ll‘M—A
‘1‘

- -um

-"v- w":-

 

H

L . “a?“ .3”;

if

 

:7

. v
\
'1.

L»

 

 

THESIS

:1
I I)’T>"7\

5/6,;ut’all

This is to certify that the
dissertation entitled

EFFECTS OF CONTROL AND FLEXIBILITY IN
WORKPLACE DESIGN ON EMPLOYEE’S JOB
SATISFACTION, GROUP COHESIVENESS AND
PERCEIVED PERFORMANCE

presented by

So Young Lee

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Human Environment and Design

 

 

 
   

Major rofessor's Signature

2%“ a1 34/ [4%

Date

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

 

W'
Michigan State
University

 

 

PLACE IN RETURN BOX to remove this checkout from your record.

TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

NOV 0 8 200

{17:31qu

UT'

 

I

V'ZJUVU

DEC 1 (I 2008

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65-p.15

 

EFFECTS OF CONTROL AND FLEXIBILITY IN WORKPLACE DESIGN ON
EMPLOYEE’S JOB SATISFACTION, GROUP COHESIVENESS
AND PERCEIVED PERFORMANCE
By

So Young Lee

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Human Environment and Design

2004

ABSTRACT

EFFECTS OF CONTROL AND FLEXIBILITY IN WORKPLACE DESIGN ON
EMPLOYEE’S JOB SATISFACTION, GROUP COHESIVENESS
AND PERCEIVED PERFORMANCE

By
So Young Lee

In today’s changing workplace, the importance of the role of the physical
environment in supporting organizational goals and changing needs is increasingly
acknowledged. However, many workplace designs have not yet successfully supported
these changing needs, and the balance in working alone and working with others is often
neglected. At the individual level, control over the physical environment is expected to
help employees to cope with change, experience a more stable environment, mitigate the
negative effects of distraction and disturbance from the environment, and thus to have
higher job satisfaction and performance. On the other hand, at the interpersonal or group
level, a flexible use of the workplace is expected to enable employees to have more
instant interactions, to do effective collaborative work both in their proximal and broader
work areas, and thus to have higher group cohesiveness and higher performance.

In this study, four hundred and nine questionnaires were collected from
employees of three manufacturing corporations in Michigan. Approximately 28% of
respondents were intensively engaged in engineering while others (24%) intensively
worked for marketing/sales or finance/accounting.

The results indicate that control over the physical environmental features of the
workplace positively influenced job satisfaction. However, the influence was less than

the influence that control over work had on job satisfaction. With respect to physical

dimensions, being able to control social contacts was the important predictor of job
satisfaction, along with personalization, and determination of organization and
appearance of the workplace. It was found that control played a mediating role in
reducing the negative effects of distraction on performance. According to individual work
styles, those who valued a quiet workplace had a negative effect from distraction on
performance, but those who were less concerned with quietness, the effect of distraction
on performance was positive. The findings also supported the hypothesis that flexibility
is positively related to task group cohesiveness but not to social group cohesiveness.
General flexibility and convertibility were significant predictors of task group
cohesiveness. No significant relationship was found between openness and group
cohesiveness. In order to measure the physical environmental aspects, as well as control
and flexibility, from the user’s perspective, the study investigated the gaps between what
employees perceive and what they expect from their workplace. Based on linear
projection, there was no substantial difference between gap measures and perception
measures to explain work environment satisfaction and job satisfaction. Scatter plots
revealed different patterns between the two measures.

The present study found control and flexibility in the workplace to be important.
Some practical implications were suggested emphasizing that control and flexibility
allow limited resources to be used in ways that permit individuals to work effectively
within a changing work environment. Finally, this dissertation gives suggestions for
future studies including the use of diverse methodological approaches and unexplored

areas.

Copyright by
SO YOUNG LEE
2004

ACKNOWLEDGEMENTS

I would like to express my appreciation to my major professor and dissertation
director, Dr. Dana Stewart for her encouragement and commitment. I appreciate my
committee member’s advice and assistances. Acknowledgements are given to Dr. Susan
Mireley for her guidance of my doctoral program and studies. I am very grateful to Dr.
Lawrence for his help and support throughout my doctoral program. I am also thankful
to Dr. Douglas Stewart for his advice and willingness to serve on my committee. I
greatly appreciate Dr. John Hinks for his insightful advice and recommendations for my
doctoral study.

Acknowledgements are given to Mr. Derthick, Mr. Goslin, and Mr. Weiss for
help with the data collection and for sharing their experiences. I would also like to thank
my friends, Judy Fogle and Dhammika for their support.

I would like to express my appreciation to my mother and father, parents—in law
and family for their love and support. I am very thankful to my son for his love and
understanding. Finally, with words, I cannot express my thanks to my husband for his

understanding, patience over years, encouragements, support, and love.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii
LIST OF FIGURES ........................................................................................................... ix
CHAPTER I INTRODUCTION ........................................................................................ 1
Importance of Study ........................................................................................................ 8
Research Objectives ........................................................................................................ 9
Research Questions ....................................................................................................... 10
CHAPTER II LITERATURE REVIEW ....................................................................... 12
Level of Analysis and Classification in the Workplace ................................................ 12
Roles of Control in Job Satisfaction and Performance ................................................. 16
Effects of Negative Ambient Features on Work outcomes .......................................... 23
Relationships between Flexibility and Interpersonal /Group works. ............................ 25
Satisfaction with the Physical Environment and Job satisfaction ................................. 28
Gaps between Perceptions and Expectations Toward A Physical Milieu .................... 30
CHAPTER III CONCEPTUAL FRAMEWORK & HYPOTHESES ............................. 35
Person-Environment Transactions-Environmental Probabilism Theory .................... 35
Brunswik’s Lens Theory ............................................................................................... 36
The Stress-Behavior Model and Moderating Process ................................................... 38
Models of Control in the Workplace ............................................................................ 42
Models of Flexibility in Workplace Design ................................................................. 45
Gaps between Perceptions and Expectations ................................................................ 47
CHAPTER IV METHODS ............................................................................................. 50
Sample .......................................................................................................................... 51
Instrumentation ............................................................................................................. 5 1

vi

Data Collection ............................................................................................................. 54

Data Analysis ................................................................................................................ 57
CHAPTER V RESULTS ................................................................................................ 62
Demographic Characteristics ........................................................................................ 62
Work Activities ............................................................................................................. 63
Facilities Characteristics ............................................................................................... 64
Effects of Control on Job Satisfaction and Perceived Performance ............................. 67
Reliability Tests and Confimratory Factor Analysis .................................................... 71
Interactive Model of Control, Distraction, and Performance ....................................... 73
Mediating Effect of Control .......................................................................................... 78
Effects of Flexibility on Group Cohesiveness .............................................................. 82
Gaps between Perceptions and Expectations ................................................................ 88
Discussion ..................................................................................................................... 92
CHAPTER V1 SUMMARY AND CONCLUSION ..................................................... 103
Summary of Research Objectives and Literature Review .......................................... 103
Summary of Methods .................................................................................................. 105
Summary of Findings .................................................................................................. 105
Limitation of the Study ............................................................................................... 109
Implication for Future Studies .................................................................................... 110
Practical Implications and Suggestions ...................................................................... l 12
APPENDICES ................................................................................................................ 1 15
Appendix A: Questionnaire ........................................................................................ 116
Appendix B: Regression Assumptions ....................................................................... 123
Appendix C: Covariance Matrices .............................................................................. 128
Appendix D: Command Files for LISREL ................................................................. 136
REFERENCES ............................................................................................................... 141

vii

LIST OF TABLES

Table 1. Proposed Work Environment Classifications ..................................................... 13
Table 2. Levels of Analysis, Facets of the Environment, Process, and Outcomes ........... 15
Table 3. Work Environmental Dimensions and Attributes ............................................... 55
Table 4. Summary of Sample Characteristics ................................................................... 62
Table 5. Summary of Job Responsibilities ....................................................................... 63
Table 6. Summary of Characteristics of Work Activities. ................................................ 64
Table 7. Summary of Facilities Characteristics ................................................................ 65
Table 8. Office Type Arrangement by Job Type .............................................................. 67
Table 9. Regression Analysis of Effects of Control on Job Satisfaction .......................... 68
Table 10. Regression Analysis of Effects of Control on Perceived Performance ............ 70
Table 1 1. Reliability and Confirmatory Factor Analysis .................................................. 72
Table 12. Parameter Estimates for Confirmatory Factor Analysis for Contol Models 74
Table 13. Results for Structural EquatiOn Model. ............................................................ 76
Table 14. Chi—square Difference Test for Low and High degree of Control Groups ....... 77
Table 15. Results of SEM for Comparison for the Models A & B ................................... 80
Table 16. Chi-square Difference Test by Individual Workstyle ....................................... 82
Table 17. Regression Analysis of Flexibility on Task Group Cohesiveness .................... 83
Table 18. Summary of Path Analysis .............................................................................. 85
Table 19. Regression Analysis of Distraction on Job Satisfaction ................................... 87
Table 20. Summary of Differences between Perceptions and Expectations .................... 89
Table 21. Variation explained by Perceptions and Expectations toward Workplace ....... 90
Table 22. Summary of Hypotheses Tests ....................................................................... 108

viii

LIST OF FIGURES

Figure 1. Brunswik's Lens Model ..................................................................................... 38
Figure 2. Eclectic Model of Environmental Stress in the Workplace ............................... 41
Figure 3. Effects of Control on Job Satifaciton and Performance .................................... 42
Figure 4. Model of Control, Distraction and Performance ............................................... 43
Figure 5. Mediating Role of Control ................................................................................. 44
Figure 6. Effects of Flexibility on Group Cohesiveness ................................................... 46
Figure 7. Model of Effects of Control and Flexibility in the Workplace ......................... 47
Figure 8. Gap Patterns between Perceptions and Expectations ........................................ 48
Figure 9. Relationship between Perception and Expectaiton in Satisfaction .................... 49
FigureIO. Model of Control, Distraction and Performance .............................................. 75
Figure 11. Mediating Effect of Control Model A ............................................................. 79
Figure 12. Mediating EffeCt of Control Model B. ........................................................... 79
Figure 13. Path Analysis Results ...................................................................................... 86
Figure 14. Linear and Nonlinear relationship in Perception and Gap Measures ............ 101
Figure B 1. Normal Probability Plots of the Standardized Residuals . ............................ 123

Figure B2. Scatterplots between Control variables and Standard Residuals of Job
Satisfaction .................................................................................................... 124

Figure B3. Scatterplots between Flexibility variables and Standard Residuals of Task

Group Cohesiveness ...................................................................................... 126

Figure B4. Scatterplots between Distraction variables and Standard Residuals of Job
Satisfaction .................................................................................................... 1 27

ix

CHAPTER I INTRODUCTION

The workplace is expected to continue changing rapidly. Information technology
and Internet use for information sharing enormously affect work activities and
workplaces (Challenger, 2000). A majority of today’s office workers are knowledge
workers, managerial, professional, and technical; a shift from clerical tasks (U .8.
Workers by Occupation by US. Bureau of the Census; Robertson,1999) and, in large
organizations, a collaborative work style is most prevalent with only a small percentage
of people working in isolation (Barber, 2001). Such new trends impact people’s attitudes
toward work and the nature of work. The trend is toward a more diverse workforce on
many levels, a breakdown of boundaries, a weakening of hierarchy, simplification, the
passionate pursuit of leisure and personal freedom and control (Barber, 2001). As
organizational environments have changed rapidly, it is more necessary than ever before
that the workplace support the changing needs of organizations, employees and work
processes simultaneously and provide an environmental setting in which employees work
effectively and efficiently. Spreckehneyer (1993) raised the question of how office design
can be effective in reducing worker stress and environmental instability in response to a
high degree of change.

The importance of the role of the physical environment in supporting
organizational goals and changing needs is increasingly acknowledged. Although many
researchers investigated the consequence of workplace changes or the relationships
among physical environmental features and work behaviors, a relatively small amount of

research has been conducted to deal with interrelationships between the broad range of

variables (O’Neill,1994; Donald, 1994). Pinpointing this problem, Pugsley and Haynes
(2002) addressed how little had been done in office design to reflect changing dynamics.

In spite of expectations that open office design enhances communication and
work efficiency, and reduces operating costs (Zalensny & Farace, 1987), traditional open
offices present problems such as distractions, a lack of privacy and high noise levels
(Brooks & Kaplan, 1972; Hedge, 1982; Sundstrom, Town, Brown, Forman, & McGee,
1982), decreased performance (Oldham & Brass, 1979) and lower satisfaction
(Spreckelmeyer, 1993; Brennan, Chugh & Kline, 2002). At the individual level,
workstations and immediate work areas are too ofien uncontrolled and disturbed to be
effective workplaces. Traditional office approaches fail to fulfill the needs for
individuals to work undisturbed and the needs for instant interactions among groups.
Although open office space is considered to be a flexible accommodation, once office
layout had been prepared, the office studied remained relatively static for several years
and failed to be used in flexible ways at either the personal or the organizational level
(Hedge, 1982). In changing workplaces, people often neglect the balance between
working alone and working with others (Brenner & Cornell, 1994; Olson, 2002).
Robertson (2000) addressed that one of the greatest dilemmas was the conflict between
placing people in their own workplace and enhancing communication through the
implementation of team coincidently.

Therefore, in response to these problems and the new changing needs of
organizations, some workplace strategies have been introduced. According to
Spreckelmeyer (1993), one strategy is to increase the quality of the ambient features to

compensate the worker for the increased level of disruption and to let office occupants

have a higher degree of self-determination and personal control in response to change.
Other specific alternative workplace options, such as dispersed work sites, flexible work
hours and locations, adaptable space and furniture, team focus, unassigned or shared
resources, and integrated resources were also introduced (Austin, Bain, Heath, Ratekin,
Reilly, Richert, & Ross, 2001). In a general context, these strategies are to provide more
control for employees and more flexibility in their work environments.

Increased control over the proximal environment provides more stable work
environment for the individuals to cope with negative sources such as distractions and
frequent environmental changes. Control is an essential feature of homeostasis (Fisher,
1989), and “Mankind has a characteristic tendency toward self-govemance or self-
deterrnination” (Angyal, 1949). Perceived control predicts important aspects of
motivational, cognitive and emotional functioning (Skinner, Chapman, & Baltes, 1988;
Lee, Ashford, & Bobko, 1990). The consequences of low control include withdrawal
(Abramson, Seligrnan, & Teasdale, 1978), decreased performance (Bazerrnan, 1982),
stress, reactance, depression (Greenberger, Strasser, Cummings, & Dunham, 1989),
absenteeism, and intentions of turnover (Spector, 1986). Higher control has been shown
to increase employee satisfaction, which is more important as demographic makeup
changes in labor populations (Challenger, 2000). Moleski and Lang (1982) emphasized
the importance of user’s needs. They stated that to be responsive to organizational
changes, user needs research and corporate space planning must be redefined to
recognize the importance of “freedom of choice” in determining the behavioral patterns
and involvement in the planning and research process by individuals and work groups.

Froggatt (2001) proposed that organizations take advantage of giving individuals control

over their workplaces, enumerating individual differences in working patterns within
changing workplaces. The idea that individual control leads to beneficial behavioral
outcomes and enhances the quality of life, often leads to recommendations of individual
controls for lighting, temperature and ventilation (Veitch & Newsharn, 2000).

At the individual level, control over the physical environment is expected to help
employees to come up with changes, feel more stable in their environment, better
mitigate the negative effects such as distraction and disturbance from the environment,
and thus lead to better job satisfaction and higher performance.

On the other hand, at the group level, flexible workplaces need to facilitate
individual work and collaborative work simultaneously. In the changing workplace, a
new type of flexible environment is needed to support a plethora of working preferences
(Pugsley & Haynes, 2002). The concept of flexibility has been emphasized more recently
for the following reasons. First, providing workspaces with the flexibility to adapt to
change is the most critical factor in supporting new work processes and technology (GSA,
1998). Frequent changes of organizational structure and equipment updates require a
more flexible physical environmental structure. Second, a flexible work environment can
easily accommodate various changing activities and works. Third, in order to keep up
with rapid external changes, organizations have increasingly turned to some version of
work teams such as cross-fimctional teams and self-managing work groups (DeMatteo,
Eby, & Sundstrom, 1998). Flexible office design allows employees to shift from working
as individuals to working as teams or vice versa. To support organizational objectives and
attain business success, workplaces need to be versatile for both individual and teamwork

requests. Spaces that enable employees to have more instant communications have been

emphasized (Becker, 1990). Another benefit of flexible design may be reducing the cost
of moves. The average churn rate1 is 44% (IFMA Benchmarks 111 Report, 1997), not only
because a company changes rapidly or its organization moves frequently, but also
because space cannot support the changing needs of a company. According to the
characteristics of the organization, it might be hard to reduce the churn rate, however, the
cost of churn can be reduced (Thatcher, 1995), since flexibility is considered to be a
solution for changing demands. Once a physical environment is built, if it is not flexible,
the rigid structure may lead to increased cost for renovation or reinstallation to
incorporate the organization’s changes. As it has been considered that facilities are the
largest expenses after pay-roll (Cotts, 1998), the cost of the physical environment that is
not flexible to be renovated or modified may be bigger than commonly expected.

In summary, at the interpersonal and group level, a flexible use of space is
expected to enable employees to have more instant interactions, promote collaborative
work both in their proximal and broad work areas, and thus lead to better group outcomes
such as improved group cohesiveness and higher performance. Therefore, if employees
possess a high control over their proximal work environment and high flexibility in their
workspace use, it is expected that they will not only work more effectively as individuals
but will also have more informal and instant interactions in their group work, and thus, a
better balance between individual and group work with less concern over distractions or
disruptions.

Though the concept of control has been explored intensively in the areas of job

design and psychology, and flexibility within the workspace has been strategically

 

I The total number of employee work space moves made in the last year divided by the total number of office
employees in the facility multiplied by 100.

emphasized in workplace design, there is yet a paucity of research done regarding control
and flexibility.

A general review shows inconsistent research findings on the effects of the
physical environment on individual and group work outcomes. This may be due to the
difficulty Sundstrom (1986) indicated in proving the effects of the physical
environmental features, especially on group and organizational outcomes. There is a lack
of conceptualization or measurement of the work environment when measuring the
effects of the physical environment on people at work (Carlopio, 1996). According to
Marans and Spreckelmeyer (1982), evaluation studies in office design have failed to
specify the criteria used to determine the degree to which an environment is successful,
or to provide carefully developed conceptual links between the physical environmental
attributes and various levels of worker responses to those attributes. Even if criteria are
specified, valid and reliable measures are rarely used. In a large amount of empirical
research on the effects of the physical environments on workers, physical aspects were
limited to ambient features. Few of the instruments measuring the effects of physical
environmental features have been used over time or used repeatedly to verify their
validity and reliability. In summary, only a few studies deal with the broader dimensions
of the physical environmental features (Hedge, 1982; Sundstrom, Town, Brown, Forrnan,
& McGee, 1982; Zalesny & Farace, 1987; Carlopio, 1996; Brennan, Chugh, & Kline,
2002) and the common structure of these features was not fully established.

Under these circumstances, the concept of control over the physical environment
and the flexible use of space have hardly been explored and empirical studies rarely have

been conducted. Although control and flexibility are strategically emphasized, from the

view of a practical space planner or a facility manager, the concepts of control and
flexibility are relatively abstract and lack specifics. In some research, control over
physical environmental features has been dealt with as a part of the antecedents of the
concept of control including aspects such as control over work activities, work schedule
autonomy or decision making (MacLaney & Hurrell, 1988; Greenberger etal.,1989;
Sargent & Terry,l998; Ganster, 1989). Becker (1991) emphasized the importance of
control in the workplace and specified that research needs to be done inquiring into the
types of environmental control preferred by employees and types of involvement in the
planning and design in the workplace. Based on a literature review, the results of
studying the effects of control are inconsistent.

Therefore, it is necessary to conduct studies to examine whether physical
conditions can provide Opportunities or possibilities for users to be able to decide various
physical work environment options or to use the physical work environment in various
ways and to use their physical environment in flexible ways under their control. To
understand the impacts of control and flexibility, those concepts should be examined with
other aspects of the physical environment that might play roles in mitigating the
relationships. Too often, empirical studies deal with separated concepts or simple
relationships between the physical elements instead of viewing these as integrated and
interrelated relationships.

Although a large amount of research relies on perception, empirical research
related to physical environmental aspects lacks of methods to reflect the employee’s
perspective. There may be large gaps between employees who use the space and a

planner who makes decisions or designs the space. For the right decision in workplace

design and management, the concepts of control, flexibility and antecedents of physical

aspects should be reviewed from the perspective of the users.

Importance of Study

There are both practical and theoretical reasons for conducting this study. On a
theoretical level, this study investigated interrelationships among variables and attributes
of the physical workplace environment such as control over physical environmental
features and flexible use of workspace and work outcomes, which have not been
extensively explored before. Consequently, exploring the interrelationships of the
physical environment with employee behavior will lead to an understanding of the
dynamic nature of the workplace and provide a structure of analysis for the physical work
environment. This will contribute to the establishment of a more reliable approach for
such studies.

Moreover, results of confirmatory factor analysis of the measurement can reveal a
factor structure and relationships among physical environmental features, and other
organizational environmental features. The testing of results adds theoretical knowledge
in workplace research by understanding the multiple relationships. With analysis and
classification from previous studies, this study can add a different and unique dimension
of the physical environmental characteristics on employee satisfaction and work outcome.

This study is important in the development of a model showing a relationship
between the physical office environment and behaviors and relationships leading to work

outcomes. Such a model can lead to more predictive work habits and enable fiiture

studies to explain more definitive relationships between the physical environment
features, work patterns, and profitable outcomes.

From a practical perspective, the results of research in this area will provide a
facility manager or planner with a better understanding of employee behavior and the
requirements leading to better implementation of the management process, human
resources and space utilization. Improved workspace can directly and indirectly
influence an employee’s job satisfaction, productivity, and well-being. User satisfaction
is recognized as an important factor in the success of a certain firm or an organization and
is regarded as one of the indicators of performance since higher satisfaction improves
morale and reduces voluntary turnoverz. This study will contribute to reducing gaps
between the facility manager and users, and the designer and manager, by identifying
desirable features employees want to have from the work environment. Research results
will provide evidence of how innovative work patterns and physical work environment
solutions can be implemented so that a facility manager can utilize research results to be

proactive in strategic planning in order to cope with changing needs in the workplace.

Research Objectives
The overall purpose of this study is to examine the effects of control over the
physical environment by individuals, and to examine the effects of flexibility on
individual and group work outcomes in the workplace. In order to reach this goal, several

more specific objectives were developed to guide the research.

 

2 A study using meta analysis found an inverse relationship between job satisfaction and voluntary turnover

intention being consistent with previous research (Dole & Schroeder, 2001).

1)

2)

3)

4)

5)

6)

To develop conceptual structures of control, flexibility, and other factors
including distraction, work space adequacy and openness in the physical work
environment and to test the causal relationships in the proposed constructs
(factor structure and indicators) of models for the effects of control and
flexibility.

To investigate the effects of individual control both on the perception of
negative physical environmental features such as distractions and on positive
work outcomes such as job satisfaction and individual performance.

To examine the impact of a flexible workplace on group cohesiveness.

To identify interrelationships of factors of physical environmental aspects that
affect work outcomes such as employee’s job satisfaction, task group
cohesiveness, and perceived performance.

To test the proposed models possessing predictive power in showing the
relationships between physical workplace characteristics and work outcomes.
To investigate the gap between employee’s perceptions and expectations toward
control over the physical work environment, flexibility, and other physical

environmental features in the workplace.

Research Questions

The key questions for the research project are:

1)

Does individual control over the physical workplace affect on job satisfaction
and perceived performance? Does control play a role in mediating negative
effects of distractions fi'om the work environment? Is there a difference due to

degree of control or individual work styles?

10

2) Does flexibility in the physical environment influence task group cohesiveness
and social group cohesiveness? With flexibility, are openness and accessibility
associated with group cohesiveness?

3) What kinds of multiple effects or relationships among the factors that affect
work outcomes do exist? Is the proposed model acceptable?

4) What discrepancies exist between employee perceptions and expectations
toward the workplace in terms of control over the physical environments,
flexibility, and work space adequacy? Which aspects do employees feel should
be improved? Is the gap measure more appropriate for explaining satisfaction

with the workplace or job satisfaction than simple perception measure?

Based on these research objectives, a literature review was undertaken and is

presented in chapter 11. Following the review of literature, the conceptual framework and

hypothesis are discussed in chapter 111.

11

CHAPTER II LITERATURE REVIEW

This chapter brings together previous research findings relevant to the research
objectives and variables, and includes a discussion of how physical environmental factors
affect user satisfaction, group work and performance in the workplace, focusing on the
interactive nature between built environments and human behaviors. Concepts of control
in human behaviors, antecedents of control over physical environments and multi-faceted

aspects of flexibility are also reviewed.

Level of Analysis and Classification in the Workplace
Existing research shows a lack of structure in analyzing the physical work

environmental dimensions in the workplace. This may be due to the broad dimensions of
the physical elements, their overlapping and interdependent nature, or the nature of
difficulty in verifying physical environmental effects with organizational characteristics.
This leads to a paucity of reliable and replicable measurement in workplace design.
Table 1 summarizes the proposed analysis structure of work environment dimensions
including physical environmental dimensions by several researchers. According to the
purpose of the research, the dimensions vary. Carlopio (1996) examined the dimensions
in relation to work environment satisfaction. Evans, Johansson, and Carrere (1994)
examined the characteristics that affect health and well-being. Hartkopf and Lofiness
(1985) developed a list of performance qualities in office enviromnents for improving
satisfaction and productivity. Crouch and Nimran (1989) investigated inhibitory and

facilitative features of office environments from sixty five managers and presented results

12

Table 1. Proposed Work Environment Classifications

 

Sources

Work environment dimensions

 

Carlopio (1996)

Environmental design: lighting, adequacy lighting, air quality, walk
surfaces, work atmosphere, types of facilities, cleanliness

Facilities: restrooms, recreation, eating size, eating cleanliness,
eating pleasantness

Work organization: schedule, work amount, activity/movement pace,
flexibility, work system design, time to complete work,
information quality, information move and store

Equipment and tools: number of tools/machines, efficiency of tools,
effectiveness of tools, effectiveness of machines, efficiency
of machines

Health and safety: accident avoidance, accident reporting, accident
investigation, safety precautions, hazard warnings, hazard
control, handle hazardous materials safety training
received, safety training available, safety training of others,
fire prevention system

 

Evans,
Johansson, &
Carrere (1 994)

<Physical>
Ambient: air quality, chemical hazards, noise, temperature, vibration

Layout and arrangement of space: enclosure (openness), proximity
to others, traffic patterns, boundaries, proxemics

Architectural design: lighting, color, modifiability of space furniture,
legibility, natural elements, meeting place

Ergonomic factors: equipment design, safety hazard, machine
pacing, physical load, physical restraint, automation,
computerized work

<Psychological>
Structural: occupation status, labor market
Interpersonal: social relationships, leadership style, discrimination

Organizational: role conflict and ambiguity, clarity of norms/rules,
monitoring/evaluation, shift work, wage structure, over
time, part time/ full time, opportunities for advancement,
hierarchical structure

Task parameters: mental workload, decision latitude (control),
monotony, possibility to learn new things, fractionation,
interruption, technological dependence, social interaction
requirements

13

Continued.

Sources Work environment dimensions

Hartkopf & Functional /spatial quality

Loftness (1985) Individual space layout quality: usable space, layout efficiency,
access, ergonomics, flexibility, occupancy controls

Aggregated space layout quality: proximities, access,
compartmentalization, flexibilities/growth

Building siting: layout quality: access, public interface/image,
indoor and outdoor relationships

Quality of convenience and services: sanitary, fire safety, security,
electrical, information technology

Thermal quality: air temperature, humidity, air speed, occupancy
factors

Air quality: fresh air, air distribution. restriction of mass pollution
Acoustic quality: sound source, sound path, occupancy factors

Visual quality: ambient light levels, task light levels, contrast and
brightness ratios

Building integrity: quality of mechanical/ structural properties,
quality of physical / chemical properties, and visible
properties

Crouch & Nimran Social interactions: vertical support, horizontal support, quality of
(1989) relationship, accessibility

Physical & ambient conditions: lighting, space, atmospheric
conditions, visual outlook

Utilities: equipment, technology, photocopier, etc
Information & communication

Workplace experiences: privacy, distraction

Hedge (1982) Physical factors:

Ambient environment; temperature, ventilation, lighting,
electrostatic

Office layout, office furnishing, disturbances/distractions

Psycho-social factors: Health, privacy, social environment,
proprietary attitudes

14

with the categories presented in Table 1. Hedge (1982) developed a classification to

investigate employee reactions to open plan office environments and employee attitudes.

