Content uploaded by Christina Bodin Danielsson
Author content
All content in this area was uploaded by Christina Bodin Danielsson on Jul 01, 2018
Content may be subject to copyright.
https://doi.org/10.1177/0013916518759146
Environment and Behavior
1 –32
© The Author(s) 2018
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0013916518759146
journals.sagepub.com/home/eab
Original Article
Office Employees’
Perception of Workspace
Contribution: A Gender
and Office Design
Perspective
Christina Bodin Danielsson1 and Töres Theorell2
Abstract
In this exploratory study, we investigated the relationship between office
design and employee perception of its contribution to job satisfaction,
comfort, and performance. The study includes 4,352 employees in seven
different office designs. Associations between workspace satisfaction and
perceived access to supportive facilities (ancillary spaces for concentrated
work and for different meetings) were also investigated since these
factors may be related to employees’ workspace satisfaction. Multivariate
linear and logistic regression analyses were performed separately for
men and women with adjustments for age and education. Supplementary
correlation analyses were performed between workspace satisfaction
and perceived access to supportive facilities. Results showed differences
between employees’ workspace satisfaction in studied office designs.
Those with the lowest ratings of access to supportive facilities reported
the lowest degree of satisfaction. The best results were found in cell-
offices and the worst ones in hot-desking offices. Gender differences
were also observed.
1The Royal Institute of Technology, Stockholm, Sweden
2Stress Research Institute, Stockholm University, Sweden
Corresponding Author:
Christina Bodin Danielsson, KTH School of Architecture, The Royal Institute of Technology
(KTH), Osquars backe 9, 100 44 Stockholm, Sweden.
Email: christina.bodin.danielsson@bredband.net
759146EABXXX10.1177/0013916518759146Environment and BehaviorBodin Danielsson and Theorell
research-article2018
2 Environment and Behavior 00(0)
Keywords
office design, defining features, contribution of workspace, supportive
facilities, employee satisfaction
Introduction
A full-time employee spends approximately 40% of waking hours at work,
which is, for a majority of the workforce in the Western World today, an
office (e.g., Duffy, 1997; Larsson, 2012). Research has shown that the work
environment plays a major role for employees’ health and well-being
(Theorell, 2000). An important dimension of this is the psychosocial envi-
ronment comprising various factors, such as social relationships at work and
organizational factors, i.e., those that relate to the structure and functioning
of the organization, such as lack of resources, malfunctioning categoriza-
tion, repeated organizational changes, excessively long working hours, and
lack of career advancement opportunities (De Raeve, Jansen, van den
Brandt, Vasse, & Kant, 2008; Tett & Meyer, 1993). For example, research
has shown that social support from colleagues has positive effects on
employees’ general health and sick leave (Fusilier, Ganster, & Mayes, 1986;
Stansfeld, Head, & Marmot, 2000), as well as preventive effects in relation
to depression, anxiety disorders (Netterstrøm et al., 2008), and stress-related
ill-health (Frese, 1999). Good relationships with management are also
important to employee health (M. Kivimäki et al., 2003; Nyberg, Westerlund,
Magnusson Hanson, & Theorell, 2008). Negative job characteristics include
excessive work demands, poor job control, and monotonous work (e.g.,
Allebeck & Mastekaasa, 2004; Karasek & Theorell, 1990; Vahtera, Pentti, &
Uutela, 1996).
That there are gender differences with regard to the relationship between
work environments on one hand and employee health and well-being on the
other hand is well documented. Part of this is explained by the fact that men
and women are exposed to different risk factors (Kivimäki et al., 1997;
Krantz, 2003). Occupational health research has shown that women report
more stress, poorer health (Krantz, Berntsson, & Lundberg, 2005), and more
sick leave (Niedhammer, Bugel, Goldberg, Leclerc, & Guéguen, 1998).
Gender differences have also been found with regard to social relations at
work and patterns of communication with colleagues, with women being
more relation and less instrumentally oriented than are men (Anderson &
Martin, 1995) but also giving and receiving more social support to and from
colleagues at work than men (Plaisier et al., 2007; Winter, Roos, Rahkonen,
Martikainen, & Lahelma, 2006).
Bodin Danielsson and Theorell 3
Our knowledge of the relationship between the physical work environ-
ment and employees’ health and well-being is less extensive than that of the
relationship between psychosocial or organizational factors and employee
health. In the past decade, however, the field has gained more interest since
researchers have found indications that the office environment may well
have an influence on various health outcomes among employees (e.g., Bodin
Danielsson & Bodin, 2008; Bodin Danielsson, Chungkham, Wulff, &
Westerlund, 2014; Meijer, Frings-Dresen, & Sluiter, 2009; Pejtersen,
Feveile, Christensen, & Burr, 2011). Bodin Danielsson and Bodin (2008)
found, for example, that office type per se has a significant influence on
employees’ health status. In their study, the best health was found among
employees in cell-offices, that is, individual offices, and flex-offices, that is,
an activity-based office type with nonpersonal workstations and good access
to supportive work environments, while the poorest health was found in the
traditional open plan offices. Support for this less beneficial influence of
traditional open plan offices on employees’ health has also been found in
other studies. For example, Pejtersen et al. (2011) have shown an elevated
risk of sick leave in workspaces shared by six people or more compared with
other workspaces. Also Bodin Danielsson et al. (2014) have shown that the
number of short sick leave spells, less than or equal to one week, was signifi-
cantly higher among employees in traditional open plan offices in compari-
son with other office types.
Gender differences with regard to influences of the physical environment
have not been well researched, in general or in terms of the physical work
environment. According to the sparse research in the field, some differences
have, however, been identified. For example, it has been found that women
report more problems with noise disturbance than do men (Kaarlela-
Tuomaala, Heleniusa, Keskinen, & Hongisto, 2009), and that among female
employees with high job complexity, reported problems with noise correlate
with sick leave (Fried, Melamed, & Ben-David, 2002). In terms of perception
of environmental information, there are also indications that there are gender
differences with women being more perceptive to details, for example
(d’Astous 2000). Regarding office architecture, there are indications that
office type per se has a stronger influence on female employees than on male
employees, for example, with more influence on sickness absenteeism among
women (Bodin Danielsson et al., 2014). This, in turn, may be related to gen-
der differences in social relations at work, with women being more relational
and less instrumental in their communication with colleagues than are men,
as previously described. In addition, women have been described as more
attuned to inter-personal relationships than are men (Addis & Mahalik, 2003;
Gabriel & Gardner, 1999).
4 Environment and Behavior 00(0)
A review of the office research indicates that traditional open plan offices
are less beneficial for employees’ health and well-being. Why this would be
the case is, however, not clear, although different hypotheses have been pre-
sented. One hypothesis is that the risk of infections increases with number of
employees sharing workspace (Pejtersen et al., 2011). Another hypothesis is
that it is due to a high level of exposure to environmental stimuli in tradi-
tional open plan offices, for example, noise disturbances (Bodin Danielsson
& Bodin, 2009). Noise, considered the major environmental stressor in
offices with open plan layouts (e.g., Sundstrom, 1986), has various implica-
tions for office employees. In laboratory studies, noise has been shown to
have a negative impact on participants’ motivation and cognitive perfor-
mance, as well as on fatigue (Jahncke, Hygge, Halin, Green, & Dimberg,
2011). Motivational deficits due to noisy work conditions were also identi-
fied in another experimental study, although no significant differences in
stress levels were identified (Evans & Johnson, 2000). Despite the negative
influence of noise on employees’ self-rated health, negative influences do
not necessarily have to exist for other outcomes that are important to employ-
ees’ well-being at work. In a study by P. J. Lee, Lee, Jeon, Zhang, and Kang
(2016), for example, no negative influence of noise was found on office
employees’ job satisfaction. On the other hand, degree of speech privacy had
such an influence. The concept of privacy, which includes acoustics, speech,
and visual privacy is of relevance in office design due to its relation to per-
sonal control (Haans, Kaiser, & de Kort, 2007; Kupritz, 1998) and distrac-
tion (Y. Lee & Brand, 2005). The significantly higher risks of privacy
problems identified in traditional open plan offices (Bodin Danielsson &
Bodin, 2009) may, thus, be associated with the lower job satisfaction and
environmental satisfaction among employees in these office types (Bodin
Danielsson & Bodin, 2008, 2009).
