
Feature Metric VC 2 VC 3
IS-5 Elliptical
clauses
Elliptical clauses
clause count
0,34 0,61
t(24)=-2,98, p<,01
Table IS-5/10
94
IS-6 Words per clause
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-6 Words per
clause
word count
clause count
4,04 5,37
t(20)=-4,16, p<,01
Table IS-6/1
Feature Metric VC 1.2 ST
IS-6 Words per
clause
word count
clause count
4,27 5,37
t(18)=-3,46, p<,01
Table IS-6/2
Feature Metric VC 2 ST
IS-6 Words per
clause
word count
clause count
4,62 5,37
t(17)=-3,22, p<,01
Table IS-6/3
Feature Metric VC 3 ST
IS-6 Words per
clause
word count
clause count
4,79 5,37
t(13)=-1,49, n.s.
Table IS-6/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-6 Words per
clause
word count
clause count
4,04 4,27
t(27)=-0,82, n.s.
Table IS-6/5
95
Feature Metric VC 1.1 VC 2
IS-6 Words per
clause
word count
clause count
4,04 4,62
t(27)=-1,40, n.s.
Table IS-6/6
Feature Metric VC 1.1 VC 3
IS-6 Words per
clause
word count
clause count
4,04 4,79
t(19)=-2,14, p<,05
Table IS-6/7
Feature Metric VC 1.2 VC 2
IS-6 Words per
clause
word count
clause count
4,27 4,62
t(27)=-0,55, n.s.
Table IS-6/8
Feature Metric VC 1.2 VC 3
IS-6 Words per
clause
word count
clause count
4,27 4,79
t(27)=-1,45, n.s.
Table IS-6/9
Feature Metric VC 2 VC 3
IS-6 Words per
clause
word count
clause count
4,62 4,79
t(15)=-1,07, n.s.
Table IS-6/10
96
IS-7 Words per period
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-7 Words per
period
word count
period count
12,35 12,82
t(17)=-0,14, n.s.
Table IS-7/1
Feature Metric VC 1.2 ST
IS-7 Words per
period
word count
period count
7,08 12,82
t(19)=-3,05, p<,01
Table IS-7/2
Feature Metric VC 2 ST
IS-7 Words per
period
word count
period count
7,03 12,82
t(13)=-6,19, p<,01
Table IS-7/3
Feature Metric VC 3 ST
IS-7 Words per
period
word count
period count
8,09 12,82
t(14)=-4,71, p<,01
Table IS-7/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-7 Words per
period
word count
period count
12,35 7,08
t(23)=1,73, n.s.
Table IS-7/5
97
Feature Metric VC 1.1 VC 2
IS-7 Words per
period
word count
period count
12,35 7,03
t(16)=2,69, p<,02
Table IS-7/6
Feature Metric VC 1.1 VC 3
IS-7 Words per
period
word count
period count
12,35 8,09
t(17)=2,19, p<,05
Table IS-7/7
Feature Metric VC 1.2 VC 2
IS-7 Words per
period
word count
period count
7,08 7,03
t(19)=1,13, n.s.
Table IS-7/8
Feature Metric VC 1.2 VC 3
IS-7 Words per
period
word count
period count
7,08 8,09
t(20)=0,41, n.s.
Table IS-7/9
Feature Metric VC 2 VC 3
IS-7 Words per
period
word count
period count
7,03 8,09
t(18)=-1,00, n.s.
Table IS-7/10
98
IS-8 Punctuation marks
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-8 Punctuation
marks
commas
clause count
0,02 0,18
t(8)=-5,87, p<,01
question marks
clause count
0,02 0
t(14)=1, n.s.
dots
clause count
0,02 0
t(15)=0,63, n.s.
exclamation marks
clause count
0,52 0,02
t(14)=1,33, n.s.
full stops
clause count
0,13 0,40
t(13)=-3,61, p<,01
punctuation (total)
clause count
0,71 0,61
t(14)=0,69, n.s.
Table IS-8/1
Feature Metric VC 1.2 ST
IS-8 Punctuation
marks
commas
clause count
0,08 0,18
t(13)=-3,47, p<,01
exclamation marks
clause count
0,28 0,02
t(13)=-3,97, p<,01
full stops
clause count
0,26 0,40
t(17)=-2,21, p<,05
punctuation (total)
clause count
0,68 0,60
t(16)=0,37, n.s.
Table IS-8/2
99
Feature Metric VC 2 ST
IS-8 Punctuation
marks
parenthesis
clause count
0,03 0
t(18)=0,33, n.s.
commas
clause count
0,02 0,18
t(7)=-6,31, p<,01
question marks
clause count
0,20 0
t(15)=2,02, n.s.
exclamation marks
clause count
0,15 0,02
t(16)=1,72, n.s.
full stops
clause count
0,24 0,40
t(15)=-3,56, p<,01
punctuation (total)
clause count
0,65 0,61
t(21)=0,62, n.s.
Table IS-8/3
Feature Metric VC 3 ST
IS-8 Punctuation
marks
commas
clause count
0,14 0,18
t(14)=-2,59, p<,05
question marks
clause count
0,07 0
t(8)=2,75, p<,05
exclamation marks
clause count
0,06 0,02
t(8)=1,81, n.s.
full stops
clause count
0,25 0,40
t(14)=-2,10, n.s.
punctuation (total)
clause count
0,54 0,61
t(12)=-0,77, n.s.
Table IS-8/4
100
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-8 Punctuation
marks
commas
clause count
0,02 0,08
t(20)=-2,40, p<,05
exclamation marks
clause count
0,52 0,28
t(14)=0,87, n.s.
full stops
clause count
0,13 0,26
t(25)=-1,57, n.s.
punctuation (total)
clause count
0,71 0,68
t(14)=0,57, n.s.
Table IS-8/5
Feature Metric VC 1.1 VC 2
IS-8 Punctuation
marks
question marks
clause count
0,02 0,20
t(25)=-1,26, n.s.
exclamation marks
clause count
0,52 0,15
t(16)=0,87, n.s.
full stops
clause count
0,13 0,24
t(28)=-0,77, n.s.
punctuation (total)
clause count
0,71 0,65
t(16)=0,48, n.s.
