International Journal of Facility Management, Vol 1, No 2 (2010)

Font Size:  Small  Medium  Large
Untitled Document

A model of workplace environment satisfaction, collaboration experience,
and perceived collaboration effectiveness: A survey instrument

Ying Hua


This research paper introduced a survey instrument for evaluating workplace occupant satisfaction with work environment‘s support for collaboration, occupant satisfaction with collaboration-related work experience, and occupant perceived collaboration effectiveness. The survey questions examined a range of shared spaces in workplace and highlighted the need for workplace to support both individual and collaborative work to support organizational performance. The questionnaire was then tested in a field study. Exploratory and confirmatory factor analyses revealed a two-factor model of occupant satisfaction with collaborative work environment, with the two factors as perceived facilitation from the work environment for collaboration and perceived distraction from the work environment because of interactive behaviors. Structural equation modeling explained the relationship among workplace environment satisfaction, collaboration-related work experience, and perceived collaboration performance. The development and field test of this survey tool enable evaluations of workplace settings, as well as support workplace design and reengineering, and innovative workplace management for collaboration performance.

KEYWORDS: Workplace collaboration, shared spaces, work environment satisfaction, work experience, perceived effectiveness, survey, factor analysis, structural equation modeling


White-collar work today has become increasingly interactive and collaborative, and the value of collaboration to organizational performance has been increasingly recognized (Lawler, 2001; Nadler et al., 1997). Intended to reduce real estate costs and/or energy for commuting, alternative work styles, such as home offices, virtual offices, and telecommuting, are gradually transforming physical workplaces into a place to meet and interact rather than a place for carrying out individual tasks.

This paper presents a questionnaire that assesses collaborative work environments through occupant satisfaction with workplace spatial settings (including both individual workstations and shared spaces, with the emphasis on the latter), occupant satisfaction with their collaboration-related work experience, and occupant perceived collaboration effectiveness. The goal for developing this survey instrument is to enable the exploration the effect of spatial settings on office workers’ work environment satisfaction, their collaborative work experience, and perceived work performance. The survey instrument was tested in a field study, the data collected from which were used for preliminary validation of the instrument. Furthermore, the data collected identifies key spatial characteristics of workplace that support collaboration effectiveness and offers a theoretical model for evaluating, designing, and managing workplace. The model highlight effective workplace designs that effectively deal with the tension between the need for individual, concentrated work and interactive group work.


Several aspects of workplace interaction and collaboration have been studied.

2.1. Communication

Communication is highly functional for work and occurs often in a workplace. According to Clark and Wilkes-Gibbs’s (1986) principle of least collaborative effort, people base their conversations on as little combined effort as possible. Because face-to-face conversation is the most direct and easy route to communicate, it is also the most powerful medium for coordinating collaboration in a workplace, especially in those highly interdependent groups that need intensive interactions to carry out their tasks. The powerful effect of face-to-face conversation is demonstrated by a number of studies. These effects are believed to derive from the commitment conveyed through face-to-face communication, the bonds and social contracts that accrue, and the group identity that is enhanced by direct interactions (Kiesler & Cummings, 2002). This rich channel of communication, which includes verbal and non-verbal cues, such as facial expressions, gestures, and body language, has a positive impact not only on communication effectiveness but also the communication experience.

Another important form of interaction at work is spontaneous communication, which occurs through impromptu encounters between co-workers as they move around a workplace. It accounts for a large proportion of informal social interaction and networking at work. In fact, repeated encounters, even without conversation, help to promote the awareness of co-workers and to foster office relationships. Informal communication is highly valued for collaboration at work (Kraut et al., 1990; Peponis, 2004); organizations are trying many different strategies to increase the likelihood of informal interactions between co-workers. Spatial settings could affect the possibility and frequency of these encounters as well as the likelihood that chance encounters will lead to conversations that benefit group work. McCoy (2000) demonstrated that spatial organization permits and encourages not only informal communication but also the extended collaboration necessary for high levels of creative teamwork. A good workplace design helps to increase the possibility of chance encounters through a balance between individual workstations and collaborative spaces, for example, meeting rooms and meeting spaces, shared spaces for office equipments, coffee or other services and amenities, circulation areas, as well as outdoor break areas, that are strategically laid out.

2.2. Interruptions and Distractions

Unfortunately, spontaneous communication comes at a cost. While impromptu interactions facilitate the quick flow of information through knowledge-work processes (Davenport et al., 2002), there are occupant concerns about the interruption of concentrated work, especially in an open-plan work environments.
The ability to concentrate at work remains particularly important for knowledge workers, whose tasks are likely to be complex and time-sensitive. Frequent interruptions or distractions keep them from reaching a peak state of performance. According to DeMarco and Lister (1999), reaching a state of peak performance or, in their words, “the state of flow” requires at least 15 minutes to ramp up concentration; during this immersion time, people are particularly sensitive to interruptions and distractions. When interrupted, one must restart the immersion process. DeMarco and Lister claimed that frequent interruptions can increase frustration and reduce productivity. The need for a distraction-free environment also differs with task complexity.