The scope of this research is intermediate work environments for individuals and

groups. Control and flexibility in workplace design mainly covers physical work

environmental characteristics and psychological aspects of employees.

Interestingly, Sundstrom (1986) proposed varying levels of analysis with

psychological or physiological processes and outcomes (see Table 2). He showed a

relationship of the physical environmental elements and outcomes.

Table 2. Levels of Analysis, Facets of the Environment, Process, and Outcomes

 

 

 

 

Level of Facets of Physical Key Processes Outcomes
Analysis Environment
Individual Ambient conditions Adaptation Satisfaction
Workers Temperature Arousal Performance
Air quality Overload
Lighting Stress
Music Fatigue
Work Stations Attitudes
Color
Equipment
Chair
Floor space
Supporting Environment
Hallways, Restrooms
Work areas, etc
Interpersonal Work spaces Self-identity Adequacy of
Relationships Differentiation Status communication
Room layout Regulation of immediacy Group
Seating arrangements Self-presentation formation
Furniture Choices in Group cohesion
Building layout communication
Inter-work-space proximity Regulations of interaction
Enclosure of workspaces
Gathering places
Organizations Buildings Congruence of Organizational
Separation of work units organizational structure effectiveness

Differentiation of work units

and physical environment

Source: Sundstrom, E. (1986). Work Places, Cambridge University Press, p8.

15

At the individual level, work outcomes are job satisfaction and performance,
while at the interpersonal or group level, work outcomes are better communication, group
formation and group cohesion. In this study, control is related to the individual level and
proximal individual environment. Flexibility is related to a broader work environment

including the individual workspace regarding interpersonal relationships.

Roles of Control in Job Satisfaction and Performance

In psychology, control is one of the most thoroughly explored and cited concepts
(Baum & Singer, 1980). Yet, the term ‘control’ is not uniformly defined in the literature.
Fisher (1990) defined control as mastery over the environment and said “individuals with
control can act to change or reverse situations which are disliked.” Perceived control by
employees has been heavily studied in job design and participatory decision making
(Spector, 1986). The concept of locus of control3 is different from that of perceived
control. Locus of control is the individual’s expectancy that some individuals attribute
the cause and the control of an event, to be within their power or outside of their abilities
(Roberts, Lapidus, & Chonko, 1997).

Some research showed a positive association between high work control and job
satisfaction, work performance and psychological well-being (Sargent & Terry, 1998;
Greenberger, Strasser, Cummings, & Dunham, 1989; MacLaney & Hurrell, 1988).

Spector (1986) conducted meta-analysis relating perceived control variables to employee

 

3 Rotter (1966) conceptualized locus of control as to be a relatively stable propensity to locate causality for
outcomes either in oneself or in the external environment (Greenberger et a1, 1989). For further review, see

page 32 of Greenberger and his colleagues’ paper (1989).

16

outcome variables. From the results of the reviewed studies, Spector (1989) found that
high levels of perceived control were positively related to a high level of motivation,
performance, job satisfaction, involvement, and commitment. And employees who
perceive a high level of control at work were absent less, had fewer intentions of quitting,
and were less likely to quit. Bachman, Smith and Slesinger (1966) investigated the
relationship among control, performance and satisfaction and measured control over an
office as the “general amount of influence over the way the office is run.” The finding
showed that control was associated with satisfaction but not with performance.

While the importance of control has been recognized for well being, motivation,
satisfaction, and performance, few studies have investigated the concept of control in
terms of physical environment aspects. Investigating the effects of work control on job
satisfaction, McLaney and Hurrell (1988) included physical aspects as a major
component of control. They used multidimensional measures of work control to assess
its influence on task outcomes. The construct of control consisted of task control,
decision control, control over the physical environment, and resource control. They
found that control over the physical environment and resource control accounted for a
variance in job satisfaction regardless of the type of job demands, however, decision
control did not.

In control over the physical aspects, Allen and Greenberger (1980) mentioned
“In the case of the built environment, an individual can experience an increase in the
sense of control by altering, modifying or transforming it in some manner.” Through
personalization of the individual workplace, altering the exterior or interior of a building

or even drastic acts of destruction, people gained control (Allen & Greenberger, 1980).

17

People with a greater degree of control over their immediate environment were more
satisfied than those having a diminished amount of control (Marans & Spreckelmeyer,
1982). Roelofsen (2003) indicated that if users could exercise greater control over the
indoor working environment including temperature, ventilation, and lighting, they had
higher self reported productivity and improvement in involvement in their works and
morale.

Paciuk (1989) investigated the role of personal control in the thermal
environment. She measured the following key variables: 1) two descriptors of objective
environmental properties, the thermal factors and potential control made available by the
situational context, 2) a behavioral measure of instrumental control, 3) a measure of
degree of perceived control, 4) a measure of perceived thermal sensation, and 5) three
outcome measures, thermal comfort, acceptability and satisfaction with the thermal
environment. Perceptions of thermal control were measured with 11 items such as “ I
have complete control of thermal conditions at my workspace,” “I am able to set thermal
conditions at my workspace according to my wishes,” and “It is hard to know how and
when to make use of the different available alternatives for the regulation of thermal
condition at my workspace.” Exercised control consisted of three modes, adjustment,
adaptation and self-removal. Exercised control was measured by the frequency of actions
such as setting controls for the central HVAC system, opening/closing windows,
opening/closing ventilation vanes, adding/removing items of clothing or complaining to
the building’s manager. There was negative association between the perceived control
and the exercised control —adaptive behaviors (The path coefficient between them was -.2,

while path coefficients between the exercised control-manipulative behavior and

18

perceived control was not calculated). Perceptions of control were found to moderate
the effects of other environmental and behavioral variables as well as to have a strong
influence in shaping thermal control and satisfaction, while exercised control was found
to wield certain negative effects on satisfaction. Considering the fact that the operational
definition of ‘exercised control’ was the relative frequency with which workers engaged
in several types of thermally-related behaviors to regain thermal comfort when needed, it
appears that the more uncomfortable with the thermal environment employees were in
their workstation, the more frequently they engaged in thermal related behaviors.
Exercised control measured the degree of uncomfortable thermal environment rather than
control in the experimental setting.

O’Neill (1994) found that adjustability as the concept of control was positively
associated with work outcomes such as communication, environmental satisfaction and
perceived performance based on correlation analysis and path analysis. In his research,
adjustability was related to level of environmental control. Adjustability was measured
the degree to which the work space supports organization of work materials, ease of
adjustment of storage and display features, and ease of rearranging furnishing. The items
were “How frequently do you adjust the position of storage units,” “How satisfied are
you with how well your work space lets you organize your materials,” and “How
satisfied are you with the ease of rearranging the furniture in your workspace.”

Veitch and Gifford (1996) investigated the effects of perceived control over
lighting on performance with an experimental design using undergraduate student
subjects. They measured lighting control, environmental control and session control

respectively with such items as, “I had some control over the lighting in this study,”

19

“Being able to control my environment makes me feel better,” and “ I had some control
over during this session.” Subjects with the choice reported more perceived control,
however, contrary to the researcher’s expectations, subjects with more perceived control
performed more poorly and more slowly than subjects not given a choice. Consistent
with this, Veitch and Newsham (2000) found that there were no short-term benefits of the
perceived control over lighting. They explained it might result from too high quality of
environmental conditions. In experimental design, in order to gain control, subjects
might transfer efforts from doing their given tasks to gaining control, resulting in lower
performance. However, in reality, through experience from controlling environmental
conditions under repetitive daily activities, employees might put less effort into gaining
control and increase performance by adjusting the lighting according to tasks and needs.
Results should be viewed with caution since, in some cases, providing too much control
can overwhelm people (Barnes, 1981). Especially in highly demanding jobs, additional
choices concerning the physical environment could contribute to overload (Wineman,
1982). Nevertheless, a large amount of research shows negative outcomes associated _
with a perceived decrease in control (Greenberger, Strasser, Cummings, & Dunham,
1989)

McCarrey et al. (1974) found that people in open offices perceived themselves
to have inadequate control over input to and from the environment. They reported a lack
of general privacy, a lack of personal privacy and a lack of confidentiality of
communications. Lower satisfaction in open offices is due to employee’s perceived lack
of control over input to and from the environment (McCarrey, Peterson, Edwards, & Von

Kulmiz, 1974).

20

In general, control over the physical environment is related to job satisfaction,
though some differences in findings exist due to differences in measures of perceived
control (Lee, Ashford, & Bobko, 1990). Measurements are inconsistent since many
authors are developing or adapting instruments for their particular purposes (Spector,
1986). Therefore, for the physical aspects of control, it is necessary to develop the
construct of control in the workplace.

For another important aspect, personal control apparently plays a moderator role
to help people cope with stress and negative environmental stimuli. Likewise, control
over the physical environment is expected to moderate the effects of negative
environmental sources such as distractions and disturbances on employee’s job
satisfaction and performance. Gatchel (1980) supposed that if residents in nursing homes
had some sense of control, it would alleviate some of negative reactions and lead to a
better adjustment. Similarly, Langer and Saegert (1977) investigated the aversive effects
generated by a high density condition - shopping at a crowded supermarket. Providing
cognitive control significantly ameliorated the aversiveness of the crowded condition.
Yet in spite of studies finding that individual differences may moderate the effects of the
physical environmental features, such as the way desks and other equipment are arranged
in the work area (Greenberger et a1, 1989), little empirical research has dealt with the
moderating role of personal control over the physical environment.

Interestingly, there are other moderate variables playing roles in the effects of
control as a moderator of stressors. Lee, Ashford, and Bobko (1990) found that the
effects of control on job satisfaction and performance were different to specific types of

people. They proposed “When perceived control is low, people with high levels of Type

21

A behavior may be distracted from concentrating on their performance tasks as they
divert effort toward strategies to regain control.” In a Carlopio and Gardner study (1992),
avoidance of distraction, interruption, and noise became more important for employees
whose jobs were more complex and afforded them more control over social contacts.
Paciuk (1989) proposed that factors influencing control are both personal and situational
since personal differences might increase/decrease vulnerability to negative stimuli
(stressor), and situational factors such as organizational support/constraints are associated
with a certain stressors.

Therefore, the following contention is proposed. When control over the physical
environment is low, people are likely to be distracted from concentrating on their work,
thus the consequence is lower performance. The consequences of low control may be
greater for employees who are used to working in quiet areas than for those who care lass
about noises, or for employees whose jobs are more complex than for those whose jobs
are simpler.

In the present research, control over the physical workplace can be defined as an
individual choice to change the workplace based on individual or interpersonal needs. In
the working environment, negative work environmental aspects come from various
sources. Control over the physical workplace is expected to provide a more stable
environment for the immediate individual work area under a changing environment, as
well as to mitigate or moderate the effects of negative stimuli on job satisfaction,
performance or other employee’s reactions, such as an inclination to work alone, to work

in enclosed areas, or to avoid meeting people.

22

Effects of Negative Ambient Features on Work Outcomes

Related to control, ambient features appeared to cause major stress sources and
complaints for employees in work place environments. Severely uncomfortable
environments can result in a loss of control and helplessness (Gactel, 1980). Sundstrom
explained this mechanism as follows. “If the environment contains enough distractions,
it may create overload by occupying more of a person’s capacity than can be spared from
the task while still maintaining effective performance.” Consequences of negative
ambient quality are loss of control, lack of privacy, disturbance and distraction. In
obverse, control over the physical environment is expected to relieve the negative effects
of disturbance, distraction, etc.

Open offices are popularly incorporated by organizations to increase
communication; however, there are more distractions in open plan offices than in
conventional private closed offices (Brookes & Kaplan, 1972; Canter, 1972; Hedge,
1982). In general, studies of the ambient features in office environments including noise,
lighting, temperature, and existence of windows suggest that such elements of the
physical environment influence employee attitudes, behaviors, satisfaction and
performance (Crouch & Nimran, 1989; Veitch & Gifford, 1996; Larsen, Adams, Deal,
Kweon & Tyler, 1998).

Among the ambient features, noise is the major source of distraction for
employees needing to concentrate or talk on the telephone (Marans & Spreckelmeyer,
1982). It was found that noise was related to distraction and openness (Sundstrom, Burt,
& Kamp, 1980). As an ambient stressor, noise comes from telephone ringing, people

talking, ventilation systems, and office equipment (Sundstrom, Town, Rice, Osbourn &

23

Brill, 1994). Kjellberg et al. (1996) investigated annoyance, distraction and symptoms by
various noise sources upon office workers, industry workers and laboratory workers.
They found that distraction was most strongly related to the degree of self-control of
noise and noise predictability, and annoyance was significantly related to the sound level.
Telephone signals and conversation were large indicators of distraction. Ferguson and
Weisman (1986) investigated the effects of distraction on satisfaction with the workplace.

They measured distraction with eight items such as “concentration frequently distracted,”

9’ ‘6 9’ “

“frequently armoyed by noise, workplace usually quiet, noise level too high,” and
“ability to concentrate” (standardized coefficient alpha was .94). They found that
distraction was negatively related to satisfaction with the workplace but not with job
satisfaction based on path analysis.

Other distraction sources may include flicker form fluorescent lights,
uncomfortable temperature, or poorly designed seating (Sundstrom, 1986). Sundstrom et
a1. (1994) investigated the effects of noise on enviromnental satisfaction, job satisfaction
and performance before and after the renovation. They identified disturbance by office
noise with time to be related to environmental satisfaction and job satisfaction in the
work environment. However, contrary to this expectation, performance ratings were not
related to any source of noise. Sundstrom (1986) supposed that responses to noise not
only rely on its physical properties such as degree of loudness, but also on an individual’s
sense of control. Through experimental design, it was found that noise influenced

performance (Banbury & Berry, 1998). In the research, four test conditions (speech,

office noise with speech, office noise without speech and quiet) were assigned. And

24

performance was measured by accuracy and mean time to complete the tasks for two
office related tasks, speech memory for prose and mental arithmetic

Yet other studies found that open office noise can be stressful (Evans & Johnson,
2000) and that open plan offices may have high levels of distraction and disturbance and
low privacy levels (Hedge, 1982). If considered in isolation, each ambient feature may
not have predictable effects on performance; however, repeated distraction from a
collection of physical environmental features may be negatively associated with
performance.

With an increase of knowledge workers, distraction might be more important for
knowledge workers and their work performance. Aronsson (1989) found that new
technology was associated with an increase in intensity demands such as attention,
mental effort, and concentration.

Although some research has found significant effects of noise, distraction or
disturbance on dependent measures of j ob satisfaction and performance (Sundstrom et al,
1994; Stone & Irvine, 1994, Banbury & Berry, 1998), other studies have failed to confirm .

a direct relationship between them (Ferguson & Weisman, 1986).

Relationships between Flexibility and Interpersonal /Group work
In the present study, it is assumed that if people have a higher degree of
flexibility and freedom to choose or change their workplace according to differing needs,
they will communicate better, leading to a higher degree of group cohesion. Although in
much of the literature the importance of flexible space involves support for changing

organizational structures, facilitation of instant team communication, and coping with a

25

variety of work patterns (Berndt, 2000); there is little evidence that shows a relationship
between flexible (or rigid) use of workspace and relevant group outcomes such as work
effectiveness, group cohesiveness, and performance. Even the meaning of the term
‘flexibility’ has many facets and is frequently interpreted differently in contemporary
discussions of the faltering economy and changes in the labor market. In particular, it
could be defined as labor market flexibility, freedom of employers to reduce wages
(Casey, Keep, & Mathe, 1999), and flexibility in timing and location, etc. Flexibility is
considered as “a common to all solution” (Olson, 2002) or as balancing often conflicting
individual requests and group requests within the same workplace. In general, flexibility
in space design covers three aspects; expansibility, convertibility and versatility (Pena,
1987)

However, this study focuses on the flexible use of furniture and work areas.
Flexibility in the workplace can be defined as the capacity to use, adjust and rearrange in
various ways to accommodate frequent shifts from individual work to collaborative work,
and vice versa. Flexible use of space is expected to facilitate instant communication
among work groups. This assumption is based on optimal spacing: sociopetal spaces that
bring people together and sociofugal spaces that separate people (Osmond, 1957; Bell,
Fisher, Baum & Greene, 1990). If individual workstations are flexible enough to
accommodate shifts from sociofiigal arrangement to sociopetal arrangement, it might be
easier for workers to have more interactions.

Another physical aspect that appeared as a factor affecting group activities or
outcome in the literature is the openness (enclosure) or accessibility of employees in open

offices that create a more cohesive social environment (Hedge, 1982). Removing high

'26

partitions or walls increases openness and allows for the flexible arrangement of furniture,
but it does not always ensure the flexible use of spaces, instant meetings and interactions.

A relatively small amount research has been conducted to investigate the effects
of the physical environment on interpersonal or group works. As the emphasis on
teamwork grows, organizational performance can be related not only to individuals but
also to teams and groups (DeMatteo, Eby, & Sundstrom, 1998). As one example of this
shift in dependent measures to group level phenomena, physical enclosures for team
work areas may be associated with group cohesiveness (Sundstrom, 1986).

Zahn (1991) examined the effects of physical proximity on exposure and
communication and found that the physical distance between offices reflected a chain of
command and status distance, and was associated with reduced face-to-face
communication.

Brennan et a1. (2002) investigated the effects of physical environmental changes
on team member relations. Team member relation was defined by the relationship with
other members of the team such as “ I really feel I am part of my team,” “My team
members approach me when needed,” and “ It is easy to ask advice from anyone here”.
According to Brennan and his colleagues, employees who shifted from traditional offices
to open offices experienced a number of disturbances and distractions instead of better
communication, resulting in decreased team relations. Similarly, insufficient privacy in
office interferes with the development of close interpersonal relationship (Oldham &
Rochford, 1983). Evans and his colleagues (1994) inferred the reason that workers in
overload environments develop social withdrawal is that they feel unable to hold

confidential conversation, or that direct feedback from supervisors was restrained.

27

Group cohesiveness is one of the indicators of successful group work or team
work. The definition of group cohesiveness is the perception of togetherness or sharing
within the organizational setting, including the willingness of members to provide
material aids (Koys & DeCotis, 1991). Dimensions that are associated with group
cohesion are group sprit, morale, social support, communication/cOOperation within the
team, the extent of work load sharing, and participation in decision making and goal
interdependence (Carless & DePaola, 2000). Task cohesion is the extent of “motivation
toward achieving the organization’s goals and objectives”(Widmeyer, Brawley, & Carron,
1985, p17). Social cohesiveness refers to “motivation to develop and maintain social
relationships within the group” (Carless & DePaola, 2000). Task cohesion is more
closely related to work performance than interpersonal cohesion (Mullen & Copper,
1994; Carless & DePaola, 2000). Carless and DePaola (2000) found that task cohesion
was moderately related to job satisfaction and supervisor ratings of team performance,
and strongly related to team member ratings of team effectiveness, and found that
individuals don’t distinguish between individual- and group- level task cohesion. Chang
and Bordia (2001) found that task cohesion was the predictor of self-rated performance in

longitudinal design but social cohesion was not the antecedent of performance.

Satisfaction with the Physical Environment and Job satisfaction
In the workplace, it is expected that employees who possess higher satisfaction
with the physical environment are likely to have better work outcomes. Carlopio (1996)
found that employee’s satisfaction with their work environment is related to their job

satisfaction and indirectly related to organizational commitment and turnover intention.

28

Thus, primary focuses in the following are ‘what kind of aspects contribute to physical
work satisfaction’, and ‘what relationships exist among these aspects including control
and flexibility.’

Another set of attributes that may be directly related to work effectiveness, job
satisfaction and performance in the workplace setting is workspace adequacy. Workspace
adequacy referred to adequacy of the workspace in terms of equipment, size, lighting,
temperature, equipment, and proximity to service area and co-workers (Zalesny & Farace,
1987)

For workspace adequacy, the composite view is more suitable than the factor
view since the items are physical environmental components. “The composite view
assumes that facets are components of the composite construct while the factor view
assumes that facets are manifestations of latent constructs (Law & Wong, 1999).” It is
necessary to examine whether the composite or the factor view is suitable for workspace
adequacy.

In general, previous research found consistently that satisfaction with the
physical environment was positively related to job satisfaction. Investigators have
demonstrated that the physical environmental quality affects job perception, attitudes, and
job satisfaction (Zalesny, Farace, & Hawkins, 1985; Sundstrom, 1994). Ferguson and
Weisman (1986) also found that satisfaction with the workplace was positively related to

job satisfaction.

29

Gaps between Perception and Expectations Toward A Physical Milieu

A majority of physical attributes and variables were measured based on
perception. This part of the literature review focuses on the importance of perception and
expectation in measuring physical environmental features and on methodological
approaches in relevant areas. Because control and flexibility are relatively less studied,
there is a lack of information about how much control and flexibility employees have
compared to how much they think they should have.

Many environmental satisfaction studies aim to find out the importance of
various properties or components in defining the degree of environmental satisfaction/
dissatisfaction or to specify more complex relationships between people and the
environment. For instance, attitudes and socio-demographic variables influence
evaluative hierarchies of environmental aspects, thus contribute to the relationship
between the persons and the environment (Bonnes & Secchiaroli, 1995).

The approach in the majority of research conducted to investigate satisfaction
with the physical environment is based on perception by directly asking a question such
as “all things considered, how satisfied are you with your primary workplace? ” without
asking opinions or expectations toward workplaces (Marans & Spreckelmeyer, 1981;
Marans & Yan, 1989; Lantrip, 1993; Spreckelmeyer, 1993, Sundstrom, Towm, Rice,
Osborn, & Brill, 1994). Unless people indicate severely low satisfaction levels, it is hard
to provide specific managerial direction from satisfaction results. For instance, in the
Pugsley and Hyanes’ study (2002), satisfaction with storage is below the mean of office
design (68%) and satisfaction of flat screen PC monitors is much higher than the mean

level of ICT (61%). How should a viewer interpret this? Does the higher satisfaction

30

level mean no action is required or should more flat screen monitors be provided because
people are satisfied with the type of monitor? In Carlopio and Gardner’s study (1992),
satisfaction with the work environment in general, satisfaction with the work site, and
satisfaction with health and safety were respectively 3.72, 3.47, and 3.65( “ 1 very
dissatisfied to 5 very satisfied). It is hard to get specific direction from these results.
Frequently satisfaction results may be interpreted with the researcher’s perspective rather
than reflect the actual user’s perspective. It would be more helpful to show the gaps of
the present features and future features expected from the user’s perspective.

Environmental evaluation is redefined as a function of the ‘congruence’ people
perceive between their own needs and aims and the opportunities offered by the
environment rather than as a response directly attributed to the objective properties of the
latter (Bonnes & Secchiaroli, 1995).

In a similar context, researcher has investigated incongruence between what
environment is and what environment should be to fulfill the required function of
people’s needs. Szigeti and Davis (2002) developed a building evaluation form", which
incorporates customer satisfaction needs, and customer functional needs. “Functionality”
is defined according to the levels of functionality and service of a facility from
stakeholders’ views. Serviceability is rated according to the levels of capability. By
comparing supply aspects (serviceability) with demand aspects (required functionality),
people can easily assess the building performance or capability and specify which aspects
should be improved. It provides specific managerial application information in such a

way that shortfall aspects need to be changed to save later cost.

 

4 ASTM standards on Whole Building Functionality and Serviceability

31

Brackertz and Kenley (2003) developed a facilities management tool to measure
service performance in government facilities. In this tool, people can score on a Likert-
type scale for functional requirement (1 :not needed to 5:essential) and actual performance
(1: doesn’t meet, to 5: exceeds). In this study, the functional requirement and the actual
performance scales were independent and the gap approach was not used.

Performance of the physical environment has been defined by confirmation
between the needs of users and a current diagnosis of the environment. Lantrip (1993)
measured perceived differences between what is needed by an inhabitant and what is
provided by the environment regarding the effects of constraints on human movement on
productivity in an office environment. In Lantrip’s study (1993), various methods such
as video records, questionnaires, and analysis of floor plans were employed to measure
actual spatial characteristics, environmental characteristics and worker characteristics.

However, in these approaches above mentioned, subjects who evaluated the
present environmental quality and those who decided future required enviromnent quality
are different. There is a lack of perspective that users are ‘customers’ of a built
environment. Thereby, a large of amount of research deals with perception but leaves out
the expectations of users.

In marketing research, satisfaction is related to confirmation or disconfirrnation
of expectation (Smith & Houston, 1982; Parasuraman, Zeithaml, & Berry,l988).
Parasurarnan et a1. (1985) developed ‘SERVQUAL’ to measure customer’s appraisal
with expectation together toward service quality. Interesting propositions among major
propositions are “The quality that a consumer perceives in a service is a function of the

magnitude and direction of the gap between expected service and perceived service” and

32

“Consumers typically rely on experience properties when evaluating service quality.”
Against some proposition from which SERVQUAL was developed, Cronin and Taylor
(1992) named ‘SERVPERF,’ performance based simple measures of service quality
without expectation aspects. If people have the highest or maximum level of expectation
toward every aspect, it may be worthless to ask how much they expect. In Cronin and
Taylor’s (1992) results, simple performance measurement appeared superior in terms of
explaining variation with less effort. In the service area, people can compare several
service providers, then they might expect the highest ratings for each desirable aspect.
Still the findings raise some questions. If consumers have built a relationship with a
certain type of service provider and have built up expectations over time, the results
might be different. Especially for the physical environment, people, the users of the
space, have had cumulative experiences over longer periods. Although quality in other
workplace environments might be the norm, the other environment cannot be substituted
into the workplace of people who daily use and reside in their present employment
environment. For instance, employees in offices have experiences and expectations
based on their residency experiences in the physical environment. Other office
environments can be their norm but employees rarely expect implausible features toward
their work environment. However, this inquiry has been rarely explored in research
related to the physical environmental features. Thus, it is valuable to investigate the gaps
between perceived environmental features and expected environmental features by
occupants.

This approach can tell whether cumulated experiences in an environmental

setting affect expectations and appraisal of the environment. Measuring the gaps is

33

especially useful in managerial application. People can easily define which aspects are
under expectation level and should be improved. Therefore, in order to measure the
satisfaction of employees toward their workplace, it is useful to measure their

expectations collectively.

34

CHAPTER III CONCEPTUAL FRAMEWORK & HYPOTHESES

This chapter proposes conceptual frameworks of this study regarding control and
flexibility in workplace design and the interactive nature of physical environmental
features. The conceptual frameworks are founded on the perspective of the Transactional

theory and rest on the Lens theory, Stress and eclectic model.

Person-Environment Transactions-Environmental Probabilism Theory

The aim of transactional perspective is to re-compose the dichotomy between the
person and the environment and suggest dynamic interaction and interdependenCe. As an
environment influences a person, there is a simultaneous reciprocity of influence of the
person on the environment (Bonnes & Secchiaroli, 1995). The transactional perspective
(Ittelson, 1973; Stokols & Altman, 1987) to person-environmental relationships is to free
from both perspectives, ‘objectivist; environmental determinism’ and ‘subjectivist;
environmental possibilism.’ On the one hand, ‘environmental determinism’ assumes that
the environment determines behavior absolutely, on the other hand, ‘environmental
possibilism’ ascribes such a large role to individual choice that it is hard to make
predictions regarding environmental influence in behavior (Bell, Fisher, Baum & Greene,
1990). Environmental probabilism (Porteus, 1977) assumes that while an organism may
choose a variety of responses in any environmental Situation, there are probabilities
associated with specific instances of design and behavior. Environmental probabilism

incorporates transactional perspective.

35

As office environments are integrated systems of people, activities, organizational
and personal relationships and technology that take place in dynamic settings rather than
static settings (Goodrich, 1982), the proposed models for this study have incorporated a
perspective based on the environmental probabilism theory. The probabilistic model of
behavior is a compromise. According to Sprout (1965), the probabilistic approach is
appropriate for the problem of relating environmental factors to psychological behavior.
Perceived physical environmental features affect states including values, preferences,
attitudes, etc., and actions including choices, decisions, etc. The proposed models of the
present study include both physical environmental characteristics and individual control
aspects that illustrate an transactional interrelationship between the physical environment
and the behaviors of users, rather than focusing on only absolute environmental effects on

behaviors or on only a willingness to change the physical environments.

Brunswik’s Lens Theory
This research adopted the Brunswik’s lens model theory (1956), especially in the

perception process. Brunswik’s approach was applied to Stewart’s (1987) study to
develop an observer-based assessment of visual air quality. “Each of the stimuli
emanating from the environment might be assigned a weight or probability based upon its
usefulness in supporting accurate perception... Not only do certain environmental
stimuli differ in their objective usefulness (ecological validity), but individuals may
weight them more or less appropriately (cue utilization) because of past experience,

personality or other differences. . .. Subsequent analysis may reveal that certain

36

individuals share similar weighting profiles or policies, a valuable insight that might have
been overlooked by traditional average techniques (Bell et al., 1990, p32)”.

