Publisher’s version / Version de l'éditeur:
Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.
Questions? Contact the NRC Publications Archive team at
PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.
https://publications-cnrc.canada.ca/fra/droits
L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.
Internal Report (National Research Council of Canada. Institute for Research in Construction), 2002-04-01
READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE. https://nrc-publications.canada.ca/eng/copyright
NRC Publications Archive Record / Notice des Archives des publications du CNRC :
https://nrc-publications.canada.ca/eng/view/object/?id=4657e77a-7084-482a-b02e-16e9c21f0fce https://publications-cnrc.canada.ca/fra/voir/objet/?id=4657e77a-7084-482a-b02e-16e9c21f0fce
Archives des publications du CNRC
For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.
https://doi.org/10.4224/20378854
Access and use of this website and the material on it are subject to the Terms and Conditions set forth at
Environmental satisfaction in open-plan environment: 2. Effect of workstation size, partition height and windows
Partition Height and Windows
Charles, K.E.; Veitch, J.A.
www.nrc.ca/irc/ircpubs IRC-IR-845
Effects of Workstation Size, Partition Height and Windows
by Kate E. Charles, PhD. and Jennifer A. Veitch, PhD.
Internal Report No. IRC-IR-845
Date of issue: April 2002
This internal report, while not intended for general distribution, may be cited or referenced in other publications.
Environmental Satisfaction in Open-Plan Environments: 2. Effects
of Workstation Size, Partition Height and Windows
Kate E. Charles, PhD Jennifer A. Veitch, PhD
Institute for Research in Construction National Research Council Canada
Montreal Road, Ottawa, Ontario CANADA, K1A 0R6
Internal Report No. IRC-IR-845 April 2002
Acknowledgements
This investigation forms part of the Field Study sub-task for the NRC/IRC project Cost-effective Open-Plan Environments (COPE) (NRCC Project # B3205), supported by Public Works and Government Services Canada, Natural Resources Canada, the Building Technology Transfer Forum, Ontario Realty Corp, British Columbia Buildings Corp, USG Corp, and Steelcase, Inc. COPE is a multi-disciplinary project directed towards the development of a decision tool for the design, furnishing, and operation of open-plan offices that are satisfactory to occupants, energy-efficient, and cost-effective.
The authors are grateful to the following individuals: Chantal Arsenault, Emily Nichols, Marcel Brouzes, Roger Marchand, John Bradley and Scott Norcross (data
collection); Louise Legault (research design advice); Kelly Farley (data analysis); Gordon Bazana (data management); Guy Newsham (project management, research design advice and data collection).
c. 2002 Her Majesty in Right of Canada. National Research Council Canada, Ottawa, Ontario
Environmental Satisfaction in Open-Plan Environments:
2. Effects of Workstation Area, Partition Height and Windows
Executive Summary
Open-plan office designs are characterised by modular furniture and moveable partitions that partially screen office occupants from co-workers occupying the same office space. Such designs are argued to provide a flexible working environment, to offer space and cost savings and to promote communication between office occupants.
However, research suggests that open-plan office occupants may experience a lack of both visual and acoustical privacy and an increase in the amount of unwanted distractions and interruptions. In addition, the proposed benefits regarding improved communication are often not realised. Furthermore, open-plan occupants sometimes experience
unfavourable ambient conditions, partly because of the lack of control resulting from a shared office space.
Given that open-plan office designs continue to be popular, it is important to determine the design characteristics of workstations that may influence office
satisfaction, in order to design more effective open-plan environments. Previous research has highlighted the importance of workstation enclosure and occupant density,
particularly in relation to open-plan office occupants’ satisfaction with privacy. In addition, it has been suggested that access to a window at work is beneficial to occupants.
As part of the COPE Time1 field study, we investigated the impact of three workstation characteristics (workstation area, minimum partition height and windows) on four environmental satisfaction measures (satisfaction with privacy, ventilation and lighting, and overall environmental satisfaction). Data were collected from 419 office workstations and their occupants, in three Canadian Federal Government buildings. Participants, 49% male and 51% female, ranged in age from 18 to 70 years (mean=38.6 years), and were employed in administrative, technical, professional and managerial job roles. Workstation area, minimum partition height and the presence of windows were measured on site by researchers. Occupant satisfaction ratings were collected, at the same time as physical measures, by self-report questionnaire administered using a palm-top computer.
We used hierarchical regression analyses to test the relationships between the three workstation characteristics and each of the four environmental satisfaction measures. After controlling for age, gender, job category and building, the following significant relationships were found between workstation characteristics and occupant satisfaction. Workstation area was found to positively predict satisfaction with privacy. Window was found to be a significant positive predictor of satisfaction with lighting and also negatively predicted satisfaction with ventilation. Finally, minimum partition height was negatively related to overall environmental satisfaction. Overall, the four regression models were able to explain significant, although small, amounts of variance in the four environmental satisfaction measures.
The results of the study indicated that occupants in larger workstations were more satisfied with privacy. This is probably because of the increased distance from co-workers provided by larger workstations and the increased amount of space provided within the workstation. Those occupants whose workstation contained a window were more satisfied with lighting but less satisfied with ventilation. The windows in the study sample were sealed; therefore, this latter finding is probably related to temperature variations arising from increased draught during the winter and increased heat gain during the summer. Finally, we found that those occupants with a lower minimum partition height were more satisfied with overall environmental satisfaction. The mechanisms behind this finding are less obvious, although they could be related to
improved ambient conditions and perceived space arising from workstations containing at least one lower height partition.
The relatively small amounts of variance accounted for by the regression models might be explained by the largely positive workstation characteristics found in the current study sample. That is, the average workstation area was 108.53 ft2 and the average minimum partition height was 61.63 inches; conditions which might have been too
favourable to influence occupant satisfaction greatly. In addition, alternative measures of workstation characteristics, such as average partition height, distance from co-workers or location of workstations on the floor plate, might better explain differences in occupant satisfaction.
Furthermore, the relationships between workstation characteristics and occupant satisfaction could be influenced by personal and organisational factors, such as past experience and job complexity. Future work, controlling for the influence of such factors, might help to clarify the relationships between workstation characteristics and occupant environmental satisfaction.
The relationship between workstation characteristics and occupant environmental satisfaction might also interact with the relative importance occupants place on
environmental features. As an additional analysis in the current study, we also investigated whether the three workstation characteristics influenced occupants’
importance rankings of seven environmental features. Of the twenty-one non-parametric tests conducted, only two were statistically significant. These results indicated that occupants in larger workstations and occupants in windowed workstations tended to rate access to a window as more important than did occupants in smaller workstations and windowless workstations respectively. More work is needed to determine the role of occupants’ importance rankings, particularly in relating these ratings to occupant satisfaction.
Finally, little research has examined the potential mediating effect of physical ambient conditions, such as illuminance, temperature and noise levels, on the relationship between workstation characteristics and occupant satisfaction. The current research project included the objective measurement of such ambient conditions and the investigation of these relationships is proposed as the next stage of our analyses.