In a survey of office workers, co-sponsored by the American Society of Interior Designers, Steelcase, Armstrong World Industries, and other workplace industry manufacturers, 71% of respondents found noise to be the most significant workplace distraction (Steelcase, 2007), which is a more serious counter-productive factor in an open-plan work environment as opposed to a cellular office environment. Though low acoustic privacy in open-plan work settings may create opportunities for “picking up” useful information, it is likely to expose occupants to excessive stimulation or “overstimulation” (Desor, 1972; Baum & Paulus, 1987). Individuals differ in their ability to screen the numerous stimuli in a work environment. Those with lower stimulus-screening skills respond more negatively to distractions both in attitudinal and behavioral reactions (Oldham et al., 1991).

High performance in both individual and collaborative work is both critical in today’s knowledge-based industries. A workplace’s ability to support both distraction-free individual work, and impromptu interactions (both in one’s workspace and in shared spaces) was found to have the strongest affect on both individual and team performance, and these two needs should be facilitated without compromising each other (Brill and Weidemann, 2001).

2.3. Privacy

Privacy is a well-studied subject in workplace literature. It is considered more important than other features of the physical work environment, such as size, temperature, or lighting (Farrenkopf & Roth, 1980). “The concept of privacy assumes that people try to maintain an optimal level of social contact, and dissatisfaction results from being in situations that deviate from what a person considers optimal” (Sundstrom et al., 1980). The need for privacy in a workplace is a multi-layered construct that includes the need to control access to workspace, to limit distraction and interruption, and to communicate informally with others (Rashid & Zimring, 2005).

undstrom (1986) used “speech privacy” to refer to one’s ability to hold a conversation without being overheard by others and “visual privacy” to refer to the ability to block oneself from being observed. The visual and acoustic shield provided by walls, doors, and high partitions in a workplace may contribute to the sense of privacy because they give occupants more control over their access. Characteristics of spatial settings associated with perceived privacy include workstation type (open or closed), the number of enclosure sides, the height of enclosures, and the density of the workspace (Oldham, 1988; Oldham & Rotchford, 1983; Hatch, 1987).

2.4. Group performance

Theories of group performance and collaboration effectiveness at work depict outcomes as a function of multiple factors, including task characteristics, team design, group processes (interactions such as communication and conflict that occur between group members and external others), group psychosocial traits (shared understandings, beliefs or emotions), contextual and cultural characteristics, and leadership (Cohen & Bailey, 1997). The environmental factors used in existing group performance models often refer to characteristics of the external economic and market “environment” in which the organization or company is embedded; the design factors refer to the features of the task, group or organization that can be directly manipulated by managers to create the conditions for effective performance (Cohen & Bailey, 1997; Gladstein, 1984; Gibson, 1999). The effects of the physical environment on collaboration at work tend to be overlooked in theoretical models of group performance.

However, “places” or physical environments where collaborative activity takes place can make a difference. Studies in both management and psychosocial fields recognized the impact of workplace physical environment on occupant satisfaction and performance. For example, based on an extensive interview study of over 100 academics, professionals and managers, Davenport et al. (2002) identified workplace design as one of the three major factors that influence the performance of knowledge workers and knowledge-based organizations. Allen and Henn (2007) argued that “a formal organizational structure may dictate what is supposed to happen, but whether it actually does happen is, in large part, an issue of space”. Figure 1 is a modified model, suggesting the possible role of “workplace spatial settings”. These settings are likely to influence group effectiveness through either their impact on group processes of communication and group psychosocial traits or their direct impact on the outcomes. As Peponis (2004) pointed out, spatial design supports knowledge-work by “sustaining frequent work-related informal interactions”, “allowing visual traces of thought processes to be available in the background of commonly used spaces”, and “providing settings for different work styles”. Based on extensive survey data, Brill and Weidemann of BOSTI Associates (2001) summarized ten workplace qualities that can have the strongest effect on individual and team performance. The top two are the workplace’s ability to support distraction-free individual work and impromptu interactions (both in one’s workspace and elsewhere). The researchers emphasized that each of these two needs should be facilitated without compromising the other.

Figure 1. Modified collaboration effectiveness model

(Adapted from the model by Cohen & Bailey, 1997, by adding the component of “workplace spatial settings”)


3. Hypotheses

A review of existing literature on the relationship between workplace spatial characteristics and occupant perception, behavior, work experience, job satisfaction, and performance showed that the spatial characteristics studied are mostly individual workstation variables, such as workstation density (Dean et al., 1975; Desor, 1972; Fried, 2001; May et al., 2005; Oldham, 1988; Oldham & Fried, 1987), size (Charles & Veitch, 2002), enclosure (Block & Stokes, 1989; Hatch, 1987; Hedge, 1982; Maher & Hippel, 2005; Marans & Spreckelmeyer, 1982; Oldham & Brass, 1979; Oldham & Fried, 1987; Sundstrom et al., 1982; Zalesny & Farace, 1987), and inter-workstation distance (Allen, 1977; Gullahorn, 1952; Kiesler & Cummings, 2002; Kraut et al., 1990, 2002; Oldham & Fried, 1987). Much less is known about the impact of workplace layout and shared spaces on occupant perception, satisfaction and performance.

At the same time, the independent variables examined in these studies tend to focus on individual task-related satisfaction and experience as well as the performance of individual tasks. The impact of the workspace spatial layout on collaboration-related workplace satisfaction, work experience, and effectiveness has not been sufficiently studied. Among the reasons are the lack of tools, measurements, and hypotheses. The development of reliable and easy-to-use tools is critical for user studies of shared spaces in workplaces where interaction and collaborative work constantly take place. Research efforts in this direction include studies introducing space syntax methodology to examine lines of movement and visible fields in the workplace and to explore how spatial layouts affect face-to-face interaction (Brager et al., 2003; Peponis & Wineman, 2002; Rashid et al., 2006) and workplace shared spaces typology study (Hua et al, 2010).