Table IS-8/6
101
Feature Metric VC 1.1 VC 3
IS-8 Punctuation
marks
commas
clause count
0,02 0,14
t(10)=-2,72, p<,05
question marks
clause count
0,02 0,07
t(18)=-0,10, n.s.
exclamation marks
clause count
0,52 0,06
t(14)=1,14, n.s.
full stops
clause count
0,13 0,25
t(20)=-1,69, n.s.
punctuation (total)
clause count
0,71 0,54
t(14)=0,82, n.s.
Table IS-8/7
Feature Metric VC 1.2 VC 2
IS-8 Punctuation
marks
commas
clause count
0,08 0,02
t(15)=2,81, p<,02
question marks
clause count
0,03 0,20
t(17)=-1,72, n.s.
exclamation marks
clause count
0,28 0,15
t(22)=0,06, n.s.
Table IS-8/8
102
Feature Metric VC 1.2 VC 3
IS-8 Punctuation
marks
commas
clause count
0,08 0,14
t(15)=-0,43, n.s.
question marks
clause count
0,03 0,07
t(15)=-1,28, n.s.
exclamation marks
clause count
0,28 0,06
t(15)=1,80, n.s.
Table IS-8/9
Feature Metric VC 2 VC 3
IS-8 Punctuation
marks
commas
clause count
0,02 0,14
t(9)=-2,53, p<,05
question marks
clause count
0,20 0,07
t(18)=1,35, n.s.
exclamation marks
clause count
0,15 0,06
t(20)=0,94, n.s.
Table IS-8/10
IS-9 Word types
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-9 Word types adverbs
word count
0,12 0,06
t(16)=2,10, n.s.
adjectives
word count
0,01 0,09
t(7)=-4,20, p<,01
informal
word count
0,02 0
t(14)=0,99, n.s.
Table IS-9/1
103
Feature Metric VC 1.2 ST
IS-9 Word types adverbs
word count
0,07 0,06
t(16)=1,50, n.s.
adjectives
word count
0,07 0,09
t(14)=-1,23, n.s.
informal
word count
0,06 0
t(13)=4,34, p<,01
Table IS-9/2
Feature Metric VC 2 ST
IS-9 Word types adjectives
word count
0,04 0,09
t(10)=-2,83, p<,02
informal
word count
0,07 0
t(16)=2,98, p<,01
Table IS-9/3
Feature Metric VC 3 ST
IS-9 Word types informal
word count
0,08 0
t(8)=3,02, p<,02
Table IS-9/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-9 Word types adverbs
word count
0,12 0,07
t(24)=0,89, n.s.
adjectives
word count
0,01 0,07
t(20)=-3,93, p<,01
informal
word count
0,02 0,06
t(15)=0,70, n.s.
Table IS-9/5
104
Feature Metric VC 1.1 VC 2
IS-9 Word types adverbs
word count
0,12 0,06
t(22)=1,52, n.s.
adjectives
word count
0,01 0,04
t(29)=-2,82, p<,01
greeklish
word count
0 0,13
t(16)=-2,45, p<,05
informal
word count
0,02 0,07
t(18)=-0,11, n.s.
Table IS-9/6
Feature Metric VC 1.1 VC 3
IS-9 Word types adverbs
word count
0,12 0,06
t(20)=1,44, n.s.
adjectives
word count
0,01 0,08
t(12)=-3,15, p<,01
informal
word count
0,02 0,08
t(18)=-0,23, n.s.
Table IS-9/7
Feature Metric VC 1.2 VC 2
IS-9 Word types adjectives
word count
0,07 0,04
t(24)=1,62, n.s.
greeklish
word count
0 0,13
t(16)=-2,39, p<,05
Table IS-9/8
105
Feature Metric VC 2 VC 3
IS-9 Word types adjectives
word count
0,04 0,08
t(14)=-1,13, n.s.
greeklish
word count
0,13 0,01
t(16)=2,41, p<,05
Table IS-9/9
IS-10 Adverb types
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-10 Word
types
Adverbs of time
Total adverbs
0 0,34
t(6)=-6,67, p<,01
Adverbs of manner
Total adverbs
0,28 0,23
t(12)=-0,10, n.s.
quantitative adverbs
Total adverbs
0,56 0,27
t(11)=3,05, p<,02
Table IS-10/1
Feature Metric VC 1.2 ST
IS-10 Word
types
Adverbs of time
Total adverbs
0,12 0,34
t(15)=-0,89, n.s.
quantitative adverbs
Total adverbs
0,39 0,27
t(12)=0,91, n.s.
relative adverbs
Total adverbs
0,03 0,01
t(11)=0,81, n.s.
Viewpoint and commenting adverbs
Total adverbs
0,03 0
t(11)=1,00, n.s.
Table IS-10/2
106
Feature Metric VC 2 ST
IS-10 Word
types
Adverbs of place
Total adverbs
0,33 0,16
t(13)=2,00, n.s.
Adverbs of time
Total adverbs
0 0,34
t(6)=-6,67, p<,01
Adverbs of manner
Total adverbs
0,07 0,23
t(16)=-4,34, p<,01
quantitative adverbs
Total adverbs
0,30 0,27
t(12)=0,45, n.s.
Viewpoint and commenting adverbs
Total adverbs
0,30 0
t(11)=3,07, p<,02
Table IS-10/3
Feature Metric VC 3 ST
IS-10 Word
types
Adverbs of place
Total adverbs
0,26 0,16
t(9)=0,18, n.s.
Adverbs of time
Total adverbs
0,09 0,34
t(11)=-4,40, p<,01
quantitative adverbs
Total adverbs
0,33 0,27
t(5)=1,46, n.s.
Viewpoint and commenting adverbs
Total adverbs
0,05 0
t(7)=1,00, n.s.
Table IS-10/4
107
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-10 Word
types
Adverbs of time
Total adverbs
0 0,12
t(11)=-1,90, n.s.
quantitative adverbs
Total adverbs
0,56 0,39
t(21)=1,51, n.s.
Table IS-10/5
Feature Metric VC 1.1 VC 2
IS-10 Word
types
Adverbs of manner
Total adverbs
0,28 0,07
t(12)=1,75, n.s.
quantitative adverbs
Total adverbs
0,56 0,30
t(20)=1,56, n.s.
