5 THE IMPACT OF NON VERBAL COMMUNICATION ON TEAM ...iaeme.com/MasterAdmin/UploadFolder/THE IMPACT OF...

International Journal of Management (IJM), ISSN 0976 – 6502(Print), ISSN 0976 - 6510(Online),

Volume 5, Issue 12, December (2014), pp. 36-42 © IAEME

43

THE IMPACT OF NON-VERBAL COMMUNICATION ON

TEAM PRODUCTIVITY DURING DESIGN

Dr.Wadhah Amer Hatem,

Lecturer, Baquba Technical Institute, IRAQ

Dr.Alan Kwan

Reader, Cardiff University, United Kingdom

Dr.John Miles

Professor, Cardiff School of Engineering, Cardiff University, UK

ABSTRACT

Non-verbal communication (NVC) is an important component of human communication. A

movement of the body, or some eye contact, can convey significant amounts of information. While

NVC has been studied in many applications, it is largely absent in the literature relating to the

construction industry and particularly the design process. This paper studies the non-verbal

communication in face to face meetings between team members during the design of a small

building. The study includes an evaluation of the impact of various non-verbal communication

movements observed during the design exercises, and how these relate to team productivity. The

relationship between NVC movements and the culture/ethnicity of team members, as well as the

relationship between verbal and non-verbal communication, are also explored. It is found that some

NVC movements are strongly related to productivity.

Keywords: Non-Verbal Communication, Team Productivity, Illustrator Movement, Regulator

Movements, Adaptors Movements.

1. INTRODUCTION

The communication process can be defined as the transmission of information between the

sender and one or more recipients. The information is partially transmitted by verbal communication

but body language (i.e. non-verbal communication) also is a significant factor. Non-verbal

Communication (NVC) can be used to transmit significant messages and it is sometimes faster than

verbal communication because one movement or a single glance can convey a substantial amount of

INTERNATIONAL JOURNAL OF MANAGEMENT (IJM)

ISSN 0976-6502 (Print)

ISSN 0976-6510 (Online)

Volume 5, Issue 12, December (2014), pp. 43-68

© IAEME: http://www.iaeme.com/IJM.asp

Journal Impact Factor (2014): 7.2230 (Calculated by GISI)

www.jifactor.com

IJM

© I A E M E



44

information. The components of non-verbal communication, such as body language, facial

expressions, and eye contact can be significant during face to face (FTF) communication and hence

potentially can have an effect on responses and behaviour (Gero & Tang, 2001).

There is an extensive body of research on non-verbal communication (NVC) and only the

more salient features will be discussed here. Kim & Maher (2008) have asserted NVC helps people

to express their emotional states by transmitting personal information and interpersonal attitudes, and

hence helps to organise social interaction. Knapp (1978) and Birdwhistell (1970) state that the term

of NVC is used to describe all human communication activities which do not use either written or

spoken words. In addition, social psychologists confirm that more than 65% of the total information

exchanged in FTF occurs by means of NVC. Knapp and Hall (2007) state that NVC involves three

factors: environmental conditions, physical characteristics and the behaviour of the communicators.

Hall (1984) claimed that almost 90% of communication in FTF is non-verbal while Wiener and

Mehrabian (1968) state that the percentage of NVC is 93%. Mehrabian (1981) discovered that, in

general, the message will be transmitted by the three general communication aspects in the following

proportions: 7% for spoken words, 55% for postures and gestures and 38% for pitch, volume and

intonation. Whatever is the precise percentage of the total, NVC is clearly significant feature in

human communication.

NVC is an effective method of conveying information about personal emotions without any

need for additional verbal explanation, although often NVC occurs without the transmitter being

aware of what messages they are conveying nor indeed being able to control the message content

(Guye-Vuilleme et al., 1999). Communication partners use NVC for increasing their visibility and to

clarify the points that they are trying to convey (Goffman, 1959; Gergle et al., 2004). Equally, it is

also true that some NVC expressions do not need to be associated with any verbal events (Ekman &

Friesen, 1967; Aboudan & Beattie, 1996) and are thus additional to what is being said. Additionally

the greatest use is made of NVC when the degree of interaction between the partners is at its highest

level (Stempfle & Badke-Schaub) 2002). NVC helps people to coordinate and collaborate to achieve

their team objectives (Tooby & Cosmides, 1996).

Kinesics science is used to study NVC face and body movements. It identifies five kinds of

movement: “emblems” (body movements in place of verbal phrases), “illustrators” (body

movements accompanying and reinforcing verbal phrases), “regulators” (actions relating to direction

of communication), “affect display” (facial movements to display emotions) and “adaptors”

(unconscious gestures, not necessarily directly connected with what is being said, but could be

related to negative feelings). Ekman and Friesen (1969) mention that most NVC experts consider

body movements such as eye contact, facial expression, gestures, touch communication as being the

principal components of NVC. Additional factors such as cultural differences are also important and

can be an issue during communication between people from different backgrounds. Axtell (1993)

states that there is a small number of emblems which can be used across different cultures, i.e. there

is a limited number of universal movements.

Facial expressions are an important part of human communication and some facial

expressions occur universally and are therefore understood by all. Facial expressions reflect the

personal demeanour of a person and mirror the emotions expressed in people’s comments, thus

forming a second source of information to support the spoken words (Knapp, 1978). Facial

expressions can be understood across cultural differences even when there is a language barrier.

Most people are unable to hide their feelings, and their emotions manifest themselves as certain

facial expression (Argyle, 1994; Matsumoto et al., 2005).

This paper describes an investigation into the significance of NVC on team and individual

productivity during FTF communication. The task that the teams are asked to complete involves the

partial design of a small building. The analysis of the results classifies the users’ movements into

five main categories, and studies the impact of these movements on team and individual



45

productivity. Additionally, consideration is given to the effect of the culture differences on the team

productivity. The results show that there is a strong relationship between aspects of NVC and

productivity.

2. RELATED WORK

As stated above, the literature on NVC is large and the following have been chosen to

represent the more salient features that are more immediately relevant. Sumi and Moriyama (2010)

classified body actions features for both teachers and students in classrooms during lectures

according to the taxonomy of (Ekman & Freisen, 1969) and they concluded that the emotions

exhibited by teachers have an ambiguous impact, in spite of the accepted wisdom that displaying

emotions better informs their audience. Kraut (1979) discovered that the attention of an audience

can be significantly enhanced by the speaker smiling in appropriate situations. Ekman (1997) points

out that gestures can express intention, or leak emotion, or communicate a specific cultural signal, in

the absence of language which could give additional key information to listeners.