When offices with open plan layouts are discussed, it has to be recognized
that sharing workspace with colleagues may also be beneficial for office
employees. Sharing workspace may, for example, be beneficial for the social
climate at the workplace (Hedge, 1982). This may be explained by the bene-
ficial effect that physical proximity between colleagues has on both interac-
tion and friendship at the office (Conrath, 1973), which in turn may be related
to the fact that tangible support in social networks has been found to decrease
with distance (Mok & Wellman, 2007). Also from the perspective of a man-
ager–employee relationship, there appear to be some benefits of sharing
workspace. For example, employees who see and hear their managers in the
workspace often perceive them as friendlier than other employees do (Crouch
& Nimran, 1989). Among employees in medium-sized open plan offices,
where employees and managers often share workspace, it has also been found
Bodin Danielsson and Theorell 5
that employees rate their relationship to managers better than do those in
other office types (Bodin Danielsson, Wulff, & Westerlund, 2013). Additional
positive effects of sharing workspace have been identified. Increased social
density at the office may, for example, have positive effects not only on task
facilitation and information exchange but also on job satisfaction (Szilagyi &
Holland, 1980). Proximity to coworkers in many cases seems to foster col-
laboration at the workplace, something especially critical for organizations
that rely heavily on creativity and innovation (e.g., Allen, 1997; Becker &
Sims, 2001; Hua, Loftness, Heerwagen, & Powell, 2011). However, proxim-
ity to coworkers is not always positive as the sense of having coworkers too
close can be detrimental for collaboration (Hua et al., 2011). This relates to
previously described problems associated not only with acoustics and visual
privacy but also with crowding. Crowding is an environmental stressor asso-
ciated with spatial density that may cause physiological arousal and lead to
stress symptoms that are behavioral (e.g., reduced ability to be social,
increased intention of job turnover) and physiological (e.g., high blood pres-
sure; for example, Aiello, Epstein, & Karlin, 1975; Evans, 1979; Oldham,
1988). In summary, existing office research indicates that sharing workspace
and facilities may have both positive and negative effects depending on the
issue at focus.
In terms of environmental satisfaction, there are studies that have shown
this also varies highly between employees in different office types (Bodin
Danielsson & Bodin, 2009). With regard to the transition to the more extreme
activity-based flex-office (A-FO) (i.e., non personal workstations with access
to different work environments within the office), it has been found that its
acceptance depends to a great extent on the architectural design of the office
(Brunia, De Been, & van der Voordt, 2016). In the implementation process,
the study showed that not only the management’s attitude to A-FO but also
why it was introduced played a central role for the employees’ environmental
satisfaction. Independently of that, for office employees, the perceived need–
supply fit (i.e., the fit between workers’ current work requirements and the
current office facilities) is possibly the most central factor for high environ-
mental satisfaction. In the transition to A-FO, it is even more important
according to recent research (Gerdenitsch, Korunka, & Hertel, 2017).
Nevertheless, we do not know to what degree employees in various office
designs perceive that their workspaces contribute to their job satisfaction.
Nor do we know how much satisfaction with the workspace adds to job sat-
isfaction. Such factors are, for example, the extent of satisfaction with the
contribution of workspace to comfort or performance. Nor do we know what
role employees’ perception of access to supportive facilities may play in the
extent of satisfaction with the contribution of workspace to job satisfaction.
6 Environment and Behavior 00(0)
Considering the potential importance perception of workspace contribution
may have not only for office employees’ well-being but also for their attitude
to their own organization, it is surprising how underexplored the subject is.
To our knowledge, the importance of the office design for employees’
perception of the contribution that the workspace makes for various out-
comes of employee well-being in relation to work, such as job satisfaction,
pleasantness of work environment, and so on, has not been well studied. Nor
has the potential role that an architectural feature such as access to supportive
facilities within the office, for instance, individual rooms for concentrated
work and spaces for meeting, could play for this been explored. Access to
supportive facilities within the office is an architectural feature that varies
highly between office designs. Rather than investigating employees’ percep-
tion of the contribution that workspace makes for factors important to well-
being, the focus has been on other aspects. For example, studies have been
made of differences in job satisfaction or well-being between employees
working in various office types (Bodin Danielsson & Bodin, 2008). The main
focus in this area has, however, been on how noise and visual disturbances
influence office employees’ well-being and environmental satisfaction (e.g.,
P. J. Lee et al., 2016; Sundstrom, Town, Rice, Osborn, & Brill, 1994). This is
due to the relation of these factors to privacy and personal control (Haans
et al., 2007; Kupritz, 1998; Y. Lee & Brand, 2005)—two conditions consid-
ered central for the individual’s health and well-being. Differences in envi-
ronmental satisfaction between employees in various office designs have also
been studied, although mainly in relation to noise and privacy, with some
exceptions. For example, Bodin Danielsson and Bodin (2009) investigated
differences in employees’ satisfaction with design-related factors such as
workstation and workspace design in various office types. Thus, to conclude,
satisfaction among employees with these design-related dimensions in differ-
ent office types has been investigated in various ways in office research. On
the other hand, to our knowledge, employees’ perception of the contribution
that the workspace makes to factors important to employees’ well-being and
job satisfaction has not been studied before. In addition, the potential role of
employees’ satisfaction with a defining architectural feature such as access to
supportive facilities differentiating various office types from each other has
not been studied in relation to employee satisfaction with workspace contri-
bution. This is the rationale behind the present study. In summary, our
research questions are as follows:
Research Question 1 (RQ1): Does employees’ satisfaction with the con-
tribution that workspace makes to job satisfaction, pleasantness, and doing
a good job differ between office designs? More specifically, is there (a)
Bodin Danielsson and Theorell 7
higher satisfaction among employees in cell-offices, i.e., individual
offices, due to minimal environmental disturbance; (b) higher satisfaction
among employees in office design with open plan layouts, due to optimal
social cohesion at the office (thanks to physical proximity and visual con-
tact between colleagues); and (c) variation in degree of satisfaction with
workspace contribution between office designs due to other factors. For
example, do employees in open plan layouts report higher satisfaction
with the contribution that workspace makes to comfort because of the
large spaces and good overview that they offer?
To what extent can we find such associations despite the fact that employ-
ees are more exposed to environmental stress in such offices?
Research Question 2 (RQ2): Is employees’ satisfaction with the contri-
bution workspace makes to job satisfaction, comfort, and doing a good job
associated with reported access to supportive facilities in the office?
Research Question 3 (RQ3): Are there gender differences with regard to
the previous research questions considering the established gender differ-
ences from other points of view?
Method
Participants
This exploratory study is based on the 2012 wave of the Swedish Longitudinal
Occupational Survey of Health (SLOSH), a nationally representative longitu-
dinal cohort study of work environment and occupational health conducted
every second year. SLOSH, which is voluntary, covers health and different
aspects of working life. This includes psychosocial, organizational, and phys-
ical aspects of the work environment, including questions about office design,
which makes SLOSH useful for the purpose of our study. Also background
data about participants are covered. Data are collected by paper-and-pencil
questionnaires (Hanson et al., 2011; Magnusson Hanson et al., 2009), with an
Internet questionnaire option offered in 2012. Participants for the present
study were 18,915 working subjects who had participated in the Swedish
Work Environment Survey, AMU, Arbetsmiljöundersökningen (Eng. The
Work Environment survey by the Swedish work environment authority),
2003 and 2005, and who participated in the fourth SLOSH wave in 2012.
Persons who had stopped working and who did not fill out the questions
regarding office design were excluded from our sample. The final analytic
sample, which remained after exclusion of participants who lacked
8 Environment and Behavior 00(0)
information on crucial items for our study, varied between 4,170 and 4,326
participants for different outcomes. In the sample, there were 55% women
and 45% men. The main loss (>1,000 participants) was due to participants
who did not work in an office and accordingly could not fill in information
regarding office design.
Measures
Office designs—Independent variable. The participants in our study worked in
one of seven office designs (see Table 1). These office designs are in turn
based on the seven office types identified in contemporary office designs
defined by their unique combination of architectural and functional features
(for definitions and illustrations of office types, see Bodin Danielsson, 2008;
Bodin Danielsson & Bodin, 2009; Bodin Danielsson, Bodin, Wulff, & Theo-
rell, 2015). The architectural features are physical, of which the spatial orga-
nization (i.e., plan layout) is the most prominent. On the other hand, the
functional features relate to the actual work taking place and the organization
of this (i.e., the use of the office and functions related to this). These features
are in turn determined by factors such as functional needs and technical fea-
sibility, such as ICT (Information Communication Technology). This could
also work in the opposite direction, that is, technical and functional possibili-
ties may lead to new organizations of work that affect the architectural design.
The identified office types are prototypes that can be described as “ideal”
office types, since there will always exist offices that differ from the defini-
tions or are “cross-pollinations” between certain office types. For example,
there are open plan offices that are similar to the activity-based office types
where employees have fairly good access to back-up rooms for meetings,
private conversations, or telephone calls, something the typical traditional
open plan offices lack.
The seven office designs identified in this exploratory study are (a) cell-
office, (b) shared-room office (two to three people in workspace), (c) small
open plan office (four to nine people in workspace), (d) medium-sized open
plan office (10-24 people in workspace), (e) large open plan office (≥25 peo-
ple in workspace), (f) combi-office (team-based activity-based office type),
and (g) hot-desking office (a subcategory to the activity-based office-type
flex-office). (For office designs used, see Table 1). The initial analysis of our
sample revealed that the participants in nonpersonal workstations did not
work in the office-type flex-office as expected but in a subcategory of this
labeled hot-desking office. More specifically, participants working in offices
with nonpersonal workstations did not report good access to supportive facil-
ities such as spaces for concentrated or collaborative work—These are
Bodin Danielsson and Theorell 9
Table 1. Office Designsa—Prototypes Defined by Architectural and Functional
Features.