Viewpoint and commenting adverbs
Total adverbs
0 0,30
t(11)=-3,06, p<,02
Table IS-10/6
Feature Metric VC 1.1 VC 3
IS-10 Word
types
Adverbs of time
Total adverbs
0 0,09
t(7)=-1,96, n.s.
quantitative adverbs
Total adverbs
0,56 0,33
t(10)=0,64, n.s.
Viewpoint and commenting adverbs
Total adverbs
0 0,05
t(7)=-1,00 n.s.
Table IS-10/7
108
Feature Metric VC 1.2 VC 2
IS-10 Word
types
Adverbs of place
Total adverbs
0,18 0,33
t(18)=-1,89, n.s.
Adverbs of time
Total adverbs
0,12 0
t(11)=1,91, n.s.
Adverbs of manner
Total adverbs
0,24 0,07
t(17)=2,53, p<,05
Viewpoint and commenting adverbs
Total adverbs
0,03 0,30
t(22)=-1,30, n.s.
Table IS-10/8
Feature Metric VC 2 VC 3
IS-10 Word
types
Adverbs of time
Total adverbs
0 0,09
t(7)=-1,96, n.s.
Adverbs of manner
Total adverbs
0,07 0,26
t(11)=-2,91, p<,02
Viewpoint and commenting adverbs
Total adverbs
0,30 0,05
t(14)=2,52, p<,05
Table IS-10/9
IS-11 Syntax
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-11 Syntax Subordinate clauses
clause count
0,29 0,23
t(19)=1,07, n.s.
prepositional phrases
clause count
0,25 0,36
t(20)=-0,95, n.s.
Table IS-11/1
109
Feature Metric VC 1.2 ST
IS-11 Syntax Subordinate clauses
clause count
0,12 0,23
t(18)=-2,11, p<,05
prepositional phrases
clause count
0,21 0,36
t(11)=-2,83, p<,02
Table IS-11/2
Feature Metric VC 2 ST
IS-11 Syntax Subordinate clauses
clause count
0,14 0,23
t(22)=-1,11, n.s.
prepositional phrases
clause count
0,25 0,36
t(19)=-1,93, n.s.
Table IS-11/3
Feature Metric VC 3 ST
IS-11 Syntax Subordinate clauses
clause count
0,07 0,23
t(12)=-5,05, p<,01
prepositional phrases
clause count
0,29 0,36
t(13)=-1,32, n.s.
Table IS-11/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-11 Syntax Subordinate clauses
clause count
0,29 0,12
t(21)=2,37, p<,05
Table IS-11/5
Feature Metric VC 1.1 VC 2
IS-11 Syntax Subordinate clause
clause count
0,29 0,14
t(26)=1,74, n.s.
Table IS-11/6
110
Feature Metric VC 1.1 VC 3
IS-11 Syntax Subordinate clauses
clause count
0,29 0,07
t(17)=3,93, p<,01
Table IS-11/7
Feature Metric VC 1.2 VC 3
IS-11 Syntax Subordinate clauses
clause count
0,12 0,07
t(22)=2,45, p<,05
Table IS-11/8
Feature Metric VC 2 VC 3
IS-11 Syntax Subordinate clauses
clause count
0,14 0,07
t(22)=2,62, p<,02
Table IS-11/9
IS-12 Article types
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-12 Article
types
Definite
Total articles
1,00 0,88
t(6)=4,74, p<,01
articles
word count
0,09 0,15
t(16)=-2,83, p<,02
articles
periods
1,12 1,90
t(17)=-2,18, p<,05
Table IS-12/1
111
Feature Metric VC 1.2 ST
IS-12 Article
types
Definite
Total articles
0,98 0,88
t(18)=2,24, p<,05
articles
word count
0,12 0,15
t(18)=-2,74, p<,02
articles
periods
0,84 1,90
t(18)=-3,66, p<,01
Table IS-12/2
Feature Metric VC 2 ST
IS-12 Article
types
Definite
Total articles
0,97 0,88
t(7)=4,10, p<,01
articles
word count
0,12 0,15
t(22)=-1,76, n.s.
articles
periods
0,87 1,90
t(12)=-8,18, p<,01
Table IS-12/3
Feature Metric VC 3 ST
IS-12 Article
types
Definite
Total articles
0,94 0,88
t(11)=0,34, n.s.
articles
word count
0,16 0,15
t(9)=1,62, n.s.
articles
periods
1,26 1,90
t(14)=-3,77, p<,01
Table IS-12/4
112
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
IS-12 Article
types
articles
periods
1,12 0,84
t(24)=0,31, n.s.
Table IS-12/5
Feature Metric VC 1.1 VC 2
IS-12 Article
types
articles
periods
1,12 0,87
t(15)=0,89, n.s.
Table IS-12/6
Feature Metric VC 1.1 VC 3
IS-12 articles
word count
0,09 0,16
t(21)=-3,72, p<,01
Article
types
articles
periods
1,12 1,26
t(17)=-0,47, n.s.
Table IS-12/7
Feature Metric VC 1.2 VC 3
IS-12 Article
types
articles
periods
0,84 1,26
t(19)=-1,19, n.s.
Table IS-12/8
Feature Metric VC 2 VC 3
IS-12 Article
types
articles
periods
0,87 1,26
t(15)=-3,45, p<,01
Table IS-12/9
113
IS-15 Additional terms in genitive case
Feature Metric VC 1.1 ST
IS-15 Additional terms in
genitive case
genitive case
word count
0 0,03
t(7)=-4,44, p<,01
Table IS-15/1
Feature Metric VC 1.2 ST
IS-15 Additional terms in
genitive case
genitive case
word count
0,01 0,03
t(10)=-3,54, p<,01
Table IS-15/2
Feature Metric VC 2 ST
IS-15 Additional terms in
genitive case
genitive case
word count
0 0,03
t(9)=-3,78, p<,01
Table IS-15/3
Feature Metric VC 3 ST
IS-15 Additional terms in
genitive case
genitive case
word count
0,02 0,03
t(11)=-2,02, n.s.