In examining cultural differences, Waxer (1985) has found that Americans use more hand

gestures than Canadians when expressing their emotions. Matsumoto (1992) discovered that

Americans tend to amplify their emotional expressions in NVC for emphasis, while the Japanese

tend to moderate theirs. People can also make up emotions by producing an expression of emotion

that is different from their underlying emotion (Noesjirwan, 1978). Noesjirwan (1978) noticed that

there is a big difference between Australians and Indonesians in their portrayal of emotions and

communicative behaviour. For example, in a group setting, Indonesians try to hide their

disagreement and instead they “smile and agree”. Conversely, Australians usually announce their

disagreement. Gilbert and Krull (2002] and Chen (1995) have found that Americans tend to express

their personal information in NVC more than do Chinese.

Scott and Charteris (1986) compared Europeans to South Africans and found that although

some gestures and emblems (a movement to reinforce a verbal phrase, e.g. a good bye wave) carry

the same meaning, there are some that have completely different meanings. Barakat (1973) stated

that people from Arabic countries usually use body movements and gestures in order to

communicate reactions silently. Arabs are often said to speak with their hands in addition to their

mouths and they tend to think that speech will be clearer when the total number of words is

equivalent to of the number of body movements (Aboudan & Beattie, 1996). Feghali (1997) asserted

that Iraqis and other Arabic people tend to speak loudly and at “a decibel level considered

aggressive, objectionable and even obnoxious by North Americans”. In Arabic culture speaking in a

loud voice implies strength and confidence while a soft voice implies weakness.

Some other authors have concentrated on how NVC differs across different cultures.

Yammiyavar et al. (2008) have studied the effect of cultural background on the nature of NVC for

different people in different countries such as India, Denmark and China. Their conclusion is that

the use of adapters, illustrators, and regulators was significantly different according to culture.

Vatrapu & Perez-Quinones (2006) have found in their study that where an interviewer and an

interviewee have the same cultural background, they are more able to overcome “usability issues”

during the interview (i.e. ability to “read” or “interpret” each other) as compared to when they have

different cultural backgrounds.

While facial expressions are significant, and the literature shows many detailed studies on

this aspect, most researchers examine facial expressions in isolation and without cross-references to

wider NVC movements and hence their techniques are not so readily transferable to the current

work. Brody and Hall (2000) discovered that males and females differ in their expressivity and use

of facial expression, with women tending to use facial expressions more often, and they also tend to

hide expressions of negative emotions. Ekman (1984) concluded that an important factor affecting



46

the analysis of facial expression is the difficulty of measuring the responses to emotions which

produce expressions of short duration. Frank (1997) suggested a new approach by using computer-

based methods to record these short emotions by comparing them with standard features of emotions.

Using this method it is possible to analyse many expressions in a short time.

Body movements, eye contact and changing voice tone represent various emotions which

reflect the state of a person during a conversation. Mason (2003) mentioned that the people use eye

contact during a conversation to indicate confidence sincerity and authority. Miller (1998) asserted

that the amount of eye contact used by a speaker reflected on their degree of credibility and honesty.

Wainwright (2003) states there are six tasks for eye contact: dominance, requesting for information,

controlling interaction, showing attention, giving feedback and politeness or a lack thereof.

Changing voice tone during a conversation has many functions, for example, emphasis, confirming

the importance of a passage of speech, or trying to attract the attention of the listener. Vinciarelli

(2009) postulates that changing the voice tone is used to reflect the personal state such as anger or

disagreement. Ververidis and Kotropoulos (2006) found that the voice will be in high intensity for

emotions such as happiness, anger and surprise while it will be in lower intensity when the person

feels sadness, disgust and fear. Visser & Maher (2008) provided an overview of gestures in design

and reviewed the studies related in this topic. They mentioned gestures has been studied from

various perspectives, sometimes with respect to computer support for human communication and

collaboration but also with respect to the psychology of gesture.

In the construction sector, there has been no previously reported work on the impact of NVC

on team productivity, for any stage of construction or design, and particularly relevant for this work,

in the design stage. Team productivity is affected by emotional factors or individual personality

(Rousseau et al., 2006; Von Glinow et al., 2004) and these are the aspects mainly conveyed by NVC.

A discussion between team members is strongly affected by traits relating to character, behaviour

and personal motivation (Kleinman & Palmon, 2001). Other factors such as humour, flexibility,

degree of cooperation, understanding of the problem have an effect on the decision making and the

quantity and quality of team productivity. They can be considered as significant factors while

behaviour such as general authoritarianism or related forms of domineering behaviour lead to

decreases in team productivity for many of team members, and tend to restrict the productive output

to just one or two members of the team (McHoskey, 1995; Downing & Monaco, 1986).

3. METHODOLOGY

The overall objective of the research which underlies the content of this paper is to study the

differences between people who work FTF and those who use computer mediated communication

(CMC) to undertake engineering design tasks (Hatem et al., 2011). In the current paper, only the

FTF part of the work is considered. The results have been obtained from a carefully controlled set of

experiments in which two participants are asked to collaborate and undertake a predefined design

task which involves the manipulation of an existing design using a 3D CAD package. The

experiments have been recorded using video. In addition to studying the effect of NVC on the

performance of the participants, the experiments also involved the collection and analysis of data

relating to variations caused by the cultural background of the participants.

The design task for the participants is to modify an existing 3D computer model of a

building, so this is relatively mature design rather than conceptual design. The latter has been

studied in the work of (Alel et al., 2010). The software used is Autodesk Revit Architecture, with

which none of the participants was familiar, so all of them had to be trained before they could

undertake the task. The training was a separate session of around two hours. The training was

identical for all the participants.



47

At the start of the task, the participants were given identical verbal and written instructions on

a sheet of paper. They were also not told anything about the objectives of the experiments. At the

end of the experiments, they were instructed not to discuss what had occurred with anybody else to

avoid “contamination” of potential future participants. Each experiment consisted of four separate

sub-tasks which the participants were required to complete.

During the experiment the participants sat together and used a single computer to modify the

3D model. The participants were given the freedom to choose their seating position in relation to the

computer (see Figure 1). In particular, which participant predominantly controlled the computer

mouse (designated as User1) was a matter to be settled between the two of them.

The participants were videoed for the full session and the complete computer interaction and

audio track was also recorded. Once an experiment had been completed, the various data streams

were synchronised and joined together using Video Studio 12 (see Figure 2). A detailed transcript of

each experiment is then produced.

Figure 1. An FTF session

Figure 2. A Video Studio synchronised file for the experiment, showing video and screen capture



48

The analysis of the results is undertaken by noting details such as work and non-work related

words and productivity, from the transcript of the experiments. The NVC movements such as

illustrator, regulator, etc. are also carefully recorded from observation (see recording form in

Appendix 1). The form records evidence and amount of the five main NVC movements (with many

subdivisions for each movement) separately for each user. All of these measurements were then

calculated for each individual and aggregated for the team performance.

Twenty experiments have been undertaken with 40 separate participants working in pairs.

The time allotted for undertaking the task was 35 minute, at the end of which the participants were

stopped even if the task was still incomplete. As will be shown below, much of the analysis involves

the experiments being divided into seven time intervals of five minute each.

