Human resource management researchersand practitioners recognize the importanceof the physical environment in the work-place and have called for more research onhow this environment influences employeereactions to their work (Carlopio, 1996;McCoy, 2002; Oldham, Cummings, & Zhou,1995; Sundstrom, 1987). One important di-mension of the physical environment is thespatial density of employees work areas.Studies have demonstrated that spatiallydense areas (i.e., those with little spaceavailable per person) generally have negativeeffects on employee reactions (see Oldhamet al., 1995 for a review). The purpose of thecurrent study is to extend this previous workin several ways, including examining em-ployee reactions that have received little pre-
vious research attention (e.g., tardiness andtransfer intentions) and examining the spe-cific conditions under which spatial densityhas its strongest (and weakest) relations tothese important workplace reactions.
Physical Work Environments and theSocial Interference Perspective
The social interference perspective (see Baum& Paulus, 1987; Evans, Johannson, & Car-rere, 1994; Oldham et al., 1995) suggests thatindividuals often react negatively to densework area conditions because (a) they are un-able to control their interactions with othersand (b) they have more difficulty achievingtheir immediate goals. That is, dense condi-tions tend to increase the number of un-
EMPLOYEE AFFECTIVE AND BEHAVIORALREACTIONS TO THE SPATIAL DENSITY OFPHYSICAL WORK ENVIRONMENTS
Human Resource Management, Spring 2005, Vol. 44, No. 1, Pp. 2133 2005 Wiley Periodicals, Inc. Published online in Wiley InterScience (www.interscience.wiley.com).DOI: 10.1002/hrm.20038
Douglas R. May, Greg R. Oldham, and Cheryl Rathert
This field study of a medical clinic found that employees in spatially dense work areas (i.e., thosewith little space available per person) experienced higher levels of perceived crowding, transferintentions, and tardiness, as well as lower work area satisfaction, than employees in low-densityareas. Crowding perceptions explained the relations between spatial density and the measuresof work area satisfaction and tardiness. Finally, when employees had high workloads and theirjobs required physical movement, spatial density had weaker relations to crowding perceptionsand area satisfaction than in other conditions. Implications of these findings for human resourcepractitioners are discussed. 2005 Wiley Periodicals, Inc.
Correspondence to: Douglas R. May, Department of Management, University of NebraskaLincoln, Lincoln,NE 68588-0491, (402) 472-8885, email@example.com
22 HUMAN RESOURCE MANAGEMENT, Spring 2005
wanted, unpredictable interactions employeesexperience. The presence of many others alsomay restrict freedom of movement and boostcoordination demands, making it difficult forindividuals to accomplish their activities or ob-jectives. Thus, individuals in dense conditionsmay become overloaded, report feelingcrowded and dissatisfied with their work areas,and may engage in activities that reduce thisfrustration (e.g., being late to work).
The social interference perspective alsois consistent with the argument that percep-tions of crowding explain the relations be-tween spatial density and individuals behav-iors (Baum & Paulus, 1987). Whereas spatialdensity is an objective variable reflecting theamount of space available per person, crowd-ing is an experiential state. Employees whofeel crowded may try to alleviate this per-ceived spatial restriction with certain with-drawal behaviors. Thus, spatial density is anecessary, though not sufficient, conditionfor the feeling of crowding (Stokols, 1972).Consistent with the social interference per-spective, research on residential settingsshows that crowding is associated with lessperceived control (Pandey, 1999). The cur-rent research examines whether crowding ex-plains the relations between the spatial den-sity of employees work areas and theiraffective and behavioral reactions.
Previous studies of employees in workorganizations have provided results that aregenerally consistent with the social interfer-ence perspective (see Oldham et al., 1995).For example, research has shown that em-ployees in dense conditions feel morecrowded and dissatisfied with their workareas than those in conditions of low density(Oldham, 1988; Oldham & Rotchford,1983). In addition, studies have demon-strated that density is associated withturnover intentions (Dean, Pugh, & Gunder-son, 1975) and with withdrawal from thework area during discretionary break periods(Oldham & Rotchford, 1983).