There are some difficulties in dealing with physical environmental features as
objective variables separated from human perception, not only because each element
serves multiple functions but also because peeple appreciate the same objective features
in different ways due to their perspectives, attitudes or norms.

Sundstrom (1986) stated that beyond the individual level of analysis, there is
difficulty in proving direct relationships between the physical elements and group
outcomes, and the discussion focuses on social, psychological and organizational
processes and the ways in which they incorporate the environment rather than on the
environment itself.

Marans and Spreckelmeyer (1982) hypothesized a model in which environmental
and worker characteristics act jointly to affect inhabitants’ perceptions of environmental
quality. This theory can be applied to explain the importance of subjective attributes in
workplace design and individual differences in weighting these attributes. Previous
research has shown that users’ subjective evaluations of the amount of space are a more
reliable predictor of environmental satisfaction than the actual workspace floor area (Brill,
Margulis & Konar, 1984; Marans & Yan 1989; Spreckelmeyer; 1993).

Research findings proved that perceptions of physical environmental conditions
are strongly associated with objective physical environmental conditions. Block and
Stokes (1989) found that how private participants thought their environments to be and

the manipulated privacy condition are significantly correlated ( r=.74, p<.01).

37

The importance of a study to investigate perception lies in the relationship
between people’s perceptions and other objective attributes. As an individual’s
perception or assessment of an environmental attribute is related to, but distinct from, the
objective attribute, Marans and Spreckelmeyer (1982) conclude that the primary purpose
of evaluating research is to explore such connections between specific environmental
attributes and people’s perceptions of them.

Therefore, this study investigates antecedents of latent factors of control and
flexibility regarding the physical environmental features based on employee perceptions.

Figure l. Brunswik’s Lens Model

 

Distal cues

 

 

 

 

 

 

Ambient conditions Proximal cues
Temperature
A' al't
""1“ 'y Comfort
L‘ M
. . '8 "‘8 Distraction
Setting attributes Work Stations Privacy Judgements
Works cs
pac Control
D‘fl’cr r' t"
l e" '3 '0" Flexibility
R I t
00'“ ayou Work adequacy
Building layout

 

 

 

enclosure

proximity

 

 

Amenities types Cue utilization

 

 

 

 

(Adapted from Bell, Fisher, Baum & Greene, 1990, p 32)

The Stress-Behavior Model and Moderating Process
Moderate or mediate models describe interactive relations, such as physical and
psychological factors using two or more physical factors, or two or more psychological
factors, which are altered by the presence of a second work environmental factor.

Moderator is independent (i.e. slight or no correlation) from the predictor variables while

38

mediator is interdependent (e.g. correlated) in relation to the predictor and the mediator
(Evans & Lepore, 1997). With the exogenous work factor, the antecedents can be either
physical or psychological variables, and the mediating variables can be either physical or
psychosocial work environments (Evans, Johansson, & Carrere, 1994). The role of
control as moderator or mediator will be explored in this study. Evans and Lepore (1997)
categorized four typical types of moderator functions: 1) the environment affects
behavior in opposite ways for different levels of the moderator (crossover interaction), 2)
the relationship between the environment and behavior is stronger (amplification effect)
at one level of the moderator than at another level, 3) the relationship between the
environment and behavior is weaker (attenuation effect) at one level of the moderator
than at another level, and 4) the envirOnment only has an effect in the presence of the
moderator.

As a mediator, control plays a role in reducing the effects of environmental
constraints. The effects of negative stimuli from the environment on human behavior are
more emphasized in the Constraint model than in the Eclectic model by Bell and his
cofleagues(l990)

According to the behavior constraint model of environmental stimulation,
excessive or undesirable environmental stimulation leads to arousal or a strain on the
information-processing capacity (Proshansky, Ittelson, & Rivlin,l970; Rodin &
Baum,l978). “Constraint” means something about the environment is limiting or
interfering with things we wish to do. The model consists of three basic steps: perceived
loss of control, reactance and learned helplessness. If people are exposed to unwanted

stimuli and they perceive themselves as losing control over time, the ultimate

39

consequence of loss of control is learned helplessness (Bell et al., 1990). Behavior
constraint models focus more on the side in which people’s efforts to gain control over
the constraints fail. But more complicated interrelationships are often present.

Stress can be defined in several dimensions. Roberts and colleagues (1997)
explained stress in terms of a person-environment interaction. If a poor Person-
Environment fit exists, this results in a strain, when environmental features surpass the
individual capabilities, or the environment fails to provide the necessary resources to
allow for successful performance of a required task. For instance, stress would rise, if a
workplace that is too noisy, visually distracting, and open to frequent interruptions goes
beyond the individual’s capabilities to deal with the situation; or if a workplace that is
poorly equipped, under illuminated, fixed, or unable to change its physical environmental
features keeps the individual from performing effectively. In similar context, Paciuk
(1989) defined environmental stress as disfunction between human needs and the
resources and constraints afforded by the environment. She stated that when individual
goals and expectation are frustrated by environmental conditions, stress may be a
consequence if goals cannot be readjusted or sufficient resources allocated.

Bell et al. (1990) proposed an eclectic scheme of environmental-behavior
relationships (see Figure 2). Their model is developed from several theories including
constraint and stress theory, adaptation level theory to integrate the various mediating
concepts that have been applied to environment-behavior relationships. Lantrip (1993)
summarized that the essential point of the model is that people behaviors are ruled by the
desire to satisfy needs rather than any direct influence of the environment in which their

desire may be momentary, periodic or long term goals.

40

As applied to the work environment, this model explains how people perceive
their workplace, how they deal with environmental stress or constraints, and how they
lead to satisfaction and desirable performance. Homeostasis status may be very short or
be relatively stable. The important thing is that the workplace condition needs to provide
options or means so that employees easily cope with stressors and inhibitors of doing
work by adjusting, modifying or changing their environmental conditions.

Figure 2. Eclectic Model of Environmental Stress in the Workplace

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E """""""""""""""""""""""""""""""""""" I" """"""""""" HOWCOSQSiS ................................................ Possible
5 . _ ; satisfaction after effects
. Objective j performance Locus of control..
: physrcal cnvrronmental '
conditions
: Noise. distraction T
0152:):(228. o [Witlhm Successful
I S ‘ Iarran em ,, puma. range adjustment
: pana g e ’ of stimuli
; yes or
i : to work adaptation
: 4 i
; Perception ls coping
5 ofthe (changing)
5 environment successful?
‘9 Continued
no arousal
Individual difference Outside and/or
Job responsibility. optima| range _’ Asrousal stress
_______ 705k COMP’EXICV of stimuli tress
Social condition to work overload
Organizational climate
helplessness
fatigue

 

(Adapted from Bell, Fisher, Baum & Greene, 1990, p123)

 

avoidance of
social contacts

 

In this study, perceived loss of control depends on either/ both the degree of

control the individual possesses or perceives to have or/and the degree of unwanted

environmental stimuli, for example, distraction. If people have a high degree of control

over their environment, their reactance efforts will more likely be successful. However,

41

 

 

if they have a low degree of control over their physical environment and experience a
high level of distraction, they are more likely to feel helpless.

Based on the theoretical background and literature review, conceptual frameworks

and hypotheses for this research are proposed.

Models of Control in the Workplace

The direct effects of control in the workplace on employee’s job satisfaction and

performance are shown in Figure 3.

Figure 3. Effects of Control on Job satisfaction and Performance

.®/:'0'
\-

Based on this conceptual framework, the following hypotheses are proposed.
Hla: Control over the physical work environment is positively related to job
satisfaction.
Hlb: Control over the physical work environment is positively related to perceived

performance.

The moderating effects of control on work outcomes and its intervening influence
over the relationship between negative environmental sources and work outcomes are
proposed in the conceptual framework shown in Figure 4. If employees possess a high

degree of control, the negative effects of distraction on performance, retreat or social

42

withdrawal will be less significant. However, if employees possess a low degree of
control, the negative effects of distraction on performance, retreat, or social withdrawal
will be substantial. The consequence of negative physical environmental features such as
over-crowdedness is social withdrawal (Oldham & Fried, 1987). If employees
experience distractions with a low degree of control, it is expected that they would be
more likely to avoid meeting people or will want to work alone. In the proposed model,
according to the level of control, the path relationships are expected to be different.
Based on the proposed model, hypotheses are proposed.

Figure 4. Model of Control, Distraction and Performance

+
-o—»
1
'i Work alone J

"""""""""" ’ Wm

I Work alone I

 

 

 

   

.................. ,[Avoid meetingl

 

   

 

 

 

 

Distractions

 

 
 

+ -i
‘VLAvoid meetingj

 

H2: Control over the physical environment and work is negatively related to
distractions.

H3a: Perceived distraction levels in the workplace are negatively related to perceived
employee performance.

H3b: Perceived distraction levels in the workplace are positively related to an

employee’s inclination to work alone.

43

H3c: Perceived distraction levels in the workplace are positively related to an
employee’s inclination to avoid meeting people in the workplace.

H3d: An employee’s inclination to avoid meeting people in the workplace is
positively related to an employee’s inclination to work alone.

H4: There is a difference in terms of effects of distraction on performance between

the high control group and the low control group.

To further examine the mediating role of control, the following path model is
tested. In a causal relationship among X (independent variable), Yi(mediating variable),
and Y2 (dependent variable), if a researcher constrains the path coefficient for X—> Y2 to
zero and overall fit of this constrained model is not appreciably worse than one with X—>
Y2 as free parameter, then the hypothesis on mediated relation is supported (Kline, 1998,
p132 ). It was found that noise was not negatively associated with performance
(Sundstrom et al, 1994). And this might be due to the mediating role of some other
variables in workplace. The effects of distraction on performance might be mediated by
control (see Figure 5). Based on this assumption, the following hypothesis is proposed.
Two comparable path models are to be tested.

Figure 5. Mediating Role of Control

I" ------------------------- '1
I I

H5a: Control plays a mediating role in the relation of distraction to performance.

44

Block and Stokes (1989) proposed that the distinct personality of individuals may
affect satisfaction with one’s workplace. They hypothesized that introverts would be
more satisfied in a private closed setting than extroverts. The hypothesis was not
supported. Self-rated sensitivity to noise was found to be significantly related to
distraction (Kjellberg, Landstrom, Tesarz, SOderberg, & Akerlund, 1996). Similarly, in
this study, according to employee work styles, there might be a difference in reaction
toward quietness/noise, or distraction. Therefore, it was hypothesized that there is a
difference in the path model according to solo work style.

H5b: According to the individual work style, there is a difference in the relationship

between distraction, control, and performance.

Models of Flexibility in Workplace Design
Flexible use of space and furniture arrangement enables employees to

accommodate multiple activities including impromptu meetings, and to rearrange or
adjust their workplaces. This flexibility is expected to increase interactions and result in
better communication and higher task group cohesiveness. The degree of openness or
enclosure is expected to affect the opportunity for interactions and communications. The
degree of accessibility is expected to increase communication. Since social group
cohesiveness is a measure of social relationships among colleagues outside of the
workplace, it is expected to have no relationship with flexible physical workplace
features. The relationships between the physical environmental characteristics and group
cohesiveness (task cohesiveness and social cohesiveness) are described in the conceptual

framework in Figure 6.

45

Figure 6. Effects of Flexibility on Group Cohesiveness

  

 

   

Social group
cohesiveness

 

   

 

Openness

   

 

 
 
 

 

 
 
  
  

Performance

 

Accessibility

 

 

  

Task group
cohesiveness

  

Based on assumptions from the conceptual framework, the following hypotheses are
proposed.

H6a: Flexibility in workspace use is positively related to task group cohesiveness.

H6b: The degree of openness in the workplace is positively related to task group
cohesiveness.

H6c: The degree of accessibility in the workplace is positively related to task group
cohesiveness.

H7a: Flexibility in workspace use is not related to social group cohesiveness.

H7b: The degree of openness in the workplace is not related to social group
cohesiveness.

H7c: The degree of accessibility in the workplace is not related to social group

cohesiveness.

The interactive process between physical environmental features and employee’s

work outcomes in a workplace setting is proposed in the following figure.

46

Figure 7. Model of Effects of Control and Flexibility in the Workplace

e-
K” W

\
\‘ job +
/_' atisfaction \

V \\

   
 

distractions

 

+
degree of openness \>

/
4\

‘— /

4/

Although openness is believed to increase interaction, it was also found that

  

 

 

 

openness and accessibility were positively associated with disturbances (Hedge 1982;
Brennan, Chugh & Kline, 2002). Each path will be tested based on the whole model.
Additional hypotheses are the following.

H8a: Degree of openness is positively related to distraction.

H8b: Distraction is negatively related to job satisfaction.

H8c: Job satisfaction is positively related to perceived performance.

H8d: Task group cohesiveness is positively related to perceived performance.

Gaps between Perceptions and Expectations
Satisfaction with the physical work environment might consist of several features
such as control, flexibility, and work space adequacy. If satisfaction means confirmation

or disconfirmation of user’s expectations, the gap between the perceived physical

47

environmental features and the expected physical environmental features, it is more
appropriate to measure satisfaction than just the perception level.

If people have a maximum level of expectation across the variables (fictitious
case A, see Figure 8), it is of no use to ask their expectations with perception levels
according to physical environmental features because the gaps reflect the perception
variations. If people have consistent gap between what they perceive and what they think
they should have (fictitious case B, see Figure 8), it is also useless to measure their
expectations since the gaps are consistent with physical environmental features. The
assumption of this study is that people possess various expectation levels toward physical
environmental features since they have cumulative experience in the physical

environment.

Figure 8. Gap Patterns between Perceptions and Expectations

NODAU'IOJN

 

QVO

«>01

00

 

48

Therefore, this study proposes that people have different expectation levels
according to variables based on their experiences in the workplace. The gaps are expected
to be better indicators of work environment satisfaction. A conceptual diagram is shown

in Figure 9.

Figure 9. Relationship between Perception and Expectation in Satisfaction

 

Perception
W hat ------------------------------ i
employees possess

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ i
Satisfaction Job
The gaps With satisfaction
workplace
Expectation
What

wnployees want to
possess.

 

 

 

Based on these assumptions, the following hypotheses are proposed.

H9a: Gap measures between the perceived environmental features and the expected
environmental features are stronger predictors in explaining environment
satisfaction and job satisfaction than perception based measures.

H9b: Satisfaction with the physical environment is positively related to job

satisfaction.

49

CHAPTER IV METHODS

This section presents the research design, sample selection and data collection
followed by instrument development procedure. Techniques for data analysis for testing
hypothesis and for estimating the proposed models are also discussed. In order to carry
out the objectives of this research most effectively, questionnaires were used in natural
workplace settings with cross-sectional design.

Sample

The employee population used in this study was limited to office workers in
headquarter facilities or office settings. The industry type is manufacturing and
manufacturing related. Manufacturing organizations were selected since as products
become more sophisticated and competition demands variety, frequent changes occur in
effective distribution systems that deliver physical products, service and knowledge.
Therefore, knowledge teams are necessary in product design and process design even in
the service industry (Lee, Amundsen, Nelson & Tuttle, 1997). By limiting the
workplaces to headquarter and general office settings, the differences in activities and
demands on employees were reduced. All companies participating in this study are
located in Michigan, USA.

In order to get background information regarding the physical conditions, the
researcher contacted a site administrator or a facility manager and visited facilities.
Information on the physical aspects of the office buildings such as the built year of the
building, the size of the facility, the history of renovation, the distribution method of

HVAC systems, and workstation specifications were collected.

50

According to company size, workplace strategies appeared different. Smaller, and
start up firms seek minimal renovation, open-plan designs, aesthetics of flexibility, and
diversity in the workplace portfolio (Becker, 2002). The criteria are not clear in judging
company size as small or large. Small companies are expected to have different and
unique qualities and it would be difficult to get an appropriate number of subjects for this
study from small companies. Therefore, the subjects in this study are in companies with
150 or more office employees.

Instrumentation

A structured questionnaire5 was developed. It included questions that were
adapted or directly used from the existing questionnaires developed by other researchers
cited in this study. Some items were developed and added based on the literature review
because there is a lack of empirical research that has determined factor structure of the
detailed physical work environment (Carlopio, 1996). Questionnaires were pre-tested
with small groups of people to ensure that participants would clearly understand the
questions. In order to measure expectation, the measurement items were re-worded from
the measurement of perception.

Work environmental characteristics and job characteristics were also included.
Work environmental characteristics included workspace type, a measure of the degree of
openness, completely open or more typically a full partition, and the degree of shared
space for teamwork or other common shared uses, and workspace adequacy. Workspace

types included private closed (four walls to ceiling; door), individual open and high (>54

 

5 In reviewing previous studies related to physical aspects of office environment, there are few cases to

report discriminant or convergent validity of questionnaire items and factors.

51

inch) panels/cubicle, individual open and low (<54 inch) panels/cubicle, and completely
open (no partitions). In previous studies, openness was sometimes seen as related to the
presence or absence of visual and/or acoustic barriers within a space, as well as a more
subjective sense of enclosure (Marans & Spreckelmeyer, 1982). Openness was defined
to be the overall openness or the ratio of total square footage of the office to the total
length of its interior walls and partition (Oldham & Rochford, 1983; Brennan et al, 2002),
or was measured by the number of workers within the room (Canter, 1972). In this study,
openness refers to how much people perceive their workplaces to be open. Workspace
type (workspace surrounded by walls or high partitions, or workspace surrounded by
lower or no partition) was also classified. Additionally, workspace adequacy was
measured. Workspace adequacy included features such as adequacy of size aspects of
workstations and adequacy of storage areas, and quality of tools and equipment. Items
were measured on a seven-point Likert-type scale (“1 strongly disagree to 7 strongly
agree”).

Job characteristics consisted of the degree of individual or teamwork and the
degree of complexity of the employee’s job. The degree of complexity of employee’s job
is expected to be an intervening variable in relationships between control and distraction
and between distraction and performance. Complexity of the job was measured in
several studies (Sundstrom, Town, Brown, Forrnan, & McGee, 1982; Block & Stokes,
1989)

Measurements of propensity related to ‘retreat’ were developed for this study to
measure reactions or negative effects of distraction and the effects of loss of control.

Retreat refers to either the degree to which employees want to work alone or work in an

52

enclosed area, or the degree to which employees want to avoid meeting colleagues. Items

were measured on a five-point Likert-type scale (“1=never to 5=always”).

1)

2)

3)

4)

For other major variables, operational definitions follow:
Control refers to the degree to which employees perceive the amount of control they
have over their environmental characteristics. Control factors consist of control over
the physical environment and control in the work schedule, work tasks, the amount of
timing, etc. Control over the physical environment deals with determination of one’s
own workstation organization, personalization, social contact, temperature, and
lighting. Items were measured on a seven-point Likert-type scale (“1 strongly
disagree to 7 strongly agree”).
Distraction refers to the degree to which the employee feels distracted, disturbed or
irritated by negative and unwanted sources within the workplace environment. Items
were measured on a seven-point Likert-type scale (“1 strongly disagree to 7 strongly
agree”).
Flexibility refers to the degree of flexible use of the workplace. It includes
convertibility (being flexible enough to adjust, rearrange or reorganize), versatility
(multiple functions or multipurpose), general flexibility and movable systems. Items
were measured on a seven-point Likert-type scale (“1 strongly disagree to 7 strongly
agree”).
Job satisfaction is the degree to which respondents are satisfied with their jobs. Items

were measured on a seven-point Likert-type scale (“1 strongly disagree to 7 strongly

agree”).

53

5) Group cohesiveness refers to the degree to which employees share organizational
goals or tasks (task group cohesiveness), or how well social relationship among
colleagues (social group cohesiveness) are developed and maintained. Items were
measured on a’seven-point Likert-type scale (“1 strongly disagree to 7 strongly
agree”). Group cohesiveness was measured by eight variables developed by Carless
& DePaola (2000).

6) Performance is the degree to which respondents felt they performed their work well
including quality, quantity and creativity aspects. Items were measured on a five-
point Likert-type scale (“1=never to 5=always”). The dimensions of worker
performance were drawn from Oldham’s (1988) study.

As the purpose is not to compare the mean of the variable but to use variance to
see the relationships, different scales were employed for this study. Using different scales
might contribute to decrease of the multicollinearity among factors. Table 3 is provided
to show variables of latent factors and sources.

Data Collection

A requesting cooperation letter and follow-up call, or a requesting cooperation e-
mail, was sent to a human resource manager or a facility manager in each headquarter or
general office division of several organizations in the selected type of industries. A list of
organizations was developed from the employer list of the Lansing Chamber of
Commerce, the local stock interest listings in the Lansing State Journal, a list from the
Troy the Chamber of Commerce, and the IFMA (International Facility Management

Association) Michigan directory.

54

Table 3. Work Environmental Dimensions and Attributes

 

Dimensions

Attributes

 

Job characteristics

When I'm at the office, I work alone rather than in workgroups or teams.
My job requires a lot of cooperative work with others.
My work is simple and demands little concentration.

My work requires deep thoughts and concentration.

 

 

 

Retreat I would like to work alone.
I often avoid meeting people in my workplace.
Workplace I share common work areas with others
characteristics My workplace has been moved within my office.
My workplace has been moved between offices.
My workplace is open enough to see my colleagues working.
lam able to be easily accessed fiom my colleague’s workstation.
Workspace My workstation is large.
adequacy

(Zalesny & Farace,
1987)

I have ample storage in my work area.

The quality of my equipment is more than sufficient to work effectively.

 

Control

(Ganster, 1989;
Brennan et al, 2002)

I am able to personalize my workspace.

I am able to control temperature or airflow in my office.

I am able to determine the organization/appearance of my work area.
I am able to control the lighting level in my workstation.

I am able to control the social contact with others around me.

I am able to determine the scheduling and duration of rest breaks.

I am able to control the amount and timing of contact with other people
including co-workers or clients.

I am able to choose among available methods to complete my tasks.

I am able to determine the order in which tasks are completed.

 

Distractions

(Sundstrom et al.,
1982; Hedge, 1982;
Brennan et a1, 2002)

My workplace provides an undisturbed environment so that I can
concentrate on my work.

I am easily distracted from my work.
My work environment is quiet (R).
I am able to have quiet and undisturbed time alone (R).

My work area has many visual distractions.

 

55

Dimensions

Continued
Attributes

 

Flexibility

My furniture is flexible enough to adjust, rearrange, or reorganize my
workspace.

l have informal and impromptu meeting in my private workstation*.
My furniture is fixed system (R).

My workplace serves multi-purpose functions for informal and instant
meeting.

In general, my workspace is flexible.

 

Job satisfaction

If 1 had it to do over, I would choose to work here again.
Generally speaking, I am very satisfied with my work.
I would consider taking another job.

I would recommend my job to a friend who was qualified and looking
for work.

 

Task group
cohesiveness

(Carless, & DePaola,
2000))

My colleagues are united in trying to reach its goals for performance.

My colleagues have conflicting aspirations for the group work
performance (R).

My work group does not give me enough opportunities to improve my
personal performance (R).

I am unhappy with my group’s level of commitment to the task (R).

l communicate effectively with my colleagues within the office.

 

Social group
cohesiveness

My colleagues would like to spend time together outside of work hours.

My colleagues would rather go out on their own than get together (R).

 

Eggigfss, & DePaola, My colleagues do not stick together outside of work time (R).
Performance My work is of high quality.

(Oldham, 1988)

I make mistakes or errors(R)*.
I can accomplish a great deal each day
1 do a large amount of work each day.

My work outcome is creative

 

* The variable was excluded later based on the reliability test results.

Through web search, approximate size and industry type were checked. During

the initial contact, it was determined if a company wanted to participate in the study, and

56

if the organization was appropriate for the research by asking the number of office
employees.

If the company declined to participate, it was deleted fiom the list of possible
research subjects. If the organization agreed to participate and the workplace
characteristics fit the research purpose, the researcher and a site administrator" decided on
the number of participants, and set up administrative dates for distributing questionnaires.

Three companies that fit the profile agreed to participate in the study. According
to available resources, some organizations wanted to do web surveys using their own
company’s web site to get higher response. If a company wanted to do a web survey, an
e-mail linking a web survey site was distributed to randomly selected employees. If a
company wanted to do a written paper survey, questionnaires were distributed to each
employee allowing each to return the questionnaire to a designated internal mail address
or mail box. Employees were given three weeks to return their responses. Employee
participation was voluntary. Participants were told that the study was designed to
examine their perceptions and expectations of their physical workplace and to assess the
degree to which they agree or disagree with the questionnaire statements (see Appendix
A). They were also told that their responses would be anonymous and confidential.

Data Analysis

Basic descriptive statistics were conducted using SPSS. Multivariate analysis was
conducted using SPSS 11.5 and LISREL 8.54. To find the direct effects of control,
regression analysis was conducted. In order to examine the multiple relationships

simultaneously, Structural Equation Modeling (SEM) was conducted. Statistical

 

6 According to organization, a site administrator is a human resource manager or a facility manager.

57

techniques including multivariate data analysis techniques, except for SEM, can examine
only a single relationship at a time (Hair, Anderson, Tatham, & Black, 1998). Only a few
studies (Ferguson & Weisman, 1986; Lantrip, 1993; Carlopio, 1998; Brennan, Chugh, &
Kline, 2002) examine multiple relationships regarding the physical environmental
features simultaneously using SEM. Underlying the problems in research related to
office environments, Donald (1994) discussed how the emphasis on quantitative research
has led to an imbalance with little insight or in-depth understanding of meaning and
experience as felt by those using it. In practice, the research has presented a great
number of variables independently rather than attempting to develop holistic models
(Donald, 1994). All too often, a large amount of research is faced with a set of
interrelated questions that have managerial and theoretical importance.

To overcome this limitation, it is necessary to incorporate quantitative studies that
can deal with multiple relationships. SEM can provide a Straightforward method of
dealing with multiple relationships simultaneously while providing statistical efficiency,
assess the relationships comprehensively, and provide a transition from exploratory to
confirmatory analysis (Hair et al., 1998). Therefore, it will provide for the more
comprehensive relationships regarding the impact of the general features of the physical
workplace environments, including the effects of control and flexibility.

LISREL (JOreskog & Sdrbom, 1989, 1996) is a SEM software program with EQS
(Bentler, 1995) and AMOS (Arbuckle, 1997). The Lisrel model has two parts: the
measurement model and structural equation model. The measurement model specifies
how latent variables or hypothetical constructs depend on, or are indicated by, the

observed variables, and describes the measurement properties (reliabilities and validities)

58

of the observed variables. The relationships among the observed variables, as defined by
sample data, are represented by a covariance coefficient (or it can be computed using
correlation coefficients with standard deviations). The structural model specifies the
causal relationships among the latent variables, describes the causal effects, and describes
the amount of variances that remains unexplained by the model (JOreskog & SOrbom,
1996)

According to Hair et a1. (1998), SEM is particularly useful when one dependent
variable becomes an independent variable in subsequent dependent relationships. The
transition from exploratory to confirmatory analysis corresponds to greater efforts in all
field of study toward developing a more systematic and holistic view of problems.

Therefore, confirmatory factor analysis will be conducted to test the constructs.
Using Structural Equation Model, the proposed models and hypotheses are also tested.

Before evaluating the structural or measurement model, the researcher must
evaluate the overall fit of the model to ensure that the model represents the entire set of
causal relationships well (Hair et al., 1998).

According to Bagozzi and Yi (1988), preliminary fit criteria are l) absence of
negative error variances, 2) absence of error variances not significantly different from
zero, 3) absence of correlation greater than one, 4) absence of correlation too close to one,
5) absence of factor loading too small (e.g. < about.5) or too large (e.g. >95), and 6)
absence of very large standard errors. Overall model fit is the achievement of the
following; 1) nonsignificant [ (e.g., [7 with p-value 2.5), 2) adequate statistical power

of - [7 test, 3) satisfactory goodness of fit index, 4) low root mean square residuals, 5)

59

linear Q-plot of normalized residuals with a slope greater than one, 6) satisfactory critical
N, and 7) satisfactory ratio of sample size to number of free parameters (i.e., ratio 25:1).
To measure the goodness of fit of a model, several indices are used. The criteria
for the goodness of fit are slightly different from one to the other.
A chi-square (,1; ) is a likelihood ratio statistic for testing a hypothesized model
against the alternative that the covariance matrix is unconstrained. [9 -measure is
sensitive to sample size. [7 /df is a good indicator of fit with large samples. A frequent

suggestion is that [7 /df, the ratio be less than 3 (Kline, 1998), or the range 2-3 is an
“adequate” fit (Wheaten, Muhten, Alwin & Summers, 1977). The smaller, the better it is.
When using the [7 -measure in comparative model fitting, a large difference (drOp) in [I -
measure, compared to the difference in degrees of freedom, indicates that the changes
made in the model represent a real improvement (Jdreskog & SOrbom, 1996).