The analysis allows the comparison of the performance of the individuals in each experiment

and a comparison of how each pair performed in comparison to the other teams. As the participants

have varying levels of experience from expert to novice, this allows some interesting inferences to be

drawn about how the various groups perform. The performance comparisons include the following.

1 Team and individual productivity

Productivity has been calculated by allocating points for completion of various aspects of the

task and sub-tasks. Each experiment consists of four sub-tasks and each sub-task has five sections.

Generally each sub-task is allocated 2.5 points (with 0.5 points each for the five sections in a sub-

task) so a maximum of ten points is possible for each team.

2 Number of words

The number of words for each team and each individual is regarded as an indication of the

level interaction in the team, and also indicative of whether a team member is more dominant.

Likewise the number of non-work related words gives an indication of how effective and task

focussed the participants were.

3 Correlation of NVC movements with team productivity

An observation form (Appendix 1) has been used to record all the body movements by users

during the experiments from careful analysis of the video recording. The NVC analysis is as

follows.

• Record and compute statistics for the NVC elements types such as illustrators, regulator, etc.

• Find the relationship between NVC and both team and individual productivity.

• Evaluate the link between culture and expertise and its impact on team productivity.

The video from each experiment has been processed using Corel Video Studio 12 so careful

playback can allow the spoken words and NVC to be transcribed. Table 1 gives a short extract of

such a transcription. In order make the analysis of the transcription easier, each statement by a

participant has been assigned to one of a number of categories using a coding system as given below

Table 2.



49

Table 1: Extract of a transcription

Speaker

Turn

Time

(Sec)

Line

number

Speech Comments

User1 4 1 [[ hello how are you, are you fine]] Non- work

related words

User2 10 2 [[hello thank you {ssss} I am ok ]] Non work

related words

User1 15 3

4

5

We will discuss of them step by step

and we change during our progress

look at the computer

User2 agree

and say yes

and {H.N}

User2 10 6

7

Exterior wall discussion can be divided

for many divisions

User2{E.Y.C}

to User1

User1 4 8 What is this? User2

emphasizing

User2 12 9 But this is block and this is brick, what

is your opinion

User2{Int}

User1

User1 10 10 Let us discuss each one individually

You said the brick(48) cm.

User1{TT}

User2 7 11 I think that is funny idea, is it! User2{RR}

User1 10 12 This window or{ssss) ok ok User1{WW}

Table 2: Coding system used in transcription

Code Meaning Code Meaning

{ssss} Slight pause during the

conversation {That is bad

idea}

Softly spoken words

{That is

good idea}

Words in bold said

emphatically {RR} The user was relaxed

{E.Y.C} Eye contact for the speaker {EM} The user was embarrassed

{That is

great}

The user is smiling or

laughing [[hello how

are you, you

are fine ]]

Non-work related words

{H.N} Showing agreement by head

nod {Cof} The user was confident

{yes……but

..ok}

The user was worried {WW}. {Int} One user interrupts the

other

{EE} The user was emphasizing a

point during the discussion {TT} The user was tense

4. NON-VERBAL MOVEMENTS ANALYSIS

The analysis of NVC during the experiments was undertaken using techniques based on the

work of (Efron, 1941; Exman & Friesen, 1969; Boday Language, 2010). The techniques are not new

and have been established and used by many previous authors. Table 3 describes the main types of

NVC used in the analysis. (More information on NVC movements is found in Appendix 2.)



50

Table 3: Body language classification.

No. Movement

Category

Definition

1 Emblems Emblems are defined as having a direct verbal equivalent to the

particular movement such as good bye being substituted by a wave or

hand shake.

2 Illustrators

Illustrators are defined as a group of movements that can be used to

describe a specific event or illustrate a specific idea, for example, the use

of hands during speaking. Typical examples include pointing out

something, using hands for descriptions (e.g. making gestures while

speaking), adding of emphasis to speech by movements, etc. Illustrators

can be used for many purposes, for example:

• emphasising speech or individual words by movements;

• emphasising speech by changing voice tone;

• explaining ambiguous words by body movements;

• reflecting emotion by body movement; and

• attracting attention to the speaker by movement.

3 Adaptors Adaptors are defined as movements that help the participants to adjust

to the working environment, or to satisfy some personal need, e.g. for

comfort or security. These could be necessary movements in the

progress of work, e.g. moving of upper body from talking to computer

typing. These movements could also reflect the emotional state of the

person during the conversation, e.g. wriggling on the chair, scratching,

chewing a pen, etc., and they do not necessarily have a communicative

meaning, i.e. they could be just a display of personal habit. It is also true

that a particular adaptor action in a specific instance could be placed in

another category. Some examples of adaptors include:

• a vertical head nod to accept ideas;

• a horizontal head shake to refuse ideas;

• hand(s) on cheek, chin, head, forehead, and interlocking fingers,

meaning thinking before making a decision;

• folded arms across the chest to show discomfort with the speaker’s

idea;

• hand(s) covering mouth meaning embarrassment;

• hand(s) on thigh meaning relaxation;

• touching the nose while speaking, showing uncertainty or

hesitancy;

• touching the nose during listening, showing thoughts elsewhere and

not interacting with the speaker; and

• wriggling on the chair, to mean anxiety.

4 Regulators Regulator are defined as movements which are used to control the

discussion, for example, e.g. a raised hand to interrupt the speaker or

draw attention, or waving gesture to show more speech is forthcoming,

or threatening with the index finger, etc.

5 Affect

display

Affect display are defined as motions that convey emotion, for example

smiling or laughing, etc.



51

5. EXPERIMENTAL DETAILS

As described above, each experiment involves two participants who are required to undertake

a series of sub-task involving an existing Revit Architecture model of a building (Figure 3). The

building design contains some deliberate flaws and inadequacies and the participants are asked to

address and improve specific aspects of the building. How and exactly what is to be amended is left

open to the participants.

The Revit model contains full details of the project such as geometry, materials and a bill of

quantities. Limitations and constraints have been placed within the model in order to make the

participants consider factors such as costs, time and quality. The model itself represents a small

Middle Eastern residential building consisting of two floors with three bedrooms, a kitchen, a bath

room, living rooms and a W.C. In addition to that there is a large garden surrounding the building

which is contained within a fence.

Figure 3. External view of the Revit model

The Revit model has been divided to ten sub-tasks or “worksets”. (A workset in Revit is a

specified sector of the model, each sector represents various elements of the scheme such as

electrical, Mechanical, etc. For example the “Exterior wall” workset would contain walls, windows

and doors.) Furthermore, the jobs for each workset are arranged so that they are independent of

other worksets. This has been done so that each time the experiment is run, the subjects considered

by the participants are similar and therefore comparable. It also keeps the time for each experiment

within reasonable bounds.