This paper contributes to the methodology and tool development by testing the following hypotheses based on data collected using the new survey tool:

Hypothesis 1: The measures of satisfaction with the collaborative work environment can be reduced to two factors that respectively represent satisfaction with the facilitation from work environment for collaboration (positive) and satisfaction with the level of distraction from various sources at work (negative).

Hypothesis 2: The two-factor can be related to collaboration-related work experience.

Hypothesis 3: Both the two-factor satisfaction measures and collaboration-related work experience can be related to perceived collaboration performance.


4. Methods

4.1. Sites
Field studies were carried out in eleven office buildings in eight US cities. Altogether, 27 workgroups on 22 different floors were studied. Often reflecting the divisions of departments, workgroups were defined based on information obtained both prior to the field studies and on site. A workgroup occupied either an entire floor or part of a floor. The age of the buildings studied varied, which reflects the office building stock in the US. The workgroups were selected based on the nature of their work, which required a certain level of collaboration.

4.2. Participants
All occupants in each workgroup who were present on the days of field study were eligible to participate. Taking into account an even physical distribution of samples on the floor plans, approximately 20% of the occupied workstations in each workgroup were sampled, with as many occupants being contacted as the researchers had time to approach during their time on site. The total 308 participants in this study are white-collar office workers with various job categories, including executive/managerial, professional/technical, and clerical/support. Table 1 shows characteristics of the sample used for statistical analysis. Confidentiality and voluntary participation were emphasized after explaining the study to the occupants. The participants were assured that their responses to the questionnaire would only be used in an aggregated manner for research purposes. Response rates were not formally recorded, but approximately 90% of the occupants who were invited agreed to answer the questionnaires. The numbers of participants working in open-plan cubicles and those working in closed offices by and large reflected the proportion of these two major types of individual workstations in each workgroup studied.

Table 1: Participant characteristics

N Gender % Retention Individual workstation type %
308 Female Male Mean # of months (SD) Open cubicle Closed office
52.3% 47.7% 47.7 (71.4) 82.7% 17.3%

4.3. Measures
Participants were invited to fill out a two-page, paper-based “Workplace Collaborative Environment Questionnaire” consisting of five major sections. The section investigating perceived collaboration performance was added to the questionnaire after the initial launch of the project. The purpose of the questionnaire was to assess white-collar office workers’ opinions of the quality of their workplace’s spatial environment for collaboration, their collaboration-related experience at work, and their perceived collaboration effectiveness. Questions were developed based on a review of the relevant literature and field assessments of workplace spatial design choices that facilitate interaction and collaboration. In addition to the questionnaire, systematic spatial documentation and floor plan analysis were carried out for the work environment of each workgroup studied.

4.3.1. Satisfaction with collaborative work environment
The first section of the questionnaire was intended to assess participants’ satisfaction with their work environment for their collaboration needs. Participants were asked to what extent they agreed with eighteen statements about their satisfaction with various aspects of their work environment related to interaction and collaboration (see Table 2). The statements include the amount, size, location, combination, and availability of various formal and informal collaborative spaces at work, including meeting rooms and/or spaces, common copier and/or printer areas, shared kitchen and/or coffee areas, circulation areas, and individual workstations, as well as the furnishings, tools, and technologies in these spaces. Participants were also asked about their overall satisfaction with the quality of their work environment in terms of whether or not it supported their desired interactions and collaborations for work deliverables. Responses were reported on a five-point Likert scale from “strongly disagree” (1) to “strongly agree” (5).

Table 2: Collaborative environment satisfaction items and rotated two-factor exploratory factor analysis results

Statement Factor #1: Facilitation Factor #2: Distraction
Adequate types of collaborative places at work .81  
Availability of places for collaboration needs .81  
Availability of suitable places for collaboration needs .80  
Availability of meeting spaces .79  
Good variety of collaborative places at work .79  
Arrangement and furnishing of meeting places .70  
Collaboration environment in general .68  
Tools and technology in meeting places .58  
Different-sized meeting rooms .50  
Distraction from common copier/printer areas   .77
Reluctance to talk in kitchen/coffee areas   .73
Distraction from kitchen/coffee areas   .69
Distraction from circulation areas   .60
Reluctance to talk in common copier/printer areas   .55
Distraction from meeting places   .50
Enough space in individual workstation for meetings    
Arrangement and furnishing of kitchen/coffee areas    
Enough space in circulation areas for conversations    
% of variance explained by each factor 28.3 16.1
Cronbach’s α .90 .79

Note: Factor loading cut-off = .40.


4.3.2. Spatial preference for interaction and collaboration
The second set of items in the questionnaire was designed to understand participants’ preferred spaces in their current work environment for interaction and collaboration with their co-workers. The questionnaire provided participants with a list of different spaces at work and asked them to indicate all places that they prefer to use for work-related collaboration and for casual interaction with co-workers. If they use places other than those listed, they were asked to identify those places. However, this part of the data was not used in this paper’s analyses.