Goodness-of-fit measures the correspondence of the actual or observed input
(covariance or correlation) matrix with that predicted from the proposed model. Values of
index theoretically range from 0 (poor fit) to 1 (perfect fit). If goodness of fit index (GFI)
is .9, it is considered an “acceptable” fit (McDonald & Ho, 2002).

Another widely used index is RMSEA (root mean square error of approximation).
Research reporting the results with RMSEA rely on a RMSEA of less than .05
corresponding to a “good” fit and a RMSEA of less than .08 corresponding to an
“acceptable” fit (McDonald & Ho, 2002).

In summary, goodness of fit measures concerns the assessment of the overall fit of
the model to the data. However, a researcher should be cautious going from model

specification to data, analysis, respecification and interpretation. Good fit indices alone

60

do not always mean that the model is proved. Kline (1998) recommended that such a
complex statistical procedure as SEM use must be guided by reason, and that researchers

must be careful not to modify only for the sake of improving fit.

61

CHAPTER V RESULTS

This chapter reports the results of the statistical analysis and hypotheses testing
and discusses these findings. The first part describes demographic characteristics, job
activities of participants and facilities characteristics. The following part presents the

results of hypotheses testing and model testing with discussions.

Demographic Characteristics

A total of 409 questionnaires were collected: 203 from organization ‘A’, 158 from
organization ‘B’, and 48 from organization ‘C’. For Corporation ‘A’, a web survey was
conducted in response to a request by the site administrative team because the
organization has previously had a low response rate to paper surveys. The response rate
to the web survey was 18.4 %. For Corporation ‘8’ and ‘C’, paper surveys were
conducted with response rates of 37.6 % and 40% respectively. After deleting cases that
had missing variables, 384 responses were analyzed for this study.

The demographic characteristics of the sample are shown in Table 4.

Table 4. Summary of Sample Characteristics

 

 

 

 

Characteristics %
Gender Male 62.5
Female 37.5
Age Under 25 4.0
25-34 24.8
35-44 35.2
45-54 28.6
55-65 7.4
Education High school 10.6
College degree 53.6
Master’s degree 32.9
Higher degree than master’s - 2.9

 

62

Sixty-five percent of the respondents were male. The age group of 35-44 was the
largest. Most respondents (89.4 %) possessed a college degree or higher. Compared to
US. Census Bureau statistics (2001), the educational level of the sample was much
higher and the majority age group was similar (35-54). However, in such a large
corporate environment, the education level may be in normal range.

Work Activities

The job responsibilities of respondents are shown in Table 5. Twenty-eight
percent of the respondents were extensively involved in engineering (“1” no involvement
to “7” extensive involvement). Marketing or sales was the second largest category
(12.8 %) in which respondents were extensively engaged. Managers were involved in
more than three types of job responsibilities, e.g., human resources, finance, engineering,
strategic planning. Other responsibilities are related to legal, project management, clerical
support, and other functions.

Table 5. Summary of Job Responsibilities

 

1 2 3 4 5 6 7 Total
(%) (%) (%) (%) (%) (°/o) (%) (%)

 

 

 

 

 

 

 

 

 

Marketing/sales 63.4 8.9 4.7 2.4 5.5 2.4 12.8 100
Engineering 46.3 6.1 2.9 3.1 7.9 6.0 27.7 100
Operations 56.2 5.8 8.9 10.0 9.7 5.2 4.2 100
Human resources 67.7 11.6 6.6 6.6 2.6 2.1 2.9 100
Finance/accounting 57.1 8.6 12.6 6.8 1.6 2.6 10.7 100
Purchasing 67.0 1 1.8 9.9 6.5 2.4 2.1 0.3 100
Research/development 52.5 10.8 10.2 7.6 6.6 6.0 6.3 100
Strategicjlanning 50.1 7.6 7.3 9.4 8.2 8.4 8.9 100
Other .4 71.7 0.4 0.7 0.7 1.9 23.4 100

 

1: No involvement
7: Extensive involvement

63

The characteristics of work activities are presented in Table 6. More than half of
the respondents replied that they worked in teams (58.4%). F ifty-two percent of
respondents indicated that their jobs required a lot of cooperative work. Coincidently, a
majority of them indicated that they often worked alone rather than in groups or teams.
These results indicate the necessity of balancing working alone and working in groups in
their workplace. Most of the respondents were engaged in more complex activities, which
required concentration. Forty-one percent of respondents never or rarely shared common
work areas. Sixty percent of participants did teleconferencing sometimes or more
frequently.

Table 6. Summary of Characteristics of Work Activities.

 

Never Rarely Sometimes Often Always Total

 

(%) (%) (%) (%) (%) (%)
When I’m at office, I work .5 4.4 35.9 51.0 8.1 100
alone rather than work in
group or team.
My i013 ”Wires 8 1010f .5 3.4 24.0 51.3 1.0 100
cooperative works
My work feqUifes deep .1 3.1 23.4 54.4 18.0 100
thoughts and concentration.
My work is Simple and 37.0 47.9 ‘ 10.4 3.6 1.0 100
demands little concentration.
[share common work areas 13.5 27.6 27.3 17.2 14.3 100
with others.
1 d0 teleconferencing 20.8 19.0 27.6 29.2 3.4 100

 

Facilities Characteristics
The researcher visited each facility to collect general facility information. With

demographic information, facility characteristics are important to understand physical

environmental characteristics where employees work. The size of the facilities indicated

in Table 7 only the office workplace, not the manufacturing facilities of the corporations

studied.

Table 7. Summary of Facilities Characteristics

 

6A9

‘89

6C,

 

 

 

 

 

 

Business Auto and auto parts Chemical products Textile products
Built year 1925 Building 1: 1982 1953 (owned)
Building 2: 1974
Location Urban Rural Urban
Hours of operation 6 a.m.-6 pm. 7 a.m.-6 pm. 8 a.m.-5 p.m.
Size of facility GSF: 598,000 Building 1: GSF: 28,171
Gross sq.ft. RSF: 471,500 GSF: 136,835
(GSF) USF: 439,500 RSF: 125,044
Rentable sq.ft. USF: 97,384
(RSF) Building 2:
Usable sq.ft GSF: 119,808
(USF). RSF: 109,970
USF: 88,944
Number of 1,176 (designed to Building 1: 220 170
employees have 2460) Building 2: 323 (1200 trainees)
History of New furniture Building 1 Office, bathroom,
renovation systems (1990) new carpet and public areas
some furniture (1983-1986)
HVAC system Central (steam) Central Central
Lighting General General (direct) General (direct)
Task lighting Task lighting Task lighting

 

 

The building of Corporation ‘A’ is a one story building built in 1925. The

original building was expanded, while maintaining the original facade and entrance hall.

The whole area was designed for 2,460 occupants but 1,176 employees occupied it when

this study was conducted. One building of Corporation ‘B’ consisted on three stories; its

basement was connected to that of the second building and other buildings in the complex.

The ‘Building 2’ of Corporation ‘B’ had four stories. Corporation ‘8’ has been updating
office furniture. The office furniture and arrangement were slightly different by floor.
Corporation ‘C’ was housed in a one story building, connected to its manufacturing areas.
Other small office buildings were scattered around the building. Different from the other
two buildings, private closed offices and open offices were provided together.

The average space per occupant (gross square footage per occupant) was 243, 112,
51, and 165 respectively for each building. For headquarters or offices, the means of
gross square footage per occupant were 399 and 456 based on the IFMA benchmarking
report (IF MA, 1997). Thus, all buildings in this study had high utilization rates. None of
buildings provided local unit controls for heating and ventilation.

To investigate differences in office type by job type, cross-tabulation analysis was
conducted. As presented in Table 8, an Open office with high partitions was the dominant
type (74%) of the various office types investigated in this study. In comparison,
according to the IF MA benchmark report, the ratio of open plan to private office is 1.721
(IFMA, 1997).

In this study, based on cross tab analysis, office type was differentiated by job
categories ([7:23368, df=9, P=.005). Depending on the organization, obviously
different office types were provided according to job type. For Corporation ‘A’, there
was no difference (12:3.274, df=6, P=.774). Regardless of job type, an open office with
high partitions was the dominant workplace in the organization (83.3%). For Corporation
‘8’, there was a significant difference in office type by job category (1921.4, df=9,
P=.01 1). Clerical employees had open offices without high partitions while employees in

other categories reside in open offices with high partitions. Organization ‘C’ showed no

66

difference in office type with respect to job type ([7:9715, df=9, P=.374). Organization
‘C’ had fewer participants, which may have contributed to the non-significant result.

Table 8. Office Type Arrangement by Job Type

 

Clerical/ Engineer/ Manager Other Row (%)

 

 

 

 

 

Support Professional Total (%)

Private closed N 4 10 8 2 24
row (%) 16.7 41.7 33.3 8.3 100.0

column (%) 8.0 4.5 8.8 12.5 6.4

Open individual high N 27 174 68 10 279
row (%) 9.7 62.4 24.4 3.6 100.0

column (%) 54.0 79.1 74.7 62.5 74.0

Open individual low N 14 33 12 4 63
row (%) 22.2 52.4 19.0 6.3 100.0

column (%) 28.0 15.0 13.2 25.0 16.7

Open N 5 3 3 0 ll
row (%) 45.5 27.3 27.3 0 100.0

column (%) 10.0 1.4 3.3 0 2.9

50 220 91 16 377

Column(%) 13 .3 58.4 24.1 4.2 100.0

Total(%) 100.0 100.0 100.0 100.0 100.0

 

As horizontal organizational structure becomes popular, workplace design has
reflected this trend by having few differences in office type and design regardless of job
type. In this study, it was to difficult to determine if various job types and their
corresponding workplaces followed this trend. However, according to functions and job

requirements, different strategies for workplace type are generally required.

Effects of Control on Job Satisfaction and Perceived Performance
The contribution of independent variables of control to the dependent variable of
job satisfaction was examined using linear regression analysis in order to test the

proposed hypothesis.

67

Hla: Control over the physical work environment is positively related to job satisfaction.

Nine variables, including control over works and control over physical

environmental features were entered. The standardized and unstandardized regression

coefficients are presented in Table 9.

Table 9. Regression Analysis of Effects of Control on Job Satisfaction

 

 

Independent variables B B T P R2
Constant 2.021

Personalization .085 .l l 1 2.086 .038*
Temperature/airflow control .017 .014 .292 .770
Organization/appearance control .080 .120 2.275 .023*

Lighting control -.044 -.074 -. 1535 .126 .207
Social contact .124 .184 3.404 .001 ** .187a
Scheduling and duration .051 .080 1.604 .1 10

Control contacts -.034 -.046 -.859 .391

Choice among available methods .148 .180 3.452 .001**
Determine order .121 .135 2.702 .007**

 

* p<.05, ** P<.01 a = adjusted R2

The model was significant at the .01 level, [ F (363 , 9)=10.501, p=.000], rejecting

the null hypothesis. Five of the nine independent variables were significant at the .05

level of probability. Thus, hypothesis Hla was supported. The squared multiple

correlation (R2) was .207 with an adjusted R2 = .187. The linear combination of nine

variables accounted for 20% of the variance in the overall job satisfaction. Consistent

with previous findings, work control was a predictor of job satisfaction. Control features

with respect to work activities were stronger predictors of job satisfaction than control

over physical environment features. For physical features, personalization and

determination of organization and appearance of the workplace were predictors of job

satisfaction. ‘Being able to control social contact (place oriented)’ was a predictor of job
satisfaction while ‘being able to control over the amount and timing of contact (work
oriented)’ was not. Among variables of work control, choice among available methods
was the most important predictor.

There was no evidence of serious multicollinearity. All VIF (Variance Inflation
Factor) were below 2, which was much lower than 10, and tolerance values were above .7,
which was much higher than .10 (Cohen, Cohen, West, & Aiken, 2003, p423).

The assumptions of linear regression analysis were examined. Linearity was
examined through partial regression plots. Scatter plots of the residuals against each
measured independent variable and against the predicted values indicated no systematic
relationship. The assumption regarding the residuals in the regression model was also
examined. The normality of the residuals of the dependent variable was assessed. The
normal probability plot (Figure B1 in Appendix B) showed that the distributed points
clustered around the straight line. The Kolmogorov-Smimov test indicated the residuals
were not significantly different from normal (z=1.014, p=.255).

By plotting the independent variables and the standardized residuals of the
dependent variable, independence of the error was examined. Violation will be identified
by a consistent pattern in the residuals (Hair, Anderson, Tatham, & Black, 1998, p175).
No specific pattern or clustering was found in the scatter plots with exception of the
temperature control variable.

Presence of unequal variance (heteroscedasticity) is one of the most common
assumption violations (Hair et al., 1998, p174). To detect this problem, a simple method

is to a construct a set of scatterplots, plotting residuals against each of the independent

69

variables and the predicted values (Cohen et al., 2003, p130). Distribution values in each
scatter plot were equally distributed above and below the regression line across the values
of an independent variable (See Figure BZ in Appendix B). A few outliers did not suggest
any evidence of a substantial departure from linearity or homoscedasticity. Thus, no
significant violation in the linear regression assumptions was found.

Hlb: Control over the physical work environment is positively related to perceived

performance.

To test the hypothesis Hlb, the effect of the control variables on perceived
performance were examined using regression analysis. The same independent control
variables were used to examine to explain perceived performance. The results of the
regression analysis are shown in Table 10.

Table 10. Regression Analysis of Effects of Control on Perceived Performance

 

 

Independent variables B [3 T P R2
Constant 3.582 26.056 .000
Personalization -.013 -.044 -.760 .448
Temperature/airflow control .034 .074 1.386 .166
Organization/appearance control .004 .017 .286 .775

Lighting control .003 .015 .284 .776 .045
Social contact .007 .026 .440 .660 .0213
Scheduling and duration -.015 —.061 -l.116 .265

Control contacts .006 .020 .344 .731

Choice among available methods .030 .095 1.657 .098

Determine order .048 .140 2.560 .011*

 

* p<.05 a = adjusted R2
The squared multiple correlation (R2) was .045, with an adjusted R2 = .021. The

linear combination of the nine variables accounted for only .045 proportion of the

70

variance in the overall perceived performance. The F was not statistically significant at
the .05 probability level, [ F(365, 9)=1.907, P=.050]. All T-values regarding the effects
of control over the physical environment were not significant. The determination of order
of work variable was the only variable accountable for the perceived performance. For
every one rating scale unit that determination of order of work increases, the perceived
performance is predicted to change .140 rating scale units.

The significant value of the Kolmogorov-Smimov test indicated the residuals
were not significantly different from normal (z=.598, p=.867). No evidence of serious
multicollinearity was found. Therefore, the regression results allowed rejection of the
hypothesis Hlb.

Reliability Tests and Confirmatory Factor Analysis

Reliability test results are presented in Table 11. For each latent factor,
Cronbach’s alpha coefficients are presented in parentheses. All of the alpha coefficients
were above .60, which is the minimal acceptable level according to Hair et al. (1998, p
88). Because the control factor contained more items compared to the other factors, the
reliability of the construct might be relatively weak. Missing data were deleted using
listwise deletion methods.

The validity of the measures was assessed by conducting confirmatory factor
analysis (CF A). Items with low construct loadings or low reliability were deleted from
the model. The covariance matrix for the analysis is presented in Appendix C. The results
indicated a marginal level based on Maximum Likelihood estimation (chi-square=674.82,
df=432, RMSEA=.04, GFI=.89). Factor loading of each individual indicator with its

respective construct was significant at the .01 level (p<0.01).

71

Table l 1. Reliability and Confirmatory Factor Analysis

 

Loading} Error

ru an mea r m n items"I
COHSI cts d S” e at SE T-values SE T-values

 

 

 

 

 

 

 

Control ( .6456)
Personalization .51 8.85 .74 1 1 .75
Organization/appearance control 46 8.01 .78 12.10
Social contact .48 8.37 .77 12.04
Scheduling and duration .24 4.04 .94 13.11
Control amount and timing of contacts .40 6.90 .84 12.52
Choice among available methods .49 8.52 .76 12.01
Determine order .27 4.62 .93 13.01
Flexibility (. 6440)
Convertibility .66 12.06 .56 9.81
Versatility .41 7.16 .83 12.55
General flexibility .82 14.88 .32 5.25
Movable systems .25 4.10 .94 13.02
Distractions (.7137)
Undisturbed workplace (R) .72 13.96 .48 9.41
Easily distracted .42 7.43 .82 12.58
Quietness (R) .71 13.62 .50 9.73
Quiet and undisturbed time alone (R) .72 13.85 .48 9.51
Visual distraction .32 5.55 .90 12.93
Task group cohesiveness (.7515)
United in trying to reach goals .74 14.51 .46 9.54
Conflicting aspiration (R) .59 11.46 .65 12.24
Do not give me opportunities (R) .72 13.78 .49 9.58
Unhappy with my group commitment (R) .66 13.15 .57 11.71
Communicate effectively .48 8.69 .77 12.46
Social group cohesiveness (.7326)
Spend time together .55 9.83 .70 11.73
Go out with their own (R) .75 13.71 .44 7.91
Do not stick together (R) .80 14.66 .37 6.32
Job satisfaction (.7666)
Generally satisfied .82 17.35 .33 8.57
Consider taking another job (R) .49 9.03 .76 12.40
Recommend job to a friend .73 14.62 .46 9.87
Choose to work here again .70 13.96 .51 10.63
Performance (.6188)
Quality .43 6.86 .81 l 1.75
Large amount .61 9.78 .63 9.18
Accomplishment .61 9.70 .63 9.13
Creativity .53 8.51 .72 10.57

 

2': Cronbach alphas are in parentheses. b: All loadings are significant at p<0.01.
SE: Standardized estimates

72

The modification index recommended cross loading, thus it failed to achieve good
discriminant validity. One of the indicators of task group cohesiveness, “My colleagues
have conflicting aspirations for the group work performance (R)” was recommended to

load on performance.

Interactive Model of Control, Distraction, and Performance

To examine the moderating and mediating effects of control in the relationship
between distraction and performance, the proposed models were tested using Structural
Equation Modeling. Since the whole confirmatory factor analysis (CF A) result was just
at a marginal level, CF A was conducted again for the sub-measurement model only, prior
to testing the path model. The covariance matrix for these analyses are presented in
Appendix C. Overall, since all fit indices were in the acceptable range, the measurement
model showed a good fit ([7:20069, df=97, f/df=2.06, GFI=.94, RMSEA=.054). As
presented in Table 12, all factor loadings were significant at the .01 level and there was
no negative value in errors and disturbances. Since factor loadings are all significant and
there is no significant violation of the construct, the constructs were used to examine the
path models.

Thus, the proposed model was used to evaluate the following hypotheses,

explaining the effect of control on distraction and performance.

H2 : Control over the physical environment and work is negatively related to distractions.

H3a: Perceived distraction levels in the workplace are negatively related to perceived

employee performance.

H3b: Perceived distraction levels in the workplace arepositively related to an emplcfiee’s

inclination to work alone.

73

H3c: Perceived distraction levels in the workplace are positively related to an employee’s

inclination to avoid meeting people in the workplace.

H3d: An employee’s inclination to avoid meeting people in the workplace is positively

related to an employee’s inclination to work alone.

Table 12. Parameter Estimates for Confirmatory Factor Analysis for Control Models

 

 

Parameters Estimates T-value Errors Estimates T-value

LX (1,1) .31 4.68 TD 1 .90 12.67
LX (2,1) .48 7.18 TD 2 .77 10.61
LX (3,1) .77 9.32 TD 3 .41 3.68
LX (4,1) .47 7.12 TD 4 .78 11.06
LX (5,2) .38 6.32 TD 5 .86 12.67
LX (6,2) .34 5.72 TD 6 .88 12.84
LX (7,2) .65 11.36 TD 7 .58 9.39
LX (8,2) .27 4.54 TD 8 .92 13.17
LX (9,2) .62 10.78 TD 9 .62 10.06
LX (10,2) .38 ‘ 6.46 TD 10 .85 12.65
LX (11,2) .21 3.38 TD 11 .96 13.36
LX (12,3) .72 14.36 TD 12 .48 9.78
LX (13,3) .42 7.65 TD 13 .82 12.88
LX (14,3) .70 13.78 TD 14 .51 10.27
LX (15,3) .74 14.66 TD 15 .46 9.49
LX (16,3) .35 6.28 TD 16 .88 13.14

 

The estimates are standardized.

LX (1,1)-LX (4,1 ): Performance variables: quality, quantity, accomplishment, and quality
LX (5,2)-LX (l 1,2): Control variables

LX (12,3)-LX (16,3): Distraction variables

In the initial analysis, the result indicated an acceptable fit ([7:30433, df=l31,
GFI=.92, RMSEA=.O60). As the errors and disturbances (theta-delta and theta epsilon)
were freely estimated, the model fit improved. The results are presented in Figure 10 and
Table 13. The results of SEM analysis, using Maximum Likelihood estimation, indicated
that the proposed model adequately fit (f=l98.36, df=l24, GFI=.94, RMSEA=.O40).

Standard residuals varied from 5.01 to —5.01 (median, 0).

74

As expected and as shown in Figure 10, a negative effect of control on distraction
was found, as expected [GA (4,1)=-.62, p<.01] . Therefore, hypothesis H2 was confirmed.

Figure 10. Model of Control, Distraction and Performance

 

 

 

 

 

 

 

 

 

 

pc1 pel pe2 pe3 pe4

pc2 \
pc3 11
Control - - -‘ ----- > Performance

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

wcl ‘
I
[I
wc2 -.62“ -.00 ,1
wc3 Work alone

 

 

 

 

 

 

 

/ Avoid meeting

dil

 

 

 

 

 

 

 

 

 

 

 

———> significant *p<.05, “p<.01

........ > nonsignificant

The results of the T values indicated that control had no positive effect on
perceived performance [GA (3,1)=.11, nonsignificant] , a result consistent with the

regression analysis results. Thus, hypothesis H3a was not supported.

On the other hand, hypothesis H3b was supported. Distraction was positively
related to employee’s inclination to avoid meeting people in the workplace [BE (1 ,4)=.15,
p<.01].

In addition, distraction appeared to positively influence employee’s inclination to
work alone, as expected [BE (2,4)=.13, p<.05], and which confirmed hypothesis H3c.

Finally, the results of the T-values indicated that employee’s inclination to avoid meeting

75

people in the workplace is positively related to their inclination to work alone [BE

(1,2)=.24, p<.01]. Thus, hypothesis H3d was supported.

Table 13. Results for Structural Equation Model.

 

 

 

 

Measurement
T-values Estimates T-values Estimates
LX (1,1 ) 7.59 .56 TD 1 8.61 .69
Lx (2,1) 6.42 .42 TD 2 11.62 .82
Lx (3,1) 11.73 .73 TD 3 6.36 .46
LX (4,1) 6.05 .37 TD 4 12.49 .86
Lx (5,1) 10.46 .59 TD 5 10.79 .65
Lx (6,1) 6.91 .38 TD 6 13.03 .85
LX (7,1) 3.89 .22 TD 7 13.44 .95
LY(1,1) - 1.00 TEl 0 0
LY (2,2) - 1.00 TE 2 0 0
LY (3,3) - .43 TE 3 11.74 .81
LY (4,3) 5.05 .60 TE 4 8.81 .64
LY (5,3) 5.51 .6] TE 5 8.61 .63
LY (6,3) 5.18 .49 TE 6 11.01 .76
LY (7,4) - .73 TE 7 9.60 .47
LY (8,4) 6.41 .37 TE 8 13.05 .86
LY (9,4) 11.27 .70 TE 9 10.22 .51
LY (10,4) 11.57 .73 TE 10 9.61 .47
LY (11,4) 5.82 .34 TE 11 13.14 .88
Path
TD (3,1) -333 —.21
GA (4.1) -8-97 -.62 TD (1,2) 2.60 .15
GA (3.1) 1.04 -11 TD (2,3) -294 -.15
BE (1,4) 257 -15 TD (1,5) -2.81 —.14
BE (2.4) 2.19 -13 TD (3,4) -3.64 -.18
BE (3.4) --02 --00 TD (7,6) 4.20 .21
BE (12) 4.80 .24 TE (5,8) 415 -.19

 

The estimates are standardized.
LX (1,1)-LX (7,1 ): Control variables

LY (1,1): Work alone, LY(2,2): Avoid meeting people

LY (3,3)-LX (6,3): Performance variables: quality, quantity, accomplishment, and quality
LY (7,4)-LY (1 1,4): Distraction variables

H4: There is a difference in terms of effects of distraction on performance between the

high control group and the low control group.

76

It was assumed that the causality in the relationship between distraction and
performance might be different according to the level of control. If people possess a high
degree of control, the control might be strong enough to suppress the effects of
distraction on performance, propensity to work alone or propensity to avoid meeting
people. Conversely, people possessing low degrees of control, the effects of control
would not be significant, but distraction would significantly affect performance,
inclination to work alone or propensity to avoid meeting people.

Therefore, the model was tested according to different levels of control. Factor
scores were used to determine control levels because factor scores are unbiased estimates
of the factor (Joreskog, Sorbom, duToit, & duToit, 2000, p156). The factor scores were
computed using PRELIS. A median split of factor scores of control was performed to
classify a low control group and a high control group. Assuming that there is no
difference in path, equality constraints were given in six paths. These constraints were
released one by one. The results of chi-square difference tests are summarized in Table
14.

Table 14. Chi-square Difference Test for Low and High Degree of Control Groups

 

 

Model A)? (df=1)
Equality released

1st EQ BE (1,2)=BE (1,2) 6.58 (p<.01)

2“d EQ GA (1,3): GA (1,3) .05 (non significant)

 

The models showed different patterns indicating differences in the relationship by
control level. The high control group showed a stronger positive relationship between the
inclination to avoid meeting people and the inclination to working alone (2.27) compared

to a smaller relationship (.01) in the lower control group. However, the role differing

77

control levels was not clear. Between the two groups, the path between distraction and
performance was equally constrained. After release of the equality, the fit did not

improve significantly. Thus, there was no difference in terms of the effects of distraction
on performance between the high control and low control groups. Thus, hypothesis H4
was not confirmed. Although the moderating effect of control was feasible, the roles of
control in mediating and the moderating process still remain questionable. Environment
and behavior relations are not always as straightforward, but often demonstrate multi-
causal and non-recursive association between the variables (Evans & Lepore, 1997).

Since control and distraction are interdependent, the multiple relationships among control,

distraction and performance were further examined.

Mediating Effect of Control

To examine the mediating role of control in the relationship between distraction

and performance, a path model and its alternative model were tested.

H5a: Control plays a mediating role in the relaflm of distraction to performance.
The first model (see Figure l I) examined both the direct effects of distraction on
performance and the indirect effects of distraction on performance mediated by control.
In the alternative model (see Figure 12), the path coefficient for distraction—>
performance was set to zero assuming that the effects of distraction on performance is
mediated entirely by control without the direct effect of distraction on performance. All

parameters and errors for both models are presented in Table 15.