Table IS-15/4
IS-16 Additional terms in accusative case
Feature Metric VC 1.1 ST
IS-16 Additional terms in
accusative case
accusative case
word count
0 0,01
t(6)=-5,58, p<,01
Table IS-16/1
114
Feature Metric VC 1.2 ST
IS-16 Additional terms in
accusative case
accusative case
word count
0 0,01
t(19)=-2,28, p<,05
Table IS-16/2
Feature Metric VC 2 ST
IS-16 Additional terms in
accusative case
accusative case
word count
0 0,01
t(7)=-5,04, p<,01
Table IS-16/3
Feature Metric VC 3 ST
IS-16 Additional terms in
accusative case
accusative case
word count
0 0,01
t(7)=-6,38, p<,01
Table IS-16/4
IS-17 Abbreviations
Comparison of VCs vs ST
Feature Metric VC 2 ST
IS-17 Abbreviations Abbreviations
word count
0,02 0
t(18)=1,49, n.s.
Table IS-17/1
Feature Metric VC 3 ST
IS-17 Abbreviations Abbreviations
word count
0,04 0
t(8)=1,22, n.s.
Table IS-17/2
115
Comparison between VCs
Feature Metric VC 2 VC 3
IS-17 Abbreviations Abbreviations
word count
0,02 0,04
t(8)=-1,33, n.s.
Table IS-17/3
IS-18 Metaphors
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
IS-18 Metaphors Metaphors
clause count
0 0,04
t(6)=-6,39, p<,01
Table IS-18/1
Feature Metric VC 1.2 ST
IS-18 Metaphors Metaphors
clause count
0 0,04
t(6)=-6,39, p<,01
Table IS-18/2
Feature Metric VC 3 ST
IS-18 Metaphors Metaphors
clause count
0,12 0,04
t(9)=1,24, n.s.
Table IS-18/3
Comparison between VCs
Feature Metric VC 2 VC 3
IS-18 Metaphors Metaphors
clause count
0,04 0,12
t(14)=-1,10, n.s.
Table IS-18/4
116
IS-19 Similes
Feature Metric VC 1.1 ST
IS-19 Similes Similes
clause count
0 0,02
t(6)=-3,41, p<,02
Table IS-19/1
Feature Metric VC 1.2 ST
IS-19 Similes Similes
clause count
0 0,02
t(6)=-3,41, p<,02
Table IS-19/2
Feature Metric VC 2 ST
IS-19 Similes Similes
clause count
0 0,02
t(6)=-3,41, p<,02
Table IS-19/3
Feature Metric VC 3 ST
IS-19 Similes Similes
clause count
0 0,02
t(6)=-3,41, p<,02
Table IS-19/4
IS-20 Word variety
Feature Metric VCs (average) ST
IS-20 Word variety Variety
word count
0,37 0,50
t(8)=-4,41, p<,01
Table IS-20/1
117
Statistical analysis (Collaborative Learning)
CA-1 Verbs in first plural person
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,76 0,16
t(20)=7,10, p<,01
Table CA-1/1
Feature Metric VC 1.2 ST
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,43 0,16
t(18)=4,31, p<,01
Table CA-1/2
Feature Metric VC 2 ST
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,31 0,16
t(20)=3,37, p<,01
Table CA-1/3
Feature Metric VC 3 ST
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,03 0,16
t(7)=-2,91, p<,05
Table CA-1/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,76 0,43
t(24)=3,73, p<,01
Table CA-1/5
118
Feature Metric VC 1.1 VC 2
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,76 0,31
t(29)=2,50, p<,02
Table CA-1/6
Feature Metric VC 1.1 VC 3
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,76 0,03
t(14)=9,63, p<,01
Table CA-1/7
Feature Metric VC 1.2 VC 2
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,43 0,31
t(23)=-0,48, n.s.
Table CA-1/8
Feature Metric VC 1.2 VC 3
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,43 0,03
t(14)=8,00, p<,01
Table CA-1/9
Feature Metric VC 2 VC 3
CA-1 Verbs in first
plural person
Verbs (we)
Verb count
0,31 0,03
t(15)=5,11, p<,01
Table CA-1/10
119
CA-2 Emotional clauses
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-2 Emotional
clauses
emotional clauses
clause count
0,33 0,05
t(14)=5,02, p<,01
Table CA-2/1
Feature Metric VC 1.2 ST
CA-2 Emotional
clauses
emotional clauses
clause count
0,21 0,05
t(13)=2,98, p<,02
Table CA-2/2
Feature Metric VC 2 ST
CA-2 Emotional
clauses
emotional clauses
clause count
0,22 0,05
t(16)=2,81, p<,02
Table CA-2/3
Feature Metric VC 3 ST
CA-2 Emotional
clauses
emotional clauses
clause count
0,10 0,05
t(16)=0,82, n.s.
Table CA-2/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-2 Emotional
clauses
emotional clauses
clause count
0,33 0,21
t(26)=2,05, n.s.
Table CA-2/5
120
Feature Metric VC 1.1 VC 2
CA-2 Emotional
clauses
emotional clauses
clause count
0,33 0,22
t(29)=1,72, n.s.
Table CA-2/6
Feature Metric VC 1.1 VC 3
CA-2 Emotional
clauses
emotional clauses
clause count
0,33 0,10
t(22)=3,39, p<,01
Table CA-2/7
Feature Metric VC 1.2 VC 3
CA-2 Emotional
clauses
emotional clauses
clause count
0,21 0,10
t(20)=1,52, n.s.
Table CA-2/8
Feature Metric VC 2 VC 3
CA-2 Emotional
clauses
emotional clauses
clause count
0,22 0,10
t(22)=1,61, n.s.
Table CA-2/9
CA-3 Rewarding clauses
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-3 Rewarding
clauses
rewarding clauses
clause count
0,02 0
t(14)=1,00, n.s.
Table CA-3/1
121
Feature Metric VC 1.2 ST
CA-3 Rewarding
clauses
rewarding clauses
clause count
0,16 0
t(13)=3,64, p<,01
Table CA-3/2
Feature Metric VC 2 ST
CA-3 Rewarding
clauses
rewarding clauses
clause count
0,16 0
t(15)=2,81, p<,02
Table CA-3/3
Feature Metric VC 3 ST
CA-3 Rewarding
clauses
rewarding clauses
clause count
0,06 0
t(8)=1,80,n.s.