4.3.3. Collaboration-related work experience
This measure consisted of twelve statements intended to assess participants’ satisfaction with their interactive and collaborative work experience (see Table 3). Participants were asked to what extent they agreed with these statements on a five-point Likert scale from “strongly disagree” (1) to “strongly agree” (5). Based on the literature and field experience, items assessing collaboration experience including access to necessary information, group and task awareness, collective learning, trust and commitment, perceived openness, and available channels for communication.

Table 3: Work experience items and one-factor principal factor analysis results

Statement Collaboration Experience
Line of trust .70
Feel well-informed .69
Informed in time .67
Easy to communicate face-to-face .66
Easy to get help .66
Always feel learning .64
Frequently communicate face-to-face .62
Clear picture of co-worker expertise .61
Commitment to department .61
Know department’s future plans .59
Have information needed for work .59
Collaboration valued in department .54
% of variance explained 40.1
Cronbach’s α .89

Note: Factor loading cut-off = .40.

4.3.4. Perceived collaboration effectiveness
A four-item scale of perceived collaboration effectiveness, adopted from Dabbish’s (2006) survey of communication on the job, was added to the “Workplace Collaborative Environment Questionnaire” at a later stage of this study. Participants in three buildings (a total of 84 participants on seven different floors) responded on a five-point Likert scale from “strongly disagree” (1) to “strongly agree”) (5), to the following statements: “we work together in a well-coordinated fashion”, “we have very few misunderstandings about what to do”, “we accomplish tasks smoothly and efficiently”, and “there is much confusion in the way we accomplish tasks”.

4.3.5. Demographic characteristics
Five questions recorded participants’ demographic characteristics, including gender, age, nature of work, job category, and retention. These variables were used as control variables in a multi-level regression analysis on workplace spatial impact on collaborative performance, which was not reported in this paper.

4.3.6. Spatial settings
Five spatial characteristics of individual workstations (workstation type, size, partition height, distance to co-worker and workstation density) and six workgroup-level spatial variables (distance to meeting space, common copier/printer area and kitchen/coffee area, openness of floor plan, and percentages of floor space dedicated to meetings and casual interactions) were recorded through systematic spatial documentation and floor plan analysis. Again, these spatial data were not used in the analyses reported here.

4.4. Analyses
In this study, Hypothesis 1 was tested using exploratory factor analysis and confirmatory factor analysis. Structural equation modeling was carried out to test Hypotheses 2 and 3. Hypotheses 1 and 2 were tested based on the full sample; Hypothesis 3 was tentatively tested based on the part of the sample that has data on perceived collaboration performance.

Univariate outliers with standardized scores more than three standard deviations from the mean were identified and excluded from the analyses. Multivariate outliers, which were also excluded, were those with Mahalanobis distance statistics p<.001. Univariate normality was assessed using the skewness index (<3) and the kurtosis index (<10). Full information maximum likelihood estimation was used in the presence of missing data. The data sets were also examined for multicollinearity (correlations >.80).

The model fit for confirmatory factor analysis and structural equation modeling was assessed using multiple fit indices, including Chi-square, Chi-square/degree of freedom, root mean square error of approximation (RMSEA), comparative fit index (CFI), and Bentler-Bonett Normed fit index (NFI).

5. Results

5.1. Hypothesis 1: Factor structure of environmental satisfaction measures

5.1.1. Data screening
        For the collaborative environment satisfaction data, no univariate outliers were identified; one multivariate outlier was identified and excluded from the analyses. The mean, standard deviation, skewness and kurtosis of all items are shown in Table 4. The entire sample was randomly split into two independent subsets for exploratory and confirmatory factor analysis respectively. The distribution of site, floor, gender, age, job category, and retention was equivalent in the two subsets.

Table 4: Descriptive statistics of all variables

Variable Mean StDev Skewness Kurtosis
Perceived Support (N=307)
Variety 3.13 1.11 -0.13 -1.02
Adequate 3.25 1.03 -0.29 -0.77
Available 3.16 1.03 -0.06 -0.93
Suitable 3.14 1.03 -0.15 -0.82
Diff_mtg 3.49 1.03 -0.41 -0.80
Mtg_avlb 2.78 1.07  0.36 -0.73
Arr_mtg 3.38 0.95 -0.32 -0.35
Tools_mtg 3.37 1.02 -0.35 -0.53
Support_envir 3.17 0.97 -0.47 -0.37
Distr_mtg 3.10 1.21 -0.04 -1.04
Distr_com 2.81 1.15  0.28 -0.82
Notlk_com 3.36 1.03 -0.29 -0.67
Distr_ktch 2.31 1.00  0.53 -0.13
Notlk_ktch 2.75 1.15  0.11 -0.91
Distr_circu 3.10 1.09  0.01 -0.74
Work Experience (N=278)
Informed 3.52 0.94 -0.53 -0.28
Future 3.25 1.03 -0.24 -0.64
Intime 3.20 1.00 -0.13 -0.75
Expertise 3.79 0.82 -0.54 -0.05
Learning 4.03 0.72 -0.59  0.60
Info_need 3.25 0.92 -0.08 -0.76
Help 3.89 0.76 -0.58  0.39
Trust 3.77 0.78 -0.43  -0.03
Commit 4.11 0.75 -0.61  0.22
Ease f-to-f 4.14 0.65 -0.46  0.67
Freq f-to-f 4.14 0.62 -0.19 -0.07
Collab_value 3.97 0.81 -0.40 -0.38
Perceived Performance (N=80)
Well_coord 3.98 0.66 -0.53  1.08
Few_misund 3.70 0.76 -0.52  0.16
Efficiency 3.68 0.73 -0.64  0.36
Confusion 3.76 0.85 -0.55 -0.09

5.1.2. Exploratory factor analysis
Subset 1 (N=160) was used for exploratory factor analysis in order to identify latent variables of satisfaction with collaborative work environment. The Pearson correlation of the eighteen items disproved the existence of multicollinearity. The largest correlation coefficient observed was 0.799, occurring between “I can always find a place to carry out collaborative work when I need it” and “I can always find suitable place for certain type of conversation or collaborative work”. All eighteen items were included in an exploratory factor analysis to identify the possible latent constructs.