The Physical Work Environment of HealthCare Organizations
The current study extends this previous workin four ways. First, we examine the physical
environment of an outpatient medical clinic,a setting that few studies have explored de-spite its importance to the delivery of healthcare. In particular, we focus on the reactionsof the medical receptionists who process in-coming and outgoing patients. These em-ployees are critical to the effective operationof clinics that service large numbers of pa-tients each year in the health care industry.Indeed, health care research shows that pa-tient satisfaction is linked to the satisfactionof health care employees (Meyer & Massagli,2001) and that working conditions are re-lated to medical errors in such settings(Leape, 1999; Leape et al., 1995).
Second, we examine relations betweenthe spatial density of work areas and potentialemployee reactions derived from the socialinterference perspective. Specifically, em-ployees affective (i.e., crowding perceptions,work area satisfaction, transfer intentions)and behavioral (i.e., tardiness) reactions areinvestigated in this study. No previous studyhas examined the relations between workarea spatial density and employee tardinessand transfer intentions. Such a study is par-ticularly important in a medical group clinicsetting given that medical receptionists areplaying an increasingly important role inmedical care (Patterson, Del Mar, & Najman,2000). Based on the social interference per-spective discussed above, employees whowork in areas with little space should experi-ence many unpredictable interruptions andrestrictions of their goals, which, in turn, leadthem to feel crowded and dissatisfied withtheir work areas and behaviorally withdraw.Thus, the first formal hypothesis of the studyis offered below:
Hypothesis 1: Employees in dense workareas will react more negatively (i.e., expe-rience greater crowding, lower work areasatisfaction, be late to work more fre-quently, and report stronger transfer inten-tions) than employees in low-density areas.
Third, we examine whether crowding ex-plains (i.e., mediates) relations between workarea spatial density and employee affectiveand behavioral reactions. Research in the en-vironmental psychology literature suggests
Thus,individuals indense conditionsmay becomeoverloaded,report feelingcrowded anddissatisfied withtheir workareas, and mayengage inactivities thatreduce thisfrustration (e.g.,being late towork).
Employee Affective and Behavioral Reactions to the Spatial Density of Physical Work Environments 23
The currentstudycontributes tothis limitedbody ofknowledge andmore narrowlyexamineswhethercrowdingperceptionsexplain therelationsbetween workarea densityand work areasatisfaction,tardiness, andtransfer intent.
that perceptions of crowding play a signifi-cant role in individuals reactions to theirphysical environments (see Baum & Paulus,1987); however, only a few studies have em-pirically explored the mediating effects ofcrowding and employee reactions in theworkplace. One such investigation (Oldham& Rotchford, 1983) did find that three envi-ronmental experience variables (i.e., crowd-ing, concentration, and privacy) explainedthe relations between several office charac-teristics (i.e., openness, office density, workarea density, accessibility, and darkness) andoffice satisfaction. Another study by Carlopioand Gardner (1995) found that workplaceexperiences (i.e., privacy and crowding) me-diated the relations between job level andemployees work environment satisfactionand turnover intentions. The current studycontributes to this limited body of knowledgeand more narrowly examines whether crowd-ing perceptions explain the relations betweenwork area density and work area satisfaction,tardiness, and transfer intent. Based on thetheory and research reviewed, the followinghypothesis is offered:
Hypothesis 2: Employees perceptions ofcrowding will mediate the relations be-tween work area spatial density and em-ployees affective and behavioral reactions.
Fourth, this investigation contributes tothe literature by examining the specific con-ditions under which spatial density has itsstrongest (and weakest) relations to employeereactions. Past research has explored suchmoderating factors as job level and complex-ity, organizational tenure, and stimulusscreening ability (e.g., Carlopio & Gardner,1992; Fried, Slowik, Ben-David, & Tiegs,2001; Oldham, Kulik, & Stepina, 1991). Re-search from social and environmental psy-chology suggests another condition thatmight significantly affect employees reac-tions to spatial densitythe degree to whichindividuals are required to physically moveabout the dense environment (Heller, Groff,& Solomon, 1977; Schopler & Stockdale,1977). Individuals who are required to moveabout dense work areas in order to completetheir tasks are likely to encounter more inter-
personal obstructions than those who are not.These obstructions, in turn, should reduceopportunities for activity and goal accom-plishment and cause individuals to exhibitgenerally negative reactions.