Table of Contents
Acknowledgements 2 Executive Summary 3 1. Introduction 7 2. Method 10 2.1 Participants 10 2.2 Procedure 11 2.3 Independent variables 11 2.4 Outcomes 12 3. Results 13 3.1 Descriptive statistics 133.2 Predicting occupant satisfaction 19
3.3 Additional analyses: Workstation characteristics and ranked
importance of environmental aspects 23
4. Discussion 26
5. Conclusions 29
6. References 30
Appendix A: Frequency and percentage counts for workstation characteristics 33 and ranked importance of environmental aspects
Figures and Tables
Table 1: Means, standard deviations and correlations 18
Table 2: Hierarchical regression analyses on satisfaction with privacy 20 Table 3: Hierarchical regression analyses on satisfaction with ventilation 21 Table 4: Hierarchical regression analyses on satisfaction with lighting 22 Table 5: Hierarchical regression analyses on overall environmental satisfaction 22 Table 6: Frequency and percentage counts for workstation area and ranked
importance of window access 24
Table 7: Frequency and percentage counts for window and ranked
importance of window access 25
Figure 1: Frequency distribution for workstation area 14
Figure 2: Frequency distribution for minimum partition height 14
Figure 3: Frequency distribution for window 14
Figure 4: Frequency distribution for satisfaction with ventilation 15 Figure 5: Frequency distribution for satisfaction with privacy 15 Figure 6: Frequency distribution for satisfaction with lighting 16 Figure 7: Frequency distribution for overall environmental satisfaction 16 Figure 8: The relationship between workstation area and ranked importance
of window access 24
Figure 9: The relationship between window and ranked importance of
Environmental Satisfaction in Open-Plan Environments:
2. Effects of Workstation Area, Partition Height and Windows
1. Introduction
Over the lifespan of a typical office building, it is estimated that 82% of all costs are those associated with office employees (mainly salaries and benefits), with the remaining cost being for the construction, maintenance and operation of the building (Brill, Weidemann & BOSTI Associates, 2001). It is also estimated that almost half of the US population is employed in office buildings (Giuliano, 1982; Christie, 1985) and similar estimations have been made in relation to employees from industrialised nations in general (Bloom, 1986). As “every day, people have to work in a physical environment that affects their ability and desire to work” (Goodrich, 1982; p.355), and given the proportionally small cost of changing the physical office space, it makes considerable sense to ensure that offices are designed to facilitate the comfort and satisfaction of office occupants.
The concept of open-plan office design evolved from burolandschaft, more commonly known as office landscaping, a design movement that developed in Germany in the 1960s (Hedge, 1986; Burgess, Lai, Eisner & Taylor, 1989). Although varying in form, this type of office design is characterised by modular furniture and moveable partitions which partially screen office occupants from co-workers occupying the same office space. This is in contrast to the conventional office design in which full height internal walls and doors provide separate, private office spaces.
The proposed benefits of the open-plan office have been summarised by
numerous researchers (eg. Oldham & Brass, 1979; Sundstrom, Herbert and Brown, 1982; Hedge, 1982; 1986; Cangelosi & Lemoine, 1988; Burgess et al, 1989; Jackson, Klein & Wogalter, 1997). In brief, the open-plan office is argued to be beneficial in providing flexibility which organisations can utilise in response to changes in organisational size and structure. Such designs also allow higher occupant density, thereby providing space and cost savings. Finally, advocates of the open-plan office claim that these designs enhance communication between occupants, which in turn promotes morale and
organisational effectiveness. These potential benefits are still persuasive in the modern office context and are reflected in the continued popularity of open-plan office designs.
However, despite enthusiastic support for open-plan offices, research suggests some problems associated with this type of office design. The most commonly stated disadvantage of open-plan offices is their lack of visual and acoustical privacy, coupled with an increase in distractions and interruptions (Marans & Speckelmeyer, 1982; Hedge, 1982; 1986; Cangelosi & Lemoine, 1988; Burgess et al, 1989). Mital, McGothlin & Faard (1992), for example, found that employees working in an open-plan computer office often reported being annoyed by distracting sounds such as conversation, computer and printer beeping, the arrival and departure of other people, keyboard typing and
A number of studies have demonstrated a lack of privacy in open-plan as
compared to conventional enclosed offices (eg. Brookes, 1972; 1978; Brookes & Kaplan, 1972; Sundstrom, Burt & Kamp, 1980; Hedge, 1982; Sundstrom, Herbert et al, 1982; Zalesny & Farace, 1987; Block & Stokes, 1989). Sundstrom, Herbert et al (1982) for example, found a decrease in satisfaction with privacy after employees moved from closed to open offices. Similarly, Carlopio & Gardner (1992) found that employees working in enclosed offices reported more communication privacy than did those working in either open-plan or completely open office environments. Boyce (1974) noted that lack of privacy remained a major complaint one year after moving from closed to open offices.
Acoustical privacy appears to be a particular problem for open-plan offices, and Sundstrom, Town, Rice, Osborn & Brill (1994), for example, found that phones ringing, face-to-face conversations and phone conversations were the most distracting noises for open-plan office occupants. Noise may be a more intrusive factor than other ambient conditions (Jackson et al, 1997) and speech in particular has been shown to be distracting, especially when employees are engaged in complex processing tasks (eg. Young & Berry, 1979; Jackson et al, 1997).
Privacy and distractions are important considerations as they affect employees’ ability to concentrate on quiet, focussed work, an activity in which office occupants spend an estimated 48-64% of their time at work (Brill et al, 2001). Speech and office noise can disrupt performance on some tasks (Banbury & Berry, 1998; Jackson et al, 1997) and perceived privacy and distractions have also been shown to be related to
environmental and job satisfaction (eg. Sundstrom et al, 1980; Sundstrom, 1986; Block & Stokes, 1989; Sundstrom et al, 1994)1.
In addition to privacy considerations, research suggests that the proposed communication benefits arising from open-plan office designs are not always realised. One of the purposes behind open-plan environments is that “…the layout is planned so that people who work together are near each other for efficient communication” (Sundstrom, Herbert et al, 1982; p.380). Researchers have also argued that increased opportunities for communication facilitate the formation of social relationships, which in turn affect employee morale and satisfaction (see Oldham & Brass, 1979). However, although Allen & Gerstberger (1973) reported greater ease of communication after employees moved from closed to open office environments, other research has suggested that open-plan offices negatively influence communication (eg. Oldham & Brass, 1979). In Becker, Gield, Gaylin & Sayer’s (1983) study, for example, faculty staff and students at a community college reported difficulties in interacting with each other effectively when staff members were based in open-plan as compared to closed offices.