As outlined above, the participants have been divided into groups of two people. The

participants are volunteers from different world regions and cultures such as Far East, Middle East,

Continental Europe, as well as some from United Kingdom, and are mostly PhD students. Most of

them have experience in aspect of design and the construction industry. It has been found useful to

classify the participants into groups according to their level of experience as in Table 4. Sampling in

the experiments was such that each of the five categories in Table 3 had four teams, thereby giving

an even spread of data across the range of experience.



52

Table 4: Type and level of team expertise

Expertise Experience

Level Description

4 Expert-expert Both users have a high level of experience in the design

field (typically > 5 years post Engineering graduation)

3 Expert-junior

expert

One user has a high level of experience, but the other user

has only a moderate level of experience (typically an

engineering graduate but with < 5 years of experience)

2 Expert-novice

One user has high level of experience but the other is a

novice who is not an engineering graduate, and has not any

experience in construction or design.

1 Junior expert-

novice

One of the users has moderate experience but the other is a

novice.

0 Novice-novice Both users have no experience in construction or design.

6. ANALYSIS METHOD

The analysis of the results can be classified into two groups with these being: 1) the impact of

the level of experience of the participants on NVC and productivity; and 2) how NVC varies

according to cultural background and the impact this has on productivity.

Grouping of results according to Experiences

Analysis of the results revealed that there are two distinct types of behaviour according to the

experience levels of the participants and therefore the results are presented in two categories as

described below.

• Category A (8 experiments) consists of participants who have similar level of expertise, e.g.

expert-expert or novice-novice

• Category B (the remaining 12 experiments) have participants in teams with different

experience levels e.g. expert-novice, novice-junior expert, etc.

Grouping of results according to cultural background

It so happens that many of the participants came from an Arabic background and thus it was

possible to examine the possible effect from teams being made up of members of the same cultural

background. The second grouping thus differentiates the teams as follows.

• The seven Arabic-only teams had their participants from Iraq and Kuwait.

• The non-Arabic grouping consists of 11 teams with participants from a diverse background

(e.g. United Kingdom, Malaysia, Greece, Lithuania, Nigeria, China, etc.).

Two remaining groups had pairings of an Arabic with a non-Arabic background, and so these

two are excluded from this part of the analysis.

7. EFFECT OF GROUP EXPERIENCE

It has been observed in earlier work (Hatem et al., 2012) that whether the team members had

similar levels of experience or not had an impact of how they behaved. This particular aspect is now

examined with respect to NVC, where the teams have been divided into two categories. The results

in Figure 4 show the occurrence of the five types of NVC movements for both teams with similar,

and dis-similar, experience levels. The results are plotted separately for the two participants, i.e.

User1 and User 2.



53

Figure 4 shows that the participants exhibited principally “illustrator” movements alongside

their speech, although “adaptor” movements were also noticeable. What is also very apparent that

while there is little to distinguish between the two participants when the pairings have similar level

of expertise, there is a very significant difference when the pairing have dis-similar (i.e. uneven)

experience. In the latter category, User1, the self-designated participant who naturally mostly

controlled the computer mouse and interfaced with the Revit model, is clearly the more “animated”

in the collaboration between dis-similar pairs, with high number of both illustrator and adaptor

movements, and at the same time, User2 is correspondingly more inert. The level of activity for

these two users is respectively some 25% above, and 40% below, the near-identical average of the

pairings with similar experience. Since the dis-similar pairings include both participants with all

levels of expertise, it is clear that the behaviour observed is not is related to the experience level as

such, but to the more controlling and domineering personality/behaviour of one of the participants,

and in all but two cases, the more domineering participant was the one with the more experience,

even when about half of these pairings were of complete strangers.

Figure 4. Average NVC movements in pairings with similar (Category A)

and dis-similar (Category B) level of experience

Figure 4 shows that the curves for User1 and User2 for the dis-similar grouping form upper-

and lower-bounds to the curves of the similar grouping. The participants were all volunteers and the

only aspect of planning in the pairing was in trying to achieve an even number of the different teams

(i.e. a range of expertise). Who was paired with whom was dependent also on availability at the

time. It is therefore interesting that the behaviour of a certain participant is dependent on the nature

(i.e. experience level) of the other team member, because when an expert is teamed up with a lower

experience person, they almost invariably made themselves User1 and thus their behaviour is

represented by the upper-bound curve in Figure 4, while the same expert teamed up with another

expert, would then, on average, exhibit a lower number of NVC movements. The behaviour, in

terms of NVC, of a person is thus as much affected by the experience level of the other team

members, as by their own.

It is also useful to compare these results with those for verbal communication of the same

participants in the dis-similar pairings, where User 1 was responsible for 66% of total words spoken.

This is also reflected in the results for productivity, with User1 being responsible for 67% of the total

team productivity. It would seem that the greater use of NVC was linked to the fact that User1 spoke

more and did more.



54

It contrast, where the pairings have similar levels of expertise (whether it be expert-expert or

novice-novice), the two users have very similar usage of NVC, with User 1 being responsible for

52% of the number of words, and 53% of the team productivity. The overall picture therefore for the

category A experiments, is that the behaviour and performance of the users is very similar and one

can infer that this is due to them having similar levels of expertise.

7.1. Illustrator component of NVC A finer breakdown of the different illustrator movements is shown in Figure 5. For Category

A where both users have similar experience level, User1 (averaging 192 movements) is slightly more

active than User2 (averaging 177 movements) in most types of illustrator movements, but the

different is small. About half of the movements come from intensive staring at an object, which

emphasises some aspects of the object for greater attention.

Where the two participants have dis-similar experience levels, User1 was much more active

than User2 both overall (by a factor of two) and across every type of illustrator movement. Actually,

the corresponding ratio for number of words spoken is also nearly twice (1.9), so clearly User1 made

much more use of NVC to emphasise his/her spoken words. The data here also suggest that the

illustrator component of NVC movements is directly proportional to the number of words spoken.

Figure 5 also shows that the difference between User1 and User2 is most pronounced in the first type

of action, where it is seen that User1 assumes more overall control of the teamwork by speaking

more, and controlling what points/objects are discussed.

Figure 5. Distribution of average number of illustrator actions

for different users per experiment

7.1.1. Temporal distribution of illustrators Figure 6 shows the averages for illustrator movements for User1 and User2 as distributed

over the time of the experiment. Where both participants in a team have similar levels of experience,

User1 is overall marginally more active than User2, this being most apparent in the middle part of

the experiment, but both participants are more equal in the last 15 minute.



55

Conversely, for teams where members have dis-similar levels of experience, after a fairly

similar first five minutes, there is substantial difference between User1 and User2, with User1 having

an average twice that of User2. As stated above, for all these 10 of the 12 teams, User1 turned out to

be the team member with the greater experience level. Since illustrator movements are used to

improve clarification of what is being described, then the more experienced team member had

clearly taken on the role of being the guide, instructor and leader. Indeed, in earlier study (Hatem et

al., 2012), User1 for these teams was also found to be domineering. Where the two members of a

team were of similar experience level, there is less need for one to explain to the other, and

correspondingly, the number of illustrator movements is very similar for both team members.