The current study extends this previouswork by examining whether the effects ofphysical movement on employee reactions tospatial density are amplified by the amount ofwork required of employees. Specifically, it isproposed that employees will react more neg-atively to dense conditions when their workrequires considerable physical movement andthey must complete large amounts of suchwork (i.e., high workload). Based on the so-cial interference perspective described ear-lier, individuals should encounter the mostinterferences and disruptions in such condi-tions. When individuals are required to com-plete large amounts of work that requiresphysical movement, the presence of manyothers may actively restrict the freedom ofmovement necessary for the completion ofgoals or activities (Saegert, 1978), which canlead to employees experiencing high levels offrustration and exhibiting negative reactions.
Alternatively, an information process-ing or attention perspective based on em-ployees limited cognitive capabilities (March& Simon, 1958) maintains that when workdemands are high, individuals may be forcedto detach themselves from their surround-ings and focus their limited attention on thework itself. Indeed, Wells and Matthewss(1994) review of the literature suggests thatone of the few consistent effects of arousingstressors which generalize across differentsources of stress is narrowing of attention(italics added) (p. 187). According to thisview, the stress resulting from high work de-mands may cause individuals to focus on se-lective task-relevant attributes and excludeother cues in the environment from theirlimited attention. Such a detachment hy-pothesis (Sutton & Rafaeli, 1987) suggeststhat when work demands are high, employ-ees will tend to focus their attention on thework and will ignore the social interferencespresent in the environment. As a result,when work demands are high, dense workareas should have generally weak or negligi-ble effects on employees reactions.
24 HUMAN RESOURCE MANAGEMENT, Spring 2005
This study compares and contrasts thesocial interference and detachment per-spectives described above. The interferenceperspective (Hypothesis 3a) suggests thatdense conditions should have strong nega-tive effects on employee reactions whentheir work requires substantial physicalmovement and they must complete largeamounts of such work. By contrast, the de-tachment perspective (Hypothesis 3b) sug-gests that dense conditions should haveweak or negligible effects when their work-load and physical movement are high. Basedon these perspectives, the third set of hy-potheses is offered below:
Hypothesis 3a: When employees have highworkloads and their work requires sub-stantial physical movement, spatial den-sity should have stronger negative rela-tions to employee reactions than underother combinations of workload and phys-ical movement.
Hypothesis 3b: When employees havehigh workloads and their work requiressubstantial physical movement, spatialdensity should have weaker negative rela-tions to employee reactions than underother combinations of workload and phys-ical movement.
The theoretical framework for the studyis summarized in Figure 1.
The research design employed in this studywas a cross-sectional field study that usedboth employee and supervisor question-naires and organizational archival records indata collection.
Research Setting and Participants
The research was conducted in 43 units ofan outpatient medical clinic. The partici-pants were receptionists whose responsibili-ties involved processing incoming and out-going patients. Data were collected from182 receptionists, all but one of whom wasfemale. Their ages ranged from 20 to 68,with a median of 29. The median organiza-tional tenure level was 34 months, and themedian education level was some technicalschool experience.
Data were gathered onsite by the first author.A questionnaire was administered to groups
Figure 1. Theoretical Framework for the Effect of Work Area Spatial Density on Employee Reactions.
Employee Affective and Behavioral Reactions to the Spatial Density of Physical Work Environments 25
Since thereceptionistsprimaryresponsibilitiesinvolvedprocessingincoming andoutgoingpatients, themeasure ofworkload wasdesigned to tapthe number ofcontactsbetweenpatients andreceptionists.
of receptionists. It included items that meas-ured required physical movement, transferintention, crowding, and work area satisfac-tion. Before completing the questionnaire,employees were told briefly about the natureand purpose of the research and were giventhe option of not participating. Three indi-viduals chose not to participate; the totalparticipation rate was 98 percent.