Some research has also suggested that employees may experience less favourable ambient conditions in open-plan offices. Open-plan office occupants in Hedge’s (1982) study reported dissatisfaction with ambient conditions, particularly those relating to air quality and thermal comfort. In a telephone survey conducted by Woods, Drewry & Morey (1987), employees working in open-plan offices were 1.5 times more likely to report poor air quality and to believe this negatively affected their productivity as
1
The literature on acoustical privacy, attention and distraction is currently being reviewed in detail as a separate COPE subtask.
compared to employees in closed offices. Such problems are likely to be related, at least in part, to employees’ inability to control ambient conditions and to the variability of conditions within the open-plan work space. Furthermore, Hedge (1982) provides anecdotal evidence that, because of a lack of storage space in open-plan cubicles, occupants stored items on ventilation diffusers, thereby restricting airflow.
However, despite the potential problems that have been associated with open-plan offices, it seems likely that this type of office design will remain popular in future years. Thus, it is important to determine whether there are design characteristics that may minimise the potential detriments to occupant satisfaction. Previous research has suggested that workstation density and enclosure are two factors that could influence occupant reactions to open-plan office environments.
Workstation density can be measured in terms of the number of occupants sharing an open-plan office, the distance from one occupant to another, the area of each
workstation or the area of the floorplate per person. Although Szilagyi & Holland (1980) found that increased density lead to improved friendship opportunities, information exchange and work satisfaction, the majority of research concerning density has indicated adverse occupant reactions as density increases (eg. Dean, Pugh & Gunderson, 1975; Sundstrom, Town, Brown, Forman & McGee, 1982; Marans & Speckelmeyer, 1982; Oldham & Rotchford, 1983; Oldham & Fried, 1987; Oldham, 1988). Increased workstation area, for example, has been associated with increased environmental satisfaction (Oldham & Rotchford, 1983; O’Neill & Carayon, 1993; O’Neill, 1994) and with perceived distractions (O’Neill, 1994). Similarly, Oldham (1988) found that employees moving from an open office to a similar, but lower density, open office experienced greater task privacy, communication privacy and environmental satisfaction, and reduced perceived crowding. In addition, occupants working in more crowded work areas were more likely to be dissatisfied with air quality (Woods et al, 1987).
Workstation enclosure can be measured by the number of partitions surrounding a workstation or the height of those partitions. In general, studies show that the number of enclosed sides is positively related to occupant perceptions of privacy and environmental and job satisfaction (eg. Desor, 1972; Sundstrom et al, 1980; Sundstrom, Town et al, 1982; Oldham & Rotchford, 1983; Oldham & Fried, 1987). However, although studies comparing offices with full height walls, partitions and no partitions indicate greater privacy, communication and satisfaction as enclosure height increases (eg. Sundstrom, Herbert et al, 1982; Oldham, 1988), few studies have compared the more subtle effects of open-plan office partitions of different heights (O’Neill, 1994). Of those studies which have been undertaken, O’Neill & Carayon (1993) reported that average partition height was positively related to perceived privacy, and Brill et al (1984) found that partition height was positively related to ratings of communication, privacy and job performance. O’Neill (1994), by contrast, did not find partition height to influence communication, distractions, privacy or satisfaction. In addition to these occupant reactions, O’Neill (1992; p.891) notes that “the degree of enclosure may influence the circulation of air within the work space and to some extent the ability to control the thermal environment (such as shutting a door to reduce drafts).” In his own comparison study of open, open-plan and enclosed offices, this researcher found that enclosure predicted satisfaction with temperature, although it was not significantly related to satisfaction with air quality.
The current study investigated the impact of a density characteristic, workstation area, and an enclosure characteristic, minimum partition height, on occupant
environmental satisfaction in open-plan environments. Minimum partition height, rather than average partition height, was used because it appeared that the lowest workstation partition would be the one most likely to influence satisfaction, particularly in relation to privacy.
In addition to the two workstation characteristics noted above, we also examined the influence of having an external window in the workstation. Access to a window is clearly preferred by most people (Collins 1975). Office occupants express a preference for natural rather than artificial light (Markus, 1967), and in addition to being a source of illumination, windows provide a view to the outside world (Collins, 1975; Wells, 1965). The accessibility of a view, particularly one of nature has been suggested as an important buffer to occupational stress (eg. Markus, 1967; Kaplan, 1995; Purcell et al, 1994) and is also related to occupant satisfaction (eg. Leather, Pygras, Beale & Lawrence, 1998). Open-plan office environments offer the potential to improve daylight penetration, but little work has been conducted to investigate the relationship between windows and occupant satisfaction in such contexts.
In the current study, we aimed to examine the relationship between three
workstation characteristics (workstation area, minimum partition height and windows) on four environmental satisfaction measures (satisfaction with privacy, ventilation and lighting, and overall environmental satisfaction). In assessing these four measures of environmental satisfaction we extended on previous research, which has tended to focus primarily on satisfaction with privacy or overall environmental satisfaction.
2. Method
The procedures used to collect data for the COPE T1 field studies have been described in detail in previous reports (eg. Veitch, Farley & Newsham, 2002). Thus, in this report, we focus only on those aspects of data collection relevant to our current investigations.
2.1 Participants
We collected data from 419 open-plan office workstations and their occupants. The workstations were located in three Canadian Federal Government buildings.
Although information on participation rates was not recorded, we estimated over 90% of office occupants who were approached to take part in the research agreed to do so.
The offices measured in these three buildings are due to undergo interior renovations over the next 18 to 24 months. As such, these data represent a first
timepoint, with a second period of data collection planned to be undertaken six months after the renovations.
Data were collected relatively equally across the three buildings (31%, 30% and 39% respectively). We also observed a near-equal gender split amongst office workers (49% male, 51% female). Office occupants’ age ranged from 18 to 70 years, with the
average age being 38.6 years. 36% of occupants were employed in administrative jobs, 15% in technical jobs, 42% in professional jobs, and 7% in managerial jobs.
2.2 Procedure
Informal emails were circulated, prior to data collection, to inform employees of the purpose and intent of the research. Once on site, researchers introduced themselves to occupants while they were seated at their desks and gave them the opportunity to
participate in the study. At each workstation where consent was granted, we collected data on the three workstation characteristics used in this study.
At the same time as data on workstation characteristics were collected, the
workstation’s occupant was taken to an unoccupied workstation to complete a self-report questionnaire. The questionnaire was administered using a palm-top computer (NEC Mobile ProT M 770, running Microsoft® Windows® CE) with a touch sensitive screen. The questionnaire was developed using Microsoft® Visual Basic 6 for Windows® CE, which enabled the occupant’s responses to be stored on the computer for later analyses. 2.3 Independent Variables
Workstation Area: The width and length of each workstation were measured by the researchers using an electronic tape measure (Dimension Master Plus, Calculated Industries Inc). This devise has 99.5% accuracy over a 0 to 60 foot range, and was also verified by researchers prior to the start of data collection. The width and length
measurements were used to calculate the area of each workstation, in square feet. Fifteen of the measured workstations were triangular in shape and this feature was adjusted for as appropriate in workstation area calculations2.