Table CA-3/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-3 Rewarding
clauses
rewarding clauses
clause count
0,02 0,16
t(25)=-2,23, p<,05
Table CA-3/5
Feature Metric VC 1.1 VC 2
CA-3 Rewarding
clauses
rewarding clauses
clause count
0,02 0,16
t(25)=1,88, n.s.
Table CA-3/6
122
Feature Metric VC 1.1 VC 3
CA-3 Rewarding
clauses
rewarding clauses
clause count
0,02 0,06
t(22)=-0,38, n.s.
Table CA-3/7
Feature Metric VC 1.2 VC 3
CA-3 Rewarding
clauses
rewarding clauses
clause count
0,16 0,06
t(20)=2,05, n.s.
Table CA-3/8
Feature Metric VC 2 VC 3
CA-3 Rewarding
clauses
rewarding clauses
clause count
0,16 0,06
t(21)=1,69, n.s.
Table CA-3/9
CA-4 Clauses of negation
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-4 Clauses of
negation
Clauses of negation
clause count
0 0,05
t(6)=-4,20, p<,01
Table CA-4/1
Feature Metric VC 1.2 ST
CA-4 Clauses of
negation
Clauses of negation
clause count
0,03 0,05
t(19)=-1,22, n.s.
Table CA-4/2
123
Feature Metric VC 2 ST
CA-4 Clauses of
negation
Clauses of negation
clause count
0,07 0,05
t(21)=-0,62, n.s.
Table CA-4/3
Feature Metric VC 3 ST
CA-4 Clauses of
negation
Clauses of negation
clause count
0,06 0,05
t(14)=-0,53, n.s.
Table CA-4/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-4 Clauses of
negation
Clauses of negation
clause count
0 0,03
t(13)=-1,41, n.s.
Table CA-4/5
Feature Metric VC 1.1 VC 2
CA-4 Clauses of
negation
Clauses of negation
clause count
0 0,07
t(15)=-1,77, n.s.
Table CA-4/6
Feature Metric VC 1.1 VC 3
CA-4 Clauses of
negation
Clauses of negation
clause count
0 0,06
t(8)=-2,91, p<,02
Table CA-4/7
124
Feature Metric VC 1.2 VC 2
CA-4 Clauses of
negation
Clauses of negation
clause count
0,03 0,07
t(28)=-0,42, n.s.
Table CA-4/8
Feature Metric VC 1.2 VC 3
CA-4 Clauses of
negation
Clauses of negation
clause count
0,03 0,06
t(21)=-0,72, n.s.
Table CA-4/9
CA-5 Clauses of reason
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-5 Clauses of
reason
Clauses of reason
clause count
0,29 0,02
t(15)=4,55, p<,01
Table CA-5/1
Feature Metric VC 2 ST
CA-5 Clauses of
reason
Clauses of reason
clause count
0,01 0,02
t(8)=-1,07, n.s.
Table CA-5/2
Feature Metric VC 3 ST
CA-5 Clauses of
reason
Clauses of reason
clause count
0,01 0,02
t(7)=-1,25, n.s.
Table CA-5/3
125
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-5 Clauses of
reason
Clauses of reason
clause count
0,29 0,02
t(15)=4,38, p<,01
Table CA-5/4
Feature Metric VC 1.1 VC 2
CA-5 Clauses of
reason
Clauses of reason
clause count
0,29 0,01
t(14)=4,75, p<,01
Table CA-5/5
Feature Metric VC 1.1 VC 3
CA-5 Clauses of
reason
Clauses of reason
clause count
0,29 0,01
t(14)=4,78, p<,01
Table CA-5/6
CA-6 Familiarity words
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-6 Familiarity
words
familiarity words
word count
0,33 0
t(14)=4,70, p<,01
Table CA-6/1
Feature Metric VC 1.2 ST
CA-6 Familiarity
words
familiarity words
word count
0,05 0
t(12)=4,42, p<,01
Table CA-6/2
126
Feature Metric VC 2 ST
CA-6 Familiarity
words
familiarity words
word count
0,02 0
t(16)=2,32, p<,05
Table CA-6/3
Feature Metric VC 3 ST
CA-6 Familiarity
words
familiarity words
word count
0,04 0
t(8)=2,35, p<,05
Table CA-6/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-6 Familiarity
words
familiarity words
word count
0,33 0,05
t(15)=4,18, p<,01
Table CA-6/5
Feature Metric VC 1.1 VC 2
CA-6 Familiarity
words
familiarity words
word count
0,33 0,02
t(14)=4,64, p<,01
Table CA-6/6
Feature Metric VC 1.1 VC 3
CA-6 Familiarity
words
familiarity words
word count
0,33 0,04
t(17)=3,90, p<,01
Table CA-6/7
127
CA-7 Inclusive words
Feature Metric VC 1.1 ST
CA-7 Inclusive
words
inclusive words
word count
0,01 0
t(15)=1,21, n.s.
Table CA-7/1
CA-8 Social words
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-8 Social
words
social words
word count
0,06 0
t(14)=2,56, p<,05
Table CA-8/1
Feature Metric VC 2 ST
CA-8 Social
words
social words
word count
0,01 0
t(22)=-1,30, n.s.
Table CA-8/2
Feature Metric VC 3 ST
CA-8 Social
words
social words
word count
0,02 0
t(8)=1,68, n.s.
Table CA-8/3
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-8 Social
words
social words
word count
0,06 0
t(15)=2,37, p<,05
Table CA-8/4
128
Feature Metric VC 1.1 VC 2
CA-8 Social
words
social words
word count
0,06 0,01
t(14)=2,41, p<,05
Table CA-8/5
Feature Metric VC 1.1 VC 3
CA-8 Social
words
social words
word count
0,06 0,02
t(20)=0,34, n.s.