The exploratory factor analysis with oblique rotation identified two factors with Eigenvalue greater than 1 (5.25 and 2.52), which were retained for a two-factor solution. The scree plot also supported this solution. The two factors were labeled “perceived facilitation from work environment for collaboration” and “perceived distraction from various sources at work”. The two factors together accounted for 44.4% of the total variance observed. A loading of .40 or greater was used as the criteria for items to be included in a factor; no cross-loading was observed. Nine items, which shared the conceptual meaning of perceived facilitation from work environment for collaboration, loaded significantly on the first factor; and six items, indicating perceived distraction from various sources in work environment, loaded significantly on the second factor. There were three items that did not load significantly on either factor; therefore, they were dropped from the solution. The internal consistency, indicated by Cronbach’s α, was satisfactorily high for both factors (Table 2).


5.1.3. Confirmatory factor analysis
Subset 2 (N=147) was used for confirmatory factor analysis, which was conducted based on the two-factor model from the exploratory factor analysis. The model (Model 1) consisted of 15 observed variables with two latent variables (factors). Each observed variable was hypothesized to load on one single factor, with 9 variables loading on “facilitation” and 6 variables loading on “Distraction”. The two factors were hypothesized to be correlated. In this model, there was no correlation among the measurement error variables. Overall, Model 1 contained maximum likelihood estimations of 15 factor loadings, one factor variance, and 15 measurement error variances.

Results of the CFA analysis indicated a marginal fit between the model and the data (Table 4: Model 1-Sample 2). All the factor loadings were statistically significant. Modification indices suggested adding covariance between the following measurement error variables: E_1-1 and E_1-3, E_1-4 and E_1-5, E_1-4 and E_1-8, E_1-10 and E_1-11. Model 2 was estimated accordingly (Figure 2), and the results from the CFA showed a better model fit (Table 4: Model 2-Sample 2). Using Model 2 on the full sample also demonstrated a good fit between the model and the data and thus suggested the final model of factor structure for the collaborative environment satisfaction items. Figure 1 shows the standardized output of Model 2.

Table 5: Fit statistics of the confirmatory factor analysis

  X2 X2/df  RMSEA NFI  CFI
    <3 < .10  >.90 >.90
Model 1 - Sample 2
224.0  2.517  .102  .782 .851
Model 2 - Sample 2
149.9  1.764 .072 .854 .928
Model 2 - Full sample
186.5  2.194 .062 .904 .944


Figure 2: AMOS output of CFA Model 2 with standardized estimates (Full sample N=307)


5.2. Hypothesis 2– Structural equation modeling

5.2.1. Data screening
In the collaboration experience data, 7 cases had missing data on all items and were excluded from the analysis. 18 univariate outliers and 3 multivariate outliers were identified and excluded from the analyses. Table 4 shows the mean, standard deviation, skewness and kurtosis of all variables. Pearson correlation analysis did not provide evidence of multicollinearity. The largest correlation coefficient observed was 0.704, occurring between “I feel well-informed about the current activities in my department” and “I have a clear idea of future development steps of my department”. All twelve items were included in the structural equation modeling.

5.2.2. Collaboration experience factor structure
One factor with an Eigenvalue of 4.81 was identified for the 12 collaboration-related work experience items. The factor accounted for 40.1% of the total variance observed. Using a loading of .40 or greater as the criteria, all 12 items loaded significantly on this factor (Table 3). Although accepting an additional factor could add 9.8% to the variance explained, it would cause cross-loading. Adding complexity did not significantly contribute to increasing the variance explained; therefore, the one-factor model was retained. The internal consistency of the 12 items, indicated by Cronbach’s α, was satisfactorily high at 0.89.

5.2.3. Structural equation modeling
In order to test Hypothesis 2, a structural equation model was constructed with two inter-correlated factors (“facilitation” and “distraction”, as shown in Figure 1) and unidirectional paths from each factor to the collaboration-related work experience.

Results of the confirmatory factor analysis indicated a marginal fit between the model and the data (Table 6: Model 3). All the factor loadings were statistically significant. Modification indices suggested adding covariance between the following measurement error variables: E_2-1 and E_2-2; E_2-1 and E_2-3; E_2-2 and E_2-3; and E_2-10 and E_2-11. Model 4 was estimated accordingly (Figure 3). The results showed a better model fit (Table 6: Model 4). The “facilitation” factor is significantly and positively associated with collaboration experience, with a correlation coefficient of 0.23; “distraction” is significantly and negatively associated with collaboration experience, with a correlation coefficient of -0.27.