Minimum Partition Height: The heights of all sides of each workstation were also measured using the electronic tape measure. In some instances, however, the location or height of partitions made it difficult to reliably use the electronic tape measure and, in these cases, partition height was measured manually using a standard tape measure. Minimum partition height represents the lowest of these measurements and was recorded in inches.
Window: We also noted whether an external window was present on each side of the workstation. The presence of a window was determined by at least one external window in the workstation. Those workstations in which a window was present were coded as “1”, whilst those workstations without windows were coded as “0”.
2 We also considered the issue of ‘aspect ratio’ (ie. differences in the proportional length and width of
workstations which may result in the same overall workstation area arising from different shaped
workstations). Although two workstations may have the same overall area, differences in aspect ratio may result in differences in occupants’ perceptions of workstation size, which in turn may influence occupant satisfaction.
Aspect ratio was calculated as width divided by length, with aspect ratios closer to 1 denoting a more square, rather than rectangular, workstation. We found a limited range of aspect ratios in the current sample, with 85% of workstations (excluding triangular workstations) having aspect ratios between 0.7 and 1.3. As the majority of workstations were relatively square in shape, the potential influence of aspect ratio could be discounted.
2.4 Outcomes
In addition to demographic information, including age, gender, job category and education level, occupants completed items relating to satisfaction with various aspects of the office environment. The items were based on work by Stokols & Scharf (1990) and were combined into the following composite scales, on the basis of explanatory and confirmatory factor analysis (Veitch et al, 2001).
Satisfaction with privacy: This scale comprised ten items and included occupant ratings of satisfaction with aspects such as: “the level of privacy for conversations in your office”; “the amount of noise from other people’s conversations while you are at your workstation” and “the frequency of distractions from other people”. Occupants responded on a seven-point scale, ranging from “very unsatisfactory” (1) to “very satisfactory” (7). The scale score was computed as the average of the ten items. Cronbach’s alpha internal reliability coefficient for this scale was 0.88.
Satisfaction with ventilation: Three items were combined to form this scale, namely: occupants’ satisfaction with “the overall air quality in your workspace”; “the temperature in your workspace”; and “the air movement in your workspace”. The scale score was the average of these three items. The response scale used was identical to that given above and the internal reliability of this scale was 0.83.
Satisfaction with lighting: This scale was formed from five items and also used the same response scale as those above. Occupants rated their satisfaction with: “the amount of light for computer work”; “the amount of reflected light or glare in the computer screen”; “the access to a view of outside from where you sit”; “the quality of lighting in your work area”; and “the overall lighting on the desk top”. As above, the scale score was the average of these five items. Cronbach’s alpha for this scale was 0.73.
Overall environmental satisfaction: The scale used to assess overall
environmental satisfaction was comprised of two items. The first item asked occupants: “taking into consideration all the environmental conditions in your workstation, what is your degree of satisfaction with the indoor environment in your workstation as a whole?” This item used the same seven-point response scale described above. The second item asked occupants to: “estimate how you think your personal productivity at work is increased or decreased by the environmental conditions as compared to what you usually achieve?” This item used a seven-point scale, ranging from 30% less than usual
productivity (1) to 30% more than usual productivity (7). The overall environmental satisfaction score was the average of these two items, and the internal reliability for this scale was 0.65.
3. Results
Of the 419 data cases, fifteen were excluded due to missing data or univariate outliers. These analyses were, therefore, conducted using 404 cases. An additional seventeen cases had missing data for one or both of the items comprising overall environmental satisfaction. However, rather than reducing the total sample size for all analyses, we excluded these cases only for analyses involving overall environmental satisfaction, resulting in a sample size of 387 for these analyses.
For the purposes of statistical testing, data on occupants’ gender were coded as ‘0’ for females and ‘1’ for males. In addition, due to the relatively small percentages of occupants in the technical and managerial job categories, we combined the four job categories into two, with technical and administrative jobs being coded as ‘1’ and professional and managerial categories being coded as ‘2’.
We also found education level to be significantly correlated with job category (r =.56, p=.001). As this correlation suggested a degree of overlap between occupants’ education level and job category, we chose to include only job category in further analyses.
3.1 Descriptive Statistics
The first stage of analysis was to calculate the means, standard deviations, frequency distributions and correlations for all variables used in the study.
The frequency distributions for workstation area, minimum partition height and window are shown in Figures 1 to 3. As Figure 1 indicates, workstation area ranged from 41.56 to 209.31 ft2, and the mean for this variable was 108.53 ft2. Minimum partition height ranged from 0 to 81 inches. However, as is indicated in Figure 2, the majority of workstations had a minimum partition height of 60 to 70 inches, with the mean for this variable being 61.63 inches. Finally, as is shown in Figure 3, 49% of workstations had no window, whilst 51% had a window.
The frequency distributions for the four environmental satisfaction scales are shown in Figures 4 to 7. These figures indicate good distributions of data for these four satisfaction measures. The mean score for satisfaction with lighting was slightly higher than for the other three satisfaction measures, being 4.77, as compared to 4.07 for satisfaction with ventilation, 3.97 for satisfaction with privacy and 4.02 for overall environmental satisfaction (NB. scale labels: 1=very unsatisfactory, 4=neutral, 7=very satisfactory).
0 50 100 150 200 250 AREA 0 10 20 30 40 50 60 70 80 90 Count 0.0 0.1 0.2
Proportion per Bar
0 10 20 30 40 50 60 70 80 90 MINPH 0 50 100 150 200 Count 0.0 0.1 0.2 0.3 0.4
Proportion per Bar
0 1 WINDOW 0 100 200 300 Count 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Proportion per Bar
Workstation Area (sq ft)
No Window Window Minimum Partition Height (in)
Figure 1: Frequency
Distribution for Workstation
Area
Figure 2: Frequency
Distribution for Minimum
Partition Height
Figure 3: Frequency
Distribution for Window
1 3 5 7 SAT_AIR 0 10 20 30 40 50 60 Count 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
Proportion per Bar
1 3 5 7 SAT_PRIV 0 10 20 30 40 50 60 70 Count 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16
Proportion per Bar
2 3 4 5 6 Satisfaction with Ventilation
2 3 4 5 6 Satisfaction with Privacy
Figure 5: Frequency
Distribution for Satisfaction
with Privacy
Figure 4: Frequency
Distribution for Satisfaction
with Ventilation
1 = very unsatisfactory 7 = very satisfactory
1 3 5 7 SAT_LIGHT_V 0 10 20 30 40 50 60 70 80 90 100 Count 0.0 0.1 0.2
Proportion per Bar
1 3 5 7 OVERALL_ENVS 0 10 20 30 40 50 60 70 Count 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18
Proportion per Bar
2 3 4 5 6 Overall Environmental
Satisfaction
2 3 4 5 6 Satisfaction with Lighting
Figure 6: Frequency
Distribution for Satisfaction
with Lighting
Figure 7: Frequency
Distribution for Overall
Environmental Satisfaction
1 = very unsatisfactory 7 = very satisfactory
Means, standard deviations and correlations between the variables measured in this study are shown in Table 1. This table indicates that window was significantly positively correlated with satisfaction with lighting (r = .33, p=.001) and that minimum partition height was significantly negatively correlated with overall environmental satisfaction (r = -.11, p=.05).