Table CA-8/6
CA-9 Emotional words
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-9 Emotional
words
emotional words
word count
0,18 0,02
t(14)=4,97, p<,01
Table CA-9/1
Feature Metric VC 1.2 ST
CA-9 Emotional
words
emotional words
word count
0,10 0,02
t(14)=3,30, p<,01
Table CA-9/2
Feature Metric VC 2 ST
CA-9 Emotional
words
emotional words
word count
0,07 0,02
t(16)=2,16, p<,05
Table CA-9/3
129
Feature Metric VC 3 ST
CA-9 Emotional
words
emotional words
word count
0,04 0,02
t(9)=0,58, n.s.
Table CA-9/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-9 Emotional
words
emotional words
word count
0,18 0,10
t(26)=1,80, n.s.
Table CA-9/5
Feature Metric VC 1.1 VC 2
CA-9 Emotional
words
emotional words
word count
0,18 0,07
t(27)=2,57, p<,02
Table CA-9/6
Feature Metric VC 1.1 VC 3
CA-9 Emotional
words
emotional words
word count
0,18 0,04
t(22)=3,88, p<,01
Table CA-9/7
Feature Metric VC 1.2 VC 3
CA-9 Emotional
words
emotional words
word count
0,10 0,04
t(21)=2,23, p<,05
Table CA-9/8
130
Feature Metric VC 1.2 VC 2
CA-9 Emotional
words
emotional words
word count
0,10 0,07
t(28)=0,83, n.s.
Table CA-9/9
Feature Metric VC 2 VC 3
CA-9 Emotional
words
emotional words
word count
0,07 0,04
t(23)=1,37, n.s.
Table CA-9/10
CA-10 Positive emotion
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-10 Positive
emotion
positive emotion words
emotional words
1,00 0,65
t(6)=3,93, p<,01
Table CA-10/1
Feature Metric VC 1.2 ST
CA-10 Positive
emotion
positive emotion words
emotional words
0,55 0,65
t(13)=-0,44, n.s.
Table CA-10/2
Feature Metric VC 2 ST
CA-10 Positive
emotion
positive emotion words
emotional words
0,94 0,65
t(9)=2,79, p<,05
Table CA-10/3
131
Feature Metric VC 3 ST
CA-10 Positive
emotion
positive emotion words
emotional words
0,68 0,65
t(4)=-0,30, n.s.
Table CA-10/4
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-10 Positive
emotion
positive emotion words
emotional words
1,00 0,55
t(9)=2,47, p<,05
Table CA-10/5
Feature Metric VC 1.1 VC 1.2
CA-10 Positive
emotion
positive emotion words
emotional words
1,00 0,55
t(9)=2,47, p<,05
Table CA-10/6
Feature Metric VC 1.1 VC 2
CA-10 Positive
emotion
positive emotion words
emotional words
1,00 0,94
t(8)=1,51, n.s.
Table CA-10/7
Feature Metric VC 1.1 VC 3
CA-10 Positive
emotion
positive emotion words
emotional words
1,00 0,68
t(3)=1,50, n.s.
Table CA-10/8
132
Feature Metric VC 1.2 VC 2
CA-10 Positive
emotion
positive emotion words
emotional words
0,55 0,94
t(10)=-2,01, n.s.
Table CA-10/9
Feature Metric VC 1.2 VC 3
CA-10 Positive
emotion
positive emotion words
emotional words
0,55 0,68
t(5)=0,02, n.s.
Table CA-10/10
Feature Metric VC 2 VC 3
CA-10 Positive
emotion
positive emotion words
emotional words
0,94 0,68
t(3)=1,25, n.s.
Table CA-10/11
CA-12 Use of first person plural pronouns
Comparison of VCs vs ST
Feature Metric VC 1.1 ST
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0,33 0,08
t(16)=2,31, p<,05
Us, ours
Possessive + personal
pronouns
0,34 0,10
t(18)=1,96, n.s.
Table CA-12/1
133
Feature Metric VC 1.2 ST
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0,44 0,08
t(9)=2,51, n.s.
Us, ours
Possessive + personal
pronouns
0,47 0,10
t(10)=2,38, p<,05
Table CA-12/2
Feature Metric VC 2 ST
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0,68 0,08
t(19)=6,91, p<,01
Us, ours
Possessive + personal
pronouns
0,70 0,10
t(21)=6,27, p<,01
Table CA-12/3
Feature Metric VC 3 ST
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0 0,08
t(6)=-1,97, n.s.
Us, ours
Possessive + personal
pronouns
0 0,10
t(6)=-1,90, n.s.
Table CA-12/4
134
Comparison between VCs
Feature Metric VC 1.1 VC 1.2
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0,33 0,44
t(16)=-0,69, n.s.
Us, ours
Possessive + personal
pronouns
0,34 0,47
t(15)=-0,89, n.s.
Table CA-12/5
Feature Metric VC 1.1 VC 2
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0,33 0,68
t(27)=-3,24, p<,01
Us, ours
Possessive + personal
pronouns
0,34 0,70
t(27)=-3,18, p<,01
Table CA-12/6
Feature Metric VC 1.1 VC 3
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0,33 0
t(13)=3,17, p<,01
Us, ours
Possessive + personal
pronouns
0,34 0
t(13)=3,18, p<,01
Table CA-12/7
135
Feature Metric VC 1.2 VC 2
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0,44 0,68
t(15)=-1,92, n.s.
Us, ours
Possessive + personal
pronouns
0,47 0,70
t(14)=-1,50, n.s.