78

Figure 11. Mediating Effect of Control Model A

 

 

dil

 

 

 

di2

 

 

d13

 

 

 

 

di4

 

 

diS

 

 

 

Figure 12

-.09
Distraction """" Performance

-.70**

 

 

pel

 

 

pe2

 

 

 

pe3

 

 

 

 

\

pe4

 

 

 

pcl

 

 

 

pc2

 

 

pc3

 

 

wc1

 

 

 

wc2

 

 

wc3

 

 

wc4

 

. Mediating Effect of Control Model B

 

dil

 

 

di2

 

 

di3

 

 

di4

 

 

diS

 

 

 

Distraction

Performance

 

 

pel

 

 

pe2

 

 

pe3

 

 

 

pe4

 

 

 

pcl

 

 

pc2

 

 

pc3

 

 

wcl

 

wc2

 

 

wc3

 

wc4

 

 

 

 

—’ significant *p<.05, **p<.01

-------- v nonsigniftcant

79

 

Table 15. Results of SEM for Comparison for the Models A and B

 

 

 

Parameter Model A Model B
GA (1,1) -.72 (-09) -
GA (2,1) -5.21 (-.70) -5.22 (-.71)
BE(1,1) .98 (.14) 2.23 (.21)
LX(1,1) 14.38 (.72) 14.42 (.73)
Lx (2,1) 7.66 (.42) 7.60 (.42)
LX (3,1) 13.77 (.70) 13.76 (.70)
LX(4,1) 14.66 (.74) 14.66 (.74)
LX(5,1) 6.28 (.35) 6.28 (.35)
LY(1,1) - (.31) - (.32)
LY (2,1) 4.74 (.48) 4.79 (.50)
LY (3,1) 3.88 (.77) 3.97 (.74)
LY (4,1) 4.19 (.47) 4.23 (.48)
LY (5,2) — (.34) - (.34)
LY (6,2) 4.73 (.31) 4.74 (.31)
LY (7,2) 5.22 (.67) 5.23 (.67)
LY (8,2) 3.53 (.27) 3.52 (.27)
LY (9,2) 5.17 (.63) 5.18 (.63)
LY(10,2) 4.52 (.34) 4.53 (.34)
LY(11,2) 2.99 (.20) 3.02 (.20)
TD1 9.77 (.48) 9.72 (.47)
TD2 12.88 (.82) 12.90 (.82)
TD3 10.29 (.51) 10.30 (.51)
TD4 9.50 (.46) 9.50 (.46)
TD5 13.14 (.88) 13.14 (.88)
TE1 12.67 (.90) 12.54 (.90)
TE2 10.60 (.76) 10.38 (.75)
TE3 3.66 (.41) 4.30 (.45)
TE4 11.07 (.78) 10.87 (.77)
TE5 12.85 (.88) 12.85 (.88)
TE6 12.98 (.90) 12.98 (.90)
TE7 8.87 (.55) 8.95 (.56)
TE8 13.19 (.93) 13.20 (.93)
TE9 9.80 (.61) 9.86 (.61)
TE 10 12.87 (.88) 12.87 (.88)
TB 11 13.38 (.96) 13.38 (.96)
TE(1,2) 3.13 (.17) 2.93 (.16)
TE (6,5) 5.51 (.28) 5.51 (.28)
TE (5,10) 3.00 (.15) 2.99 (.15)
TB (6,10) 3.22 (.16) 3.21 (.16)
TE(11,10) 4.27 (.21) 4.26 (.21)
TE (5,11) 2.33 (.11) 2.32 (.llL
185.76 186.01
DF 95 96
GFI .94 .94
RMSEA .051 .050

 

Standardized estimates are in parentheses.

80

The overall fit of the constrained model was not appreciably worse than model A.
As presented in Table 15, little difference between the two models was evident. Model fit
and parameter estimates from both models were adequate (f7 /df < 3, GF I >.9,
RMSEA<.1). There was no significant difference in A3”) (df=1)=.40. Thus, as
hypothesis H5a suggested, a mediated relation was supported.

Interestingly, by deleting the negative direct effect of distraction on performance
in the second model, the effect of control on performance appeared significant. This
result indicated the interrelated nature of the physical environmental features in the

workplace.

H5b: According to the individual work style, there is a difference in the relationship

between distraction, control, and performance.

It was assumed that according to employee work styles, there might be a
difference in reaction toward quietness/noise, so distraction and the relationship between
control and performance might be different. Individual work style was measured with
one indicator. The respondents were classified into two groups based upon a median split,
one used to working in. quiet areas and the other did not care that much about quietness.
Before the hypothesis was tested, an independent sample t-test was conducted. There
was no significant mean differences for the five distraction variables with respect to
individual work style at .05 level of probability.

In order to test the hypothesis, an equality test was conducted to compared the
two groups. Chi-square difference tests were performed to examine the difference in the

paths between the two groups. Assuming that there are no differences in paths, equality

81

constraints were given in three paths. These constraints were released one by one. The
results of chi-square difference tests are summarized in Table 16. In pe0ple who care
more about quietness, the effect of distraction on performance (-.22) was negative, but
the effect of distraction on performance (.35) in people who do less was positive. In
other paths, there was no significant difference. Between the two groups, other positive or
negative directions were consistent, however, the effect of distraction was significantly
different for people who care more about quietness and for employees who do less care
noise. Thus, the hypothesis was supported. This result is consistent with the findings of
Kjellberg et al. (1996) but not with the findings of Block & Stokes (1989).

Table 16. Chi-square Difference Test by Individual Workstyle

 

 

Model A12 (df=1)

Equality released

15‘ EQ GA (1,1) =GA(1,1) 5.55 (p<.01)

2“d EQ GA (2,1) = GA (2,1) .00 ( non significant)

 

Effects of Flexibility on Group Cohesiveness
With respect to antecedents of flexibility, four variables were examined in order
to examine the effects of flexibility on task group cohesiveness. Openness and
accessibility were also entered as independent variables to explain task. group

cohesiveness.

H6a: Flexibility in workspace use ismositively related to task group cohesiveness.

H6b: The degree of openness in the workplace is positively related to task group

cohesiveness.

H6c: The degree of accessibility in the workglagce is positively related to taslggroup

cohesiveness.

82

The R2 value for the model was .067 with an adjusted R2 =.052. The F was
statistically significant at .001 probability level, [ F(361, 6)=4.324, P=.000]. The
standardized and unstandardized regression coefficients are presented in Table 17.
General flexibility contributed the most to task group cohesiveness (13:. 139), followed by
convertibility, flexible enough to adjust, rearrange, or reorganize (B=.137). The score of
the Kolmogorov-Smirnov test was .937 (p=.343). Standard residuals were normally
distributed. There was no specific pattern or clustering found in the scatter plots (see
Figure B3).

Table 17. Regression Analysis of Flexibility on Task Group Cohesiveness

 

 

Independent variables B B T P R2
Constant 4.276 16.973 .000

Openness -.015 -.030 -.570 .569
Accessibility .038 .054 1.012 .312 .067
Convertibility .072 .137 2.088 037* .0528
Versatility .026 .045 .813 .417

General flexibility .084 .139 2.184 .030*

Movable systems -.034 -.061 -1.076 .283

 

* p<.05 a = adjusted R2

The results indicated that flexibility was positively related to task group
cohesiveness although the proportion of explanation for task group cohesiveness was
small. Therefore, hypothesis H6a was confirmed. No significant relationship was found
between openness and task group cohesiveness or between accessibility and task group

cohesiveness respective1y. Thus, hypotheses H6b and H6c were not supported.

H7a: Flexibility in workspace use is not related to social group cohesiveness.

H7b: The degree of openness in the workplace is not related to social group cohesiveness.

83

H7c: The degree of accessibility in the workplace is not related to socialgroup

cohesiveness.

To examine the effects of flexibility on social group cohesiveness, the four
flexibility variables were studied with openness and accessibility. This linear model was
not statistically significant at the .05 probability level, [ F(362, 6)=1.582, P=.151]. None
of the t- values were significant; thus regression coefficients were considered as being
equal to zero. There was no obvious evidence of multicollenearity (all tolerance values
were higher than .1 and all VIF were lower than 2). The score of the Kolmogorov-
Smirnov test was .748 (p=.630). Standard residuals were normally distributed. Thus,
hypotheses H7a, H7b and H7c were not rejected.

As expected, flexibility was not associated with social group cohesiveness
because social group cohesiveness refers to social relationships among colleagues outside
of the workplace rather than inside the workplace. However, the findings support the
hypothesis that flexibility in the workplace design had a positive influence on task group

cohesiveness.

H8a: ngree of openness is positively related to distraction.
H8b: Distraction is negatively related to job satisfaction.
H8c: Job satisfaction is positivelmlated to perceived performance.
H8d: Task group cohesiveness is positively related to perceivecherformance.

To test the hypothesis, the proposed model (see Figure 7) was analyzed using
SEM. In the initial run, the model was not converged using both Maximum Likelihood
estimation and Generalized Least Squares estimation. To test the hypothesis, Phi matrix

(covariance matrix of latent factors) was used to conduct a path analysis. Phi matrix was

84

generated from the initial run of the Confirmatory factor analysis. Because CF A does not

include the openness variable, H83 was not tested using path analysis. The path analysis

results are presented in Table 18 and Figure 13. The initial result indicated that the path

analysis was perfectly fit. In the initial run, the path coefficient between flexibility and

task group cohesiveness was not significant and the path coefficient between control and

task flexibility was significant.

Table 18. Summary of Path Analysis

 

 

Parameters Estimates T-value
PH (1,2) Control (--) Distraction -.60** -10.24
PH (13) Control (—) Flexibility 57” “-08
PH (23) Distraction <—) Flexibility “-44" '8-01
GA (2,1) Control ——> Job satisfaction .26** 7.32
GA (3,1) Control —-) Performance -06 --76
GA (1,2) Distraction —) Task Group Cohesiveness -.32** -6.29
GA (2,2) Distraction —> Job satisfaction -03 -85
GA (3’2) Distraction —> Performance -03 -47
GA (13) Flexibility —> Task Group Cohesiveness .l6** 13.13
GA (3,3) Flexibility —-> Performance .05 .76
BE (1,2) Task Group Cohesiveness» Job Satisfaction .72** 22.80
BE (1,3) Task Group Cohesiveness» Performance ~17* 2- 13
BE (23) Job Satisfaction —) Performance 21" 2-60

 

Standardized Total effects

Standardized Indirect effects

 

 

Control Distraction Flexibility Control Distraction Flexibility
TGS -.32 .16 - ' - -
JS .26 -.20 .ll - -.23 .11
PER .00 -.07 .10 .06 -. 10 .05
TGS J S PER TGS J S PER
TGS - - - - - -
J S .72 - - - - -
PER .32 .21 - . 15 - -

 

 

 

* p<.05, ** p<.01, Estimates are standardized.

The “theory trimming” approach was employed (Ferguson & Weisman, 1986). As

two paths were trimmed from the initial run, the path relationships showed similarity to

85

the proposed model, however, the overall fit was reduced (12:59.37, df=2, GFI=.95,
RMSEA=.269).

In the path diagram, control was negatively associated with distraction, thus the
result was consistent with previous findings of this study. No significant effect of
distraction on job satisfaction was found (T value=.85). Thus, hypothesis H8b was not
supported. A positive effect of job satisfaction on perceived performance was found (.21,
T-value=2.60). Thus, hypothesis H8c was confirmed. T value indicated that task group
cohesiveness was positively related to performance (.17, T-value=2. 13). The total effect
of task group cohesiveness on performance was .32 including an indirect effect of .15.
Therefore, the hypothesis H8d was also confirmed. The significant results in the
relationships among task group cohesiveness, job satisfaction and performance were not
changed between the path analysis results before and after trimming.

Figure 13. Path Analysis Results

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

r ------------------------------------------- I
Control Kim‘. 1
I
I
Job :
"60" ,x’” Satisfaction l
I” AL
.67** Distraction I: _______________________________ > Performance
.72** A
I
I
--44** Task group :
Cohesiveness
I
‘i Flexibility E

 

 

—-——+ significant *p<.05, “p<.01

--------- -> non significant

Path coefficients reflect only the direct effects of one variable on another in the

model, while correlation coefficients include direct effects, indirect effects, and

86

correlation between two variables and a third variable not included in the model
(Ferguson & Weisman, 1986, p99).

In order to examine the effects of openness on distraction, correlation analysis
was conducted between openness and the five distraction variables. Visual distraction
was significantly associated with openness (r =.168, p=.001). A summated distraction
indicator of five variables was also positively associated with openness (r =.111, p=.03).
Thus, hypothesis H83 was confirmed. This result was consistent with the previous
finding (Ferguson & Weisman, 1986)

The effect of distraction on job satisfaction was examined using regression
analysis. The squared multiple correlation (R2) was .100 with an adjusted R2 =.088. The
linear combination of five variables accounted for 10% of the variance in the overall job
satisfaction variable. The F was statistically significant at .001 probability level,
[F(373,5)=8.247, P=.000], allowing rejection of the null hypothesis. The standardized
and unstandardized regression coefficients are presented in Table 19.

Table 19. Regression Analysis of Distraction on Job Satisfaction

 

 

Independent variables B B T P R2
Constant 6.058 27.133 .000
Undisturbed workplace (R) -.013 -.018 -.282 .778

Easily distracted -.083 -.l 11 -2.067 .039* .100
Quietness (R) -.079 -.110 -1.781 .076 .0883
Quiet and undisturbed time alone (R) -.l 16 -.173 -2.793 .005**

Visual distraction -.007 -.011 -.207 .836

 

* p<.05, ** p<.01 a = adjusted R2
The beta coefficients indicated that all variables related to distraction were

negatively associated with job satisfaction. However, the significant value of

87

Kolmogorov-Smirnov test indicated that the residuals were significantly different from
normal (z=1.386, p=.043). The distribution values in each scatter plot were not all
equally distributed above and below the regression line across the values of an
independent variable (see Figure B4). Especially for ‘quietness’ and ‘quiet and
undisturbed time alone,’ variance increases with the independent variables. This violates
residual assumptions. Since the path analysis result did not support hypothesis H8b, and
regression results were not robust, due to violations of linear regression assumptions,

hypothesis H8b was not supported.

Gaps between Perceptions and Expectations

In order to determine employee’s expectations toward their workplace,
expectation levels were investigated with perception levels for physical environmental
features. The differences between perceptions and expectations are presented in Table 20.
Standard deviations are shown in parentheses.

The largest gap was found in temperature and air control (The mean difference
was —3.698). The results indicated that employees participating in this study expected to
have more flexibility in their workplaces to readjust, reorganize, have quiet and
undisturbed time, have more lighting control, have more adequate storage area, and have
more control over social contacts compared with the existing condition of their
workplaces. The results showed the importance of control and flexibility in the workplace.

The gaps showed which features should be improved in the workplace.

88

Table 20. Summary of Differences between Perceptions and Expectations

 

 

Items Perception Expectation
Mean (St. D) Mean (St. D)
1 My workplace is Open enough to see my 4 12 (2 088) 3 12 (1 722)

colleagues working.
2 My workplace provides an undisturbed

 

 

environment so that I can concentrate on my 3.35 (1.709) 5.64 (1.272)
work.

3 My furniture 1s flexrble enough to adjust, 3.05 (1.985) 5.32 (1.460)
rearrange, or reorganize my workspace.

4 I am able to control the soc1al contact With 394 (1.792) 5.48 (1.300)

others around me.
5 Ihave informal and impromptu meetings in my 5 52 (1 493) 5 91 (1 191)
private workstation. ' ' ° '
6 The quality of my equipment is more than
sufficient to work effectively. 4'89 (L626) 6'23 (0942)
7 My workplace serves multi-purpose functions 4 80 (1 754) 5 65 (1 283)
for informal and instant meeting. ' ' ' '

 

 

 

 

 

8 I am able to personalize my workspace. 5.18 (1.558) 5.90 (1.067)

9 I am able to control temperature or airflow 1n 1.40 (1.004) 5.10 (1.750)
my office.

10 My work environment is quiet. 3.18 (1.676) 5.46 (1.221)

11 lam ableto determine the 4.73 (1.802) 5.75 (1.126)
organization/appearance of my work area.

12 I am able to be easrly accessed from my 5.43 (1.520) 5.29 (1.393)

colleague’s workstation.
13 My workstation lS over-equ1pped for my typical 2.05 (1.246) . 3.08 (1.777)

 

 

 

 

 

 

 

needs.
14 In general, my workspace is flexible. 3.81 (1.716) 5.41 L1.094L
15 Ml fumiture is a fixed system.(R) 2.55 (L857) 450(1 .710)
16 My workstation is large. 3.85 (1.692) 4.82 Q.319)
17 I am able to have quiet and undisturbed time 3.24 (1.806) 5.79 (1.101)
alone.
18 My work area has many visual distractions. 3.10 (1.794) 2.54 (1.500)
19 I have ample storage in my work area. 4.32 (2.007) 5.95 (1.088)
20 I am able to control the lighting level in my 3.23 (1.998) 5.59 (1.329)
workstation.

 

Next, the effectiveness of gap measures and perception measures was examined.

89

H9a: Gap measures between the perceived environmental features and the expected
environmental features are stronger predictors in explaining environment

satisfaction and job satisfaction than percgtion based measures.

Table 21. Variation explained by Perceptions and Expectations toward Workplace

 

 

 

Variables Perception Gap
WS .13 WS JS
1 My workplace is open enough to see my colleagues
working.
2 My workplace provides an undisturbed environment .218*** .175**
so that I can concentrate on my work.
3 My furniture is flexible enough to adjust, rearrange, .148** .1 17*
or reorganize my workspace.
4 I am able to control the social contact with others .133* .144**
around me.

5 I have informal and impromptu meetings in my
private workstation.

6 The quality of my equipment is more than sufficient .108*** .142**
to work effectively.

7 My workplace serves multi-purpose functions for
informal and instant meeting.

8 1 am able to personalize my workspace. .122* .124*
9 1 am able to control temperature or airflow in my .105*

office.
10 My work environment is quiet. .144*
l 1 I am able to determine the organization/appearance .1 14* .144** .150**

of my work area.
12 I am able to be easily accessed from my colleague’s

workstation.

13 My workstation is over-equipped for my typical -.083*
needs.

14 In general, my workspace is flexible. .137** .151**

15 My furniture is a fixed system.(R) .1 1 l**

16 My workstation is large.

17 1 am able to have quiet and undisturbed time alone. .173*** .126* .l47**

18 My work area has many visual distractions.

19 l have ample storage in my work area. .178*** .126**

20 I am able to control the lighting level in my .1 14** .120**
workstation.
R2 .476 .173 .445 .134
Adjusted R2 .460 .161 .433 .124

 

* p<.05, ** p<.01, *** p<.001
Entries in the cells represent standardized coefficients. All nonsignificant cocffcients are omitted.
WS: satisfaction with work environment 18: Job satisfaction

90

Stepwise regression analysis was conducted for each measure. The results are
summarized in Table 21. Perception measures had a larger number of significant
predictors (10 items) among 20 items for work environment satisfaction than did gap
measures (7 items). For job satisfaction, no substantial difference between perception
measures and gap measures was found. The values of squared multiple correlation (R2)
and adjusted R2 were not considerably different. Based on this result, the gap measure
did not have greater predictive power compared to the simple perception measure.
Therefore, the hypothesis H9a was not confirmed. The results are further discussed in a

later part of this chapter.

H9b: Satisfaction with the physical environment is positively related to the job

satisfaction.

To examine the relationship between satisfaction with the physical environment
and job satisfaction, correlation analysis was conducted between openness and four job
satisfaction variables at the .001 probability level. A summated job satisfaction indicator
of four variables was also positively associated with work environment satisfaction
(r= .434, p=.000). Thus, hypothesis H9b was confirmed. This result was consistent with
previous findings. Ferguson & Weisman (1986) also found that satisfaction with

workspace was positively related to job satisfaction.

91

DISCUSSION

In this section, further interpretations of the findings are discussed for each
hypothesis.

H l a: Control over the physical work environment is positively related to job satisfaction.

Results regarding control in the workplace on employee’s job satisfaction are
consistent with previous findings. The results of this study found that control was a
predictor of job satisfaction, which was consistent with previous findings (McLaney &
Hurrell, 1988; Spector, 1989). Control over the physical environment was a weaker
predictor of job satisfaction than control over one’s work. Especially, control over
temperature and control over lighting were not significantly related to job satisfaction,
although employees indicated that they should have more control over temperature and
lighting than they had. This may be due to the building conditions of the participants
included in the study. Not all of the buildings provided a local unit or control device for
heating and ventilation. Thus, inability to control temperature, air flow or lighting did
not affect employee’s job satisfaction because they recognized these as factors outside of
their control.

The study findings emphasized that control is an important feature in the
workplace in order to obtain high quality employees and have them satisfied with their
jobs. The results showed that being able to control social contact was a predictor of job
satisfaction. In order to balance individual work and group work, employees needed to
have control over social contact. Social contact control could be achieved by providing

movable doors for individual workspaces, separate closed rooms, or hubs which

92

employees could stay temporarily and do their works. It would be especially important
for workers who require a high degree of concentration. Control over organization and
personalization were predictors of job satisfaction. Providing options for workplace

layout could increase control over the workplace.

Hlb: Control over the physical work environment is positively related to perceived

Lerfomance.

The findings indicated that control over the physical work environment was not
positively related to perceived performance. This may be due to the questionnaire design
or the complicated nature of measuring performance. For the perceived performance
variables, the results indicated that employees were more likely to evaluate their
performance as above the mid point (the data were negatively skewed). This finding may
result from the small variations available in using 5 point- scale performance variables.

Research findings in this area have not been consistent. Some researchers have
found that control regarding the physical aspects of the office environment was not
positively related to performance (Veitch & Gifford, 1996) while control over one’s work
was positively related to performance (Spector, 1989).

Although the findings did not support the hypothesis, the common belief that
control will increase performance remains questionable since actual performance may be
different from perceived performance. This could be further examined by using different

questionnaire designs or a different methodological approach.

H2 : Control over the physical environment and work is negatively related to distractions.
The findings indicated that as employees perceived higher control over their

work and work environment, they were less distracted by their workplace. This finding

93

supported the proposition that control plays a role in not only increasing job satisfaction,
but also in decreasing distraction.

As previously mentioned, open offices have been viewed as problematic,
distraction from noise and frequent interruptions are common in them (Brooks & Kaplan,
1972; Hedge, 1982; Evans & Johnson, 2000). Control over the physical environment can
directly or indirectly reduce the negative effects of distraction, reduce perceived
helplessness from distraction and compensate employees experiencing distractions in the
workplace.

H3a: Perceived distraction levels in the workplace are negatively related to perceived

employee performance.

H3b: Perceived distraction levels in the workplace are positively related to an employee’s

inclination to work alone.

H3c: Perceived distraction levels in the workplace gre positively related to an employee’s

inclination to avoid meeting pe_ople in the workplace.

H3d: An employee’s inclination to avoid meeting people in the workplace is positively

related to an employee’s inclination to work alone.

The findings did not support the hypothesis that distraction was negatively related
to perceived performance. This may result from the distraction level not being severe
enough to keep employees from working. O’Neill (1994) found that distraction was
negatively related to self-assessed performance.

As expected, if employees felt more distracted, they were more likely to want to
work alone and avoid meeting people in their workplace. If employees did not need to
work collaboratively or in teams, the propensity to work alone or avoid meeting people

would not cause problems. However, if high group cohesiveness is required for work or

94

if employees need to work intensively collaboratively, the propensity to work alone
would not be desirable.

This finding can be applied to workplace design and management. If employees
need to work individually or require concentration for their work, facility managers need
to reduce the distraction level, for instance, by dividing working zones, using higher
panels, using sound absorbing materials, raising ceiling heights, or using sound masking
devices. If high distraction levels result from the nature of work activities, facility
managers can provide work areas for employees to work alone or to avoid meeting

people when it is necessary.

H4: There is a difference in terms of effects of distraction on performgapce between the
high control group and the low control group.

According to different levels of control, there was no significant change on the
effects of distraction on performance. This may be the result of from several reasons.
There may be little distinction between the two groups since the two groups were defined
based on continuous factor scores. If control and distraction are independent of each
other, factorial design may reveal the moderating effects clearly. For instance,
comparing the performance by groups of high control-high distraction, high control-low
distraction, low control-high distraction and low control-low distraction, can explain
moderating effects clearly. If the direct effect of distraction is much stronger than the
indirect effect moderated by control, the moderating effects may appear non significant.
This inquiry was explored in the following hypothesis by testing the mediating effect of
control in the relationship.

H5a: Controlplays a mediating role in the relation of distraction to performance.

95

H5b: According to the individual work style, there is a difference in the relationship
between distraction, control and performance.

Interestingly, a mediating role of control and its relationship was found regarding
negative physical environmental features, distraction. This is a support for Sundstrom’s
(1986) proposition that performance not only is affected by noise but also by one’s sense
of control. It is evident that control relieves the negative effects of distraction on
performance.

The results indicated that individual work styles moderated the effects of ’
distraction on control and performance. For an employee who desired quietness and who
was used to working alone, the effect of distraction on performance was negative.
However, for an employee whose work style is coffee shop regular and less relied less on
quietness, distraction was not negatively related to perceived performance. This finding
emphasized the individual differences and preference toward the workplace, and the

importance of various options of the workplace conditions to support individual needs.

H6a: Flexibility in workspace use is positively related to task group cohesiveness.

H6b: The degree of openness in the workplace is positively related to task group

cohesiveness.

H6c: The degree of accessibility in the workplace is positively related to task group

cohesiveness.

It was found that employees who perceived their workspace to be more flexible,
have higher task group cohesiveness. This result emphasized the importance of
flexibility in workplace design, especially for workers who worked as teams or required

frequent collaboration to achieve group goals.

96

However, the roles of openness and accessibility in office design remain
questionable. The results did not confirm whether openness or accessibility leads to an
increase or decrease in communication and interaction, or in group cohesiveness. Since
openness was positively correlated with distraction and negatively correlated with task
cohesiveness, it may due to the reasoning in Evans and his colleague’s proposition (1994).
They proposed that insufficient privacy hampers the development of close interpersonal
relationships because workers feel unable to hold confidential conversations and have
limitations placed on feedback from supervisors (Oldham & Rochford, 1983; Evans,

J ohansson, & Carrere, 1994). Among the respondents from this study, some people
added the following comments supporting this reasoning. “F or a manager the Open air
office is not eflective. Private discussions on performance, pay, goals, etc. are difficult to
have in a public environment.” “For consumer interaction, ...manager should be able to
control the social contacts with others around him/her... confidential information, voice
should not transfer....” This indicates that one of the intentions for planning openness in
office design, which is to increase interaction and communication, fails to be fulfilled and
instead resulted in reverse effects. This result was consistent with that of Brennan and
her colleagues (2002). They found that the open office did not facilitate communication
since employees felt that open offices disallowed employees to have confidential
conversations.

In this study, accessibility referred to the degree to which an employee can be
accessed by colleagues. For future study, accessibility needs to be measured containing

broader dimensions of the concept.

H7a: Flexibility in workspace use is not related to social group cohesiveness.

97

H7b: The degree of openness in the workplace is not related to social group cohesiveness.

H7c: The degree of accessibility in the workplace is not related to social group

cohesiveness.

As expected, flexibility was not related to social group cohesiveness, i.e., social
relationships among colleagues out side of the workplace. It also found that openness and
accessibility in the physical environmental features of the workplace were not related to
social group cohesiveness. Interestingly, flexibility was positively related to task group
cohesiveness, but not to social group cohesiveness. If social cohesion refers to
interpersonal closeness in the workplace, the result might be different.

H8a: Degree of openness is positively related to distraction.

H8b: Distraction is negatively related to iob satisfaction.

H8c: Job satisfaction is positively related to perceived performance.

H8d: Task group cohesiveness is positively related to perceived pprformance.

The results indicated that the more open the workplace is, the more distractions
employees in the workplace experience. Therefore, open office design inherently faces
some degree of distraction problems compared to a closed office type. Especially for
open offices, it is important to deal with distraction. Hedge (1982) also concluded that a
workplace which has greater openness and accessibility between the staff would
unavoidably produce a variety of disturbance problems.

Regarding hypothesis H8b, there was a conflict in the findings based on the
regression analysis and the path analysis. Regression analysis indicated that distraction
was negatively associated with job satisfaction. However, path analysis failed to support

the direct effect of distraction on job satisfaction. The regression analysis result was not

98

robust since assumptions for regression analysis were deviated. Therefore, it was
concluded that the direct effect of distraction on job satisfaction was not supported.

The result of path analysis indicated that job satisfaction was positively related to
self- reported performance. It was also found that people who had higher task group
cohesiveness, rated their performance higher. This result is consistent with the findings
of Carless and DePaola (2000), where task group cohesiveness was more closely related
to work performance than was social cohesiveness.

H9a: Gap measures between the perceived environmental features and the expected

environmental features are stronger predictors in egglppning environment

satisfaction and iob satisfaction than perception-based measures.

H9b: Satisfaction with the physical environment is positively related to thegmb

satisfaction.

The research finding supported that as employees are more satisfied with their
physical work environment, they have greater job satisfaction. The result indicated strong
correlation between satisfaction with work environment and job satisfaction. This results
is consistent with the previous findings (Sundstrom et al., 1994; Ferguson & Weisman,
1986)

When using the gap measure, employees showed large discrepancies between
perception of their current status and their expectations toward control, flexibility and
workplace adequacy aspects. However, the study findings did not show superiority of the
gap measure to simple perception measures in explaining satisfaction with the workplace
environment and job satisfaction.

There are two possible interpretations regarding these results. One may be due to

the questionnaire design. To reduce the time and to increase response rates from

99

employees, the items for measuring perception and expectation were parallelized (See
Appendix A). Respondents showed a strong tendency to reply on both measures at the
same degree of agreement or disagreement. People may judge two items as similar
because they look almost identical or people may be reluctant to answer differently to the
similar questions.