Office design: Architectural features Functional features
1. Cell-office (personal room)
--The plan layout is characterized by corridors, either
a single or double corridor system
--Most equipment is in the own room
--Individual room has access to a window --Work is concentrated and independen
2. Shared-room office (2-3 people/room). Often a consequence of lack of workspace
--Workstations freely arranged in the room
--For privacy reasons, sometimes screens or other
divisional elements between workstations
--No individual window, shares with roommate(s)
--Team-based work or people with similar work
assignment share room
--Most equipment outside of room, team-based
shared rooms tend to have own equipment
Traditional open plan offices—employees share workspaces in different configurations
Found in following three subcategories:
3. Small open plan office (4-9 people share workspace)
4. Medium-sized open plan office (10-24 people share workspace)
5. Large open plan office (>24 people share workspace)
--Shared workspaces within the office
--Plan layout is open, based on an open flow of
workspaces instead of corridor systems
--Workstations freely arranged in the room or in
rows in a larger workspace
--Flexible for organizational changes
--Routine-based work
--Low level of interaction between employees
--Often no amenities at workstation
Activity-based and flexible office types—focused on flexibility and work activity
(Includes combi- and flex-office. Flex-office is not represented in this study)
6. Combi-office—teamwork, sharing of workspace and facilities
--No strict spatial definition, personal workstations
can be either individual rooms or open plan office
--Back-up rooms for work activities not suitable at
the personal workstation. Extra focus on group
activities such as project rooms (to be booked for
longer periods), team rooms, and meeting rooms
-->20% of the work in the office not at personal
workstation
--Sharing of common amenities in common spaces
--Work is both independent and with a high degree
of interactive teamwork with colleagues
--The team move around in the office on a “need
basis” using common facilities designed for this
purpose
Subcategory to office-type flex-office used in this study:
7. Hot-desking office—nonpersonal workstations, no access to back-up environments
--Open plan office
--No access to supportive facilities, for example,
back-up rooms and supplementary environments
in the office, which differentiates it from flex-offices
--No personal workstations
--Often routine-based work, with low degree
of characteristic skill discretion and decision
authority
aThe office designs presented are based on six office-types out of seven identified in contemporary office
design. One category added to the original six office-types is the seventh one labeled hot-desking office, which
is separated from the sixth office-type. For illustrations of the seven office-types identified in contemporary
architectural design practice, see Bodin Danielsson (2008) and Bodin Danielsson and Bodin (2009).
10 Environment and Behavior 00(0)
architecture features that define a flex-office. Nor did these participants
report high decision authority, which is also a functional feature of the office-
type flex-office. Therefore, in this article, we use the term office design
instead of office type.
The classification of the participants into the seven office designs in our
study was done by the participants themselves, since it is by a unique combi-
nation of specific questions in the SLOSH questionnaire that the office types/
office designs are given. The questions covered ownership of workstation,
spatial organization of number of people sharing workspace combined with
job characteristics, and access to supportive workspaces. A combination of
the two latter dimensions constituted hot-desking offices. (For information
regarding SLOSH and questions used in SLOSH2012, see http://wwww.
idear-net.net/slosh/in-english).
Dependent variables—contribution workspace makes and access to supportive
facilities. The main analyses included two categories of dependent outcome
variables that measure employee satisfaction with (a) contribution workspace
makes, and (b) access to supportive facilities at the office.
Employees’ satisfaction with the contribution that workspace makes. This was
assessed with three outcome variables for satisfaction with the contribution
workspace makes to the following:
a. Job satisfaction. “Do you think the workspace you work in contrib-
utes to job satisfaction?”
b. Pleasantness. “Do you think the workspace you work in is
pleasant?”
c. Do a good job. “Do you think your office (i.e., physical workplace)
enables you to do a good job?”
The response scales ranged from “not at all” (1) to “to a great extent” (4),
which resulted in sum scores ranging from 4 to 16, with a high score corre-
sponding to high satisfaction.
Access to supportive facilities at the office. This was assessed with three out-
come variables:
a. Individual rooms for concentrated work. “At the office, do you have
access to individual rooms for concentrated work?”
b. Spaces for small meetings. “At the office, do you have access to spaces
for small open meetings?”
c. Spaces for booked meetings. “At the office, do you have access to
spaces for booked meetings?”
Bodin Danielsson and Theorell 11
The response scales ranged from “yes, sufficiently” (1) to “no, not at all” (3).
These variables were analyzed in relation to office design by means of mul-
tiple logistic regression after dichotomization of the responses (scores 2 and
3 combined)—which means that subjects with “sufficient support” were
compared with the others.
Statistical Analyses
In the main analyses, multivariate linear regression models were used to ana-
lyze the association between office design, the independent variable, and the
first category of outcome variable: (a) satisfaction with the contribution that
workspace makes. For the second outcome category—(b) perception of
access to supportive facilities at the office—multivariate logistic regression
was, instead, used. In all the analyses, we adjusted for the background factors
age and education. Linear regression was chosen as the statistical method
because it provides maximal retention of statistical variance and because the
scales for satisfaction with contribution that workspace makes were close to
normally distributed. With regard to the variables related to access to sup-
portive facilities, dichotomization with multiple logistic regression was a
more appropriate statistical method since these variables were far from nor-
mally distributed.
In accordance with the aim of this exploratory study, we used the indepen-
dent variable office design as the main explanatory variables. As previously
described, the term office design used in the study includes six out of the
seven identified office types and one subcategory of one office type. Cell-
office was chosen as a reference category in the analyses against which the
other office designs were compared. The rationale behind this is that the cell-
office is easy to define—all other office designs mean sharing of workspace
and amenities between colleagues to varying degrees.
The results of our main analyses are presented either as means or as odds
ratios (ORs) with 95% confidence intervals for men and women separately. A
higher number for a given variable compared with cell-office for a specific
office design represents a higher likelihood, while a lower value represents
the opposite. Accordingly, a higher value for a given outcome variable for
both workspace satisfaction and access to supportive facilities indicates a
better situation in comparison with cell-office, while a lower value indicates
a worse situation.
In addition to the main analyses, a supplementary correlation analysis
between the two categories of dependent outcome variables was performed
(i.e., satisfaction with the contribution that workspace makes and access to
supportive facilities). We considered it to be of interest to examine not only
how strongly the outcome variables within the two 3-factor sets correlated
12 Environment and Behavior 00(0)
with one another but also how the two groups of outcome variables were
intercorrelated.
The computations were performed by JMP, version 11.2. For all the mul-
tivariate analyses, the program first tests the total effect of our explanatory
variable, that is, office design. Second, there are post hoc tests for each office
design after adjustment for age and education. Demographics presented in
Table A1 of the online appendix include gender, mean age, education (low/
middle/high). We also analyzed (not presented) job rank (SEI, Swedish
Socio-economic classification) and supervisory position (four grades; see
Table A1 in online appendix). Hence, only education was used as a covariate
in the multivariate statistics since collinearity precludes use of both as covari-
ates. The effect of including supervisory position as a covariate was tested
without any effect on results—with one exception for the outcome variable:
“access to rooms for booked meetings.” Results from this analysis including
supervisory position are presented in the text.
Results
Descriptive Results on Background Factors
Our descriptive results regarding background factors (age, education, job
rank, supervisory position) show that these are all significantly differently
distributed across the office designs for both genders. In Table A1 (see online
appendix), the distributions of age, gender, and education level are shown
with p values between .0006 and <.0001. (Data on job rank and supervisory
position available on request.)
The largest proportion of participants worked in cell-offices (women =
46%, men = 50%), and the smallest in combi-offices, 81 employees (2% of
both genders). Our sample consisted of more women (54.7%) than men, in all
office designs. The most unequal gender distribution was found in hot-desk-
ing offices. Regarding age, we found the oldest employee group in cell-
offices with mean age for women = 51.5 years and for men = 52.5 years,
while the youngest group was found in large open plan offices (women = 46
years, men = 48 years).
The lowest average educational level was found among female employees
in hot-desking offices and the highest in small open plan offices. Among
men, the highest educational level was reported in the three office designs—
small and medium-sized open plan offices as well as combi-offices. Despite
a higher mean educational level among women, we found men at higher job
ranks (SEI). The highest proportion of women with high job ranks worked in
cell-offices (39.4%), while for men, this was in large open plan offices
Bodin Danielsson and Theorell 13
(47.0%) and combi-offices (48.5%). The highest proportion of managers
with staff responsibility was found in cell-offices (women = 22.0%, men =
40.4%), while the lowest proportion was found in hot-desking offices (women
= 7.5%, men = 12.3%). More men than women held supervisory positions in
our sample (women = 15.5%, men = 35.4%). (Data on job rank and supervi-
sory position are only presented in the text, but tables are available on
request.)
Satisfaction With the Contribution That Workspace Makes
Table 2 presents results from multiple linear regression analyses of the impor-
tance of the different office designs for three dependent variables capturing
satisfaction with the contribution that workspace makes (to job satisfaction,
pleasantness, and doing a good job). It shows numbers of participants with
complete data and mean scores (unadjusted and adjusted for covariates),
unadjusted 95% confidence intervals, and p values corresponding to the con-
tribution of each one of the office designs to the dependent variable in mul-
tiple linear regression, including both covariates and all office designs, with
cell-office as reference group. In addition, the table shows the results of anal-
yses of variance with p value for the total regression model, and skewness
and kurtosis for the dependent variables.