Table CA-12/8
Feature Metric VC 1.2 VC 3
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0,44 0
t(8)=3,10, p<,02
Us, ours
Possessive + personal
pronouns
0,47 0
t(8)=3,15, p<,02
Table CA-12/9
Feature Metric VC 2 VC 3
CA-12 Use of first
person plural
pronouns
Us, ours
Pronouns
0,68 0
t(15)=8,18, p<,01
Us, ours
Possessive + personal
pronouns
0,70 0
t(15)=8,31, p<,01
Table CA-12/10
136
References
Bereiter Carl, (1994), Implications of postmodernism for science, or, science as
progressive discourse, Educational Psychologist, Vol. 29, Iss. 1, pp. 3-12. Available:
http://www.tandfonline.com/doi/abs/10.1207/s15326985ep2901_1#.UzNKJvl_ty0
(last accessed October 2013)
Bielaczyc, K., and Collins, A., (1999), Learning communities in classrooms: A reconceptualization of educational practice, In C. Reigeluth (Ed.)Instructional design theories and models (pp. 169–292) Mahwah, HJ: Erlbaum. Available: http://isites.harvard.edu/fs/docs/icb.topic541040.files/Bielaczyc%20and%20Collins-Learning%20Communities%20in%20Classrooms.pdf (last accessed: December 2013)
Blau R. Susan, Hall John, and Strauss Tracy, (1998), Exploring the Tutor/ClientConversation: A Linguistic Analysis. The Writing Center Journal, Volume 19, Number 1, Fall/Winter 1998. Available: http://casebuilder.rhet.ualr.edu/wcrp/publications/wcj/wcj19.1/wcj19.1_blau.pdf (last accessed October 2013)
Brooks Michael, Kuksenok Katie, Torkildson Megan K., Perry Daniel, Robinson John J.,
Scott Taylor J., Anicello Ona , Zukowski Ariana, Harris Paul and Aragon Cecilia R.,
(2013), Statistical Affect Detection in Collaborative Chat, CSCW '13 Proceedings of
the 2013 conference on Computer supported cooperative work, pp. 317-328.
Available: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.296.9852&r
ep=rep1&type=pdf (last accessed: January 2014)
Daniel Ben, Schwier A. Richard, McCalla Gordon (2003), Social Capital in Virtual
Learning Communities and Distributed Communities of Practice, Canadian Journal of
Learning and Technology/La revue canadienne de l’apprentissage et de la
technologie,[S.l.],Oct.2003.
Available:http://cjlt.csj.ualberta.ca/index.php/cjlt/article/view/85/79 (last accessed:
February 2014)
Dillenbourg P., Fischer F. (2007), Basics of Computer-Supported Collaborative Learning, Zeitschrift für Berufs- und Wirtschaftspädagogik, 21, pp. 111-130. Available: http://pileas.csd.auth.gr/file.php/47/MA8HMA_2013/DillenbourgFischer-CSCL-Basics-ARTICLE.pdf (last accessed: August 2013)
Ehrich F. J., (2006), Vygotskian Inner Speech and the Reading Process, Australian
Journal of Educational & Developmental Psychology Vol. 6, 2006, pp 12-25.
Available: http://files.eric.ed.gov/fulltext/EJ815614.pdf (last accessed: February
2014)
Emerson, Caryl, (1983), The Outer Word and Inner Speech: Bakhtin, Vygotsky, and
the Internalization of Language, Critical Inquiry, 10 (2):245, pp. 245-264. Available:
137
http://www.jstor.org/discover/10.2307/1343349?uid=3738128&uid=2&uid=4&sid=2
1103643743873 (last accessed: September 2013)
Engeström Yrjö, (2011) From design experiments to formative interventions, in Theory Psychology, October 2011, vol. 21, no. 5, 598-628. Available: http://tap.sagepub.com/content/21/5/598.abstract (last accessed: September 2013)
Engeström Yrjö, (2001), Expansive Learning at Work: toward an activity theoretical
reconceptualization, Journal of Education and Work, Vol. 14, No. 1. Available:
http://www.handover.eu/upload/library/jimaw6szeyuluh4tho6oq.pdf (last accessed:
August 2013)
Innes B. Robert, (2007), Dialogic Communication in Collaborative Problem Solving Groups, International Journal for the Scholarship of Teaching and Learning, 1(1). Available:http://digitalcommons.georgiasouthern.edu/cgi/viewcontent.cgi?article=1178&context=int_jtl&seiredir=1&referer=http%3A%2F%2Fscholar.google.gr%2Fscholar%3Fhl%3Del%26q%3DDialogic%2BCommunication%2Bin%2BCollaborative%2BProblem%2BSolving%2BGroups%26btnG%3D#search=%22Dialogic%20Communication%20Collaborative%20Problem%20Solving%20Groups%22 (last accessed: August 2013)
Jameson, J., Ferrell, G., Kelly, J., Walker, S. and Ryan, M., (2006), Building trust and
shared knowledge in communities of e-learning practice: collaborative leadership in
the JISC eLISA and CAMEL lifelong learning projects. British Journal of Educational
Technology,37: 949–967. Available:
http://onlinelibrary.wiley.com/enhanced/doi/10.1111/j.1467-8535.2006.00669.x/
(last accessed: February 2014)
Knipfer Kristin, Mayr Eva, Zahn Carmen, Schwan Stephan, Hesse Friedrich W., (2009), Computer support for knowledge communication in science exhibitions: Novel perspectives from research on collaborative learning, Educational Research Review, Volume 4, Issue 3, Pages 196-209. Available:http://www.sciencedirect.com/science/article/pii/S1747938X09000311 (last accessed: August 2013)
Kollar Ingo, Fischer Frank, Slotta James D., (2005), Internal and external collaboration
scripts in web-based science learning at schools, In Proceedings of th 2005
conference on Computer support for collaborative learning: learning 2005: the next
10 years! (CSCL '05), International Society of the Learning Sciences 331-340.
Available: http://telearn.archives-ouvertes.fr/docs/00/19/06/41/PDF/Kollar-Ingo-
2005.pdf (last accessed: February 2014)
138
Koschmann Timothy (1999), Toward a dialogic theory of learning: Bakhtin's
contribution to understanding learning in settings of collaboration, CSCL '99
Proceedings of the 1999 conference on Computer support for collaborative learning,
Article No. 38.
Available:http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.22.5920&rep=r
ep1&type=pdf (last accessed: August 2013)
Mairesse Francois, Walker A. Marilyn, Mehl R. Matthias, and Moore K. Roger, (2007),
Using Linguistic Cues for the Automatic Recognition of Personality in Conversation
and Text. Journal of Artificial Intelligence Research, 30, 2007, pp. 457-500. Available:
http://people.csail.mit.edu/francois/research/papers/personality-jair07.pdf (last
accessed: January 2014)
Maness M. Jack, (2008), A Linguistic Analysis of Chat Reference Conversations with
18–24 Year-Old College Students. The Journal of Academic Librarianship, Vol. 34, Iss.