Table 6: Fit statistics for structural equation modeling – RQ2

    <3 < .10 >.90 >.90
Model 3 – Full sample
896.0 2.827  .081 .747 .817
Model 4 – Full sample
613.7  1.961 .059 .827 .905

Figure 3: AMOS output of SEM Model 4 with standardized estimates


5.3. Hypothesis 3 – structural equation modeling
The four perceived collaboration performance items were added to the questionnaire after the initial launch of the study. Eighty-four participants responded to this part of the questionnaire. Eighty cases with complete data, no outliers and satisfactory normality (no excessive skewness or kurtosis) were used for the analysis of all variables. The Pearson correlation matrix revealed no evidence of multicollinearity. The largest correlation coefficient observed was 0.600, occurring between “We work together in a well-coordinated fashion” and “We have very few misunderstandings about what to do”.

All four items, including three positive statements and one negative statement reversed in the analysis, all loaded on a single factor that explained 45.2% of the total variance. The factor loadings were 0.721, 0.761, 0.733 and 0.412. Reliability was at a satisfactorily high level, with Cronbach’s α at 0.77.

In order to test Hypothesis 3, a structural equation model was constructed based on Model 4, with additional unidirectional paths from the two factors of satisfaction with collaborative work environment (“facilitation” and “distraction”) to perceived performance and a unidirectional path from collaborative work experience to perceived performance (Figure 4). According to the results of the structural equation model, a marginal fit was identified, possibly related to the small sample size for this part of the analysis. Model fit statistics were reported in Table 7. All paths were significant with a small correlation coefficient except for those from facilitation to perceived performance and from distraction to performance. However, eliminating these two paths did not improve the fit of the model.

Table 7: Fit statistics for structural equation modeling – RQ3

    <3 < .10 >.90 >.90
Model 5 (N=80)  667.6  1.590  .086  .594  .784



The “Workplace Collaborative Environment Questionnaire” is a survey instrument designed to understand office workers’ satisfaction with their collaborative work environment, satisfaction with collaboration experience at work, and their perceived collaboration effectiveness. The environmental satisfaction items evaluate occupant satisfaction with diverse spaces at work, including individual workstations and various shared spaces. These items were hypothesized to be meaningfully reduced to a smaller number of latent variables. The satisfaction scales were further hypothesized to relate to work experience and perceived performance.

Hypothesis 1 was supported by the results from exploratory and confirmatory factor analyses, which identified and confirmed a two-factor structure of the satisfaction items. This structure indicated both perceived facilitation of collaboration (9 items) and distraction (6 items) in the work environment. Three items did not significantly load on either factor and were thus dropped from the model. The fit of the confirmatory factor analysis model (Model 2) was good, with all paths statistically significant. The identification of the covariance between several measurement error variables (E_1-1 and E_1-3; E_1-4 and E_1-5; E_1-4 and E_1-8; and E_1-10 and E_1-11) improved the model fit. The two scales, facilitation and distraction, were negatively associated with each other,

Figure 4: AMOS output of SEM Model 5 with standardized estimates


with the coefficient of -0.41. The purpose of this 15-item satisfaction with the collaborative work environment measure was to gather evidence of occupant preference for shared spaces in the workplace. The two satisfaction scales reflected the intrinsic need for the work environment to support both interaction and concentration (Brill and Weidemann, 2001). It can be used to evaluate the extent to which both needs are fulfilled.

Hypothesis 2 was supported by results from the structural equation modeling. The model fit (Model 4) was satisfactory according to most goodness-of-fit indices, with all paths in the model statistically significant. The scale of perceived facilitation from the work environment for collaboration was positively associated with collaboration-related work experience, with the coefficient of 0.23. Perceived distraction was negatively associated with work experience, with the coefficient of -0.27.

Hypothesis 3 was not support by the data collected in this study. The model (Model 5) demonstrated a marginal fit. All paths were significant, except those from facilitation to perceived performance and from distraction to performance, with small correlation coefficients (0.07 and -0.06, respectively). The small sample site for this step of the analysis allowed only a preliminary test of the hypothesis.



Shared spaces in the workplace were recognized by researchers in the managerial and architectural fields as critical spots for interaction and collaboration (Allen & Henn, 2007; Davenport, 2002). However, despite their importance, little is known about the impact of the design and management of shared spaces on occupants and their collaboration effectiveness. Because of the insufficient understanding of shared spaces at work, workplace designers are experimenting with different approaches. Nevertheless, such practices depend heavily on fashion or faith. The lack of reliable tools remains a barrier. The development and field test of this tool will help fill this gap and support workplace design and reengineering. It can be used to evaluate particular workplace settings, generate knowledge of occupant preferences and explore supportive environments for collaboration.

On the other hand, group effectiveness is a broad concept, and measuring it is a challenge for both research and organizational practice. The performance outcomes, which can be assessed in terms of the quantity and quality of outputs, are an important part of group effectiveness, which also includes group member attitudinal outcomes, such as group member job satisfaction, trust and commitment, and behavioral outcomes such as absenteeism and turnover (Cohen and Bailey, 1997). The objective measures of group performance used in both literature and practice are summarized in Table 8 for two common types of white-collar teams.

One of the few efforts to quantify the direct impact of spatial work settings on measured group performance, Teasley et al. (2000) carried out an experiment in a Fortune 50 automobile company to test a team room setting, also known as “radical collocation”. Radical collocation is an extreme work setting, in which all team members and their artifacts are in one room for the duration of a certain project. The experiment showed that radical collocation had a positive impact on group performance. The performance measures used in the study showed that radically collocated teams produced twice as much as the company baseline in cycle time and function points per staff month. The teams in project rooms completed jobs in about a third of the amount of time it took those in standard cubicle settings. However, the change of setting was accompanied by changes in managerial strategy, which not only complicates the study’s results but also demonstrates the difficulties in measuring group performance.