Results show an overall strong association between office design and
employees’ satisfaction. For both genders, the significant association between
office design and employees’ satisfaction with the contribution that work-
space makes persists after adjustment for age and education (p < .0001). Age
per se is significantly associated with contribution of workspace satisfaction
among men only, and then only for contribution to job satisfaction and to
pleasantness.
Our post hoc analyses showed a repetitive pattern with employees in cell-
offices reporting much higher satisfaction with contribution that workspace
makes than do other employees, while, in contrast, employees working in
hot-desking offices reported lower satisfaction in this regard.
Access to Supportive Facilities at the Office
Table 3 presents results from multiple logistic regression analyses of the
importance of the different office designs for the three dependent variables
assessing perception of access to supportive facilities. This is assessed with
access to individual rooms for (a) concentrated work, and spaces for (b) small
spontaneous meetings and (c) booked meetings. It shows numbers of partici-
pants with complete data as well as percentage of those satisfied: participants
14
Table 2. Satisfaction With Workspace Contribution (To Job satisfaction, Pleasantness, and Do a Good Job).
Outcome scale
1-4aOffice design (N)
Unadjusted
(adjusted) [95% CI]
Impact of office
designbdf F p value
Skewness = −.32
Kurtosis = −.81
Contribution to job satisfaction
Male
(N = 1,966)
1 - Cell-office (Ref.) 962 3.13 (3.12) [3.08—3.18] Ref. Office design 6 47.19 <.0001
2 - Shared-room office 357 2.69 (2.70) [2.60—2.77] .048 Age 1 6.13 .013
3 - Small open plan 191 2.42 (2.43) [2.30—2.54] .003 Education 1 0.16 n.s.
4 - Medium-sized open plan 99 2.51 (2.52) [2.33—2.68] n.s.
5 - Large open plan 178 2.44 (2.45) [2.31—2.57] .012
6 - Combi-office 33 2.76 (2.75) [2.48—3.04] n.s.
7 - Hot-deskingb116 2.32 (2.24) [2.08—2.38] <.0001
Skewness = −.25
Kurtosis = −.92
Female
(N = 2,360)
1 - Cell-office (Ref.) 1,077 3.23 (3.23) [3.18—3.27] Ref. Office type 6 81.04 <.0001
2 - Shared-room office 455 2.68 (2.68) [2.60—2.76] <.0002 Age 1 31.17 n.s.
3 - Small open plan 276 2.43 (2.43) [2.32—2.55] n.s. Education 1 0.02 n.s.
4 - Medium-sized open plan 128 2.48 (2.47) [2.32—2.63] n.s.
5 - Large open plan 172 2.47 (2.46) [2.34—2.47] n.s.
6 - Combi-office 48 2.17 (2.16) [1.88—2.46] .0008
7 - Hot-deskingb204 2.22 (2.22) [2.10—2.34] <.0001
Impact of office
designc
Skewness = −.38
Kurtosis = −.59
(continued)
15
Outcome scale
1-4aOffice design (N)
Unadjusted
(adjusted) [95% CI]
Impact of office
designbdf F p value
Skewness = −.32
Kurtosis = −.81
Contribution to pleasantness
Male
(N = 1,968)
1 - Cell-office (Ref.) 990 3.01 (3.00) [2.96—3.06] Ref. Office design 6 31.17 <.0001
2 - Shared-room office 357 2.61 (2.62) [2.52—2.70] n.s. Age 1 9.63 .002
3 - Small open plan 191 2.40 (2.41) [2.27—2.53] .032 Education 1 0.02 n.s.
4 - Medium-sized open plan 101 2.54 (2.56) [2.38—2.71] n.s.
5 - Large open plan 178 2.42 (2.43) [2.28—2.55] n.s.
6 - Combi-office 35 2.55 (2.54) [2.28—2.81] n.s.
7 - Hot-deskingb116 2.23 (2.25) [2.08—2.38] <.0001
Female
(N = 2,357)
1 - Cell-office (Ref.) 1,073 3.15 (3.15) [3.10—3.20] Ref. Office design 6 58.16 <.0001
2 - Shared-room office 455 2.63 (2.63) [2.55—2.71] n.s. Age 1 0.04 n.s.
3 - Small open plan 276 2.48 (2.48) [2.37—2.60] n.s. Education 1 1.80 n.s.
4 - Medium-sized open plan 128 2.54 (2.54) [2.38—2.70] n.s.
5 - Large open plan 172 2.47 (2.48) [2.33—2.60] n.s.
6 - Combi-office 48 2.31 (2.31) [2.04—2.59] .039
7 - Hot-deskingb205 2.22 (2.22) [2.10—2.34] <.0001
Table 2. (continued)
(continued)
16
Outcome scale
1-4aOffice design (N)
Unadjusted
(adjusted) [95% CI]
Impact of office
designbdf F p value
Skewness = −.32
Kurtosis = −.81
Contribution to do a good job
Male
(N = 1,960)
1 - Cell-office (Ref.) 988 3.30 (3.30) [3.25—3.34] Ref. Office design 6 53.97 <.0001
2 - Shared-room office 357 2.90 (2.90) [2.82—2.98] .001 Age 1 2.59 n.s.
3 - Small open plan 190 2.57 (2.57) [2.46—2.68] .0002 Education 1 0.45 n.s.
4 - Medium-sized open plan 99 2.72 (2.72) [2.57—2.87] n.s.
5 - Large open plan 177 2.62 (2.63) [2.50—2.75] .009
6 - Combi-office 33 2.70 (2.69) [2.47—2.92] n.s.
7 - Hot-deskingb116 2.55 (2.56) [2.42—2.68] .001
Female
(N = 2,362)
1 - Cell-office (Ref.) 1,076 3.34 (3.35) [3.30—3.39] Ref.
2 - Shared-room office 457 2.82 (2.82) [2.74—2.89] .003 Office design 6 81.76 <.0001
3 - Small open plan 277 2.63 (2.63) [2.53—2.73] n.s. Age 1 0.36 n.s.
4 - Medium-sized open plan 128 2.60 (2.60) [2.46—2.75] n.s. Education 1 0.88 n.s.
5 - Large open plan 172 2.59 (2.58) [2.47—2.71] .031
6 - Combi-office 48 2.44 (2.43) [2.18—2.69] .006
7 - Hot-deskingb204 2.50 (2.50) [2.40—2.60] <.0001
Note. Mean values of scores with 95% CIs. Multivariate linear regression analysis. Crude (no adjustment) and adjusted model presented. Adjusted model in parentheses
controlled for education and age. Reference category (cell-office) is marked with underlined text, significances are marked in bold text. p value adjusted for education
and age. CI = confidence interval.
aResponse scales ranged from “not at all” (1) to “to a great extent” (4).
bHot-desking is a not an office type but a subgroup to the office-type flex-office.
Table 2. (continued)
17
Table 3. Perception of Access to Supportive Facilities in Office Design.
Outcome Scale
1-3aOffice design
Adjusted
(N)
% Unadjusted
(adjusted) [95% CI]
Impact of
office designb
Access to
Individual rooms for concentrated work
Male
(N = 1,959)
1 - Cell-office (986) 83.3 (83.1) [80.8—85.5] Ref.
2 - Shared-room (354) 62.1 (62.3) [57.0—67.0] n.s.
3 - Small open plan (191) 53.9 (54.1) [46.8—60.8] .025
4 - Medium-sized open plan (101) 59.4 (59.5) [49.7—68.5] n.s.
5 - Large open plan (178) 55.1 (55.1) [47.7—62.2] n.s.
6 - Combi-office (33) 60.6 (60.3) [43.7—75.3] n.s.
7 - Hot-deskingb(116) 52.6 (53.2) [43.6—61.4] .043
Female
(N = 2,349)
1 - Cell-office (1,068) 82.8 (83.0) [80.4—84.9] Ref.
2 - Shared-room (455) 53.2 (53.2) [48.6—57.7] n.s.
3 - Small open plan (276) 46.4 (46.1) [40.6—52.3] n.s.
4 - Medium-sized open plan (126) 44.4 (44.1) [36.1—53.2] n.s.
5 - Large open plan (171) 47.4 (46.8) [40.0—54.8] n.s.
6 - Combi-office (48) 37.5 (37.1) [25.2—51.6] .045
7 - Hot-deskingb(205) 33.2 (33.0) [27.1—39.9] <.0001
Spaces for special meetings
Male
(N = 1,954)
1 - Cell-office (983) 83.3 (83.4) [80.9—85.5] Ref.
2 - Shared-room (354) 68.9 (68.9) [63.9—73.5] n.s.
3 - Small open plan (190) 63.7 (63.7) [56.6—70.2] n.s.
4 - Medium-sized open plan (100) 65.0 (64.5) [55.3—73.6] n.s.
5 - Large open plan (178) 69.7 (69.2) [62.6—75.9] n.s.
6 - Combi-office (33) 60.6 (60.3) [43.7—56.3] n.s.