The other interpretation may be due to the method of analysis. Cronin and Taylor
(1992) conducted stepwise regression analysis to test whether a weighted performance —
based measure of service quality (perception only, SERVPERF) is a more appropriate
basis for measuring service quality than SERVQUAL (gaps between perception-
expectation). In this study, the Cronin and Taylor method was repeated. To decide the
predictor of the physical work environment satisfaction, linear stepwise regression
analysis was used. Linear regression analysis assumes linear relationships between
independent variables and dependent variables. If there is no straight linear relationship,
the results indicate weaker relationships between the independent variables and
dependent variables. For instance, as shown in Figure 14, once people fulfill their
expectation level (no discrepancy between the perception and expectation), the exceeding
portion (current status exceed the desired level) does not increase satisfaction any more,
thus, it does not have a straight linear relationship.

To test this assumption, scatter plots with fit lines were generated. The two
figures show the differences when a researcher uses a linear regression option vs,
quadratic regression or lowess options. The difference proves the proposition. For the

relationship between simple perception of physical environmental features and physical

100

environmental satisfaction, the linear line option and quadratic or squad option did not
produce a great amount of difference in the shape of lines.

Figure 14. Linear and Nonlinear Relationship in Perception and Gap Measure

 

 

 

 

 

 

 

 

8
7 I 1 11 u .1
6 q i) .1 1 J I1 11
5 1 ti 1 U n d (J O
C ,
.9
'6
£9 41 *1 11- n i 1 (:1 ii
.‘2
E ,.
3 31 i L) D r G U 1
C
8
c 21 U 3 r) o
O
.L':
>
C
q) 1 1 L1 L1 ti
.2
5
g 0
0 1 2 3 4 5 6 7 a perception
8
7 I :1 11 (1 n U
6 I I 11 t‘ u (i U C: 0 l) 1
ii
51 11 ii ri F ”p.--“ 4“ ii _'_i _, - .. U
C
.9
8 44 ii 11 1 11 11 U U
“—-
.‘L’
‘6
m 3 I .. ..i U rt 3 a C1 (1 0
a
C
“5’
c 2 q l (l (:1 r) i 1 O
O
.2
>
C
q; 1 1 o 1. 1* L
x
S
g 0
-8 -6 .4 -2 o 2 4 6 gap

However, for the gaps, there are big differences in the shape of the linear line
option and in that of the quadratic or lowess option. In the non linear line plot, the line
shows that the increase of gaps is in proportion to that of satisfaction until the point at
which the gap is 0. After that point, the value of Y (satisfaction) does not increase in

proportion to that of X (gaps). Therefore, this result indicates that there is a weakness in

101

the conclusion Cronin and Taylor (1992) made. In their conclusions, based on linear
analysis, a simple perception measure explains more of the variation to predict service
quality.

Since the gaps do not have straight linear shapes, it is not valid to compare the
results based on linear regression analysis. Furthermore, this result indicated that
physical environmental features that are lower in quality than people expect are strongly
related to satisfaction, while the physical environmental features which surpass the
expectation do not affect satisfaction notably.

,In a simple graphical presentation, the gaps model is of use to present which
physical aspects should be improved. The issue of which is a better predictor of work
environment satisfaction is different from that of which aspects improve the work
environment. This study did not deal with thorough and in-depth comparisons for
perception measures and gap measures regarding the objective physical environmental
features. Instead, search in-depth comparisons are left for future study since there were

limitations to investigating the objective physical environmental characteristics.

102

CHAPTER VI SUMMARY AND CONCLUSION

This concluding chapter summarizes the objectives of the study, conceptual
models, research methods, and the main findings of the study. It concludes with a
discussion of the limitations of the study and suggestions for future studies. Implication

for practice in workplace design and management are also presented.

Summary of Research Objectives and Literature Review

In today’s changing workplace, the importance of the role of the physical
environment in supporting organizational goals and changing needs is increasingly
acknowledged. However, many workplace designs have not yet successfully supported
these changing needs, since they do not resolve the conflict of placing people in
individual workplaces while at the same time expecting them to work collaboratively.
Therefore, the objective of this study was to examine the effects of control and flexibility
on employee’s out comes in the workplace. The purposes were 1) to investigate the
effects of control (including control over the physical environment by individuals),
whether control plays a role in mitigating the negative effects of distraction and
disturbance in the work environment, and to lead to better job satisfaction and higher
performance at the individual level, and 2) to investigate the effects of flexible use of the
workplace on group cohesiveness at an interpersonal level.

Control has been intensively explored in the field of psychology and in job design.
However, control over the physical environmental features has not been investigated in a

comprehensive way in its relationships with work control. Some studies regarding control

103

over physical environmental features limit the scope to lighting or the thermal
environment. Furthermore, a very few studies have dealt with a moderating or mediating
role of control in physical milieu. For instance, density in residential design was
investigated to examine the moderating role of control. In the workplace, noise, a
stressful source, has been investigated intensively among ambient environmental features,
however, with a lack of interrelationship with other physical environmental aspects.
Therefore, based on the literature review, it was concluded that the effects of control over
both work dimensions and physical environmental dimensions need to be studied to find
the influences on work outcomes and on mediating processes in the workplace.

While teamwork and group work have been emphasized in the workplace, few
studies have been conducted to investigate the relationship between the physical
environmental and interpersonal relationships within the workplace. Few studies have
explored how the physical environment might support not only individual work, but also
group work. Flexibility has been emphasized in workplace design and in workplace
management, but little empirical study has been conducted.

Besides the effects of control and flexibility, this study dealt with how important
these features were to employees. Research into satisfaction with the physical
environment has focused primarily on how satisfied employees are with certain features
of the physical environment instead of measuring perceptions and expectations toward

the physical environmental features.

104

Summary of Methods
In order to carry out the objectives of this research most effectively,

questionnaires were used in a natural setting with cross- sectional design. A questionnaire
was developed for this study and was pre-tested. Research participants were office
employees in manufacturing corporations located in Michigan. Organizations with more
than 150 employees were sought. An organizations list was developed from the ‘largest
employer’ list from the Lansing Chamber of Commerce, the local stock interest listings in
the Lansing State Joumal, a list from the Troy Chamber of Commerce and the IF MA
(International Facility Management Association) Michigan directory. Formal letters and
follow-up calls or e-mails were sent to human resource managers or facility managers.
Three companies agreed to participate in this study. A total of 409 questionnaires were
collected though web survey and written paper survey and 384 responses were analyzed
for this study. Data were analyzed using SPSS and Lisrel based on the proposed

hypotheses.

Summary of Findings
The majority of respondents indicated that they worked individually as well as
collaboratively. Approximately 28% of the participants were intensively involved in
engineering, 12% of the respondents work for marketing/sales, and 10% of the
respondents work in finance. An open office with high partitions was dominant in
proportion for office type. Sizes of buildings where participants work varied (square
footage per occupant varied from 51 to 243). All buildings provided central heating and

ventilation systems. None of the building had a history of major renovation.

105

The results of each hypothesis tested are summarized in Table 22.

Multiple methods were employed to test each hypothesis. For simple relationships,
linear regression analysis and correlation analysis were conducted. SEM was also used to
test the hypothesis in the proposed models.

The reliability test of the variables indicated that most constructs achieved good
or acceptable reliability. A confirmatory factor analysis was performed for the constructs
with multiple indicators. The results of CFA showed a marginal acceptable level.
Therefore, for each model, CFA was conducted again for the latent factors of the sub
model. The proposed model had an adequate level of goodness of fit.

For hypothesis tests, T-values and chi-square differences were used. For group
analysis by different levels of control and by individual work styles, chi-square difference
test were performed.

Based on the results of the study, control had a positive effect on job satisfaction.
For physical dimensions, ‘personalization,’ and ‘determination of organization and
appearance in the workplace’ were significantly related to job satisfaction. ‘Control over
social contacts’ was also a predictor of job satisfaction. Control over temperature and
lighting control were positively related to satisfaction with the work environment but not
to job satisfaction. It was not found that control was positively related to perceived
performance.

The results indicate that control plays a mediating role in the relationship between
distraction and perceived performance. Different levels of control did not produce
significant differences in the negative effects of distraction on performance. It was found

that individual work styles affected the effects of distraction. Individual work style was

106

measured with a semantic differential scale (‘go to an office or a private and quiet area’
versus ‘coffee shop regular’). In people who care more about quietness, the effect of
distraction on performance was negative, but the effect in people who do less about
quietness was positive.

As expected, flexibility was positively related to task group cohesiveness and not
related to social group cohesiveness. Among antecedentsof flexibility, the regression
result indicated that general flexibility and convertibility were significantly related to task
group cohesiveness. There was no significant relationship found between openness and
group cohesiveness including both task group cohesiveness and social group
cohesiveness. Accessibility was also not related to group cohesiveness for either task
group cohesiveness or social group cohesiveness.

Participants in this study expected that they should have more control over social
contacts, temperature and lighting control, and should have more flexibility in their
workplaces to readjust, reorganize, and to have quiet and undisturbed time. Participants
also expected that they should have more adequate storage area compared with the
existing condition of their workplaces.

Against the proposition, it was not found that gap measures were superior
predictors for explaining work environmental satisfaction or job satisfaction to simple
perception measure. However, scatterplots showed some explanation for possible flaws in
the conclusion based on linear projection. The gaps model is useful for managerial
implication because the methods present which physical aspects should be improved.
The issue of which is a better predictor of work environment satisfaction is different from

that of which aspects improve the work environment.

107

Table 22. Summary of Hypotheses Tests

 

 

 

 

 

 

 

 

 

 

Hypothesis Method Results

Hla: Positive effects of control on job satisfaction Regression Supported

Hlb: Positive effects of control on performance Regression Not supported

H2: Negative effects of control on distraction Regression Supported

H3a: Negative effects of distraction on performance SEM Not supported

H3b: Positive effects of distraction on avoidance SEM Supported
meeting people

H3c: Positive effects of distraction on working alone SEM Supported

H4: Difference in effects of distraction on SEM Not supported
performance by control levels

H5a: Existence of mediating role of control SEM Supported

H5b: Difference in the relationship by solo work style SEM Supported

H6a: Positive effects of flexibility on task group Regression Supported
cohesiveness

H6b: Positive effects of openness on task group Regression Not supported
cohesiveness

H6c: Positive effects of accessibility on task group Regression Not supported
cohesiveness

H7a: No effects of flexibility on social group Regression Supported
cohesiveness

H7b: No effects of openness on social group Regression Supported
cohesiveness

H7c: No effects of accessibility on social group Regression Supported
cohesiveness

H8a: Positive relationship between openness and Correlation Supported
distraction.

H8b Negative effects of distraction and job Path analysis Not supported
satisfaction.

H8c: Positive relationship between job satisfaction and Path analysis Supported
performance

H8d: Positive effects of task group cohesiveness and Path analysis Supported
perceived performance.

H9a: Gap measures are stronger predictors than Regression Not supported

perception based measures

H9b: Positive relationship between work environment Correlation

satisfaction and job satisfaction

108

Supported

Limitation of the Study
This research used a questionnaire to investigate employee perceptions and

expectations toward physical environmental features in their workplace settings. Physical
environmental characteristics were inferred from employees’ perceptions since data were
collected by self-reporting. Bames (1981) enumerated some problems in using self
reporting measures to assess environments. On one hand, if the effects of the
environment are subtle, people are unaware of the influence of physical aspects. On the
other hand, self-reporting measures are subject to all the biases associated with inaccurate '
recall of experience (Taylor & F iske,1978; Barnes,198 l ). Although it is a weakness of
the study, the research findings show that perception of the physical aspects such as
lighting control and temperature control reflect a very low degree of control, which was
not significantly deviated from the objective physical environmental condition. There
might be questionable differences between the degree of control expressed and the actual
degree of control people possessed. Objective measures of physical aspects can provide
in-depth understanding of the physical environmental features, however, individual’s
appreciation for the same features might be quite different. Therefore, self-reported

assessments and measurements of objective environmental features need to compliment

 

the interactive nature of built environments.

This study dealt with multiple relationships in the workplace. Interpretation for
causal relationships should be based on the context and proposed model. Some
organizational characteristics may influence the relationships. The results will be valid
only for the established model and contexts. In spite of this weakness, the model suggests

a comprehensive approach in dealing with complicated and interwoven physical aspects.

109

 

The study is limited to three selected Michigan firms and the industry type is
limited to manufacturing corporations. The results should be interpreted with
characteristics of those organizations and participating employees. It requires careful
consideration for generalizations of the results based on the demographic profiles of
employees, the physical settings of offices investigated and the corporation
characteristics.

The sc0pe of flexibility in this study is very narrow. It is mainly related to
workplace characteristics, and how flexibly individuals can switch their workplace use
for either individual work or group work. Since there are few empirical studies, it is
meaningful to investigate the effects of flexibility, but the measurement of flexibility

needs to also be developed for different levels.

Implication for Future Studies

For future studies, several research directions need to be sought. First, the
analysis frame should be developed more thoroughly for the workplace. This study
primarily focused on control and flexibility, and included other physical and behavioral
variables associated with control and flexibility. To develop the analysis frame, focus
groups or other qualitative approaches might be conducted to reveal the new changing
workplace characteristics.

Second, the measurement needs to be improved to cover the comprehensive
nature of physical environmental aspects and to produce more reliable and predictable
results. The measurement used in this study was developed based on previous studies and

literature. Because of a lack of reliable tools to measure the proposed concepts, the

110

present construct of control and concept of flexibility need to be improved further to
reflect more facets of control and flexibility as well as to get more reliable results. With
physical dimension, behavioral characteristics associated with a certain physical setting
should be measured properly. Teamwork and individual work characteristics should be
explored more regarding the physical characteristics. In this study, individual work style
was measured with one indicator. Instead of directly asking questions, several behavioral
indicators need to be developed to measure a propensity toward a certain physical setting.
As mentioned earlier, multiple methods might be used to measure the concept. For
instance, for physical dimensions, both objective and subjective measures can be used
together.

Third, a different research design might reveal interesting facts regarding the
effects of control and flexibility. Longitudinal design can reveal effects that cannot be
caught in cross sectional design. For instance, by comparing employee’s reactions before
and after physical environmental changes, a researcher would reveal how people gain
control COping with the changes or how different levels of control aspects affect the
employees in their individual work and team work.

This study dealt with the effects of control on job satisfaction and performance. It
would be intriguing to identify physical environmental stress sources and investigate the
moderating role of control. The distinction between moderating role and mediating role
of control needs to be further elucidated.

The areas for teamwork should be further identified. As this study was based on
individuals focusing on balancing individual needs and interpersonal needs, collaborative

activities were not fully explored. Various collaborative activities and team- based work

111

should be described. A focus group method might reveal the shared thoughts of a group
regarding what kinds of physical environmental features should be provided to suppOit
their collaborative work. Some inquires still remained, such as what kind of flexibility in
space use can be given to individuals; flexibility in which areas, for which specific job
responsibilities, doing traditional team based work vs Internet based meetings.

Finally, the results indicate the effectiveness of measuring both perception and
expectation toward physical environmental features. The gaps represented the major
weakness in physical environmental features and showed which aspects should be
improved from the user’s perspective. The remaining area is to develop more
comprehensive physical environmental dimensions that influence satisfaction. These will
include the essential issues of designing, planning and managing the physical

environmental features in the workplace.

Practical Implications and Suggestions

Implications for workplace managerial practice and workplace design and
planning are suggested from this study.

Control over physical issues is related to the organizational culture and to
workplace options. Paciuk (1989) emphasized careful use of control action in the
workplace because if occupants fail to appreciate the efficiency of control devices, or if
they have a lack of required information, or a lack of understanding of energy use and
the consequence of alternative control, the benefits of control may be impaired. Control
cannot be achieved in a straightforward method. For instance, temperature control is

associated not only with local control devices, but also with direct sunlight. Fixed

112

windows and uncontrollable sunlight may severely add thermal discomfort. In order for
employees to be able to determine the organization and appearance of their workplaces,
workplace solutions should be based on required needs rather than on a set of fixed
system furniture solutions. To control social contacts, sliding doors or semi transparent
doors might be added for individual work stations, especially for managers or
professionals who deal with confidential matters. The results indicate that it is necessary
to provide workplace options that allow employees to control social contacts.

People inherently try to gain control. It is interesting to see how people gain
control, or try to gain it, from the following sentence one of the respondents voluntarily
added. . We do adjust light and temperature in the cubicle space, although we are
not allowed to do so. Some people unscrew the fluorescent lights in the ceilings over
their desks, others buy desk lights, heaters, and fans, which are against the fire code. We
also have small refrigerators and coflee pots tucked on or under our desks. ...” It may be
an extreme case, however it emphasizes an insightful fact that people try to make their
own work environment comfortable enough to work properly. Therefore, it is important
to provide appropriate controls to individuals through various workplace options and
workplace use guidelines.

Control issues are about utilizing limited resources to support individual needs
and satisfy occupants. Therefore, a facility manager, planner or designer should think
comprehensively and provide options that contribute to an appropriate level of control. It
is hard to satisfy every individual need, but it is feasible to allow workers to find better

ways to use their workspaces based on guided directions. Since the findings of this study

113

highlighted the importance of control and its interactive nature, a planner or a facility
manager should consider the physical workplace elements in a comprehensive way.

Flexible use of space should be designed based on individual work characteristics
and categories individuals fall into. This study focused on a narrow scope of flexibility,
the solution for individual workplaces and surrounding spaces for employees who switch
from individual work to group works, and vice versa. Flexibility in arrangement should
be carefully provided so it does not become a source of distraction.

Physical environmental features are interrelated, and require a comprehensive
approach. Improving control and flexibility for users cannot be one straightforward
solution, instead, it requires deliberately comprehensive procedures to fit the amount and
type of control and flexibility with the organizational culture, work activities, and

I'CSOUI'CCS.

114

APPENDICES

115

Appendix A: Questionnaire

 

 

Workplace Evaluation Questionnaire

Dear employee:

This survey will help us understand more about how you work and what you expect from your workplace.
This research is conducted by So Young Lee, a doctoral candidate in the Dept. of Human Environment
Design, specializing Facilities Design and Management at Michigan State University. This questionnaire is
intended to assess employees’ perceptions and expectations of their physical work environment. The
findings of this study could contribute to planning, management and improvement of workplaces from a

user’s perspective.

Your organization has agreed to cooperate with the research. Your participation is absolutely voluntary.
You may discontinue answering the questionnaire at any time or may refuse to answer certain questions.
You indicate your voluntary agreement to participate by completing and returning this questionnaire.
Please try to answer each question as honestly as possible. With my best knowledge, there is little potential
risk to participate in this study. The questionnaire takes approximately 10-15 minutes for an employee to
finish. Be assured that your response will be anonymous and confidential. Your privacy will be protected to

the maximum extent allowable by law.

If you have any questions about this study, please contact the investigator, So Young Lee. If you have
questions or concerns regarding your rights as a study participant, or are dissatisfied at any time with any
aspects of this study, you may contact - anonymously, if you wish- Ashir Kumar, M.D., Chair of the
University Committee on Research Involving Human Subjects (UCRIHS) by phone: (517) 355-2180, fax:
(517) 432-4503, e-mail: ucrihsfiimsu.edu., or regular mail:202 Olds Hall, East Lansing, MI 48824. Thank

you for your cooperation.

Investigator: So Young Lee

1232 Woodcrest lane #102, East Lansing, MI 48823, Tel: (517) 332-0651, E-mail: leesoZfimsuedu
Advisor: Dr. Dana Stewart, Professor

204 Human Ecology building, Michigan State University Ml 48824-1030

Tel: (517)432-2738, E-mail: dgstewaga:msu.edu

Ph. D. Program in Human Environment and Design

College of Human Ecology, Michigan State University

 

116

 

Section I General Workplace -. _-_ , I. ,_ . 1
Please check the appropriate box.

Your workplace is
El Closed to the public
[:1 Open & accessible to the public

My current office is best described as:

[:1 Private closed (four walls to ceiling; door)

Cl Individual open and high (>54 inch) panels/cubicle
Cl Individual open and low (<54 inch) panels/cubicle
[:1 Open office (no partitions)

Your company offers :

U Work at home options

Cl Telecenter options

Cl Hoteling options: Use hotel as an workplace for a certain period of time.
[:1 Satellite options

Cl None of them

Section I11 .YoUrjob,_;activities ‘ .

Please check the appropriate box.

Your current job:

Cl Clerical/Support (for example, Administrative Assistant; Data Entry; Reservations)
Cl Engineer/Fechnical/Professional

Cl Manager (for example, Director; Supervisor; Vice President)

Cl Other (please describe):

 

Please, write down the approximate time you spend working.

 

Your regular work hours: from to
Work hours per week in this office: hours
Work hours per week outside of the office: hours

 

117

S
-\_u

Please circle the appropriate number.

Your job responsibilities: No Involvement Extensive involvement
\lr \l/
Sales/Marketing ..................................................... 1 2 3 4 5 6 7
Engineering ......................................................... l 2 3 4 5 6 7
Operations ........................................................... 1 2 3 4 5 6 7
Human Resources ................................................... 1 2 3 4 5 6 7
Finance/Accounting ................................................ 1 2 3 4 5 6 7
Purchasing ........................................................... l 2 3 4 5 6 7
Research/ Development ............................................ l 2 3 4 5 6 7
Strategic Planning ................................................... 1 2 3 4 5 6 7
Other ....................... 1 2 3 4 5 6 7

 

Please indicate yes or no by checking the appropriate box.
Do you work as team?
Cl Yes
C] No
-) If yes, do you have a designated shared space for your team when you do team work?
D Yes
Cl No
9 If yes, is your team work area fully enclosed? Please circle one of the numbers.

No, not at all 1 2 3 4 5 6 7 Yes

Please circle the appropriate number.
Your solo work style:

Go to an office or a private and quiet area ...1 2 3 4 5 6 7 ...coffee shop regular

Here are a few statements about people ’3 jobs. Please indicate how often each statement is true for your job by circling
the numbers.

Never Rarely Sometimes Oflen Always

 

 

 

 

 

 

 

When I'm at the office, I work alone rather than in 1 2 3 4 5
workgroups or teams.

My work is simple and demands little concentration. 1 2 3 4 5
My job requires a lot of cooperative work with others. 1 2 3 4 5
I take work home because it is difficult to corrrplete at work. 1 2 3 4 5
I share common work areas with others 1 2 3 4 5
I do teleconferencing. 1 2 3 4 5

 

118

 

 

 

 

 

 

 

 

Never Rarely Sometimes Often Always
My work requires deep thvopghts and concentration. 1 2 3 4 S
I would like to work alone. 1 2 3 4 5
I often avoid meeting people in my workplace. 1 2 3 4 5
My workplace has been moved within my office. 1 2 3 4 5
My workplace has been moved between offices. 1 2 3 4 S
I have moved between teams 1 2 3 4 5

 

Section Ill Physical enirironmental features of workplace

The following set of statements related to your feelings and opinions about your workplace. In each set, please show

the extent to which you agree or disagree the statement and show the extent to which you think your workplace should
possess the features described by each statement. You may circle one of the numbers. If you strongly agree, circle the
number 7. If you strongly disagree the statement. circle the number I . If your feelings are not strong, circle one of the
numbers in the middle. There are no right or wrong answers-all we are interested in is a number that best shows your

perception about your work environment.

 

Strongly Strongly
disagree age}
My workplace is open enough to see my colleagues working. 1 2 3 4 5 6 7

My workplace should be Even enouLhto see my colleagues working. 1 2 3 4 5 6 7

 

My workplace provides an undisturbed environment so that I can 1 2 3 4 5 6 7
concentrate on my work.

My workplace should provide an undisturbed environment so that I 1 2 3 4 5 6 7
can concentrate on my work.

 

My fimiiture is flexible enough to adjust, rearrange, or reorganize my 1 2 3 4 5 6 7
workspace.

My furniture should be flexible enough to adjust, rearrange, or 1 2 3 4 5 6 7
reorganize my workspace.

 

 

 

I am easily distracted from my work. 1 2 3 4 5 6 7
I should not be easily distracted from my work. 1 2 3 4 5 6 7
I am able to control the social contact with others around me. 1 2 3 4 5 6 7
I should be able to control the social contact with others around me. 1 2 3 4 5 6 7
I have informal and impromptu meetings in my private workstation. l 2 3 4 5 6 7
I should be able to have informal and impromptu meeting in my private 1 2 3 4 5 6 7

workstation.

 

 

119

Strongly Strongly

 

 

 

 

disagree agree

My work schedule is flexible. I 2 3 4 5 6 7
My work schedule should be flexible. I 2 3 4 5 6 7

I am able to determine the scheduling and duration of rest breaks. 1 2 3 4 5 6 7

I should be able to determine the scheduling and duration of rest breaks. 1 2 3 4 5 6 7
The quality of my equipment is more than sufficient to work effectively. 1 2 3 4 5 6 7
The quality of my equipment should be more than sufficient to work 1 2 3 4 5 6 7
effectively.

My workplace serves multi-purpose functions for informal and instant I 2 3 4 5 6 7
meeting.

My workplace should serve multi-purpose functions for informal and l 2 3 4 5 6 7
instant meeting.

 

 

 

 

 

 

 

 

 

 

 

 

I am able to personalize my workspace. 1 2 3 4 5 6 7
I should be able to personalize my workspace. l 2 3 4 5 6 7
I am able to control temperature or airflow in my office. I 2 3 4 5 6 7
I should be able to control temperature or airflow in my office. 1 2 3 4 5 6 7
My work environment is quiet. I 2 3 4 5 6 7
My work environment should be quiet. I 2 3 4 5 6 7
Strongly Strongly
disagree age;
I am able to determine the organization/appearance of my work area. 1 2 3 4 5 6 7
I should be able to determine the organization/appearance of my work 1 2 3 4 5 6 7
area.
I am able to control the amount and timing of contact with other people 1 2 3 4 5 6 7
including co-workers or clients.
I should be able to control the amount and timing of contact with other 1 2 3 4 5 6 7
people including co-workers or clients.
I am able to be easily accessed from my colleague’s workstation. I 2 3 4 5 6 7
I should be able to be easily accessed from my colleague 's workstation. 1 2 3 4 5 6 7
My workstation is over-equipped for my typical needs. 1 2 3 4 5 6 7
My workstation should be over- equipped for my typical needs. 1 2 3 4 5 6 7
In general, my workspace is flexible. I 2 3 4 5 6 7
In general, my workspace should be flexible. I 2 3 4 5 6 7
I am able to choose among available methods to complete my tasks. 1 2 3 4 5 6 7
I should be able to choose among available methods to complete my 1 2 3 4 5 6 7
tasks.
My furniture is a fixed system. 1 2 3 4 5 6 7
My furniture should be fixed fitem. l 2 3 4 5 6 7

 

 

120

Strongly Strongly

 

 

 

 

 

 

 

 

disagree agree
I am able to determine the order in which tasks are completed. 1 2 3 4 5 6 7
I should be able to determine the order in which tasks are completed. 1 2 3 4 5 6 7
My workstation is large. 1 2 3 4 5 6 7
My workstation should be large. 1 2 3 4 5 6 7
I am able to have quiet and undisturbed time alone. 1 2 3 4 5 6 7
I should be able to have quiet and undisturbed time alone. 1 2 3 4 5 6 7
My work area has many visual distractions. l 2 3 4 5 6 7
My work area should have many visual distractions. 1 2 3 4 5 6 7
I have ample storage in my work area. 1 2 3 4 5 6 7
I should have ample storage in my work area. 1 2 3 4 5 6 7
I am able to connol the lighting level in my workstation. I 2 3 4 5 6 7
I should be able to control the lighting level in my workstation. 1 2 3 4 5 6 7
I am able to determine the order in which tasks are completed. 1 2 3 4 5 6 7
I should be able to determine the order in which tasks are completed. 1 2 3 4 5 6 7

 

Please circle the appropriate number.
How are you satisfied with your workplace in general?

very unsatisfied 1 - 2 -- 3 - 4 — 5 -- 6 -- 7 very satisfied

Section IV ‘ Toward outcomes of your general work activities

This survey deals with your work outcomes. Please show the extent to which you think how well you work as
described by each statement if you strongly agree, circle the number 7. If you strongly disagree the statement,
circle the number I . If your feelings are not strong, circle one of the numbers in the middle, there are no right or
wrong answers — all we are interested in is a number that best shows your evaluation about your works.

Strongly Strongly
disagree agree

 

My colleagues would like to spend time together outside of work 1 2 3 4 5 6 7
hours.

 

 

 

 

My colleagues are united in trying to reach its goals for 1 2 3 4 5 6 7
performance.
Generally speaking, I am very satisfied with my work. 1 2 3 4 5 6 7
I would consider taking anothepjob. 1 2 3 4 5 6 7
My colleagues have conflicting aspirations for the group work 1 2 3 4 5 6 7
performance.