For all teams, to varying degree, there is an initial low number of illustrators, which is then

followed by a rapid increase in the second interval, leading to a peak of movements around the half-

way point before a significant decrease in movement in the last 15 minute. In earlier study (Hatem et

al., 2012), the same pattern is also observed for the temporal distribution of the number of words

spoken. This pattern is consistent with the observation that initially, after social greetings, the team

spend time reading and studying the brief, and hence the first few minutes are somewhat quiet. The

teams thereafter enter a stage of highly active working, in which the number of illustrator

movements (and number of words) are at a high level, and this lasts for 15-20 minute. After this, it

is observed that most teams seem to begin to get distracted, or to enter a lull with an appearance of

self-satisfaction on progress, or to disagree about the work. For the remainder of the time, most

teams then increase their activity just before the end of the time limit. These four observed phases

are each reflected in the number of illustrator movements (and number of words).

Figure 6. Temporal distribution of illustrator actions for team members



56

7.1.2 Team productivity and illustrator movements Apart from evaluation of the processes in the design collaboration, it is also important to

evaluate the product of the collaboration. In this design, productivity is accumulated through

completing the different tasks to improve/correct the model. It is therefore useful to compare the

progress of productivity in the experiments, with the accompanied processes over the duration of the

experiments. Figure 7 shows the number of illustrator movements (in each 5-minute interval)

superimposed on the productivity achieved in those intervals.

It is clear from Figure 7 that there is a close correspondence between shape of the bar chart

for productivity and the curves for illustrator movements. Both of these start low, and increase to a

high level over the “main production” stage, decreasing in the “distraction” stage, before increasing

again in the “final push” stage. Spearman’s rank correlation (Storch & Francis 1999) has been

calculated to show correlation values of 0.72 and 0.82 for the similar, and dis-similar, groupings

respectively, which shows a reasonable correlation between illustrator movements and productivity.

It can be argued therefore that illustrator movements in FTF collaborative design are desirable in that

they accompany, and even promote, productivity.

It should be noted that while the four stages in the experiments have been identified in

Section 7.1.1 arising from the level of activity, they are now seen to equally apply to the level of

productivity. There is therefore a correspondence between the rate of illustrator movements and the

rate of productivity.

It is also notable that the group with similar experience level have slightly higher number of

illustrator movements (369, compared to 331) and higher productivity (6.8, compared to 6.3), but the

difference is small. In earlier work (Hatem et al., 2012) it was found that productivity was largely

influenced by the level of experience. In this work, both groupings have a range of experience such

that in effect, the average experience for the two groups is actually similar, and hence the level of

experience has little impact in Figure 7. Therefore, the lack of any big difference between results for

the two groups in Figure 7 show that whether the teams are made up of members with similar or dis-

similar level of experience has little effect on the temporal distribution of both illustrator movements

and productivity.

7.2. Adaptor component of NVC Figure 8 shows the average distribution of the different adaptor actions recorded in the

experiments. For teams made up participants with similar experience levels, it is User2 (41) who

made slightly greater use of adaptors than User1 (31). The main actions are those relating to “hand

on cheek/chin/forehead” (i.e. thinking), “changing position in chair” (i.e. principally re-positioning

for work, but sometimes for personal comfort) and “vertical head nod (i.e. agreement). Interestingly,

there is almost no use of the “head shake”, which is a fairly abrupt signal of disagreement, in this

group. All of these point to the pairings working well together and of similar strength personalities.

Generally User2 exhibits signs of being slightly more active in terms of thinking, evaluating and

decision making.

On the other hand, for User1 displays much greater number of adaptor actions (75) than

User2 (38) in the pairings with dis-similar experience level. This is most pronounced in the action of

“changing position in chair” (29.5 vs 6.8), which, in these experiments, are principally due to re-

positioning of the body, e.g. as a participant moves from the computer to face the other user, etc.

While it can be expected that User1 might re-position more



57

Figure 7. Temporal distribution of productivity with illustrator movements

frequently, since they are the controller of the computer mouse (and hence the computer), it is also

notable that the User1 in the other grouping (i.e. teams of similar experience level) in fact showed

fewer such action than the User2. The much larger number of body re-positionings for User1, in

teams with dis-similar experience, is in fact due the User1 (who is typically the more experienced of

the pair) not only doing, but also explaining, the work, i.e. dominating the work. This is consistent

with the much higher occurrence of User2 passively agreeing by the use of head-nods, User2 also

shows embarrassment by covering of the mouth while speaking, User1 has nearly 21 times more

head-shakes (i.e. a coarse rejection gesture) than the other participant, and User1 has more thinking

actions (28 vs 13).

Figure 8. Distribution of average number of adaptor actions




58

The observations concerning User1 are further reinforced in Figure 9 which shows the

distribution of adaptor movements over time. User1 of the teams with dis-similar experience level

again stands out prominently by displaying many more adaptor movements, while the other three

curves overlap each other to the point that they are not particularly distinguishable. Furthermore,

while User1 typically has higher number of adaptors, most of these occur in the main-production

stage. User1’s adaptor movements are thus related to work, even though the dominating

characteristics might arguably also be having a negative impact on the work.

The impact of these movements can be deduced by observing the productivity over time (see

Figure 10). Generally, it can be seen that there is good direct correspondence between the number of

adaptor movements and the level of productivity, with Spearman’s rank coefficients of 0.89 and 0.85

for teams with similar, and dis-similar, level of experience respectively. In this respect, the adaptor

movements are similar to the illustrator movements in that they both accompany, and possibly

positively affect, productivity. This is perhaps more noticeable from observing the data due to User1

of teams with dis-similar experience level, since this User1 singularly dominates the number of

adaptor movements, and this happens most especially in the last two-thirds of the “main production”

stage (i.e. time periods 15 and 20 minute), and it is in these two periods that the highest

productivities are observed.

Figure 9. Temporal distribution of adaptor actions for team members



59

Figure 10. Temporal distribution of productivity with adaptor movements

7.3. Regulator component of NVC The illustrator component of NVC are movements which accompany, support or enhance

verbal communication, but they are generally not made in place of verbal communication. On the

other hand, regulator movements can be made independently of any accompanying verbal

communication. Figure 11 shows the number of regulator movements, for the various types of

regulator actions observed in these experiments. For all users, the main regulator action is the use of

a hand to interrupt the speech of the other speaker. This was used more by the group with similar

experience (evenly between the two users) but in the group with dis-similar experience level, it was

User1 who prominently interrupted with a hand action while User2 was speaking. This is consistent

with the earlier observation of the more experienced member in an uneven team (User1) being

domineering. Although the other regulator actions were infrequent, it is notable that User1 in the

dis-similar experience grouping was also prominent in the use of the hand to both reject User2’s

suggestions, and cut short User2’s speech. Again, this is consistent with domineering behaviour.