7 - Hot-deskingb(116) 56.9 (57.6) [47.8—65.5] .019
(continued)
18
Outcome Scale
1-3aOffice design
Adjusted
(N)
% Unadjusted
(adjusted) [95% CI]
Impact of
office designb
Female
(N = 2,345)
1 - Cell-office (1,067) 84.3 (84.5) [81.9—86.3] Ref.
2 - Shared-room (452) 58.9 (58.9) [54.3—63.3] n.s.
3 - Small open plan (276) 57.2 (57.0) [51.3—62.9] n.s.
4 - Medium-sized open plan (126) 64.3 (63.9) [55.6—72.1] n.s.
5 - Large open plan (171) 64.9 (64.3) [57.5—71.7] n.s.
6 - Combi-office (48) 43.8 (43.3) [30.7—57.7] .007
7 - Hot-deskingb(205) 41.5 (41.2) [34.9—48.3] <.0001
Spaces for booked meetings
Male
(N = 1,956)
1 - Cell-office (985) 88.1 (87.3) [86.0—90.0] Ref.
2 - Shared-room (353) 42.2 (42.7) [37.2—47.4] .0002
3 - Small open plan (191) 38.7 (39.2) [32.1—45.8] <.0001
4 - Medium-sized open plan (101) 51.5 (52.2) [41.9—61.0] n.s.
5 - Large open plan (177) 52.0 (52.5) [44.7—59.2] n.s.
6 - Combi-office (33) 51.5 (51.0) [35.2—67.5] n.s.
7 - Hot-deskingb(116) 37.9 (38.7) [29.6—47.0] .0008
Female
(N = 2,350)
1 - Cell-office (1,069) 83.7 (83.6) [81.4—85.8] Ref.
2 - Shared-room (454) 38.3 (38.2) [34.0—42.9] n.s.
3 - Small open plan (277) 29.6 (29.9) [24.5—35.2] .0002
4 - Medium-sized open plan (127) 38.6 (38.7) [30.6—47.3] n.s.
5 - Large open plan (171) 43.9 (44.1) [36.6—51.4] n.s.
6 - Combi-office (48) 22.9 (23.0) [13.3—36.5] .004
7 - Hot-deskingb(204) 28.9 (28.2) [23.1—35.5] .0002
Note. Mean values of scores with 95% CI and impact of office design. Multivariate linear regression analysis. Crude (no adjustment) and adjusted
model presented. Adjusted model in parentheses controlled for education and age. Reference category (cell-office) is marked with underlined text,
significances are marked in bold text. p value adjusted for education and age. CI = confidence interval.
aResponse scales ranged from “yes, sufficiently” (1) to “no, not at all” (3).
bHot-desking is a not an office type, but a subgroup to the office-type flex-office.
Table 3. (continued)
Bodin Danielsson and Theorell 19
(unadjusted as well as adjusted for covariates), unadjusted 95% confidence
intervals and p values corresponding to the contribution of each one of the
office designs to the dependent variables in multiple logistic regression
including both covariates, and all office designs with cell-office as reference
group.
Results show that perception of access to supportive facilities varied sig-
nificantly between employees in different office designs and that women in
most office designs, overall, were less satisfied than were men with access to
supportive facilities.
For individual outcomes, we find, with regard to access to spaces for con-
centrated work, the highest level of satisfaction among employees in cell-
offices (83% for both genders) and the least satisfaction in hot-desking offices
(women = 33%, men = 53%). This pattern repeated itself for the two other
outcomes as well. Accordingly, in cell-offices, employees’ satisfaction with
access to spaces for small spontaneous meetings was at its highest (women =
84%, men = 83%), while the lowest level of satisfaction was found in hot-
desking offices (women = 42%, men = 57%). The corresponding findings for
access to spaces for booked meetings show the highest level of satisfaction in
cell-offices (women = 84%, men = 88%) and the lowest in hot-desking offices
(women = 29%, men = 38%). However, when supervisory position is included
as a covariate in a separate analysis, the results change markedly. (Results
only presented in text.) These results show that men in large open plan offices
differed significantly from those in cell-offices (p < .01), whereas no differ-
ences are observed between shared-room offices and small open plan offices
when compared with cell-offices. Hot-desking offices remain significantly
different from cell-offices (p < .05). Among women, the results are sustained
in small open plan offices, hot-desking offices, and combi-offices, while after
adjustment for supervisory position, shared-room offices differ significantly
(worse) from cell-offices (p < .01).
Correlations Between Contribution of Workspace and Access to
Supportive Facilities
The correlation analyses within each domain revealed strong correlations
between items, especially within the domain contribution that workspace
makes, as expected (see Table 4). The question was, however, how well the
two domains correlated with each other. Results revealed correlations
between .31 and .46, that is, 10% or more of the variance could be explained
by correlations between the two domains. The lowest degree of correlation
was found between the domains contribution that workspace makes and
access to supportive facilities for the item “Access for spaces for booked
20 Environment and Behavior 00(0)
meeting,” while the highest degree of correlation was found for the item
“Access to individual rooms for concentrated work.”
Discussion
Our purpose was to investigate the role that office design may have for
employees’ satisfaction with the contributions that office design makes to job
satisfaction and related outcomes. Such relationships contribute to their well-
being. Moreover, we set out to explore underlying environmental factors that
could play a role for this satisfaction, such as employees’ perception of access
to supportive facilities at the office—that is, back-up rooms of various kinds.
The rationale behind this was that access to such facilities could play a role in
employees’ satisfaction with the contribution that the workspace makes in
different office designs—since this could be related to noise and privacy
issues. This is something that has been found to differ significantly between
various office types (Bodin Danielsson & Bodin, 2009). After all, the inten-
tion of supportive facilities such as back-up rooms is not only to provide
Table 4. Correlations Between the Domains of (a) Satisfaction With the
Contribution That the Workspace Makes and (b) Access to Supportive Facilities.
Outcomes 1 2 3 4 5 6
Workspace contribution
Contribution to
1. Job satisfaction —
2. Pleasantness .86 —
3. To do a good job .83 .79 —
Access to supportive facilities
Access to
1. Individual rooms for
concentrated work
.46 .41 .46
2. Spaces for small meetings .38 .35 .37 .58 —
3. Spaces for booked meetings .31 .31 .31 .45 .71 —
Note. Sample N = 4,372 participants. All correlations are highly significant (p ≤ .001). Degree
of correlation between variables with
=lowest degree of correlation (0.31-0.46).
=middle degree of correlation (0.45-0.71).
=highest degree of correlation (0.79-0.86).
21
Table 5. Graphical Overview of the Differences1 in Satisfaction with the Contribution that the Workspace Makes and Perception of
Access to Supportive Facilities Distributed Between Different Office Designs Split by Gender.
Outcome Cell-office (Ref.) Shared-room Small open-plan Med. sized open-plan Large open-plan Combi-office Hot-deskinga
f=1073-77
m=962-90
f=455
m=357
f=191
m=276
f=191
m=276
f=125-8
m=99-101
f=48
m=33-5
f=48
m=33-5
Contribution to:
Job satisfaction
Men ■ ■ ■ ■ ■ ■ ■
Women ♦-♦-♦ ♦-♦-♦ - ♦-♦-♦
Pleasantness
Men ■ ■ ■ ■
Women ♦♦-♦-♦
To do a good job
Men ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■
Women ♦-♦ ♦♦-♦ ♦-♦-♦
f=1073-77
m=962-90
f=452-55
m=353-54
f=276-77
m=190-91
f=126-27
m=100-1
f=171
m=177-78
f=48
m=33
f=48
m=33
Access to:
Ind. rooms for concent. work
Men ■ ■
Women ♦♦-♦-♦
Spaces for spont. meetings
Men ■
Women ♦-♦ ♦-♦-♦
Spaces for booked meetings
Men ■ ■ ■ ■ ■ ■ ■ ■ ■
Women ♦-♦-♦ ♦-♦ ♦-♦-♦
Note. Synthesis is based on linear multivariate analysis adjusted for age and education. Cell-office was used as reference category, f=female, m=male. a = hot-desking is a not an office type,
but a subgroup to the office type flex-office. 1 = differences in strength of significance between office designs, not in highest and lowest reported value. When no symbol is presented
there are no significant outcomes for the specific variable. 1 marker= p ≤ .05, 2 markers= p ≤ .01, 3 markers= p ≤ .001. □, ■ (squares) = men, ◊, ♦ (diamonds) = women, □, ◊ (white
symbol) =significantly positive value, ■, ♦ (black symbol) = significantly negative value
22 Environment and Behavior 00(0)
privacy for conversations and concentrated work but also to reduce distur-
bances in meetings and collaborative work.
One office design stands out negatively both in terms of satisfaction with
the contribution that workspace makes and with sufficient access to support-
ive facilities—the hot-desking office (see graphic overview in Table 5). This
statistical negative association with the hot-desking office concerns both men
and women, although the outcomes are more negative for women. In addition
to the hot-desking office, there are two other office designs that also stand out
negatively with regard to the two domains investigated—the small open plan
office and the combi-office. There was, however, a clear gender difference
between these two office designs. In small open plan offices, it was exclu-
sively men who were significantly less satisfied than other men in terms of the
contribution that the workspace makes. In contrast, women were significantly
less satisfied with this aspect in combi-offices compared with other women.