However, Table 1 also shows that age, gender and job category were each significantly correlated with at least one workstation characteristic and/or environmental satisfaction measure. Whilst interesting in their own right, for the current study these significant correlations have the potential to mask, distort or otherwise influence the relationships between workstation characteristics and environmental satisfaction
(Tabachnick & Fidell, 2001). As such, it was important for us to conduct further analyses that controlled for the potential effects of these demographic variables.
These further analyses are presented in the next section. However, prior to this we also conducted one-way analysis of variance tests, to determine whether there were significant differences between buildings in relation to the demographic, workstation characteristic and environmental satisfaction measures used in this study. These analyses indicated that age and gender did not significantly differ by building. However, we found that there were differences between buildings in terms of job category (F=18.59, p=.001). We also found that all three workstation characteristics differed significantly between buildings (workstation area: F=52.03, p=.001; minimum partition height: F=10.88, p=.001; window: F=26.69, p=.001). Finally, these analyses indicated that there were significant differences across buildings in relation to satisfaction with ventilation (F=4.11, p=.05), satisfaction with lighting (F=4.36, p=.05) and overall environmental satisfaction (F=6.86, p=.001), although significant differences were not observed with respect to satisfaction with privacy.
To determine whether building could be predicted from occupant satisfaction, a direct discriminant function analysis was also performed. Using the classification procedure in the analysis, a total of 58% of all cases were classified correctly. For each building, 66%, 57% and 53% of cases, respectively, were classified correctly.
Overall, the above analyses indicated that there were differences between
buildings on a number of the measured variables. As such, it was also important for us to control for the effects due to building in further statistical analyses.
mean standard deviation 1 2 3 4 5 6 7 8 9 1. age 38.67 10.69 2. gender 0.52 0.50 .064 3. job category 1.49 0.50 .121* .253*** 4. w/s area 108.53 29.95 .119* .075 .236***
5. min partition height 61.63 16.74 .057 .080 .115* .389***
6. window 0.52 0.50 .179*** .088 .006 .305*** .063
7. satisfaction w/t privacy 3.97 1.14 -.117* .054 -.026 .037 -.029 -.029
8. satisfaction w/t air quality 4.07 1.43 -.001 .199*** .084 -.057 -.065 -.082 .413***
9. satisfaction w/t lighting 4.77 1.18 -.001 .062 -.085 .053 -.025 .329*** .485*** .419***
10.overall environmental satisfaction
4.02 1.31 -.029 .047 -.117* -.025 -.110* .061 .603*** .471*** .491***
* p=.05, **p=.01, ***p=.001
3.2 Predicting Occupant Satisfaction
The correlations outlined above indicated two significant relationships between workstation characteristics and occupant environmental satisfaction. However, as mentioned above, the correlation matrix and the analyses comparing buildings also indicated that workstation characteristics and occupant environmental satisfaction varied as a function of age, gender, job category and building. These relationships have the potential to influence relationships between workstation characteristics and
environmental satisfaction. We therefore conducted hierarchical regression analyses in order to gain a more detailed view of the relationship between workstation characteristics and environmental satisfaction, controlling for the effects of these extraneous variables.
Regression analyses are a set of statistical techniques that test the strength and direction of relationships between several predictor variables and a dependent variable. In the current study, predictor variables include both workstation characteristics and the extraneous variables that needed to be controlled for (ie. demographics and building). The dependent variables in this study were the four environmental satisfaction measures. Regression techniques are advantageous in allowing several predictor variables to be assessed at once, and indicate the relationship between each predictor and the dependent variable whilst controlling for all other predictor variables in the model. Furthermore, this technique provides a standardised result, so that the relative predictive ability of variables can be determined.
We conducted these analyses using hierarchical regression techniques. This type of regression analysis allows predictor variables to be entered into the model in a series of steps, thereby providing information on the additional benefit gained from entering additional predictor variables into each model.
Four hierarchical regression models were evaluated, one for each of the environmental satisfaction measures detailed above. In each regression, predictor variables were entered in the following three steps:
1. In the first step, the ability of extraneous variables to predict environmental satisfaction was assessed, by entering building, age, gender and job category into the model. As building was a non-ordered categorical variable, the established approach of entering building effects as two dummy variables (denoted a” and “building-b”) was used (Tabachnick & Fidell, 2001).
2. In the second step we assessed the predictive ability of the two main
workstation characteristics of interest to our present study, namely: workstation area and minimum partition height.
3. In the final step, we entered the final workstation characteristic, window, in order to assess the additional benefit of including this variable in the model.
In addition to the cases excluded in earlier analyses, three further cases were found to be multivariate outliers and so were excluded from all regression analyses. The results of the hierarchical regression analyses are shown in Tables 2 to 5. For each analysis, the standardised beta weight, ß , indicates the strength and direction of the relationship between each predictor variable and environmental satisfaction, holding the influence of other predictors constant. The total amount of variance in environmental satisfaction explained for by each model is shown as the “total R2” and the additional
variance accounted for by each subsequent step is shown as the “change in R2”. The statistical significance of beta weights and explained variance are shown in Tables 2 to 5 with asterisks. The results of the hierarchical regression analyses are summarised below. Satisfaction with privacy: In the first step of this regression model, age was found to significantly predict the dependent variable, with younger occupants being more satisfied with privacy than older occupants. The predictor variables entered in the first step did not explain a significant amount of the variance in satisfaction with privacy. After entering workstation area and minimum partition height in step 2, the age effect still remained significant, in addition to a significant building effect. At this stage of the regression model, neither workstation area nor minimum partition height were found to significantly predict satisfaction with privacy. Entering these two workstation
characteristics did not significantly improve the total amount of variance in satisfaction with privacy accounted for by the regression model. However, after entering window as a predictor variable in the final step of the regression model, building, age and
workstation area were all found to be significant predictors of satisfaction with privacy. The beta weight for workstation area was positive, indicating that those occupants with larger workstations reported greater satisfaction with privacy than did those occupants with smaller workstations. Window was not found to be a significant predictor of this dependent variable. Overall, the predictor variables were able to explain a small but significant amount of variance in satisfaction with privacy (R2 =.04, p=.05).