Figure 11. Distribution of average number of regulator actions




60

Although the number of regulator movements is small, there is a significant message when

that number is compared to the productivity, see Figure 12. Unlike Figure 7, the curves for these

NVC movements tend to run counter to the bar chart for productivity. There is generally a reduction

in regulator movements in the main “production stage” where productivity is high, and a sharp

increase in regulator movements in the “distraction stage” where productivity is low. Unlike the

illustrator movements which enhance and support verbal communication, and are a positive

influence on productivity, the regulator actions observed are almost entirely ones that disrupt verbal

communication, reduce agreement, and delay decision taking, all of which in turn reduce

productivity.

Although both groups have a similar pattern of regulator movements across the four stages of

an experiment, the teams with dis-similar experience level have a continually high level of regulator

movements, compared to teams with similar experience level. This would suggest that teams with

similar level of experience (whether expert-expert or novice-novice) are able to work more

harmoniously.

Figure 12. Temporal distribution of productivity with regulator movements

7.4. Affect display movements Overall, there were few movements displaying emotions, and generally, User1 showed more

display movements, and especially for teams with dis-similar experience level (see Figure 13) This

was most pronounced in the category of “laughing or smiling” which shows User1 to have been

more at ease and relaxed.



61

Figure 13. Distribution of average number of affect display actions


8.0. Effect of Similarity in team cultural background

The results in this section are presented to show how cultural background has an effect on the

usage of NVC and productivity. It is known (Yammiyavar et al., 2008) that NVC is culture

dependent, and it is surmised that where a team is drawn from participants of a similar background,

they would be more at ease with, and thus display more, NVC (Vatrapu & Perez-Quinones, 2006).

Since NVC is also such a significant component of communication, then it is reasonable to expect

that teams with more free-flowing NVC (i.e. teams from similar cultural background) would also

consequently have the better productivity.

Of the participants who took part, the largest ethnic group turned out to be Arabic

participants, and it was thus convenient to sort the teams according to those that had only Arabic

participants, and those with only non-Arabic participants. It is also convenient that, despite the

several nationalities involved, the Arabic group can be considered a single cohesive cultural

grouping (as opposed to, say, an Asian, or a European grouping) due to the commonality of socio-

religious and language background. The Arabic group contains 7 teams and the non-Arabic group

consists of 11 teams. The remaining two teams were of mixed ethnicity and so are not included in

this analysis. Table 5 shows the results in terms of productivity for each group, where it can be seen

that there is only a small difference between the averages of team productivity (6.63 points vs 6.35)

for the two groups.



62

Table 5: Group culture and team productivity

Arabic

group

Productivity

(points)

Non-Arabic

Group

Productivity

(points)

1 8.25 5 5.50

2 6.75 7 8.50

3 5.00 8 6.00

4 8.50 9 3.625

6 7.75 12 5.625

10 3.25 13 6.875

19 6.875 14 8.375

16 7.00

17 6.125

18 6.50

20 5.75

Total 46.375 69.875

Average 6.63 6.35

In the above, no account is yet taken of the level of expertise within the two groups.

However, all the preceding results in this paper have shown the level of expertise to be a significant

factor so the results in Table 5 should also be interpreted in view of the level of expertise as well. To

achieve this, the various levels of expertise have been allocated as shown in Table 4. It is recognised

that this is a relatively coarse ranking and thought was also given to using “years of experience” as a

more precise measure, but it was difficult to determine what exactly was each participant’s years of

relevant experience and so it was decided to use the approach presented here which avoids the

potential illusion of higher precision.

Figure 14. Team Productivity for teams with similar and dis-similar background

Figure 14 shows the relationship between commonality of cultural background and

productivity, across the range of experience level, where there is clearly a higher productivity from

teams with greater experience. A linear best-fit line has also been provided for each group to help

compare between the two. It is clear when level of experience is taken into account, while there is

little difference for teams with high level of expertise (expert-expert teams score around 8.3/10

regardless of whether the team members have a common background). However, while a decreasing

level of experience is expectedly accompanied by a decreasing productivity, surprisingly, the teams

with a common cultural background (Arabic Group) have a steeper decrease (slope of 1.281) than



63

teams from a dis-similar background (slope of 0.661). It would be expected that having a similar

background would aid communication and collaboration, and thus provide an advantage for

productivity, but the current results do not show this. These results might have been distorted by

having too selective a small group (i.e. Arabic group) or that, for these purposes, Arabic participants

from different nations could not actually be considered to originate from a similar background, or

that two Arabic participants who would naturally speak to each other in Arabic found speaking in

English for these experiments more limiting than two non-Arabic (non-English speaking)

participants would.

Table 6: Average non-verbal movements in the two cultural groupings

Movements Arabic Group Non-Arabic Group

Illustrators 381 339

Regulators 31 33

Adaptors 105 91

The difference in productivity in these two groups is also not apparent from consideration of

the NVC movements. Table 6 gives the average number of NVC movements per experiment made

by the two groupings, and there is no significant difference between the two. Overall, the Arabic

group made about 12% more NVC movements than the non-Arabic group, and that is predominantly

in illustrator movements (11% more) which form the bulk of all the movements observed (>70%).

When NVC is considered across the different levels of experience for the two groups (see Figure 15)

the pattern is very similar. This is best seen in the best-fit linear lines for each of the three NVC

movements for both groups, which lie close to each other. Apart from the slightly higher level of

movements in the Arabic group, there is actually little difference in the behaviour between groups

from similar and dis-similar background.

Figure 15. Frequency of the three most frequent NVC movements for the different teams with

similar and dis-similar background (note that the frequencies for Regulators and Adaptors are plotted

with exaggerated scale to improve clarity)



64

Aside from the similarity in NVC movements between the Arabic and non-Arabic groups, it

is interesting to note that there is a definite relationship in NVC with level of expertise. There is an

increase in Illustrator and Adaptor movements with increase in expertise, and a corresponding

decrease in Regulator movements. This could be because with a higher level of experience, the

participants are more confident of what to do, and are thus more active (more Illustrators) and

assured (fewer Regulators).

9. VERBAL AND NON-VERBAL COMMUNICATION

The verbal and non-verbal characteristics have been examined for the two groups with

similar (Arabic) and dis-similar (non-Arabic) cultural background. Table 7 shows the average

number of words per experiment spoken by each of the two groups. The ratio of words to illustrator

movements are 5.68 and 6.11 for the Arabic and non-Arabic groups respectively, which shows that

although the Arabic group spoke more, their number of illustrator movements is not proportionately

as high. Table 7 also shows that, since the number of work-related works is about the same for the

two groups, the increase in the total number of words spoken by the Arabic group is entirely due to

their greater use of social conversation (i.e. non-work related words), which arguably requires few

illustrator movements. A more meaningful ratio would be based on the number of work-related

(instead of total) words, and the ratios are thus 5.19 and 5.79 for Arabic and non-Arabic respectively,

which tell us, word-for-word, Arabic workers employ more illustrator movements.