Comparing the results of the two domains—contribution that workspace
makes and access to supportive facilities—in the graphic overview in Table
5, we also observe some differences. For satisfaction with the contribution
that workspace makes, there was significant discontent in all office designs
except medium-size open plan offices in comparison with cell-offices. The
degree of dissatisfaction varied both between the office designs and the out-
come variables. The highest dissatisfaction was found among employees in
hot-desking offices for both domains. With regard to access to supportive
facilities, there appears, overall, to have been less dissatisfaction with this
domain than with the contribution that workspace makes. This discontent
was also more unevenly distributed between the different office designs stud-
ied and for two of them—the medium-sized and large open plan office—we
found no significant dissatisfaction in comparison to the cell-office at all.
Beyond indicating a higher dissatisfaction with the contribution that work-
space makes, our results also indicate that the satisfaction with this is a broad
concept that is not easy to grasp. When aiming to measure architectural fea-
tures important for employees’ satisfaction with the contribution that work-
space makes to job satisfaction, to pleasantness, and to doing a good job, we
found that the items used in our study have no absolute association with
employees’ satisfaction with the contribution that workspace makes. Our
results indicate that the items used may not be precise enough to “catch”
dimensions important for the design of a supportive office design. Despite
this lack of precision, in office designs where employees reported poor access
to supportive facilities, that is, in hot-desking offices followed by combi-
offices and small open plan offices, we found that employees tended to report
more dissatisfaction with the contribution that workspace makes. There are,
however, exceptions from this. For example, in shared-room offices where
male and female employees reported low satisfaction with the contribution to
Bodin Danielsson and Theorell 23
job satisfaction and doing a good job, we found that only men reported insuf-
ficient access to spaces for booked meetings.
Research Questions in Relation to Results
Did the results of this exploratory study correspond with the hypotheses pre-
sented? First (RQ1a), we hypothesized that the satisfaction with contribution
that workspace makes would be higher in cell-offices due to less exposure to
environmental disturbance in comparison to office designs where workspaces
and facilities are shared—as a consequence of the established risks of noise
exposure and privacy problems in the latter (Bodin Danielsson & Bodin,
2009; Haans et al., 2007; Jahncke et al., 2011; Kupritz, 1998; Y. Lee & Brand,
2005). In line with RQ1a, our results show more satisfaction with the contri-
bution that workspace makes among employees in cell-offices than among
employees in other office designs. Employees working in medium-sized
open plan designs are content with the contribution that workspace makes in
their offices—at least they are not significantly less satisfied than cell-office
employees are on any of the items used to measure satisfaction with the con-
tribution that workspace makes. Why there is less discontent with contribu-
tion of workspace in this regard in this traditional open plan office than in
small and large open plan offices is not clear. Other studies, however, have
found this office design to stand out positively in some other regards. A recent
study showed, for example, that the manager–employee relationships func-
tioned in a significantly better way in this specific traditional open plan office
than in other office designs (Bodin Danielsson et al., 2013). Our results show
that, of the three open office types, the small open offices stand out as more
negative in terms of both employees’ satisfaction with the contribution that
workspace makes and the perception of access to supporting facilities, but
this problem was almost exclusively found among men.
Employees in hot-desking offices, who report markedly higher levels of
dissatisfaction with the contribution that workspace makes, share workspace
and facilities with colleagues—as do employees in several other office
designs. There are, however, also other defining architectural and functional
features of this office design that in our opinion may play a critical role in
explaining the poor results in comparison with the traditional open plan
offices. Hot-desking office means nonpersonal workstations, but unlike the
flex-office, the main office type, this office design does not compensate the
lack of personal workstations with good access to back-up rooms of various
kinds. Additionally, the hot-desking office is not defined by functional fea-
tures important to personal control such as high decision authority or skill
discretion, job characteristics important to employees’ well-being, and job
satisfaction. (For definitions of office designs see Table 1.)
24 Environment and Behavior 00(0)
The fairly high dissatisfaction with workspace contribution identified in
shared-room offices among both genders may be indirectly related to the
sharing of workspace and facilities as a result of workspace density. The fact
is that this office design is often selected because of limited space at the
office; hence, larger cell-offices or meeting rooms are converted into shared-
room offices. In accordance with this, the fairly high dissatisfaction with
workspace contribution could be due to other problems with ambient factors
such as temperature and lighting in shared-room offices than in other office
designs (Bodin Danielsson & Bodin, 2009).
Second (RQ1b), in contrast to RQ1a, we hypothesized that sharing work-
space and facilities have positive associations with employees’ satisfaction
with workspace contribution as a result of good social cohesion that comes
with working in close proximity to one another. No support for this theory
was found in our results. In contrast, what we found for both genders was less
satisfaction with workspace contribution in office designs where employees
share workspace and facilities than in cell-offices.
Third (RQ1c), we presented an alternative hypothesis suggesting that
satisfaction with workspace contribution varies within office designs
depending on the dimension of workspace contribution. This hypothesis
was supported by our results in that we found less dissatisfaction with
workspace contribution to pleasantness in all office designs than we did for
contribution to job satisfaction and to doing a good job. It could indicate
that it is harder for the office design to contribute to the two latter dimen-
sions—these are determined more by psychosocial and organizational fac-
tors than by workspace design. In addition to this, our results also indicate
that satisfaction with pleasantness is possibly an indicator of satisfaction
with the workplace, overall, as we find high dissatisfaction with pleasant-
ness in hot-desking offices where employees report high dissatisfaction
independently of outcome variable.
To conclude, our exploratory results only partly support the different
hypotheses presented in RQ1a-c regarding satisfaction with workspace con-
tribution. We find more support for our first research question (RQ1a) than
the second (RQ1b). We also find support for the hypothesis presented in
RQ1c as employees’ satisfaction varies both internally between dimensions
related to satisfaction with workspace contribution and between the different
office designs. However, our results show that cell-office employees are,
overall, more satisfied with the workspace contribution of their office design
than are employees in other office designs. An internal comparison between
the three dimensions of workspace contribution shows a higher dissatisfac-
tion with workspace contribution to performance (i.e., doing a good job) than
with the other two. The least dissatisfaction was found with the contribution
that workspace makes to pleasantness.
Bodin Danielsson and Theorell 25
Our second research question (RQ2) investigated the possible association
between perception of access to supportive facilities and employees’ satisfac-
tion with workspace contribution. Our results support to a great extent our
hypothesis, in that office designs in which employees reported the most dis-
satisfaction with access to supportive facilities were also where the most dis-
satisfaction with workspace contribution was found—as clearly presented in
the graphics of Table 5. Only in two cases—in shared-room offices and in
large open plan offices—was there no correspondence between high dissatis-
faction with workspace contribution and insufficient access to supportive
facilities in the office. That there is an association between satisfaction with
workspace contribution and access to supportive facilities also finds support
in our correlations analysis, which shows that 10% or more of the variance is
explained by such associations (see Table 4). We observed that correlations
between items used to assess access to supportive facilities varied, which indi-
cates that some of the supportive facilities measured in our study were more
important to employees’ satisfaction with workspace contribution than were
others.
Our final research question (RQ3) investigated possible gender differ-
ences in the office design’s association with employees’ satisfaction with
workspace contribution and perception of access to supportive facilities. The
results of our study show gender differences both with regard to how satisfied
men and women are with the contribution that workspace makes and with
regard to perceived access to supportive facilities in the different office
designs studied. According to our results, small open plan offices are more
negative for men than for women. Why this would be the case is not known.
A speculation is that it may be related to the stronger social control and group
identity developed in a smaller group, something possibly less positive for
men than for women since it has been found that women rely more on social
relations with colleagues than do men (Plaisier et al., 2007; Winter et al.,
2006). It is harder for subgroups to develop in smaller groups with fewer than
12 people (Hare, 1981), which could buffer negative effects of social control
in smaller groups. This may also add to the potentially negative relationship
to the smaller group size in this office design.
In contrast, among women, we find that another office design—the combi-
office—is more negative than other office designs. Why the combi-office
stands out as being less beneficial for women than for men in terms of satis-
faction with the contribution that workspace makes and access to supportive
facilities is not clear. A speculation is that it could be the result of the focus
on teamwork in this office design. Teamwork depends on interpersonal rela-
tionships that hold an inherent risk of conflicts. An increased risk for con-
flicts in teamwork has also been shown (Gladstein, 1984; Jehn, 1995).
Women are possibly more sensitive to conflicts than are men as a result of the
26 Environment and Behavior 00(0)
greater importance women put into social relationships at work, as previously
described (Plaisier et al., 2007; Winter et al., 2006). This may explain the
worse conditions reported among women than men in combi-offices in our
study.
Limitations and Strengths
The major strength of this study is its large and nationally representative
study base. At the same time, a large survey covering various aspects such as
work environment, health, and general life situation like ours holds certain
limitations such as limited space to address each topic, for example, office
environment. Hence, neither all aspects nor details of the topic can be cov-
ered, which may lead to misunderstandings of the purpose of the questions
on, for example, office environments. It may explain why (a) critical items
for characterization of the office design were sometimes not filled in, and (b)
the responses were not always consistent. This, in turn, could affect the clas-
sification of exposure and result in erroneous estimations of the effect of
office design on outcomes studied.