1, January 2008, pp. 31-38. Available:
http://ucblibraries.colorado.edu/facultyprofiles/files/publications/ADmanessj/Mane
ss-Lingustic%20Analysis%20of%20Chat%20Reference.pdf (last accessed: January
2014)
McMillan W. David, Chavis George M. David (1986), Sense of Community: A
Definition and Theory, Journal of Community Psychology, Volume 14. Available:
https://communities.autodesk.com/india/sites/default/files/secure/docs/McMillanC
havis---psychological-Sense-of-community.pdf (last accessed February 2014)
Patterson Glenys, (2005), Collaboration/Competition Crossroads: National/
Supranational Tertiary Education Policies on a Collision Course, Tertiary Education
and Management, December 2005, Volume 11, Issue 4, pp 355-368. Available:
http://link.springer.com/article/10.1007/s11233-005-0980-7# (last accessed:
February 2014)
Pérez-Sabater Carmen, (2012), The Linguistics of Social Networking: A Study of
Writing Conventions on Facebook. Linguistik online, 56, 6/2012. Available:
http://www.linguistik-online.de/56_12/perez-sabater.html (last accessed: February
2014)
Rovai P. Alfred (2002), Sense of community, perceived cognitive learning, and persistence in asynchronous learning networks, The Internet and Higher Education, Volume 5, Issue 4, Pages 319-332. Available: http://dx.doi.org/10.1016/S1096-7516(02)00130-6 (last accessed: February 2014).
Roussos L. Petros, Tsaousis Giannis (2006), Applied Statistics in Social Sciences,
Ellinika grammata, Athens. (In Greek)
139
Santaholma Marianne Elina, (2005), Linguistic representation of Finnish in a limited domain speech-to-speech translation system. In: Proceedings of the 10th Conference of the European Association for Machine Translation (EAMT), pp. 226-234. Available: http://archive-ouverte.unige.ch/download/unige:3434/ATTACHMENT01 (last accessed: February 2014).
Socolov N. A., (1972), Inner Speech and Thought, Plenum, New York, pp. 46-122.
Spanger Phillip, Masaaki Yasuhara, Ryu Iida and Takenobu Tokunaga, (2009), A
Japanese corpus of referring expressions used in a situated collaboration task.
Proceedings of the 12th European Workshop on Natural Language Generation, pp.
110-113. Available: http://delivery.acm.org/10.1145/1620000/1610213/p110-
spanger.pdf?ip=46.103.197.48&id=1610213&acc=OPEN&key=4D4702B0C3E38B35%
2E4D4702B0C3E38B35%2E4D4702B0C3E38B35%2E6D218144511F3437&CFID=3101
44203&CFTOKEN=44019478&__acm__=1396213753_20a6e0a4545ac35f73204e2ab
ea994f0 (last accessed: December 2013)
Stahl Gerry (2000), A Model of Collaborative Knowledge-Building, In B. Fishman & S.
O'Connor-Divelbiss (Eds.), Fourth International Conference of the Learning Sciences,
(pp. 70-77), Mahwah, NJ: Erlbaum. Available:
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.97.8816&rep=rep1&type
=pdf (last accessed: July 2013)
Stahl Gerry (2004), Building Collaborative Knowing, What We Know About CSCL, Computer-Supported Collaborative Learning Series, Volume 3, 2004, pp 53-85. Available: http://link.springer.com/chapter/10.1007/1-4020-7921-4_3 (last accessed: February 2014)
Stahl G., Koschmann T., Suthers D., (2006), Computer-supported collaborative learning: An historical perspective, In R. K. Sawyer (Ed.), Cambridge handbook of the learning sciences (pp. 409-426), Cambridge, UK: Cambridge University Press. Available: http://gerrystahl.net/cscl/CSCL_English.htm (last accessed: August 2013)
Veermans Marjaana and Cesareni Donatella, (2005), The nature of the discourse in
web-based Collaborative Learning Environments: Case studies from four different
countries. Computers & Education, Vol.45, Iss.3, November 2005, pp. 316–336.
Available:https://www.tlu.ee/~kpata/haridustehnoloogiaTLU/discoursecontentcomp
arison.pdf (last accessed: December 2013)
Vekyri Ioanna (2007), Learning as knowledge building, in Panteliadou Souzana,
Botsas Giorgos, Learning disabilities, basic concepts and features, University of
Thessaly , Department of Special Education, Volos. (In Greek) Available:
http://www.specialeducation.gr/files4users/files/pdf/teyxos_a.pdf (last accessed
February 2014)
Vygotsky S. Lev, (1978), Mind in society: The development of higher psychological processes, Cambridge, Mass.: Harvard University Press.
140
Vygotsky S. Lev, (2008), Thought and Language, Gnosi, Athens, pp. 378-436. (In Greek)
Warschauer Mark, (1997), Computer-Mediated Collaborative Learning: Theory and
Practice, The Modern Language Journal, Volume 81, Issue 4, pp. 470–481, Winter
1997.
Available:http://moodle.bracu.ac.bd/pluginfile.php/2542/mod_resource/content/1/
Warschauer-M_Computer-mediated%20collaborative%20learning-
theory%20and%20Practice.pdf (last accessed: December 2013)
Wells Gordon, (2002), The Role of Dialogue in Activity Theory, Mind, Culture, and
Activity ,Vol. 9,Iss. 1,pp.43-66. Available:
http://www.tandfonline.com/doi/abs/10.1207/S15327884MCA0901_04#.UzNOnfl_t
y0 (last accessed: August 2013)
Wiley Norbert, (2006), Inner Speech as a Language: A Saussurean Inquiry, Journal for
the Theory of Social Behaviour, Volume 36, Issue 3, pp. 319–341, September 2006.
Available: http://cdclv.unlv.edu//pragmatism/wiley_speech.pdf (last accessed:
November 2013)
Zhang Jianwei, Scardamalia Marlene, Lamon Mary, Messina Richard and Reeve
Richard, (2007), Socio-cognitive dynamics of knowledge building in the work of 9-
and 10-year-olds. Educational Technology Research and Development, April
2007, Vol. 55, Iss. 2, pp. 117-145. Available:
http://medicina.iztacala.unam.mx/medicina/Socio-
cognitive%20dynamics%20of%20knowledge%20building.pdf (last accessed:
November 2013)