Table 8. Objective group performance measures for project teams and management teams
(Adapted from Cohen & Bailey, 1997)

Type of teams Team characteristics Objective measures of team performance
Project team
  • Produce one-time outputs, such as a new product or service;
  • Involve considerable application of knowledge, judgment and expertise;
  • Multi-discipline, cross-functional
  • Adherence to budgets;
  • Adherence to schedules (time to complete the project);
  • Innovation;
  • Project quality;
  • Overall performance or efficiency
Management team
  • Coordinate and provide direction to sub-units;
  • Responsible for the overall performance of a business unit;
  • Collective expertise, integrating disparate efforts and sharing responsibility
  • Return on equity/assets;
  • Sales growth;
  • Total return to shareholders
  • Decision quality


Group performance is not only difficult to measure, but it is also context dependent. Indices used to describe collaboration are often different from case to case and across different industries, organizations, and job types. As a result, it is difficult for a study of multiple workplace settings to compare collaboration performance between organizations. Finding performance measures that are relevant for various tasks at different organizations remains a challenge. In some cases, self-rated and supervisor-related performance has been used (Oldham, 1988), but they have been used mostly for individual tasks. Based on empirical studies, Gittell (2002) claimed that perceived relational coordination effectiveness has a significant positive association with measured quality performance.

Gittell’s claim reflects the role that frequent, timely, accurate, and problem-solving-based communication plays in the process of coordination. Based on Gittell’s study, and facing the practical difficulty of identifying and obtaining common performance measures for various organizations, this study adopted self-perceived coordination effectiveness as a collaboration performance indicator, but integrating objective group performance measures into the tool remains a research goal.



Allen, T.J. (1977). Managing the Flow of Technology. MIT Press: Cambridge, MA.

Allen, T., & Henn, G. (2007). The Organization and Architecture of Innovation: Managing the Flow of Technology. Elsevier: Oxford, UK.

Block, L. & Stokes, G. (1989). Performance and satisfaction in private versus nonprivate work settings. Environment & Behavior, Vol.22, 277-297.

Brager, G., Heerwagen, J., Bauman, F., Huizenga, C., Powell, K., Ruland, A., & Ring, E. (2000). Team Spaces and Collaboration: Links to the Physical Environment. University of California, Berkeley, Center for Built Environment.

Brill, M., Weidemann, S. & the BOSTI Associates. (2001). Disproving Widespread Myths about Workplace Design. Kimball International: Jasper, IN.

Carlopio, J.R. (1996). Construct validity of a physical work environment satisfaction questionnaire. Journal of Occupational Health Psychology, Vol.1, No.3, 330-344.

Carlopio, J.R. & Gardner, D. (1992). Direct and interactive effects of the physical work environment on attitudes. Environment & Behavior, Vol.24, No.5, 579-601.

Charles, K. E., & Veitch, J. A. (2002). Environmental satisfaction in open-plan environments: 2. Effects of workstation size, partition height and windows.Report for National Research Council of Canada, Institute for Research in Construction.

Cohen, S.G., & Bailey, D.E. (1997). What Makes Teams Work: Group Effectiveness Research from the Shop Floor to the Executive Suite. Journal of Management, Vol.23, No.3, 239-290.

Davis, T. R. V. (1984). The influence of the physical environment in offices. The Academy of Management Journal, Vol.9, No.2, 271-283.

Davenport, T.H., Thomas, R.J., & Cantrell, S., (2002). The Mysterious Art and Science of Knowledge-Worker Performance, MIT Sloan Management Review, Fall 2002, 23-30.

Dean, L. M., Pugh, W. M., & Gunderson, E. (1975). Spatial and perceptual components of crowding: Effects on health and satisfaction. Environment and Behavior, Vol.7, 225–236.

Desor, J.A. (1972). Towards a psychological theory of crowding. Journal of Personality and Social Psychology, 21, 79-83.

Evans, G. W., Johansson, G. & Carrere, S. (1994). Psychosocial factors and the physical environment: Inter-relations in the workplace. In C. L. Cooper & I.T. Robertson (Eds.), International review of industrial and organizational psychology, Vol.9, pp1-30, John Wiley: New York, NY.

Fried, Y., Slowik, L. H., Ben-David, H. A. & Tiegs, R. B. (2001). Exploring the relationship between workspace density and employee attitudinal reactions: An integrative model. Journal of Occupational and Organizational Psychology, Vol.74, 359-72.

Gladstein, D.L. (1984). Groups in context: A model of task group effectiveness. Administrative Science Quarterly, 29, 499-517.

Gibson, C.B. (1999). Do they do what they believe they can? Group efficacy and group effectiveness across tasks and cultures. Academy of Management Journal, Vol.42, No.2, 138-152

Gonzalez, R., & Griffin, D. (2001). Testing parameters in structural equation modeling: Every “one” matters. Psychological Methods, 6, 258-269.

Gullahorn, J. (1952). Distance and friendship as factors in the gross interaction matrix. SociometryVol.15, No.1/2, 123-134.

Hatch, M. J. (1987). Physical barriers, task characteristics, and interaction activity in research and development firms. Administrative Science Quarterly, Vol.32, No.3, 387-399.