Additionally, as in most large surveys, we only use self-reported data. This
means that common method bias may arise (Podsakoff, MacKenzie, Lee, &
Podsakoff, 2003), although this problem should not be overestimated
(Spector, 2006). Also, our study is cross-sectional, that is, participants were
not studied over time. This is a limitation since causal interpretations cannot
be made. However, this limitation primarily concerns outcomes where a
long-term effect can build up over time, for example, health outcomes, which
are not included in the present study.
The choice of covariates can always be discussed. Age was chosen because
of its importance for attitudes and satisfaction with the work situation, which
in turn may influence satisfaction with the office. Education is an important
covariate in and of itself that also influences job rank, health, and socioeco-
nomic status. Since all analyses were sex-stratified, gender was not used as a
covariate. Initial analyses showed no impact of job rank nor labor market
sector; thus, these were not included as covariates.
Conclusion
This exploratory study suggests that choice of office design, defined by spe-
cific architectural and functional features, is related to employees’ satisfaction
with the contribution that the workspace makes to job satisfaction, comfort,
and doing a good job. The results suggest that employees’ satisfaction with
contribution of workspace is highly related to employees’ perception of access
to supportive facilities, an architectural feature that differentiates studies of
Bodin Danielsson and Theorell 27
office designs. The picture is consistent from one point of view: Out of seven
studied office designs, we find one office design standing out as markedly bet-
ter—cell-office—and one as markedly worse than the others—hot-desking
office. For all studied outcomes, cell-office employees are significantly more
satisfied than are other employees. The negative influence of hot-desking con-
cerns both men and women, although the influence is stronger on women with
regard to perception of supportive facilities. Other interesting gender differ-
ences are revealed in the study. We find that men, overall, report less satisfac-
tion than do women with office designs in which employees share open
workspaces and facilities. These results may be interpreted to mean that shar-
ing workspace and facilities is less positive for men than for women, while
teamwork, instead, is more critical to women due to the inherent risk of con-
flicts that teamwork holds. This may possibly be due to gender differences in
interpersonal strategies, which in turn are associated with the defining features
of office designs in which the employees work. The very negative perception
of hot-desking offices is, however, almost gender equal, indicating that there
are some fundamental problems with this office design from an employee
perspective. We believe the problems may be related to the defining architec-
tural and functional feature of hot-desking offices that does not support the
employee’s personal control—physical, psychological, or organizational.
Since this is an exploratory study, it should only be viewed a first step in inves-
tigating what role different architectural features in office design could play
for factors important for office employees’ welfare—in terms of job satisfac-
tion, satisfaction with pleasantness in the work environment, and so on. Our
starting point was to investigate office employees’ satisfaction with contribu-
tions that workspace makes in different office designs and what role access to
some supportive facilities within the office may have for this. The next step is
a more precise study design that goes more into detail of studied outcomes and
other dimensions possibly important for these. One should also study how
these, in turn, may be related to organizational and psychosocial outcomes
important to employees’ health and well-being.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research,
authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research,
authorship, and/or publication of this article: This research was supported by Formas,
the Swedish Research Council for Sustainable Development (Young Mobility Grant
259-2011-1580).
28 Environment and Behavior 00(0)
References
Addis, M. E., & Mahalik, J. R. (2003). Men, masculinity, and the contexts of help-
seeking. American Psychologist, 58, 5-14.
Aiello, J. R., Epstein, Y. M., & Karlin, R. A. (1975). Effects of crowding on electro-
dermal activity. Sociological Symposium, 14, 42-57.
Allebeck, P., & Mastekaasa, A. (2004). Chapter 5. Risk factors for sick leave—gen-
eral studies. Scandinavian Journal of Public Health, 32(Suppl. 63), 49-108.
doi:10.1080/14034950410021853
Allen, T. J. (1997, September). Architecture and communication among product devel-
opment engineers (Working Paper No. 165-97, Sloan Working Paper No. 3983).
Cambridge. MA: The International Center for Research on the Management of
Technology, Massachusetts Institute of Technology (MIT).
Anderson, C. M., & Martin, M. M. (1995). Why employees speak to coworkers and
bosses: Motives, gender, and organizational Satisfaction. Journal of Business
Communication, 32, 249-265.
Becker, F., & Sims, W. (2001). Offices that work balancing communication, flex-
ibility and cost. Ithaka, NY: Cornell University (International Workplace Studies
Program).
Bodin Danielsson, C. (2008). Office experiences. In H. Schifferstein & P. Hekkert
(Eds.), Product experience (pp. 605-628). San Diego, CA: Elsevier.
Bodin Danielsson, C., & Bodin, L. (2008). Office-type in relation to health, well-
being and job satisfaction among employees. Environment and Behavior, 40,
636-668. doi:10.1177/0013916507307459
Bodin Danielsson, C., & Bodin, L. (2009). Differences in satisfaction with office
environment among employees in different office types. Journal of Architectural
and Planning Research, 26, 241-257. Retrieved from http://www.jstor.org/sta-
ble/43030872
Bodin Danielsson, C., Bodin, L., Wulff, C., & Theorell, T. (2015). The relation between
office type and workplace conflict: A gender and noise perspective. Journal of
Environmental Psychology, 42, 161-171. doi:10.1016/j.jenvp.2015.04.004
Bodin Danielsson, C., Chungkham, H. S., Wulff, C., & Westerlund, H. (2014). Office
design’s impact on sick leave rates. Ergonomics, 57, 139-147. doi:10.1080/0014
0139.2013.871064
Bodin Danielsson, C., Wulff, C., & Westerlund, H. (2013). Is perception of leadership
influenced by office environment? Journal of Corporate Real Estate, 15, 194-
212. doi:10.1108/JCRE-03-2013-0008
Brunia, S., De Been, I., & van der Voordt, T. J. M. (2016). Accommodating new ways
of working: Lessons from best practices and worst cases. Journal of Corporate
Real Estate, 18, 30-47. doi:10.1108/JCRE-10-2015-0028
Conrath, C. W. (1973). Communication patterns, organizational structure, and man: Some
relationships. Human Factors, 15, 459-470. doi:10.1177/001872087301500503
Crouch, A., & Nimran, U. (1989). Office design and the behavior of senior managers.
Human Relations, 42, 139-155. doi:10.1177/001872678904200203
Bodin Danielsson and Theorell 29
d’Astous, A. (2000). Irritating aspects of the shopping environment. Journal of
Business Research, 49, 149-156. doi:10.1016/S0148-2963(99)00002-8
De Raeve, L., Jansen, N. W. H., van den Brandt, P. A., Vasse, R. M., & Kant, I. J.
(2008). Risk factors for interpersonal conflicts at work. Scandinavian Journal of
Work, Environment & Health, 34, 96-106. doi:10.5271/sjweh.1223
Duffy, F. (1997). The new office (2nd ed.). London, England: Conran Octopus.
Evans, G. (1979). Behavioral and physiological consequences of crowding in humans.
Journal of Applied Social Psychology, 9, 27-46.
Evans, G., & Johnson, D. (2000). Stress and open-office noise. Journal of Applied
Psychology, 85, 779-783. doi:10.1037/0021-9010.85.5.779
Frese, M. (1999). Social support as a moderator of the relationship between work
stressors and psychological dysfunctioning: A longitudinal study with objec-
tive measures. Journal of Occupational Health Psychology, 4, 179-192.
doi:10.1037/1076-8998.4.3.179
Fried, Y., Melamed, S., & Ben-David, H. (2002). The joint effects of noise, job
complexity, and gender on employee sickness absence: An exploratory study
across 21 organizations—The CORDIS study. Journal of Occupational and
Organizational Psychology, 75, 131-144. doi:10.1348/09631790260098181
Fusilier, M. R., Ganster, D., & Mayes, B. T. (1986). The social support and health
relationships: Is there a gender difference? Journal of Occupational Psychology,
59, 145-153. doi:10.1111/j.2044-8325.1986.tb00220.x
Gabriel, S., & Gardner, W. L. (1999). Are there “his” and “hers” types of interde-
pendence? The implications of gender differences in collective versus relational
interdependence for affect, behavior, and cognition. Journal of Personality and
Social Psychology, 77, 642-655.
Gerdenitsch, C., Korunka, C., & Hertel, G. (2017). Need–supply fit in an activity-
based flexible office: A longitudinal study during relocation. Environment and
Behavior. Advance online publication, 1-26. doi:10.1177/0013916517697766
Gladstein, D. L. (1984). Groups in context: A model of task group effectiveness.
Administrative Science Quarterly, 29, 499-517. doi:10.2307/2392936
Haans, A., Kaiser, F. G., & de Kort, Y. (2007). Privacy needs in office environments.