Table 7: Type of words in two groups

Average number of

words/movements

Arabic Group Non Arabic group

Total words 2164 2071

Illustrator movements 381 339

Work-related words 1976 1962

Non-work related words 188 109

10. CONCLUSIONS

An empirically based study of the nature and impact of NVC for people who are

collaborating on a typical construction industry design task has been undertaken. The results

presented are averages for a reasonably large number of experiments (when compared to sample size

in the literature) and hence are acceptably robust. The manner in which the experiments have been

conducted is rigorous and care has been taken to make them as scientifically valid as possible.

1. The analysis of the results for the 20 experiments resulted in the identification of two clear

categories of behaviour according to similarity of experience level. There were no significant

differences in the number of NVC movements between the participants in teams where the

members are of similar experience (Category A, 8 experiments, see Figure 4). Conversely,

there is a very big difference in the number of NVC movements (and individual productivity)

when the participants making up a team have different levels of experience (Category B, 12

experiments), directly resulting in the more experienced participant dominating the execution

of the task.

2. All participants exhibited more illustrator and adaptor movements than other type of

movements.



65

3. There is a relationship between the verbal and non-verbal communication. In Category A,

User1 spoke only about 8% more words than User2 and at the same time, there was no

significant differences between them in NVC movements. For Category B, User1 spoke about

twice as much as User2, and there was also significant differences in NVC movements

between them.

4. For all participants, about half of the illustrator movements come from the action of staring

at/intensive study of an object. From observation of the distribution of productivity and the

number of NVC movements over time, there was good justification to see the collaboration as

a progression over four phases of work (“initial study”, “main production time”, “distraction,

lull, disagreement” and “final push”, see e.g. Figure 6).

5. For all teams, there is close correspondence between the level of productivity and the amount

of illustrator movements over the whole of the seven time-periods in each 35-minute

experiment. It is clear illustrator movements are desirable for good team productivity.

6. The pattern for adaptor movements was similar to illustrator movements: both users have

similar number in Category A, but User1 used about twice the number of movements

compared to User2 in category B. The main actions found were “changing position on chair”

(principally by User1 in Category B), “hand on chin, cheek and forehead”, and “vertical head

nod”. Similarly, as for illustrator movements, there is also a close correspondence between the

level of productivity and the number of adaptor movements over the different time-periods in

each experiment, showing that adaptor movements are also useful in achieving good

productivity.

7. Conversely, regulator movements correlated negatively with productivity, where the main

action was “using hand to interrupt of the other speaker”. Both users displayed similar amount

of regulator movements in Category A, but User1 showed just over three times more regulator

movements than User2 in Category B.

8. Team productivity was found to be greatly more affected by experience level than by cultural

differences in the team. While the teams made of Arabic people had on average slightly higher

productivity (by about 4.4%) than the non-Arabic group, productivity for the former was also

twice more dependent on the level of experience, than for the latter group.

REFERENCES

1. Aboudan, R., Beattie, G. (1996). Cross-cultural similarities: The deep relationship between

gestures and speech which transcends language barriers. Semiotica, 111 (3/4), 269-294.

2. Alel, M., Kwan, A., Miles, J. (2010). Gains and losses of information when collaborating

remotely over a network communication in the construction industry. In W. Tizani (Ed.),

computing in Civil & Building engineering (pp.6).

3. Nottingham, UK: Proc. ICCCBE XVII.Argyle, M. (1975). Bodily Communication. New

York: International Universities Press Inc.

4. Argyle, M. (1994). The Psychology of Interpersonal Behaviour, fifth Ed. London: Penguin

Books.

5. Axtell, R. E. (1993). The Bestselling Guide to International Behaviour, third Ed. New York:

John Wiley and Sons, Inc.

6. Barakat, R. A. (1973). Arabic gestures. The Journal of Popular Culture, 6, 749-792.

7. Birdwhistell, R.L. (1970). Kinesics and Context, Essays on Body Motion Communication.

Philadelphia: University of Pennsylvania Press.

8. Body language. (2010). Body language guide to reading body language signs and gestures-

nonverbal communication, Available at:http://www.businessballs.com/body-

language.htm#arms-body-language [Accessed 21 April 2010].



66

9. Brody, L. R., Hall, J. A. (2000). Gender, emotion, and expression. In M. Lewis, J.M.

Havilland-Jones (Ed.), Handbook of emotions, second Ed. (pp.338-349). New York: Guilford

Press.

10. Chen, G. M. (1995). Differences in self-disclosure patterns among Americans versus

Chinese. Journal of Cross-Cultural Psychology, 26 (1), 84-91.

11. Downing, L., Monaco, N. (1986). In-group/out-group bias as a function of differential contact

and authoritarian personality. The Journal of Social Psychology, 126 (4), 445-452.

12. Efron, D. (1941). Gesture and Environments. Morningside Heights, New York: King's Crown

Press.

13. Ekman, P., Friesen, W. V. (1967). Head and body cues in the judgement of emotion: A

reformulation. Perceptual Motor Skills, 24 (3), 711-724.

14. Ekman, P. (1984). Expression and the nature of emotion. In K. Scherer, P. Ekman (Ed.),

Approaches to Emotion (pp.319-343). Lawrence Erlbaum, Hillsdale, New Jersey: John Wiley

and Sons.

15. Ekman, P., Friesen, W.V. (1996). The repertoire of nonverbal behaviour: Categories, origins,

usage and coding. Semiotica, 1, 49-98.

16. Ekman, P. (1997). Should we call it expression or communication? European Journal of

Social Sciences, 10 (4), 333-359.

17. Feghali, E. (1997). Arab cultural communication patterns. International Journal of

Intercultural Relations, 21(3), 345-378.

18. Frank, M. G. (1997). Some thoughts on FACS, dynamic markers of emotion, and baseball. In

P. Ekman, E.L. Rosenberg (Ed.), What the Face Reveals: Basic and Applied Studies of

Spontaneous Expression Using the Facial Action Coding System (FACS) (pp.239-342).

New York: Oxford University press.

19. Gergle, D., Millan, D.R., Kraut, R.E., Fussell, S.R. (2004). Persistence Matters: Making the

Most of Chat in Tightly-Coupled Work. In Proceedings of the Conference on Human Factors

in Computing Systems, Austria, pp.431-438.

20. Gero, J.S. & Tang. H. (2001). The differences between retrospective and concurrent protocols

in revealing the process-oriented aspects of the design process. Design Studies, 22, 283-295.

21. Gilbert, D. T., Krull, D. S. (2002). Cross-cultural patterns in emotion recognition:

Highlighting design and analytic techniques. Emotion, 2 (1), 75-84.

22. Goffman, E. (1959). The Presentation of Self in Everyday Life. New York: Doubleday

Anchor Books Inc.

23. Guye-Vuilleme, A., Capin, T.K., Pandzic, S., Magnenat, N.T.,Thalmann, D. (1999)

Nonverbal communication interface for collaborative virtual environments. Virtual Reality, 4

(1), 49-59.