 

My colleagues would rather go out on their own than get together. 1 2 3 4 5 6 7

 

 

My colleagues do not stick together outside of work time. 1 2 3 4 5 6 7

 

121

Strongly Strongly

 

 

disagree agree
I would recommend my job to a friend who was qualified and l 2 3 4 5 6 7
looking for work.
My work group does not give me enough opportunities to l 2 3 4 5 6 7

improve my personal performance.

 

I am unhappy with my group’s level of commitment to the task. 1 2 3 4 5 6 7

 

I communicate effectively with my colleagues within the office. 1 2 3 4 5 6 7

 

 

If I had it to do over, I would choose to work here again. 1 2 3 4 5 6 7

 

The following statements are related your work performance. Please circle the numbers that reflect how you rate your
own work performance on each of the following dimension.

Never Rarely Sometime Ofien Always

 

 

 

 

 

 

My work is of high qualifl. 1 2 3 4 5
I make mistakes or errors. 1 2 3 4 5
I do a large amount of work each day. 1 2 3 4 5
I can accomplish a great deal each day. l 2 3 4 5
My work outcome is creative. 1 2 3 4 5

 

Section V Demographic Information
Please check the appropriate box.

Your gender: Cl Male Cl Female

Your age: [:1 Under 25
Cl 25-34
1:] 35-44
1:] 45-54
El 55-64
[:1 65 or over

Your education: 1:1 High school degree/ G.E.D.
Cl College degree
[:1 Master’s degree

[:1 Higher degree than master’s

Thank you for your cooperation!!!

122

Appendix B: Regression Assumptions

Figure Bl. Normal Probability Plots of the Standardized Residuals

 

 

 

 

 

 

 

 

 

Bla: Job satisfaction (control) B lb : Performance (control)
100 ‘ 1.00

.75- .75<
'8 .50 ‘ '8 50 I
‘- h
a. o.
E E
:1 :1
g .25 I g .25 I
9 £2
0 U
Q) d)
a 9
LL] 0.00 . . LIJ 0.00 i . ' '

0.00 .25 .50 .75 1.00 0.00 .25 .50 .75 1.00

Blc: Task Group Cohesiveness (flexibility) Bld: Job Satisfaction (distraction)

 

1.00

 

1.00

 

75 ‘ .75 l

 

 

 

 

 

 

 

'8 50‘ g .50-

L' h

0- n.

E E

5 5
.25I .25-

E E

8 8

o” 3

53 0.00 Lu 0.00 1 . '
0.00 .25 .50 .75 1.00 0-00 25 -50 .75 1.00

123

Figure B2. Scatter plots between Control variables and Standard Residuals of Job

 

 

 

 

 

 

 

 

 

 

 

 

 

Satisfaction
B2a :Personalization B2b : Temperature control
3 ‘ 3
2 ., - 2 a
. 5 ‘ a
,, H :1
1 . E 5 i g E 1 .
.‘ U — E '.
r‘ i ‘ S 7
a 0 “ E '1 E ii 0 E1 ‘
5 i - a i =
g -1 r =. E i -1. 5 _
[I L. j, E , H F
'0 F. =
g -2 _ * :~ -2 ‘
E = 5
{‘3 -3 -3 ‘-
C
m
(75 4 _ - - g -4 . _
r 1 2 3 4 5 6 7 8 ( 1 2 3 4 5 6 7 t
BZc : Control organization/appearance B2d : Lighting control
3 .. 3
2 - 1 .. 2 i c 1
: :- = a a ' ‘ . E
1 _ : 1 S E 1 _' E =.
7 E i g g i “ i t ; =
5 — c E i _ E :I 3 q
o a J: a 4 a » 0 - i a = r. . -
- = ‘ ' I E g 2 t ' =.
i i i :1 : E N " r
-1 _ E 3 ‘ -1 , 5 f :‘
- . ’1 > “ i * -
_2 -. , _2 i . , .
-31 —3 "
4 p .. 4 .
( 1 2 3 4 5 6 7 8 r 1 2 3 4 5 6 7 8
BZe : Social contacts 82f : Determine scheduling
3 7 3
2 s ' 2
: i . . i F - .
‘ s . é a i . , ;
1 a F 7 u 1 ' “ = L ’ i i
g n ‘ a _ ~ * s , : :
0 ~ 5 i ! ' 01 g - ' g 5 I E
‘1 g =1 z‘ “ :: E a
F E i" i - "I ‘ 7 : 7 E
-1 n 1 ~ 1.. ,5. _ -1 l ' 5 l E
t 4 . g E ,, -’ ', r
-3 7 ". ‘3
-4 r g .4 _
( 1 2 3 4 5 G 7 8 ( 1 2 3 4 5 6 7 t

 

 

 

 

124

Figure B2. Scatter plots between Control variables and Standard Residuals of Job
Satisfaction (Continued)

 

 

 

 

 

 

B2g : Control amount & timing of contact 82h : Choice among available methods
3 3
2' ~ g? 2‘ ,. _
’3 I i: :3 . " 1 .3 :
a ‘ .4 g E 3 7 E ‘1
11 l El 5 1 11 E E E
£1135.»
a 0. 5 a. g 3—4— 4- ~ — o. :1; i i 9 f
3 = i i “ i H ‘3. ..i g E g a
2 it t ! .=. ' Ti 3. .» L - g 3
8 -1‘ _. é .—'. b F ' -1‘ ' l" g 1' i
m - 4 _ a r' ~' r . s *-
v 3 ‘ ‘ = . -. F
El -2‘ ' " '2 "
8 -3 -3.
C
(U
a 4 _ ' _ _ . - 4 _ - - _ _ g _.

 

 

BZi : Determine order

 

3
21 f‘. ..
E . = L
. P, a E
I, 1 I j
1 1 i g
g
01 E 3 ._ . I .E
3 1' E
4 i: 5 .3. a g
- 4 - r L- 1.1 I 'L.‘
1 7 t 1; B .2
E .7. :1
-2 4 H is
11 - '
d _. E1
-34 r
.4

 

 

 

125

Figure B3. Scatterplots between Flexibility variables and Standard Residuals of Task
Group Cohesiveness

BB a : Convertibility B3 b :Versatility

 

 

 

3 3
21 . n 1- _ 2+
. i l ,1 1- + , ' i“. : : .:.
11 .- 5 :1 ' g 1‘ 11 3: 5 .3 E g 5
~ 5 '~. 1:. = 7- : 73 =. E n
'5 5 .' =‘ " :1 1'1 9 i 5
a 0 - * g i :' .-, 0‘ a H '3 §
3 i ’, E i" 1 " 5 11 I E L 1;
E g ' ' {I 4 'q H H a l "‘
' ~. 11 : H "‘ __ -
8 -11 .2 a ' : a E -11 '-—: ‘ 1: ‘: U z
m y. {-2 .1 : 2 ~ '1 0' : E g ~ 11
U r‘ U - -; : '2
.3 ‘21 -2* ' ‘
g {-
'0 -3' - —.
S 3. 1
a -4 _ _ f 1 1 .4 1

 

 

 

 

 

 

OH

‘11

oo
o

o 1 2 3 4 5 q 5 5 :1
B3 c : General flexibility B3 (1 : Movable system

on
0')
MI

 

 

 

 

 

 

 

3 3
2‘ 1‘ 1.. 11 2‘
t. " l 1.
‘ Q U I 4': i i E I. " E'
n- d 1. C . 1.
1‘ g E = 23 1 1‘ . a 1 _
i 5 E 5 S 5 _ 1. E h :7 3
0‘ 5 1:. i l E 01 .i ; e
.2 l E : a g y .. F‘ ‘1 d
! Fl' E i E i E ’1' _, '7’.
-1‘ " . I a 1 -1J : 3 : 1_~ :
'2‘l _2‘ : ‘ ’
-3‘ ’ L3 ‘3‘
.4 - " -4 *-

 

NI

m
o
.5

0123456 234567
B3 e : Openness B3 f : Accessibility

 

 

3 3

21 1 , 1 2I

-.4...

i

1 1 11 '3 ‘ " E =
' ' — ,- x 7 1 L .
11 -.- 1 3 2‘ E i 3 11 , t 5 E
s E = : E . 5 » E '=' 9 l
o u I " .- 1 D Q
=. E ‘ 5 g = . g a 1'; i u
0‘ 5 i E = i '1‘- °‘ ‘ " E 1 5' a l
'- H‘ ‘2' a I : l ”' : a- E
:1 1 .= a E 1. 3 '_ . 1 i
-11 1 ti :1 j . 1:1 : ‘ 1 U ; i i
3 ~ ' W ‘1 n 11 1
.. 3 1] F, E

I
N

1

Lu .1
'
N

 

 

 

 

 

 

126

Figure B4. Scatter plots between distraction variables and standard residuals of job

 

 

 

 

 

satisfaction
B4 a : Undisturbed workplace (R) B4 b : Easily distracted
3 3
21 a , ¥ 2‘ ‘z I. 5‘ f.
‘- 3 a . E ~ r; T,

1 1 3 2' a 5 1' ' E i' i : I
_ i , , ._ .L‘ as- _ - '_.__. ,_ __ _ B 01 d 5 E I 3
‘3 O E l l 5 i _ E g g E ‘
‘3 1 l - C = L H -1 4 g .5 i. if =
w ' ' 7 .- = : - a = a
CI L, _' p g t: g .- s E e n
E -2 . = " 3 ‘ '2‘ i ;
E ‘ 'i t .9
8 -3 -3i
5 i.
:5 4 4 - - . - . - -

 

 

 

oiééiééi801234567
B4 c : Quietness (R) B4 (1 : Quiet and undisturbed time (R)

 

 

 

 

 

 

 

 

 

3 3
-. l ”I F, . U. I. :-' g ,i
1‘ u a a i g I. 1‘ H i g i a E a
i. ' 3 i a‘ l l 3 E
1. E1 : = . | a.
o‘irlai oi-i-ai-rti
' : : a c a a a :a‘ E 1! 3 i4
-1i F a F ; . g 4‘ u t fa ; «a i; i
3' * . a ‘“ 3 : 5
-2 _: . H -21 ~ Li a , ,1
-31 U _3J il
-4 _ _ _ i _ T _ .4
0 1 2 3 4 5 6 7 8 o 1 2 3 4 5 6 7
B4 e : Visual distraction
3
2' ii i .. ll
u 3 i. "
i - a E .
1 i l a a a i
5 ~ z a i
0‘ -5 i E S F t’ 5
g I 4- ‘ E
a E :
-1‘ F! g a f7 " c _.
‘5‘ = 5 ‘=’ ’
-2. H '; ‘
-3. I
4 v I i U i

 

 

 

127

Appendix C: Covariance Matrices

Office environment - A CONFIRMATORY FACTOR ANALYSIS

Covariance Matrix to be Anaiyzed

c1 c2 c3 c4 05 06
01 2.42
c2 1.15 3.28
c3 0.53 0.52 3.21
04 0.74 0.51 0.24 3.53
c5 0.46 0.60 1.28 0.60 2.65
c6 0.56 0.74 0.57 0.45 0.59 2.14
07 0.40 0.36 0.16 0.46 0.23 0.59
951 0.26 0.26 0.02 0.03 -0.03 0.35
952 0.46 0.24 0.30 0.24 0.13 0.35
953 0.37 0.18 0.28 -0.12 0.12 0.32
t91 0.19 0.34 0.49 0.10 0.24 0.35
tg2 0.38 0.38 0 56 0.11 0.35 0.31
t93 0.49 0.57 0 53 0.18 0.38 0.61
tg4 0.53 0.57 0.55 0.19 0.49 0.45
t95 0.28 0.39 0.21 0.19 0.14 0.36
j1 0.47 0.50 0.57 0.34 0.37 0.53
j2 0.28 0.57 0.51 0.33 0.17 0.44
j3 0.64 0.69 0.63 0.56 0.45 0.58
j4 0.60 0.73 0.51 0.52 0.36 0.66
p1 0.01 0.08 0.02 0.00 -0.04 0.06
p2 -0.01 0.01 0.00 -0.22 -0.03 0.03
p3 0.07 0.12 0 15 -0.02 0.14 0.10
p4 -0.01 0.02 0.11 0.13 0.13 0.20
d1 -0.52 -0 43 -1.02 -0 52 -0 73 -0.26
d2 -0 04 -0 23 -0.46 -0 09 -0 32 -0 13
d3 -0 52 -0 55 -1.06 -0.49 -0 61 -0.39
d4 -0.63 -0.58 -1.28 -0.57 -1 00 -0.47
d5 -0 42 -0 54 -0.50 -0 04 -0 50 -0.13
f1 0 94 0 86 0.83 0 09 0 46 0.55
f2 0 83 0 61 0.51 0 45 0 36 0.78
f3 1.00 1.05 0.85 0.41 0.64 0.85
f4 0.14 0.35 0.06 -0.07 -0 20 -0.10
Covariance Matrix to be Analyzed
c7 951 952 953 t91 t92
07 1.75
951 0.01 2.13
952 0.13 0.79 1.98
953 0.17 1.06 1.33 2.56
t91 0.36 0.56 0.54 0.60 1.84
t92 0.26 0.31 0.67 0.84 1.07 2.67
t93 0.36 0.30 0.72 0.82 0.88 1.08
tg4 0.45 0.38 0.56 0.58 1.02 1.17
t95 0.32 0.15 0 35 0.23 0.52 0.40
j1 0.42 0.25 0.43 0.47 0.93 0.73
j2 0.25 0.06 0.35 0.33 0.41 0.69
j3 0.58 0.43 0.61 0.57 0.88 0.75
j4 0.42 0.22 0.53 0.29 0.77 0.71
p1 0.09 0.08 0.06 0.03 0.10 -0.08
p2 0.09 0.15 0.06 0.14 0.19 0.01

128

p3
p4
d1
d2
d3
d4
d5
f1
f2
f3
f4

t93

0.12 0.05 0.12
0.01 0.09 0.05
-0.34 0.09 -0.32
-0.21 0.12 -0.18
-0.20 -0.02 —0 45
-0.11 -0.04 -0.37
-0.12 0.05 -0.08
0.04 0.28 0.36
0.30 0.04 0.12
0.24 0.37 0.35
-0 34 -0.06 0.24
Covariance Matrix to be Analyzed
t93 tg4 t95
2.81
1.16 2.35
0.43 0.48 1.21
1.13 0.92 0.51
1.10 0.74 0.38
1.19 0.93 0.63
1.02 0.89 0.65
0.08 0.02 0.12
0.17 0.17 0.11
0.16 0.16 0.16
0.28 0.09 0.12
-0.48 -0.51 -0.12
-0 30 -0.49 -0.27
-0.55 -0.55 -0.26
-0.61 -0 60 -0.22
-0.23 -0.51 -0 11
0.47 0.49 0.21
0.34 0.20 0.16
0.39 0.47 0.29
0.27 0.01 -0.05

a l I l I
OOOOOOOCOOOCON
0)
0'1

.10

00000
O
on

.04
-0.
-0.03
0.03
-0.01
-0.05
0.01
0.02
-0.12

Covariance Matrix to be Analyzed

000000000000
0
.s

129

-0.
-0,
-0.
-o_
-0.

-0.

-0.

I I I I I
00000000000

.10

.11

I I u I a
OOOOOOOOOOOOOOOQ
N
N

I u I I u o
OCOOOOOOOOO
OJ
U1

1 l l I I
COOOOOOOOOOOO-‘N
N
on

O-‘dON
.5
(O

-0.79
-0.32
-0.69
-0.11

f1 -0.30 -0.74 -0.88 -0.43 3.90

f2 0.27 -0.32 -0.43 -0.05 0.93 3.07
f3 -0.34 -0.97 -0.76 -0.17 1.91 0.91
f4 -0.14 -0.27 -0.42 -0.20 1.49 0.17

Covariance Matrix to be Analyzed

f3 f4
f3 2.94
f4 0 33 3 42

Control factor analysis 1

Covariance Matrix to be Analyzed

pe1 pe1 pe3 pe4 pc1 pc2
pe1 0.27
pe1 0.11 0.45
pe3 0.08 0.17 0.44
pe4 0.09 0.15 0.19 0.73
pc1 0.01 -0.05 0.09 0.00 2.46
p02 0.07 -0.01 0.11 0.01 1.14 3.31
pc3 0.00 -0 02 0.14 0.11 0.54 0.52
wc1 0.00 -0.20 -0.02 0.12 0.68 0.55
ch -0.04 -0.05 0.13 0.13 0.44 0.59
wcS 0.06 0.00 0.11 0.20 0.62 0.74
wc4 0.09 0.10 0.14 0.04 0.43 0.34
di1 0.05 0.18 -0.11 0.00 -0.56 -0.46
di2 -0.05 -0.05 -0.31 -0.16 -0 10 ~0.22
di3 -0.02 0.07 -0.19 -0.04 -0.53 -0.56
di4 0.04 0.04 -0.18 -0.13 -0 66 -0.58
di5 -0.01 0.07 -0.13 0.13 -0 43 —0.50

pc3 wc1 wc2 wc3 wc4 di1
pc3 3.22
wc1 0.28 3.48
wc2 1.33 0.60 2.68
wc3 0.56 0.44 0.54 2.15
wc4 0.18 0.43 0.27 0.58 1.81
di1 -1.04 -0.51 -0.78 -0.29 -0.33 2.93
di2 -0.45 -0.05 -0.29 -0.19 -0.21 0.94
d13 -1.08 -0.46 -0.62 -0.40 -0.21 1.50
di4 -1.27 -0.53 -1 02 -0.48 -0.11 1.63
di5 -0.61 -0.04 -0.60 -0.10 —0.13 0.72

d12 d13 di4 d15
d12 2.67
di3 0.88 2.77
di4 0.79 1.53 3.29
d15 0.63 0.56 0.94 3.23

130

Control-distraction-performance

Covariance Matrix to be Analyzed

alo av01 pe1 p92 pe3 pe4

alo 0.60
avoi 0.17 0.73

pe1 0.04 0.00 0.27

pe2 0.04 -0.03 0.11 0.45

p93 0.00 -0 05 0.08 0 17 0.44

pe4 0.03 -0.11 0.09 0.15 0.19 0.72
di1 0.19 0.13 0.05 0.17 -0.11 -0.01
diZ 0.13 0.20 -0.05 -0.05 -0.32 -0.16
di3 0.18 0.12 -0.02 0.06 -0.19 -0.04
di4 0.14 0.08 0.04 0.03 -0 18 -0.14
di5 0.09 0.11 0.00 0.06 -0.13 0.12
pc1 -0.12 -0.16 0.01 -0.05 0.09 0.01
p02 -0.11 -0.19 0.07 -0.01 0.11 0.02
pc3 -0.09 -0.12 0.00 -0.02 0.14 0.11
wc1 0.00 -0 02 0.00 -0.21 -0.02 0 12
wc2 -0.04 0.00 -0.04 -0.04 0.14 0.14
wc3 -0.18 -0.13 0.06 0.00 0.10 0.20
wc4 -0.05 -0.11 0.09 0.10 0.13 0.04

di1 d12 d13 di4 d15 pc1
di1 2.94
di2 0.94 2.68
d13 1.50 0.88 2.78
di4 1.63 O 78 1.52 3 30
di5 0.70 0.63 0.54 0.92 3.19
pc1 -0.56 -0.10 -0.52 -0.65 -0.41 2.45
p02 -O.46 -0.22 -0.55 -0.58 -0.48 1.12
p03 -1.04 -0.44 -1.08 -1.28 -0.60 0.53
wc1 -0.53 -0.06 -0.47 -0.54 -0.07 0.70
wc2 -0.76 -0.28 -0.61 -1.01 -0.58 0.44
woS -0.31 -0.19 -0.41 -0.49 -0.10 0.63
wc4 -0 33 -0 21 -0.21 -0.11 -0.13 0.42

Covariance Matrix to be Analyzed

pc2 pc3 wc1 wc2 wc3 wc4
pc2 3.31
p03 0.51 3.23
wc1 0.56 0.29 3.49
wc2 0.60 1.34 0.61 2.67
wc3 0.75 0.56 0.44 0.55 2.16
wc4 0.33 0.17 0.43 0.28 0.58 1.81

131

Effects of control (high control group)

Covariance Matrix to be Analyzed

alo av01 pe1 pe2 pe3 pe4
alo 0.56
avoi 0.24 0.71
pe1 0.03 0.03 0.28
pe2 -0 02 -0 06 0.11 0.46
pe3 -0.03 -0.05 0.08 0.17 0.41
pe4 -0.03 -0.15 0.08 0.20 0.17 0.70
di1 0.08 0.14 0.07 0.19 -0.07 0.06
di2 0.16 0.16 -0.07 -0.03 -0 23 -0.10
di3 0.17 0.23 0.01 -0.06 -0.17 0.05
di4 0.01 0.22 0.04 -0.01 -0.13 -0.17
dis 0.04 0.04 -0.04 0.02 -0.11 0.14
pc1 -0.10 -0.16 0.06 0.01 -0.01 0.01
pc2 -0.12 -0 12 0.08 0.05 0.09 0.05
pc3 0.02 _ -0 19 -0 11 -0.02 0.07 -0.06
wc1 -0.20 -0.19 -0 04 -0.11 -0.05 0.12
wc2 -0.01 -0.01 -0.06 -0.05 0.10 0.09
wcS -0.14 -0.09 0.11 0.11 0.13 0.25
wc4 -0.07 -0.08 0.05 0.10 0.12 0.01
Covariance Matrix to be Analyzed
di1 d12 dl3 di4 d15 pci
di1 3.04
di2 1.06 2.74
d13 1.61 0.97 3.07
di4 1.64 0.91 1.52 3.49
di5 0.77 0.62 0.49 1.18 3.15
pc1 0.03 0.17 -0 22 0.04 -0.04 1.35
p02 0.08 -0 19 -0 13 0.02 -0.15 0.33
pc3 -0.61 -0.36 -0.65 -0.68 -0.12 -0.14
wc1 -0.28 0.08 -0.08 -0.06 -0.04 0.16
wc2 -0.53 -0.14 -0 30 -0.60 -0.29 -0.26
wc3 -0.05 -0.16 -0.03 —0.19 -0.13 0.18
wc4 ~0.09 -0 19 -0.01 0.08 -0.05 0.04
Covariance Matrix to be Analyzed
pc2 pc3 wc1 wc2 wc3 wc4
pc2 1.81
p03 -0.26 2.67
wc1 0.02 -0.34 2.26
wc2 -O.20 0.59 0.17 2.47
wc3 0.15 0.00 0.27 0.04 1.11
wc4 -0.02 -0.29 0.17 0.13 0.14 1.08

132

Effects of control (low control group)

alo
avoi
pe1
p62
pe3
pe4
di1
diZ
di3
di4
di5
pc1
pc2
pc3
wc1
ch
wc3
wc4

Covariance Matrix to be Analyzed

Covariance Matrix to

di1
di2
di3
di4
di5
p01
p02
p03
wc1
wc2
ch
wc4

00000000010

-0.

05

000000000
0
O)

-0.11
-0.01

0.04
-0.01
-0.08
-0.06

0.09

Analyzed

2.13
1.03
0.42
-0.13
-0.18
-0.88
-0.35
-0.30
-0.13
0.02

Covariance Matrix to be Analyzed

p02
p03
wc1
ch
wc3
wc4

2.
.00
.99

0
0

-0.
-0.

67

03
14

4.01

-0.32
0.09

133

0.43
-0.41
-0.12
-1.02
-0.33
-0.59

0.08

0.28

-0.
-0.
-0.
-0.
-0.

-0.
-o:

-0.

.18
.46
.44
.79
.15
.59
.22
.02

-0.04
~0.21
-0.09
-0.05

I!
OOOCOOOO
...L
(D

-0.03

-0.04
-0.16
-0.03

2.05

distraction-control-performance (individual work style coffeeshop)

Covariance Matrix to be Analyzed

pe1 pe1 pe3 pe4 pc1 pc2
pe1 0.28
pe1 0.13 0.46
pe3 0.09 0.27 0.46
pe4 0.14 0.29 0.25 0.77
pc1 0.14 0.09 0.19 0.05 2.38
p02 0.11 -0.14 0.04 0.12 1.20 3.32
pc3 0.07 0.26 0.20 0.37 0.32 0.37
wc1 0.00 -0.01 0.02 0 13 0.52 0.74
ch 0.07 0.14 0.17 0 21 0.35 0.53
wcS 0.04 -0.07 0.13 0.15 0.49 1.13
wc4 0.15 0.18 0.22 0.08 0.47 0.46
di1 0.00 -0.03 0.08 0.00 -0.16 0.02
di2 -0.12 -0.23 -0 28 -0.33 0.18 -0.37
di3 -0 14 -0.19 -0.20 -0.28 -0.48 -0.47
di4 0.03 -0.11 -0.06 -0.20 -0.34 -0.28
di5 0.04 -0.04 -0.11 0.05 -0.18 -0.18
Covariance Matrix to be Analyzed
pc3 wc1 ch wc3 wc4 dl1
p03 3.09
wci -0.10 3.25
wc2 1.09 0.55 2.39
wc3 0.45 0.67 0.53 2.19
wc4 0.02 0.77 0.22 0.55 1.97
di1 -0 82 -0.45 -0 38 -0.24 -0.04 2.45
di2 -0 52 0.12 -0 25 -0.28 -0.16 0.76
di3 -1.21 -0.10 -0.54 -0.60 -0.11 0.88
di4 -1.28 0.02 -0.53 -0 17 0.12 1.00
di5 -0.33 -0.13 -0.67 0.30 0.21 0.41
Covariance Matrix to be Analyzed
d12 di3 dl4 d15
di2 2.67
di3 0.69 2.47
di4 0.60 1.06 2.78
di5 0.49 -0.01 0.53 2.96
distraction-control-performance (individual work style quiet)
Covariance Matrix to be Analyzed
pe1 pe1 pe3 pe4 pc1 pc2
pe1 0.28
pe1 0.10 0.43
pe3 0.07 0.11 0.41
pe4 0.06 0.10 0.16 0.72
pc1 -0.07 -0.18 0.02 -0.02 2.46
pc2 0.03 0.02 0.11 -0 06 1.05 3.23
pc3 -0.06 -0 18 0.08 -0 05 0.60 0.55
wc1 0.00 -0.33 -0.05 0.12 0.74 0.48
wc2 -0.12 -0 17 0.12 0.09 0.52 0.59
wc3 0.04 0.03 0.07 0.20 0.68 0.43

134

wc4 0.06 0.04 0.09 0.02 0.38 0.24
di1 0.08 0.29 -0.22 -0.02 -0.77 -0.69
di2 0.01 0.04 -0.34 -0.09 -0.13 -0.01
d13 0.07 0.18 -0.22 0.09 -0.56 -0.57
di4 0.06 0.13 -0.24 -0.13 -0.77 -0.76
di5 -0.01 0.08 -0.16 0.16 -0.61 -0.61

pc3 wc1 ch wc3 wc4 dl1
pc3 3.23
wc1 0.51 3.66
wc2 1 48 0.72 2.76
wc3 0.50 0.37 0.54 1.99
wc4 0.24 0.26 0.34 0.52 1.66
di1 -1.25 -0.58 -0.96 -0.33 -0.46 3.19
di2 -0.41 -0.09 -0.30 -0.05 -0 24 0.99
di3 -1.00 -O.68 -0 69 -0.25 -0.23 1.79
di4 -1.32 -0.83 -1 27 -0.69 -0.22 1.98
di5 -0.82 -0.08 -0.64 -0.22 -0.26 0.84

Covariance Matrix to be Analyzed

d12 d13 di4 d15
di2 2.58
di3 0.90 2.89
di4 0.85 1.77 3.47
d15 0.64 0.80 1.14 3.20

135

Appendix D: Command Files for LISREL

ACONFIRMATORY FACTOR ANALYSIS

DA Nl=32 NO=4OO

la

cl c2 c3 c4 c5 c6 c7 gsl gsZ gs3 tgl tg2 tg3 tg4 tg5j1j2j3j4 pl p2 p3 p4 d1 d2 d3 d4 d5 fl f2
f3f4

Ra fi=cfall.dat

MO nX=32 NK=7 lx=fu.fi td=sy,fi
frlxl11x2llx31lx4llx511x6l1x71lx821x921x1021x1131x1231x133lx143
frlx1531xl641xl741x184lx194lx2051x2151x2251x2351x2461x256lx266lx
2761x286

fr1x2971x307lx31 71x327
frtdl1td22td33td44td55td66td77td88td99td1010

frtdll lltd1212td13 13td1414td1515td1616td16l6td17l7

frtd 18 18td l9 l9td2020td 21 21 td22 22 td23 23 td24 24td25 25

frtd 26 26 td 27 27 td28 28 td29 29td3030td 31 31 td 32 32
frtd3239td53td12td1113td67tdl4td421tdl718td1819tdl3lStdll17