Table 2: Hierarchical Regression Analyses on Satisfaction with Privacy
Step Satisfaction with Privacy (n=401)
ß ß ß 1. building-a .063 .087 .137* building-b .105 .152* .186** job category -.002 -.013 -.012 age -.113* -.126* -.114* gender .072 .069 .075 2. workstation area .114 .158*
min partition height -.047 -.055
3. window -.100
Change in R2 .024 .009 .007
Total R2 .024 .033 .040*
Satisfaction with ventilation: The first step in this regression model indicated that building was a significant predictor of satisfaction with ventilation. Together, the control variables entered in this step accounted for a significant amount of variance in the
dependent variable. Neither workstation area nor minimum partition height were significant predictors of satisfaction with ventilation when entered in the second step. However, following their addition to the regression model, both building and gender were found to significantly predict this dependent variable, with males reporting greater
satisfaction with ventilation than females. The building and gender effects remained after introducing window as a predictor variable in the final step of the regression. In addition, window was found to be a significant predictor of satisfaction with ventilation and
accounted for a 1.6% (p=.01) increase in explained variance. The beta weight for window was negative, indicating that occupants with a window in their workstation reported lower satisfaction with ventilation than did occupants without a window in their workstation. The overall regression model was able to explain 8.6% (p=.001) of the variance in satisfaction with ventilation.
Table 3: Hierarchical Regression Analyses on Satisfaction with Ventilation
Step Satisfaction with Ventilation (n=401)
ß ß ß 1. building-a .075 .068 .143* building-b .188** .167* .218** job category .085 .087 .088 age -.028 -.021 -.002 gender .183 .188*** .196*** 2. workstation area -.009 .056
min partition height -.051 -.063
3. window -.149**
Change in R2 .068*** .003 .016**
Total R2 .068*** .070*** .086***
* p=.05, **p=.01, ***p=.001
Satisfaction with lighting: The first step of the regression model using
satisfaction with lighting as the dependent variable indicated that building was a
significant predictor. When area and minimum partition height are added to this model in step 2, workstation area was found to be a significant predictor of satisfaction with
lighting, with those occupants in larger workstations reporting greater satisfaction with lighting than those occupants in smaller workstations. However, when the window variable was added in step 3, workstation area was no longer significant whilst window proved to be a significant predictor of satisfaction with lighting. The beta weight for this effect was positive, indicating that those occupants with a window in their workstation were more satisfied with lighting than those occupants without a window in their
workstation. The addition of window in step 3 resulted in a 9% (p=.001) improvement in explained variance. Overall, this model accounted for 13.8% (p=.001) of the variance in satisfaction with lighting.
Table 4: Hierarchical Regression Analyses on Satisfaction with Lighting
Step Satisfaction with Lighting (n=401)
ß ß ß 1. building-a .075 .111 -.069 building-b .138* .209** .086 job category -.076 -.092 -.093 age .009 -.012 -.058 gender .088 .085 .065 2. workstation area .171** .014
min partition height -.069 -.041
3. window .356***
Change in R2 .029* .020* .090***
Total R2 .029* .049** .138***
* p=.05, **p=.01, ***p=.001
Overall environmental satisfaction: The first step of the final regression model
indicated a significant building effect, which remained throughout the analysis. Both workstation area and minimum partition height were found to significantly predict overall environmental satisfaction, when they were included in the regression model in step 2. The beta weight for workstation area was positive, indicating that those occupants with larger workstations reported greater overall environmental satisfaction as compared to those occupants with smaller workstations. The effect for minimum partition height was negative, suggesting that those occupants with a smaller minimum partition height experienced greater overall environmental satisfaction as compared to those occupants with a larger minimum partition height. However, when the window variable was
entered into the regression model in step 3, the effect for workstation area dropped below significance (ß =.13, p=.061). A negative significant effect remained in relation to
minimum partition height, but window was not found to be a significant predictor of overall environmental satisfaction. Overall, this model was able to explain 6.3% (p=.01) of the variance in overall environmental satisfaction.
Table 5: Hierarchical Regression Analyses on Overall Environmental Satisfaction
Step Overall Environmental Satisfaction (n=384)
ß ß ß 1. building-a .067 .089 .080 building-b .191* .228*** .222** job category -.092 -.103 -.103 age -.031 -.038 -.041 gender .075 .078 .077 2. workstation area .135* .127
min partition height -.117* -.115*
3. window .019
Change in R2 .045** .017* .000
Total R2 .045** .063*** .063**
The results of the regression analyses are discussed shortly. However, prior to this we conducted additional analyses to investigate the effects of workstation
characteristics on occupants’ rated importance of environmental aspects.
3.3 Additional Analyses: Workstation Characteristics and Ranked Importance of Environmental Aspects
We asked occupants to rank order the importance of seven aspects of the office environment, namely: lighting; air quality/ventilation; temperature; sound and noise; privacy; workstation size; and window access. Occupants were asked to assign a rank from “most important” (1) to “least important” (7) to each environmental aspect, using each rank only once.
We used the Goodman-Kruskal Gamma non-parametric test to investigate whether the rank assigned to each environmental aspect was influenced by each of the three workstation characteristics. This test assesses the relationship between two ordered categorical variables. As this test can only be conducted using categorical data, we formed categories for both workstation area and minimum partition height as follows. Workstation area was separated into five categories, representing less than 75 sq ft, 76-100 sq ft, 101-125 sq ft, 126-150 sq ft and greater than 151 sq ft respectively. Minimum partition height was also binned into five categories, representing less than 48 in, 49-57 in, 58-63 in, 64-68 in, and greater than 69 in. Window was a categorical variable by definition.
Twenty-one separate tests were conducted to assess the influence of each workstation characteristic on the rated importance of each environmental aspect (ie. 3 workstation characteristics x 7 environmental aspects). Each test computed the frequency of occupants assigning each level of importance to the environmental aspect, across each level of the workstation characteristic, and determined whether there was a significant trend across the frequencies. The tests were conducted using a sample size of 364, as 40 occupants either did not complete the importance ranking questions or had to be excluded due to incorrect completion of these items.
Two significant results were found and are shown in tables 6 and 7. For greater clarity, these results are also shown diagrammatically, in Figures 8 and 9 respectively. Both of these results concern the importance ranking assigned to “window access” and indicate that both workstation area and the window variable significantly affected the rank assigned to this environmental aspect. In the case of workstation area, it was found that occupants in larger workstations tended to rank access to a window as more
important than did occupants in smaller workstations. The significant result in relation to the window variable indicated that those occupants with a window in their workstation tended to rank access to a window as more important than did occupants without a window in their workstation3.
The results of these analyses, and the earlier regression analyses, are discussed in the next section.