24. Hall, E. T. (1984). The Dance of Life: The Other Dimension of Time. New York, Garden

City: Anchor Press.

25. Hatem, W. A., Kwan, A., Miles, J. (2011). Comparison of face to face and computer

mediated collaboration. In Proceeding of the European Group for Intelligent Computing in

Engineering (eg-ice), Twente University, Netherlands, 2011.

26. Hatem, W. A., Kwan, A., & Miles, J. (2012). Comparing the effectiveness of face to face and

computer mediated collaboration. Advanced Engineering Informatics,

doi:10.1016/j.aei.2012.01.001.

27. Kim, M. J. & Maher, M. L. (2008). The impact of tangible user interfaces on spatial

cognition during collaborative design. Design Studies, 29(3), 222-253.

28. Kleinman, G., Palmon, D. (2001). Understanding Auditor-Client Relationships: A multi-

Faceted Analysis. Princeton, New Jersey: Markus Weiner Publishers, Inc.



67

29. Knapp, M.L. (1978). Non Verbal Communication in Human Interaction, second Ed. New

York: Holt Rinehart and Winston.

30. Knapp, M.L., Hall, J. A. (2007). Nonverbal Communication in Human Interaction, fifth Ed.

Wadsworth: Thomas Earning.

31. Kraut, R. E. (1979). Social and emotional messages of smiling: An ethological approach.

Journal of Personality and Social Psychology, 37 (9), 1539-1553.

32. Mason, R. (2003). Teach Yourself, Speaking on Special Occasions. London: Hodder &

Stoughton. Matsumoto, D. (1992). American-Japanese cultural differences in the recognition

of universal facial expressions. Journal of Cross-Cultural Psychology, 23 (1), 72- 84.

33. Matsumoto, D., Yo, S. H., Hirayama, S., Petrova. G. (2005). Development and validation of a

measure of display rule knowledge: The display rule assessment inventory. Emotion, 5 (1),

23-40.

34. McHoskey, J. (1995). Narcissism and Machiavellianism. Psychological Reports, 77 (3), 755-

759.

35. Mehrabian, A. (1981). Silent Message: Implicit Communication of Emotion and Attributes,

second Ed. Wadsworth: Belmont Press.

36. Miller, P. W. (1998). Nonverbal Communication: What A Researcher Says to the Teachers,

third ed. USA: National Education Association.

37. Noesjirwan, J. (1978). A rule-based analysis of cultural differences in social behaviour:

Indonesia and Australia. International Journal of Psychology, 13 (4), 305-316.

38. Rousseau, V., Aubé, C., Savoie, A. (2006). Teamwork behaviours: A review and an

integration of framework. Small Group Research, 37 (5), 540-570.

39. Scott, P. A., Charteris, J. (1986). Gesture identification: Southern African ratings compared

with European Responses. International Journal of Psychology, 21 (6), 753-768.

40. Sumi, K., Moriyama. J. (2010). Analyses of nonverbal behaviour of technology teachers in

the classroom. Hyogo University of Teacher Education, Saga University, Japan, Available at:

http://www.iteaconnect.org/Conference/PATT/PATT14/sumi.pdf [Accessed 15 April 2010].

41. Stempfle, J. & Badke-Schaub, P. (2002). Thinking design team. An analysis of team

communication. Design studies, 23(5), 473-496.

42. Storch, H., Francis, W. Z. (1999). Statistical Analysis in climate Research. Cambridge:

Cambridge University Press.

43. Tooby, J., Cosmides, L. (1996). Friendship and the banker’s paradox: Other pathways to the

evolution of adaptations for altruism. Proceedings of the British Academy, 88 (5), 119-143.

44. Vatrapu, R., Perez-Quinones, M. A. (2006). Culture and usability evaluation: The effects of

culture in structured Interviews. Journal of Usability Studies, 1(4), 156-170. Ververidis, D.,

Kotropoulos, C. (2006). Emotional speech recognition: Resources, features, and methods.

Speech Communication, 48 (9), 1162-1181.

45. Vinciarelli, A. (2009). Capturing order in social interactions. IEEE Signal Processing

Magazine, 26 (12), 133-137.

46. Visser, W. & Maher, M. L. (2011). The role of gesture in designing. Artificial Intelligence for

Engineering Design, Analysis and Manufacturing, 25, 213-220.

47. Von Glinow, M. A., Shapiro, D. L., Brett, J. M. (2004). Can we talk and should we?

Managing emotional conflict in multicultural teams. Academy of Management Review, 29

(4), 578-592.

48. Wainwright P. G. (2003). Teach Yourself, Body language. London: Hodder Headline.

49. Waxer, P. H. (1985). Video ethology: Television as a data base for cross-cultural studies in

nonverbal displays. Journal of Nonverbal Behaviour, 9 (2), 111-120.

50. Wiener, M., Mehrabian, A. (1968). Language within Language: Immediacy, A Channel in

Verbal Communication. New York: Appleton-Century-Croft.



68

51. Yammiyavar, P., Clemmensen, T., Kumar, J. (2008). Influence of culture background on non-

verbal communication in a usability testing situation. International Journal of Design, 2(2),

31-40.

52. M.Sudarsan and Ponramalingam, “Significance of The Training on Enhancing Effectiveness

of Communication Skills In the It Industry” International Journal of Management (IJM),

Volume 4, Issue 4, 2013, pp. 1 - 8, ISSN Print: 0976-6502, ISSN Online: 0976-6510.

Appendix 1

Observation form

Category Movement User1 User2

Emblems Shake hand

Illustrators Staring at a particular object

Explicit eye contact with team members

Pointing at/out a particular object

Using hands to act out/illustrate a description

Moving in chair to new location

Hand movement to reinforce a description

Changing voice/tone to emphasise/enhance a point

Acting out/illustrating a description said by

someone else (i.e. while as a listener)

Total

Adaptor Vertical head nod

Horizontal head shake

Resting cheek/chin on hand

Interlocking hands

Hand on head/forehead

Touching nose

Hand on thigh

Hand covering mouth

Changing body position on the chair

Crossing arms over the chest

Total

Regulators Holding up a hand (to interrupt/interject)

Hand waving (to signify refusal/rejection of what

is being said)

A “thumb up” (to support/agree)

Using hand to stop other people speaking

Using hand to ask other people to wait

Waving index finger (to signify threatening)

Total

Affect

display

Smile or laughing

Looking around (puzzled, speechless, “a bit lost”)

Staring at ceiling

Total

5 THE IMPACT OF NON VERBAL COMMUNICATION ON TEAM ...iaeme.com/MasterAdmin/UploadFolder/THE IMPACT OF...

Documents

Transcript of 5 THE IMPACT OF NON VERBAL COMMUNICATION ON TEAM ...iaeme.com/MasterAdmin/UploadFolder/THE IMPACT OF...