European Psychologist, 12, 93-102. doi:10.1027/1016-9040.12.2.93
Hanson, L. L. M., Åkerstedt, T., Näswall, K., Leineweber, C., Theorell, T., &
Westerlund, H. (2011). Cross-lagged relationships between workplace demands,
control, support, and sleep problems. Sleep, 34, 1403-1410. doi:10.5665/
SLEEP.1288
Hare, A. P. (1981). Group size. The American Behavioral Scientist, 24, 695-708.
doi:10.1177/000276428102400507
Hedge, A. (1982). The open plan office: A systematic investigation of employee
reactions to their work environment. Environment and Behavior, 14, 519-542.
doi:10.1177/0013916582145002
Hua, Y., Loftness, J. H., Heerwagen, J., & Powell, K. M. (2011). Relationship
between workplace spatial settings and occupant-perceived support for collabo-
ration. Environment and Behavior, 43, 807-826.
30 Environment and Behavior 00(0)
Jahncke, H., Hygge, S., Halin, N., Green, A. M., & Dimberg, K. (2011). Open-plan
office noise: Cognitive performance and restoration. Journal of Environmental
Psychology, 31, 373-382. doi:10.1016/j.jenvp.2011.07.002
Jehn, K. A. (1995). A multimethod examination of the benefits and detriments of intra-
group conflict. Administrative Science Quarterly, 40, 256-282. doi:10.2307/2393638
Kaarlela-Tuomaala, E., Heleniusa, R., Keskinen, E., & Hongisto, V. (2009).
Effects of acoustic environment on work in private office rooms and open plan
offices—Longitudinal study during relocation. Ergonomics, 52, 1423-1444.
doi:10.1080/00140130903154579
Karasek, R., & Theorell, T. (1990). Healthy work: Stress, productivity, and the recon-
struction of working life. New York, NY: Basic Books.
Kivimäki, M., Head, J., Ferrie, J. E., Shipley, M. J., Vahtera, J., & Marmot, M. G.
(2003). Sickness absence as a global measure of health: Evidence from mortal-
ity in the Whitehall II prospective cohort study. British Journal of Management,
327(7411), Article 364. doi:10.1136/bmj.327.7411.364
Kivimäki, M., Vahtera, J., Thompson, L., Griffiths, A., Cox, T., & Pentti, J. (1997).
Psychosocial factors predicting employee sickness absence during economic
decline. Journal of Applied Psychology, 82, 858-872. doi:10.1037/0021-
9010.82.6.858
Krantz, G. (2003, October). Gender differences in health -a response to stress? Paper
presented at the Stress Research from Biology to Society (Doctoral Course)
Department of Psychology, Stockholm University.
Krantz, G., Berntsson, L., & Lundberg, U. (2005). Total workload, work stress
and perceived symptoms in Swedish male and female white-collar employees.
European Journal of Public Health, 15, 209-214. doi:10.1093/eurpub/cki079
Kupritz, V. W. (1998). Privacy in the work place: The impact of building design.
Journal of Environmental Psychology, 18, 341-356. doi:10.1006/jevp.1998.0081
Larsson, M. (2012). Lönerapport år 2012: Löner och löneutveckling år 2001-2011
efter klass och kön [Report on salaries year 2012: Salaries and development of
salaries year 2001-2011 stratified on class and gender]: LO (Landsorganisationen
i Sverige Arbetslivsenheten).
Lee, P. J., Lee, B. K., Jeon, J. Y., Zhang, M., & Kang, J. (2016). Impact of noise on
self-rated job satisfaction and health in open-plan offices: A structural equation
modelling approach. Ergonomics, 59, 222-234. doi:10.1080/00140139.2015.10
66877
Lee, Y., & Brand, J. L. (2005). Effects of control over office workspace on percep-
tions of the work environment and work outcomes. Journal of Environmental
Psychology, 25, 323-333. doi:10.1016/j.jenvp.2005.08.001
Magnusson Hanson, L., Theorell, T., Bech, P., Reugulies, R., Hyde, M., Oxenstierna,
G., & Westerlund, H. (2009). Psychosocial working conditions and depressive
symptoms among Swedish employees. International Archives of Occupational
and Environmental Health, 82, 951-960. doi:10.1007/s00420-009-0406-9
Meijer, E. M., Frings-Dresen, M. H., & Sluiter, J. (2009). Effects of office inno-
vation on office workers’ health and performance. Ergonomics, 52, 1027-1038.
doi:10.1080/00140130902842752
Bodin Danielsson and Theorell 31
Mok, D., & Wellman, B. (2007). Did distance matter before the Internet? Interpersonal
contact and support in the 1970s. Social Networks, 29, 430-461. doi:10.1016/j.
socnet.2007.01.009
Netterstrøm, B., Conrad, N., Bech, P., Fink, P., Olsen, O., Rugulies, R., & Stansfeld,
S. (2008). The relation between work-related psychosocial factors and the
development of depression. Epidemiologic Reviews, 30, 118-132. doi:10.1093/
epirev/mxn004
Niedhammer, I., Bugel, I., Goldberg, M., Leclerc, A., & Guéguen, A. (1998).
Psychosocial factors at work and sickness absence in the Gazel cohort: A pro-
spective study. Occupational and Environmental Medicine, 55, 735-741.
doi:10.1136/oem.55.11.735
Nyberg, A., Westerlund, H., Magnusson Hanson, L., & Theorell, T. (2008).
Managerial leadership is associated with self-reported sickness absence and sick-
ness presenteeism among Swedish men and women. Scandinavian Journal of
Public Health, 36, 803-811. doi:10.1177/1403494808093329
Oldham, G. R. (1988). Effects of changes in workspace partitions and spatial density
on employees reactions: A quasi-experiment. Journal of Applied Psychology, 73,
253-258.
Pejtersen, J. H., Feveile, H., Christensen, K. B., & Burr, H. (2011). Sickness absence
associated with shared and open-plan offices—A national cross sectional ques-
tionnaire survey. Scandinavian Journal of Work, Environment and Health, 37,
376-382. doi:10.5271/sjweh.3167
Plaisier, I., de Bruijn, J. G. M., de Graaf, R., ten Have, M., Beekman, A., & Penninx,
B. W. J. H. (2007). The contribution of working conditions and social support to
the onset of depressive and anxiety disorders among male and female employees.
Social Science & Medicine, 64, 401-410. doi:10.1016/j.socscimed.2006.09.008
Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003).
Common method biases in behavioral research: A critical review of the litera-
ture and recommended remedies. Journal of Applied Psychology, 88, 879-903.
doi:10.1037/0021-9010.88.5.879
Spector, P. E. (2006). Method variance in organizational research: Truth or urban legend?
Organizational Research Methods, 9, 221-232. doi:10.1177/1094428105284955
Stansfeld, S., Head, J., & Marmot, M. (2000). Work-related factors and ill-health:
The Whitehall II study (Contract Research Report 266/2000). Sudbury, UK:
Health and Safety Executive.
Sundstrom, E. (1986). Work places: The psychology of the physical environment in
offices and factories. New York, NY: Cambridge University Press.
Sundstrom, E., Town, J. P., Rice, R. W., Osborn, D., & Brill, M. (1994). Office noise,
satisfaction, and performance. Environment and Behavior, 26, 195-222.
Szilagyi, A., & Holland, W. (1980). Changes in social density: Relationships with func-
tional interaction and perceptions of job characteristics, role stress, and work satis-
faction. Journal of Applied Psychology, 65, 28-33. doi:10.1037/0021-9010.65.1.28
Tett, R. P., & Meyer, J. P. (1993). Job satisfaction, organizational commitment,
turnover intention and turnover: Path analyses based on meta-analytic findings.
Personnel Psychology, 46, 259-293. doi:10.1111/j.1744-6570.1993.tb00874.x
32 Environment and Behavior 00(0)
Theorell, T. (2000). Working conditions and health. In L. F. Berkman & I. Kawachi
(Eds.), Social epidemiology (pp. 95-117). New York, NY: Oxford University
Press.
Vahtera, J., Pentti, J., & Uutela, A. (1996). The effect of objective job demands on reg-
istered sickness absence spells; do personal, social and job-related resources act
as moderators? Work & Stress, 10, 286-308. doi:10.1080/02678379608256809
Winter, T., Roos, E., Rahkonen, O., Martikainen, P., & Lahelma, E. (2006). Work
family conflicts and self-rated health among middle aged municipal employ-
ees in Finland. International Journal of Behavioral Medicine, 13, 276-285.
doi:10.1207/s15327558ijbm1304_2
Author Biographies
Christina Bodin Danielsson is an associate professor at the Royal Institute of
Technology (KTH), Stockholm. She holds a doctoral degree in office design and
works both in design practice and in research. She works as a practicing architect
specialized in human–environment interaction and office design at Brunnberg &
Forshed Architects and as a researcher at the School of Architecture, KTH. She is also
affiliated to the Stress Research Institute, Stockholm University. Her research inves-
tigates the office environment’s influence on employees and organizations, aiming to
find environmental factors important for the welfare for both parties. She applies both
quantitative and qualitative methods.
Töres Theorell, medical doctor, is a professor emeritus at the Karolinska Institute
and the Stress Research Institute, Stockholm University. With a background in clini-
cal medicine, social medicine, and stress research, he was the director of the National
Institute for Psychosocial Factors and Health in the years 1995 to 2006. Most of his
research deals with stress mechanisms. He has, for example, developed the Demand-
Control Model together with Robert Karasek.