3
Table 6: Frequency and Percentage Counts for Workstation Area and Ranked Importance of Window Access
< 75 sq ft (%) 76-100 (%) 101-125 (%) 126-150 (%) > 151 (%) Total (%) 1 most important 2 (0.55) 14 (3.85) 13 (3.57) 13 (3.57) 7 (1.92) 49 (13.46) 2 2 (0.55) 8 (2.20) 14 (3.85) 6 (1.65) 6 (1.64) 36 (9.89) 3 4 (1.10) 5 (1.37) 10 (2.75) 2 (0.55) 3 (0.82) 24 (6.59) 4 5 (1.37) 7 (1.92) 15 (4.12) 7 (1.92) 6 (1.65) 40 (11.00) 5 3 (0.82) 12 (3.30) 14 (3.85) 2 (0.55) 5 (1.37) 36 (9.89) 6 7 (1.92) 10 (2.75) 16 (4.40) 11 (3.02) 3 (0.82) 47 (12.91) 7 least important 27 (7.42) 43 (11.81) 42 (11.54) 15 (4.12) 5 (1.37) 132 (36.26) Total % 50 (13.74) 99 (27.20) 124 (34.07) 56 (15.39) 35 (9.62) 364 (100.00) Goodman-Kruskal Gamma = -0.240, p=.05 n=364 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% <75 76-100 101-125 126-150 >151 Workstation Area (sq ft) Percentage count 7 least important 6 5 4 3 2 1 most important Figure 8: The Relationship between Workstation Area and
Table 7: Frequency and Percentage Counts for Window and Ranked Importance of Window Access
No Window (%) Window (%) Total (%)
1 most important 17 (4.67) 32 (8.79) 49 (13.46) 2 16 (4.40) 20 (5.50) 36 (9.89) 3 12 (3.30) 12 (3.30) 24 (6.59) 4 18 (4.95) 22 (6.04) 40 (10.99) 5 9 (2.47) 27 (7.42) 36 (9.89) 6 28 (7.69) 19 (5.22) 47 (12.91) 7 least important 77 (21.15) 55 (15.11) 132 (36.26) Total % 177 (48.63) 187 (51.37) 364 (100.00) Goodman-Kruskal Gamma = -0.228, p=.01 n=364 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% no window window Window in Workstation Percentage Count 7 least importance 6 5 4 3 2 1 most importance
4. Discussion
In this study, we investigated the effects of three workstation characteristics (workstation area, minimum partition height, and windows) on four environmental satisfaction measures (satisfaction with privacy, ventilation and lighting, and overall environmental satisfaction). We also examined whether the three workstation
characteristics influenced occupants’ ratings of the relative importance of seven office environment features.
Using hierarchical regression analyses, in which we controlled for the effects of age, gender, job category and building, we found that: workstation area significantly predicted satisfaction with privacy; window significantly predicted satisfaction with lighting and satisfaction with ventilation; and minimum partition height significantly predicted overall environmental satisfaction. In addition, both workstation area and window were significantly associated with occupants’ ratings of the relative importance of access to a window.
The results of this study indicate that occupants in larger workstations were more satisfied with privacy than were occupants in smaller workstations. This finding supports previous research (eg. Oldham & Rotchford, 1983; O’Neill & Carayon, 1993), and is probably because larger workstations increase the distance between the occupant and their co-workers, and also increase the amount of space available to the occupant.
As might be expected, occupants with workstations that incorporated a window were more satisfied with lighting. However, we also found that these occupants were less satisfied with ventilation, as compared to occupants in windowless workstations. The windows in the study sample were sealed; therefore, this finding is probably the result of temperature variations arising from increased draught during the winter and increased heat gain during the summer.
Finally, we found that those occupants with a lower minimum partition height were more satisfied with overall environmental satisfaction. The mechanisms behind this finding are less obvious; however workstations containing at least one lower height partition might provide occupants with an improved sense of space, and may also improve ambient conditions (through, for example, better air flow or access to overhead lighting).
However, although the above findings were statistically significant, we found that the overall amount of variance in occupant satisfaction that could be explained by
workstation characteristics was relatively small (between 4 and 14%). One reason for this may be the largely favourable workstation characteristics found in the current study sample. For example, although workstation area ranged from 41.56 to 209.31 ft2, the average workstation area was 108.53 ft2. Therefore, in many cases, workstation area might have been too large to influence occupant responses greatly. In addition, whilst minimum partition height varied from 0 to 81 inches, the majority of workstations had a minimum partition height of 60-70 inches, and this restricted range is likely to have influenced our results. Previous research on open-plan offices has often compared open and closed environments, in which the characteristics of spaces differ more dramatically. It is likely that the effects of workstation variations within open-plan environments are more subtle and more difficult to detect. Such considerations suggest that future work
should attempt to include a wider ranges of workstation characteristics, either through the use of a greater number of different buildings or through examining the effects of
changes in workstation characteristics over time (as is currently planned for the post-renovation phase of the COPE field study).
An alternative argument is that different measures of workstation characteristics might better explain differences in occupant environmental satisfaction. For example, in previous studies researchers have used alternative measures of enclosure, such as average partition height or number of partitions. Similarly, the effects of workstation area might differ from density measures that take into account the position of workstations relative to each other, such as the distance from one occupant to the next, the number of
employees within a specified radius of the target occupant, or area on the floorplate per person. Workstation characteristics not featured in our current study, for example the amount of storage space or location on the floorplate relative to exhaust outlets, might also be related to occupant environmental satisfaction. Future work could, therefore, benefit from the inclusion of alternative measures of workstation characteristics.
A number of researchers have highlighted the potentially complex relationships between the physical office environment, occupants’ perceptions of those environments and their reactions towards them. This complexity is highlighted in a study by O’Neill & Carayon (1993). Here, perceived enclosure accounted for 43% of the variance in
satisfaction with privacy, whereas physical enclosure (measured as workstation area and average partition height) accounted for only 8%; suggesting that perceptions of enclosure were formed from more than simply the physical characteristics of workstations.
It has been argued, for example, that occupant responses to office environments are influenced by personal and organisational factors. Job complexity has previously been related to occupant satisfaction, with occupants preferring more privacy when completing more complex tasks (ie. Hedge, 1982; Sundstrom, Town et al, 1982; Block & Stokes, 1989; Fried, Slowik, Ben-David & Tiegs, 2001). This is likely to be because complex tasks require more focused concentration, and so more private environments provide a buffer against distractions and interruptions. Related to job complexity, job level has also been associated with occupant satisfaction (eg. Johnson, 1970; Hedge, 1982; Carlopio & Gardner, 1992; 1995). Here, managers are typically found to be more sensitive to privacy and disturbances, primarily because higher level jobs are assumed to be more complex and demanding.
Other factors, such as organisational tenure, experience of alternative office environments, abilities to screen out distracting stimuli, and personal needs for privacy, have also been associated with occupant satisfaction (eg. Hedge, 1982; Oldham, 1988; Block & Stokes, 1989; Jackson et al, 1997; Fried et al, 2001). In addition, as noted by both Sundstrom (1987) and Marans & Spreckelmeyer (1982), occupant responses are also likely to be affected by how an individual’s own workstation compares to those of their co-workers, and to the workstation they feel they deserve to have.
Whilst it is not the principal aim of the COPE project to investigate the above factors, it is important that they be considered, in order to control for their potential influence on the relationships between physical environmental parameters and occupant satisfaction. In the current study, we controlled for effects attributable to age, gender, job category and building. Future studies might benefit from the inclusion of further personal
and organisational factors. Using the current COPE field study data, we plan to
investigate job category in greater detail, to determine whether the relationship between workstation characteristics and occupant environmental satisfaction differs depending on job category. Support for job category as a moderator would suggest that workstation design should be tailored to the job in order to achieve the most favourable levels of occupant satisfaction.