Hedge, A. (1982). The open-plan office: A systematic investigation of employee reactions to their work environment. Environment & Behavior, Vol.14, No.5, 519-542.

Heerwagen, J. H., Kampschroer, K., Powell, K. M. & Loftness, V. (2004). Collaborative Knowledge Work Environments. Building Research & Information, Vol.32, No.6, 510-528.

Hua, Y., Loftness, V., Kraut, R., & Powell, K. (2010). Workplace collaborative space layout typology and occupant perception of collaboration environment. Environment and Planning B: Planning and Design, 37, 429-448.

Ilozor, B. D., Lover, P. E. D. & Treloar, G. (2002). The impact of work settings on organizational performance measures in built facilities. Facilities, Vol.20, No.1/2, 61-68.

Kiesler. S. & Cummings, J. (2002). What Do We Know about Proximity and Distance in Work Groups? A Legacy of Research. In P. Hinds & S. Kiesler (Eds.), Distributed Work (pp57-80). MIT Press: Cambridge, MA.

Kraut, R. E., Egido, C. and Galegher, J. (1990). Patterns of contact and communication in scientific research collaboration. In J. Galegher, R. E. Kraut & C.

Egido (Eds.), Intellectual Teamwork, Lawerence Erlbaum and Associates Publishers: Hillsdale, NJ.

Kraut, R. E., Fussell, S. R., Brennan, S. E. & Siegel, J. (2002). Understanding Effects of Proximity on Collaboration: Implications for Technologies to SupportRemote Collaborative Work. In P. Hinds & S. Kiesler (Eds.), Distributed Work (pp137-162). MIT Press: Cambridge, MA.

Lawler, E. (2001). Organizing for High Performance: Employee Involvement, TQM, Re-engineering and Knowledge Management in the Fortune 1000.

Jossey-Bass: New York, NY.

Loftness, V., Aziz, A., Mondazzi, M., Moustapha, H., Yoon, H. C., Atkinson, M., Bennett, L., Blumberg, C., Chapman, M., Heerwagen, H., Horn, D.,Kampschroer, K., & Pero, K. (2002). Collaborative Work Settings. Report for the GSA Productivity Protocol Development – The Drivers and Values of Collaboration.

Mercer, A. (1979). Office environments and clerical behavior. Environment and Planning, Vol.6, 29-39.

Maher, A. & C. von Hippel (2005). Individual differences in employee reactions to open-plan offices. Journal of Environmental Psychology, Vol.25, 219-29.

Marans, R. W. & Spreckelmeyer, K. F. (1982) Evaluating open and conventional office design. Environment and Behavior, Vol.14, No.3, 333-351.

May, D. R., Oldham, G. R. & Rathert, C. (2005). Employee affective and behavioral reactions to the spatial density of physical work environments. Human Resource Management, Vol.44, No.1, 21-33.

Nadler, D., Tushman, M., & Nadler, M. B. (1997). Competing by Design: The Power of Organizational Architecture. Oxford University Press: UK.

Oldham, G. R. (1988). Effects of changes in workspace partitions and spatial density on employee reactions: A quasi-experiment. Journal of Applied Psychology, Vol.73, No.2, 253-358.

Oldham, G. R. & Brass, D. J. (1979). Employee reactions to an open-plan office: A naturally occurring quasi-experiment. Administrative Science Quarterly, Vol.24, No.2, 267-284.

Oldham, G. R. & Fried, Y. (1987). Employee reactions to workspace characteristics. Journal of Applied Psychology, Vol.72, No.1, 75-80.

Oldham, G. R.,Kulik, C. T. & Stepina, L. P. (1991). Physical environments and employee reactions: Effects of stimulus-screening skills and job complexity. The Academy of Management Journal, Vol.34, No.4, 929-938.

Oldham, G. R. & Rotchford, N. L. (1983). Relationships between office characteristics and employee reactions: A study of the physical environment. Administrative Science Quarterly, Vol.28, No.4, 542-556.

Peponis, J. (2004). Space syntax. InformeDesign Newsletter, Vol.4, Issue 12.

Peponis, J. & Wineman, J. (2002). Spatial structure of environment and behavior. In R. B. Bechtel & A. Churchman (Eds.) Handbook of Environmental Psychology. Wiley: Hoboken, NJ.

Rashid, M., Kampschroer, K., Wineman, J. & Zimring, C. (2006). Spatial layout and face-to-face interaction in offices – a study of the mechanisms of spatial effects on face-to-face interaction. Environment and Planning B: Planning and Design, Vol.33, 825-844.

Sundstrom, E., Burt, R.E., & Kamp, D. (1980). Privacy at work: Architectural correlates of job satisfaction and job performance. The Academy of Management Journal, Vol.23, No.1, 101-117.

Sundstrom, E., Herbert, R.K., & Brown, D.W. (1982). Privacy and Communication in an open-plan office: A case study. Environment and Behavior, 14, 379-92.

Teasley, S., Covi, L., Krishnan, M. S., and Olson, J. S. (2000). How does radical collocation help a team succeed? In Proceedings of CSCW’00 (p339-346). ACM Press: New York, NY.

Zalesny, M. D. & Farace, R. V. (1987). Traditional versus open offices: A comparison of sociotechnical, social relations, and symbolic meaning perspective. The Academy of Management Journal, Vol.30, No.2, 240-259.

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

ISSN: 2150-3303