LK
control sgc tgc jobs perf dis flex

pd
OU rs mi me=ml ad=off

 

CONTROL MODEL (CONFIRMATORY FACTOR ANALYSIS)
DA N1=16 NO=400

la

pe1 pe1 pe3 pe4 pcl pc2 pc3 wcl wc2 wc3 wc4 di1 di2 di3 di4 di5

ra fi=cf~con1.dat

mo lx=fu,fi td=sy,fi nx=l6 nk=3

frlxlllx211x31|x4llx521x621x721x821x92 1x1021x112
frlx1231x1331x1431x1531x163
frtdl1td22td33td44td55td66td77td88td99td1010tdll ll
frtd1212td13 l3tdl4l4td1515td1616
frtd65td12td811td1011

lk
perfo control distraction

path diagram
OU rs mi me=ml ad=off

136

MODEL OF CONTROL -DISTRACTION-PERFORMANCE

DA Nl=18 NO=4OO

la

alo avoi pe1 pe2 pe3 pe4 di1 di2 di3 di4 diS pcl pc2 pc3 wcl wc2 wc3 wc4

ra fi=cf-con2.dat

mo ly=fu,fi lx=fu,fi be=fu ph=sy,fi ga=fu,fite=sy,fitd=sy,fi nx=7 nk=l ny=ll ne=4
frlyl11y221y331y43ly531y63ly741y841y941y104ly114
frlxl11x211x311x411x511x611x71
frtdl1td22td33td44td55td66td77
frte33te44te55te66te77te88te99telOlOtell11
frbel4be24be34be12

frga4lga3l

frphll

frpsl 1p522ps33ps44

frtd31

frtdl2td23td15td34te58td76

Ie

alon avoi perfor distraction
1k

control

path diagram
OU rs me=m1 ad=off mi sc

MULTI SAMPLE ANALYSIS - EFFECTS OF CONTROL (HIGH CONTROL GROUP)
DA NG=2 Nl=18 NO=200

la

alo avoi pe1 pe2 pe3 pe4 dil di2 di3 di4 diS pcl pc2 pc3 wcl ch wc3 wc4

ra fi=high.dat

mo ly=fu,fi lx=fu,fi be=fu ph=sy,fi ga=fu,fi te=sy,fi td=sy,fi nx=7 nk=l ny=11 ne=4
frlyl 1 ly221y33ly43ly53ly631y741y841y94ly1041y114
frlxl11x211x311x411x511x611x71

frtd] ltd22td33td44td55td66td77
frte33te44te55te66te77t688te99te 1010te ll 11

fr ph 1 1

frpsl 1p522ps33ps44

frbel4be24be34be12

fr ga 4 1 ga 3 1

le

alon avoi perfor distraction
1k

control

OU me=ml se tv mi ad=off

137

Effects of control (low control group)

DA Nl=18 NO=200

la

alo avoi pe1 p62 p63 pe4 dil di2 di3 di4 di5 pcl pc2 pc3 wcl wc2 wc3 wc4
ra fi=low.dat

mo ly=fu,fi 1x=fu,fi be=fu ph=sy,fi ga=fu,fi te=sy,fi td=sy,fi nx=7 nk=l ny=l l ne=4
frlyl lly221y33ly43ly531y631y74ly841y941y104ly114
frlxl 1|x21|x31|x411x511x611x7l

frtdl ltd22td33td44td551d66td77
frte33te44te55te66te77te88te99te 10 lOte 11 11

fr ph l 1

frps11p522ps33ps44

frbel4be24be34be12
frga41ga31

le
alon avoi perfor distraction
1k

control

qual3lga3l
qua14lga4l
eqbel 14bel4
eqbe124be24
eqbel34be34
leqbe112be12

DU

138

DISTRACITON-CONTROL-PERFORMANCE (MEDIATING ROLE)
DA N1=l6 NO=400

la

pe1 pe1 pe3 pe4 pcl pc2 pc3 wcl wc2 wc3 wc4 di1 di2 di3 di4 diS

ra fi=cf-con1.dat

mo ly=fu,fi lx=fu,fi be=fu ph=sy,fi ga=fu,fi te=sy,f1 td=sy,fi nx=5 nk=1 ny=l 1 ne=2
frlyl11y211y311y41ly521y621y721y821y921y1021y112
frlxl 11x21 1x3 llx411x51

frtdl 1td22td33td44td55
frtel1t622te33te44te55te661e77te88te99te1010tell11
fr be 1 2

figaZl

fr ga 1 1

fr ph 1 1

fr ps 1 1 ps 2 2

frte65te11 10te5 11 tel 2te 1 2te5 10te610
le

perfor control

lk

distraction

path diagram
OU rs me=ml ad=off mi sc

 

COMPARISON 2: DISTRACTION-CONTROL—PERFORMANCE
(individual work style coffeeshop)

DA NG=2 N1=16 NO=200

1a

pel pe1 pe3 pe4 pcl pc2 pc3 wcl wc2 wc3 wc4 dil di2 di3 di4 diS

ra fi=indcoffe.dat

mo ly=fu,fi lx=fu,fi be=fu ph=sy,fi ga=fu,fi te=sy,fi td=sy,fi nx=5 nk=l ny=1 1 ne=2
frlyllly21ly31ly4]ly521y621y721y821y921y1021y112
frlxl 11x21 1x31 lx411x51

frtdl ltd22td33td44td55
frtel1t622te33te44te55te66te77te881699te1010tell11
fr be 1 2

fr ga 2 1 ga l 1

fr ph 1 1

fr ps 1 1 ps 2 2

1e

perfor control
1k

distraction

path diagram

139

OU rs me=ml ad=off sc

distraction-control-performance (individual work style quiet)

DA Nl=16 NO=200

la

pel pe1 pe3 pe4 pc1 pc2 pc3 wcl wc2 wc3 wc4 dil d12 di3 di4 diS

ra f1= —indquiet. dat

mo 1y=fu, f1 lx= fu, f1 be=fu ph=sy,f1 ga=fu,f1 te=sy,f1 td=sy,f1 nx=5 nk=l ny=l 1 ne=2
frlyl lly211y31ly4lly521y621y721y821y921y102ly112
frlx111x211x31lx411x51‘

frtdl ltd22td33td44td55
frtel1t622te33te44te55te66te77te88te99te1010tell11
fr be 1 2

fr ga 2 1 ga l 1

fr ph l 1

fr ps 1 1 ps 2 2

1e

perfor control
1k

distraction

!Ean121ga21
Eanlllgall
EQbelleelZ

OU

PATH ANALYSIS

data N1=6 no=400

labels; control tgc jobs perf dis flex
KMATRIX

l

.53 1

.60 .81 l

.18 .32 .33 1

-.6O -.39 -.39 -.11 l

.67 .30 .37 .13 -.441

select; 2 3 415 6

model ny=3 nx=3 be=sd ps=di

pd
ou se Ad=offtv ef sc mi

140

REFERENCES

Abramson, L.Y., Seligman, M.P., & Teasdale, J .D. ( 1978). Learned helplessness in
humans: Critique and reformulation. Journal of Abnormal Psychology, 87, 49-74.

Allen, V. L., & Greenberger, D. B. (1980). Destruction and Perceived control, Iii A.
Baum, & J. E. Singer (Eds.), Advances in Environmental Psychology, vol. 2,
Applications of personal control. New Jersey: Lawrence Erlbaum Associates.

Angyal, A. (1949). Foundations for a science of personality, New York: Commonwealth
Fund.

Arbuckle, J. L. (1997). Amos user ’3 guide. Chicago: Small waters.

Aronsson, G. (1989). Changed qualification demand in computer-mediated work. Applied
Psychology: An international Review, 38, 57-71.

Austin, J., Bain, A., Heath, P., Ratekin, J ., Reilly, E., Richert, E., & Ross, C. (2001).
Alternative workplace, In Teicholz, E.(eds.), Facility design and management
handbook, New York: McGraw-Hill

Bachman, J. G., Smith, C. G., & Slesinger, J. A. (1966). Control, performance and
satisfaction: an analysis of structural and individual effects, Journal of
Personality and Social Psychology, 4(2), 127-136.

Bogazzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal
of the Academy Marketing Science, 16 (1), 74-94.

Banbury, S., & Berry, D. C. (1998, August). Disruption of office-related tasks by speech
and office noise. British Journal of Psychology, 89(3), 499-519.

Barber, C. (2001, December). The 21St century workplace, People and workplace, 2-13.
GSA Office of Government Policy, Office of Real Property.

Barnes, R. D. (1981 ). Perceived freedom and control in the built environment. In J. H.
Harvey (eds), Cognition, Social Behavior, and the Environment, 409-422. NJ:

Erlbaum.

Baum, A., & Singer, J. E. (1980). Advances in environmental psychology: v01. 2,
Application of personal control. New Jersey: Lawrence Erlbaum Associates.

Bazerman, M. H. (1982). Impact of personal control on performance: Is added control
always beneficial? Journal of Applied Psychology, 67, 472-479.

Becker, F. (2002). Improving organizational performance by exploiting workplace
flexibility. Journal of Facilities Management, l(2),154-62

141

(1991).Workp1ace planning, design, and management, In E. H. Zube, & G. T.
Moore (Eds.), Advances in Environment, Behavior and Design: vol. 3, 115-151.
New York: Plenum Press.

(1990). The Total Workplace-Facilities management and the elastic
organization.

Bell, P., Fisher, J., Baum, A., & Greene, T. (1990). Environmental Psychology, 3rd ed.
Holt, Rinehart and Winston Inc.

Bentler, RM. (1995). Theory and implementation of EQS: A structural equations
program. Los Angeles: BMDP Statistical Software.

Bemdt, A. (2000, Sep/Oct ). Flexibility in e-commerce office planning. Facility
Management Journal, 40-45. IF MA

Block, L. K., & Stokes, G. S. (1989). Performance and satisfaction in private versus
nonprivate work settings, Environment and Behavior, 21(3), 277-297.

Bonnes, M., & Secchiaroli, G. (1995). Environmental Psychology, a psycho-social
introduction. London: Sage publications.

Brackertz, N., & Kenley, R. (2003). Moving towards an integrated facilities management
tool to evaluate facilities for service performance in local government. In the
Proceedings of the 2002 global symposium, applying and extending the global
knowledgebase, 168-182. CIB W70: Facilities management and maintenance.

Brennan, A., Chugh, J., & Kline, T. (2002). Traditional versus open office design, a
longitudinal field study. Environment and Behavior, 34(3), 279-299.

Brenner, P., & Cornell, P. (1994). The balance between privacy and collaboration in
knowledge worker teams. In the Proceedings of the Fourth International
Symposium on Human Factors in Organizational Design and Management,
Stockholm, Sweden. Retrieved November 8, 2000, from
http://www.steelcase.com/knowledgebase/balance.htm

Brill, M., Margulis, S., & Konar, E. (1984). Using office design to increase productivity.
Buffalo, NY: Workplace Design and productivity.

Brookes , M. J ., & Kaplan, A. (1972). The office environment: Space planning and
affective behavior. Human Factors, 14 (5), 373-391.

Brunswik, E. (1956). Perception and the representative design of psychological
experiments. Berkeley: University of California Press

Canter, D. (1972). Reaction to open plan offices. Built Environment, 1 (October 1972),
465-467

142

Carless, S.A., & DePaola, C. (2000). The measurement of cohesion in work teams, Small
group research, 31(1), 71-88.

Carlopio J. R. (1996). Construct validity of a physical work environment satisfaction
questionnaire, Journal of Occupational Health Psychology, 1(3), 330-344.

Carlopio, J. R., & Gardner, D. (1992). Direct and interactive effects of the physical work
environment on attitudes. Environment and Behavior, 24(5), 579-601.

Casey, B., Keep, E., & Mathe, K. (1999). Flexibility, quality and competitiveness.
National Institute Economic Review, 168, 70-8l.London

Challenger, J. A. (2000, Sep/Oct). 24 trends reshaping the workplace, The Futurist, 35-
41 .Washington

Chang, A., & Bordia, P. (2001 ). A multidimensional approach to the group cohesion-
group performance relationship. Small Group Research, 32(4), 379-405.

Cohen, J ., Cohen, R, West, S.G., & Aiken, LS. (2003). Applied Multiple
Regression/Correlation Analysis for the behavioral Sciences, 3rd ed. Mahwha, NJ:
Lawrence Erlbaum.

Cotts, L. (1998). The Facility Management Handbook, 2"d ed. New York: Amacom

Crouch A., & Nimran, U. (1989). Perceived facilitator and inhibitors of work
performance in an office environment. Environment and Behavior, 21, 206-226.

Cronin, J. J ., & Taylor, S. A. (1992). Measuring service quality: a reexamination and
extension, Journal of Marketing, 56, 55-68.

DeMatteo, J ., Eby, L., & Sundstrom, E. (1998). Team Based Rewards: Current empirical
evidence and directions for future research, Research on Organizational Behavior,
20, 141-183

Dole, C., & Schroeder, R. G. (2001). The impact of various factors on the personality, job
satisfaction and turnover intentions of professional accountants. Managerial
Auditing Journal, 16(4), 234-245.

Donald, 1. (1994). The structure of office worker’s experience of organizational
environments. Journal of Occupational and Organizational Psychology, 67,
241-258.

Evans, G. W., & Cohen, S. (1987). Environmental stress. In D. Stokols and I. Altman
(Eds.), Handbook of Environmental Psychology, 2, New York: John Wiley.

Evans, G. W., & Johnson, D. (2000). Stress and open-office noise. Journal of Applied
Psychology, 85, 779-783.

143

Evans, G. W., Johansson, G., & Carrere, S. (1994). Psychological factors and the
physical environment: Inter-relations in the workplace. In C. N. Cooper & I. T.
Roberston (Eds.), International review of industrial and organizational
psychology: vol. 9, 1-29. Chichester, UK2Wiley.

Evans, G. W., & Lepore, S. J. (1997). Moderating and mediating process in
Environment-Behavior research, In G.T. Moore & R. W. Marrans (Eds.),
Advances in environment, behavior and design :vol. 4, Toward the integration of
theory, methods, research, and utilization. New York: Plenum Press.

Ferguson, G. S., & Weisman, G. D. (1986). Alternative approaches to the assessment of
employee satisfaction with the office environment, In J. D. Wineman (Eds.),
Behavioral issues in office design, 85-108. New York: Van Nostrand Reinhold,

Fisher, S. (1989). Stress, control, worry prescriptions and the implications for health at
work: a psychological model. In S.L. Sauter, J .A. Hurrel & C.L. Cooper (Eds),
Job Control and Worker Health, 205-236. Chishester: Wiley.

Fisher, S. (1990). Environmental Change, control and vulnerability. In S. Fisher & C. L.
Copper (Eds), On the move: The psychology of change and transition, 53-65.
Chishester: John Wiley & Sons, Ltd.

Froggatt, C. C. (2001). Work naked. San Francisco, Jossey-Bass

Ganster, D. C. (1989). Worker control and well being: A review in the workplace. In S.L.
Sauter, J. A. Hurrel & C. L. Cooper (Eds), Job Control and Worker Health, 3-24.
Chishester, Wiley.

Gatchel, R. J. (1980). Perceived control: A review and evaluation of therapeutic
implications. In A. Baum & J. E. Singer (Eds.), Advances in environmental
psychology :vol. 2, Applications of personal control. New Jersey: Lawrence
Erlbaum Associates.

Greenberger, D. B., Strasser, S., Cummings, L.L., & Dunham, R. B. (1989). The impact
of personal control on performance and satisfaction. Organizational Behavior and
Human Decision Processes, 43, 29-51.

Goodrich, R. (1982). The perceive office: The office environment as experienced by its
users, Environment and Behavior, 14(3), 353-378.

GSA (1998). The integrated workplace, a comprehensive approach to developing
workspace. U.S. GSA (General Service Administration) Office of Government
Policy, Office of Real Property.

Hair, J., Anderson, R., Tatham, R., & Black, W. (1998). Multivariate data analysis, 5th
ed. New Jersey: Prentice Hall.

144

Hartkopf, V., & Loftness, V. (1985). The development of intelligent systems towards
total building performance, In M.E. Dolden & R. Ward (Eds.), Proceedings of
The Architectural Research Centers Consortium Workshop on The impact of the
work environment on productivity, 158-171. Washington: Architectural Research

Centers Consortium, Inc

Hedge, A. (1982). The open-plan office, a systematic investigation of employee reaction
to their work environment. Environment and Behavior, 14(5), 519-42.

IFMA (1997). Benchmarks 111 Report, p 36. IFMA (International Facility Management
Association)

Ittelson, W. H. ( 1973). Environment perception and contemporary perceptual theory, in
W. H. lttelson (Ed.), Environment and Cognition, Academic Press, NewYork,

Jackson, SE. (1989). Does job control control job stress?. In S.L. Sauter, J .A. Hurrel &
C.L. Cooper (Eds), Job Control and Worker Health, 25-53. Chishester, Wiley.

Joreskog, K., & Sorbom, D. (1996). LISREL 8: User’s reference guide. Lincolnwood,
IL.; Scientific Software International.

J oreskog, K., Sorbom, D., du Toit, S.,& du Toit, M.(2000). LISREL 8: New Statistical
Features. Lincolnwood, IL.; Scientific Software International

Kaplan (1983). A model of person environment compatibility. Environment and Behavior,
15, 311-332

Kjellberg, A., Landstrom, U., Tesarz, M., Soderberg, L., & Akerlund, E. (1996). The
effects of nonphysical noise characteristics, ongoing task and noise sensitivity on
annoyance and distraction due to noise at work. Journal of Environmental
Psychology, 16,123-136.

Kline, R. B. (1998). Principles and practices of structural equation modeling. New York:
The Guilford press.

Koys, D. J ., & DeCotis, TA. (1991) Inductive measures of psychological climate,
Human Relations, 44(3), 265-285.

Langer, E. J ., & Saegert, S. (1977). Crowding and cognitive control. Journal of
Personality and Social Psychology, 34,175-182

Lantrip, D. B. (1993). Predicting satisfaction with the office environment by measuring
constraints to worker activities, In Proceedings of the Human factor and
Ergonomics Society 3 7” Annual Meeting, 489-493.

Larsen, L., Adams, J., Deal, 8., Kweon, B., & Tyler, E. (1998). Plants in the Workplace,
The effects of plant density on productivity, attitude, and perceptions.
Environment and Behavior, 30(3), 261-281.

145

Law, K. S., & Wong, C. (1999). Multidimensional constructs in Structural equation
analysis: an illustration using the job perception and job satisfaction constructs.
Journal of Management, 25 (2), 143-160.

Leather, P., Pyrgas, M., Beale. D., & Lawrence C. (1998). Windows in the Workplace,
Sunlight, View, and Occupational stress. Environment and Behavior, 30(6),
pp739-762.

Lee, C., Ashford, S. J. & Bobko, P. (1990). Interactive effects of “Type A” behavior and
perceived control on worker performance, job satisfaction, and somatic
complaints, Academy of Management Journal, 33(4), 870-881.

Lee, Q., Amundsen, A. E., Nelson, W., & Tuttle, H. (1997). Facilities and workplace
design, An illustrated guide. Georgia: Engineering & Management Press.

Marans, R.W., & Spreckelmeyer, K. F. (1982). A conceptual model for evaluating work
environments, Environment and Behavior, 14(3), 333-351.

Marans, R.W., & Yan, X. (1989). Lighting quality and environmental satisfaction in open
and enclosed offices. Journal of Architectural and Planning Research, 6, 118-131.

McDonald, R.P. & Ho, M. R. (2002). Principles and Practice in Reporting Structural
Equation Analyses, Psychology Methods, 7(1), 64-82.

MacLaney, M. A., & Hurrell, J. J. (1988). Control, stress, and job satisfaction in
Canadian nurses. Work and Stress, 2, 217-224.

McCarrey, M. W., Peterson, L. Edwards, S. & Von Kulmiz, P. (1974). Landscape office
attitudes: Reflections of perceived degree of control over transactions with the
environment. Journal of Applied Psychology, 59, 401-403.

Moleski, W. H., & Lang, J. T. (1982). Organizational needs and human values in office
planning. Environment and Behavior, 14(3), 319-332.

Mullen, B., & Copper, C. (1994). The relations between group cohesiveness and
performance: an integration. Psychological Bulletin, 115, 210-227.

Oldham, G. R. (1988). Effects of changes in workspace partitions and spatial density on
employee reactions: A quasi-experiment. Journal of Applied Psychology, 73(2),
253-258.

Oldham, G. R., & Brass, D. J. (1979). Employee reactions to an open-plan office: A
naturally-occurring quasi-experiment. Administrative Science Quarterly, 24, 267-
284.

Oldham, G. R., & Fried, Y. (1987). Employee reactions to workspace characteristics.
Journal of Applied Psychology, 72, 75-80.

146

Oldham, G. R., & Rotchford, N. L. (1983). Relationships between office characteristics
and employee reactions: A study of the physical environment. Administrative
Science Quarterly, 28, 542-556.

Olson, J. (2002). Research about office workplace activities important to US business-
and how to support them. Journal of Facilities Management, 1(1), 31-47.

O’Neill, M. (1994). Work space adjustability, storage, and enclosure as predictors of
employee reactions and performance. Environment and Behavior, 26(4), 504-526.

Osmond, H. (1957). Function as the basis of psychiatric ward design. Mental Hospitals
(Architectural Supplement), 8, 23-29.

Parasuraman, A. Zeithaml, V. A. & Beny, L. (1985). A conceptual model of service
quality and its implications for future research. Journal of Marketing, 49, 41-50.

(1988). SERVQUAL: A multiple-item scale for measuring consumer
perceptions of service quality, Journal of Retailing, 64 (1), 12-40.

Paciuk, M. (1989). The role of personal control of the environment in thermal comfort
and satisfaction at the workplace, Unpublished dissertation. The University of
Wisconsin-Milwaukee.

Pena, W. (1987). Problem seeking. Washington DC: American Institute of Architects
Press.

Porteous, J. (1977). Environment and Behavior. Reading, MA: Addison-Wesley.

Proshansky, H. M., Ittelson, W. H., & Rivlin, L. G. (1970). Environmental Psychology;
Man and his Physical Setting. New York: Holt, Rinehart and Windston.

Pugsley, D., & Haynes, B. (2002). An alternative use of space in government office
accommodation, Facilities, 20, 1/2, 34-40.

Roberts, J. A., Lapidus, R. S., & Chonko, L. B. (1997). Salespeople and stress: the
moderating role of locus on control on work stressors and felt stress. Journal of
Marketing Theory and Practice, 5(3), 93-108.

Robertson, K. (2000). Work transformation: integrating people, space and technology,
Facilities, 18,10/11/12, 376-382.

(1999). Work Transformation, Planning and Implementing the New
Workplace. New York: HNB Publishing

 

Rodin, J ., & Baum, A. (1978). Crowding and helplessness: Potential consequences of
density and loss of control. In A. Baum & Y. Epstein (Eds.), Human response to
crowding, 389-401. Hillsdale, NJ: Erlbaum.

147

Roelofsen, P. (2003). The impact of office environments on employee performance: The
design of the workplace as a strategy for productivity enhancement. Journal of
Facilities Management, 1(3), 247-264.

Rotter, J. B. (1966). Generalized expectancies for internal versus external control of
reinforcement. Psychology Monographs, 8O (1, Whole No. 609).

Sargent, L. D., & Terry, D. J. (1998). The effects of work control and job demands on
employee adjustment and work performance. Journal of Occupational and of
Organizational Psychology, 71, 216-236. The British Psychology Society.

Skinner, E. A., Chapman, M., & Baltes, P. B. (1988). Control, means-ends, and agency
beliefs: A new conceptualization and its measurement during childhood. Journal
of Personality and Social Psychology, 54: 117-133.

Smith, R. A., & Houston, M. J. (1982), “ Script- based on evaluation of satisfaction with
services, in L. Berry, G. Shostack, & G. Upah (Eds.), Emerging Perspectives on
Service Marketing, Chicago: American Marketing, 59-62.

Sparks, K., Faragher, B. & Cooper, C. L. (2001). Well-being and occupational health in
the 21‘“ century workplace, Journal of Occupational and Organizational
Psychology, 74, 489-509.

Spector, P. E. (1986). Perceived control by employees: A meta-analysis of studies
concerning autonomy and participation at work, Human Relations, 39 (11), 1005-
1016

Spreckelmeyer, K. (1993). Office relocation and environmental change, a case study.
Environment and Behavior, 25(2), 181-204.

Sprout, H., & Sprout, M. (1965). The Ecological Perspective on Human Aflairs, with
special reference to international politics. New Jersey: Princeton University Press.

Stewart, T. R. (1987). Developing an observer-based measure of environment annoyance.
In H.S. Koelega (Ed.), Environmental Annoyance: Characterization,
Measurement, and Control, 213-232. New York: Elsevier.

Stokols, D., & Altman, I. (1987). Handbook of Environmental Psychology, New York:
Wiley.

Stone, N. J. (1998). Windows and Environmental cues on performance and mood.
Environment and Behavior, 30(3), 306-322.

Stone, N. J ., & Irvine, J. M. (1994). Performance, mood, satisfaction, and task type in
various work environments: a preliminary study. Journal of General Psychology,
120, 489-497.

148

Sundstrom, E. (1986). Work Places: The psychology of the physical environment in
offices and factories. Cambridge; Cambridge University Press.

Sundstrom, E., Burt, R. E. & Kamp, D. (1980). Privacy at work: Architectural correlates
of job satisfaction and job performance. Academy of Management Journal, 23(1),
101-117.

Sundstrom, E., Town, J ., Brown, D., Forman, A., & McGee, C. (1982). Physical
enclosure, type of job, and privacy in the office. Environment and Behavior, l4( 5),

543-559.

Sundstrom, E., Town, J ., Rice R., Osborn, D., & Brill, M. (1994). Office noise,
satisfaction, and performance. Environment and Behavior, 26(2), 195-222.

Szigeti, F. & Davis, G. (1997). Using the ASTM standards for whole building
functionality and serviceability for major asset and portfolio decisions. In the
Proceedings of the 2002 global symposium, applying and extending the global
knowledgebase, 168-182. CIB W70: Facilities management and maintenance.

Taylor, S. F. & F iske, S. T. (1978). Salience, attention and attribution: Top of the head
phenomena. In L. Berkowitz (Ed.), Advance in experimental social psychology,
vol] 1. New York: Academic Press.

Tetrick, L. E., & LaRocco, J. M. (1987). Understanding, prediction, and control as
moderators of the relationships between perceived stress, satisfaction and
psychological well-being. Journal of Applied Psychology, 72(4), 538-543.

Thatcher, C., & Thatcher, G. (1995). Reducing the cost of churn, Euro FM conference
proceeding. Frankfurt, Germany.

US. Census Bureau (2001). Statistical abstract of the United States:2001, Labor force,
employment and earnings. Retrieved November 14, 2003, from
http://www.census.gov./prod

Veitch, J. A. & Gifford, R. (1996). Choice, perceived control, and performance
decrements in the physical environment. Journal of Environmental Psychology,
16, 269-276.

Veitch, J. A. & Newsham, G. R. (2000). Exercised control, lighting choices and energy
use: an office simulation experiment. Journal of Environmental Psychology, 20,
219-237.

Wheaton, B., Muhten, B., Alwin, D. F., & Summers, GE (1977). Assessing reliability in
panel models. In D.R. Heise (Ed.), Sociological methodology, 85-136. San
Fransico: Jossey- Bass.

149

 

 

 

Widmeyer, W. N., Brawley, L. R., & Carron, A.V. (1985). The measurement of cohesion
in sport team: The group environment questionnaire. London. Ontario; Sports

Dynamics

Wineman, J. D. (1982). Office design and evaluation, an overview. Environment and
Behavior, 14(3), 271-298.

Wood, 8., J. (1989). Degrees of privacy affecting employees and their work tasks in
open-plan offices, Unpublished Thesis, Michigan State University.

Zalesny, M., & Farace, R. V. (1987). Traditional versus open office: a comparison of
sociotechinical, social relations, and symbolic meaning perspectives, Academy of
Management Journal, 30 (2), 240-259.

Zalesny, M. D., Farace, R.V., & Hawkins, R. K. (1985). Determinants of employee work
perceptions and attitudes, perceived work environment an organizational level.
Environment and Behavior, 17(5), 567-592.

Zahn, G. L. (1991). Face to face communication in an office setting, the effects of
position, proximity, and exposure. Communication Research, 18(6), 737-754.

150

   

ST

11111111111111111111111