In addition to the above arguments, the relationship between workstation characteristics and occupant environmental satisfaction might also interact with the relative importance that occupants place on environmental features. Sundstrom (1987), for example, argues for a “weakest link” approach, in which the aspect of the
environment that the occupant is least satisfied with becomes the most important in determining environmental satisfaction. In the current study, we found that occupants in larger workstations and occupants in windowed workstations tended to rate access to a window as more important than did occupants in smaller workstations and windowless workstations respectively. This latter finding is in contrast to previous research, which has suggested that occupants in windowless workstations tend to rate the importance of having a window higher than occupants in windowed workstations (Boubekri &
Haghighat, 1993). Thus, whilst others have argued that occupants rate aspects of the environment as more important if they do not currently have them, our findings suggest that occupants might underplay the importance of features they do not have (particularly if it is unlikely that they will obtain them in the future). Clearly, more work is needed to determine the role of occupants’ importance rankings, particularly in relating these ratings to occupant satisfaction.
Finally, we note that little work has been undertaken on the mediating role of physical ambient conditions on the relationship between workstation characteristics and occupant satisfaction. It can be argued that workstation characteristics affect ambient conditions, such as illuminance, temperature and noise levels, which in turn affect occupant satisfaction. As such, stronger relationships might be evident between ambient conditions and occupant satisfaction than were found here for workstation characteristics. This notion may be particularly important if the same workstation characteristic is
positively related to one ambient condition but negatively related to another ambient condition.
Some work has been conducted on the relationship between workstation characteristics and ambient conditions. Bauman et al (1992), for example, conducted environmental chamber experiments and concluded that variations in partition height produced only small differences in overall thermal and ventilation performance. O’Donnell & Nguyen (1990), by comparison, argue that partition height influences air velocity. Other work has suggested that partition height and the provision of a gap at the bottom of the partition does not affect air distribution or mean age of air, but does affect contaminant removal efficiency (eg. Haghighat, Huo, Zhang & Shaw, 1996). Haghighat et al (1996) also noted that the workstation nearest to the return grill tended to have the worst contaminant concentrations, because pollutants from other sources in the space were being drawn towards the return grill. Furthermore, Haghighat (1994) found that the layout of workstations in an open office space affected airflow patterns, and could therefore influence contaminant removal. In relation to acoustical parameters,
researchers have also noted the role of partition height and partition construction in determining speech privacy (eg. Moreland, 1988).
There is a clear need for the extension of work of this kind, to link workstation characteristics, ambient conditions and occupant environmental satisfaction together. We have collected data on a number of physical ambient parameters as part of the COPE field studies and intend to analyse the relationships between these parameters,
workstation characteristics and occupant satisfaction as the next stage of analyses of the COPE field study data.
5. Conclusions
In this study, we found significant, albeit small, relationships between workstation characteristics and occupant environmental satisfaction. These findings provide a
promising base from which to further explore these relationships. Potential directions for the future include obtaining data on a broader variation of workstation characteristics (either from additional buildings or from the same buildings over time), examining alternative workstation measures, investigating the role of job category in more depth, relating occupant importance rankings to occupant satisfaction, and analysing the relationships between workstation characteristics, physical ambient parameters and occupant environmental satisfaction.
6. References
Allen, E. C., & Gerstberger, P. G. (1973). A field experiment to improve communications in a product engineering department: The nonterritorial office. Human Factors, 15(5), 487-498.
Banbury, S., & Berry, D. C. (1998). Disruption of office-related tasks by speech and office noise. British Journal of Psychology, 89, 499-517.
Bauman, F. S., Faulkner, P. E., Arens, E. A., Fisk, W. J., Johnston, L. P., McNeel, P. J., Pih, D., & Zhang, H. (1992). Air movement, ventilation, and comfort in a partitioned office space. ASHRAE Transactions: Symposia, 756-780.
Becker, F. D., Gield, B., Gaylin, K., & Sayer, S. (1983). Office design in a community college: Effect on work and communication patterns. Environment and Behavior, 15(6), 699-726.
Block, L. K., & Stokes, G. S. (1989). Performance and satisfaction in private versus nonprivate work settings. Environment and Behavior, 21(3), 277-297.
Bloom, D. E. (1986). Women and work. American Demographics, 15, 25-30.
Boubekri, M., & Haghighat, F. (1993). Windows and environmental satisfaction: A survey study of an office building. Indoor Environment, 2, 164-172.
Boyce, P. R. (1974). Users’ assessments of a landscaped office. Journal of Architectural Research, 3(3), 44-62.
Brill, M., Margulis, Konar & BOSTI (1984). Using Office Design to Increase Productivity (Vol. 1). Buffalo, NY: Workplace Design and Productivity Inc.
Brill, M., Weidemann, S., & BOSTI Associates. (2001). Disproving widespread myths about workplace design. Jasper, IN: Kimball International.
Brookes, M. J. (1972). Office landscape: Does it work? Applied ergonomics, 3(4), 224-236.
Brookes, M. J. (1978). Changes in employee attitudes and work practices in an office landscape. In A. Friedmann, C. Zimring & E. Zube (eds.), Environmental Design Evaluation (pp. 35-45). London: Plenum Press.
Brookes, M. J., & Kaplan, A. (1972). The office environment: Space planning and affective behavior. Human Factors, 14, 373-391.
Burgess, M. A., Lai, J. C. S., Eisner, M., & Taylor, E. (1989). Speech privacy in open-plan offices - post occupancy. Proceedings of the 25th Annual Conference of the Ergonomics Society of Australia: Ergonomics, Technology, & Productivity, 26-29 November (pp. 351-354). Fortutide Valley, Australia: Ergonomics Society of Australia.
Cangelosi, V. E., & Lemoine, L. F. (1988). Effects of open versus closed physical environment on employee perception and attitude. Social Behavior and Personality, 16(1), 71-77.
Carlopio, J. R., & Gardner, D. (1992). Direct and interactive effects of the physical work environment on attitudes. Environment and Behavior, 24(5), 579-601.
Carlopio, J., & Gardner, D. (1995). Perceptions of work and workplace: Mediators of the relationship between job level and employee reactions. Journal of Occupational and Organizational Psychology, 68, 321-326.
Christie, B. (1985). Human Factors of Information Technology in the Office. Chichester: J. Wiley & Sons. Collins, B. L. (1975). Windows and people: A literature survey. Psychological reaction to environments
with and without windows. Washington DC: Natural Bureau of Standards.
Dean, L. M., Pugh, W. M., & Gunderson, E. K. E. (1975). Spatial and perceptual components of crowding: Effects on health and satisfaction. Environment and Behavior, 7, 225-236.