The researcher emphasized that employ-ees individual responses would be held inconfidence. Participants also were told itwould be desirable if they included on thequestionnaire their organizational identifica-tion numbers. All participants agreed to pro-vide these numbers.
The researcher typically measured thework areas of the 43 units after hours whenemployees were not present. Workload datawere obtained from organizational records,and tardiness data were obtained from su-pervisor reports.
Spatial density. This was calculated as thetotal number of square feet in each workunit divided by the number of full-time-equivalent receptionists assigned to thatunit. Thus, low scores on this measure re-flect dense conditions. Only square footageavailable to the employees (i.e., space notcovered by furniture) was included in thespatial density calculation.
Workload. Since the receptionists primaryresponsibilities involved processing incomingand outgoing patients, the measure of work-load was designed to tap the number of con-tacts between patients and receptionists.Thus, information on the average number ofpatients seen per week by each receptionistin a given work unit was obtained from orga-nizational records. The sample of workloadinformation used was from the same weekthe questionnaire data were collected.
Required physical movement. This refers tothe extent to which physical movement inthe work area was required of employees fortask completion. Three questionnaire itemswere averaged to form an index (_ = .80).
Each item was measured on a seven-pointscale. Two items used the scale (1 = very in-accurate; 7 = very accurate): The job in-volves physical movement from one place toanother in my work area in order to accom-plish the required tasks and The job isarranged so that I do not have to walk aroundmuch in my work area (reverse-scored). Thethird item: To what extent does doing yourjob require you to physically move around inyour work area? That is, do the tasks whichyou do require you to walk from one workstation to another in your work area? used aslightly different scale (1 = very little; 7 =very much).
Crowding. This refers to the degree to whichemployees feel crowded in the work unit.Four items adapted from Oldham (1988)were averaged to form an index (_ = .92):My work area has an adequate amount ofspace for the number of employees whowork in it (reverse-scored), I often feelcrowded while at work, My work areadoes not have enough space for the numberof employees currently working in it, andEmployees must work too closely togetherin my work area (1 = disagree strongly; 7 =agree strongly).
Work area satisfaction. This refers to the de-gree to which employees are satisfied withthe areas in which they work. Three itemsadapted from Oldham (1988) were averagedto form an index (_ = .88): I am satisfiedwith my work setting as a whole, In gen-eral, my work area provides a good setting inwhich to work, and Overall, I feel comfort-able in my work area (1 = disagree strongly;7 = agree strongly).
Transfer intention. This was measured by anitem adapted from the Michigan Assessmentof Organizations Questionnaire (1975): If Ihad the chance, I would take a different jobwithin [the organization] (1 = disagreestrongly; 7 = agree strongly).
Tardiness. Supervisors reported the averagenumber of days per month each receptionistunder their supervision was late to work by atleast 10 minutes. Supervisors ratings were
26 HUMAN RESOURCE MANAGEMENT, Spring 2005
obtained during the week of questionnairedata collection.
Relations Among the Variables
Correlations among the measures includedin the study are shown in Table I. The tableshows the three independent variables (i.e.,spatial density, required physical movement,and workload) were moderately correlatedwith one another. However, these correla-tions were not so high as to mitigate againstthe use of the measures as separate variablesin the substantive analyses. Correlationsamong the reaction measures were generallysmall in magnitude, with the exception ofthe association between crowding and workarea satisfaction (r = .67, p < .01). Consis-tent with past literature (Oldham, 1988),employees who felt crowded at work werealso generally unhappy with their workareas. Because of the relatively strong rela-tion between crowding and work area satis-faction, we conducted supplementary prin-cipal component factor analyses usingoblique rotation for the items in thesescales. Results of these analyses confirmedthat the items for the crowding and workarea satisfaction measures loaded on twodistinct factors with no cross-loadings above.30. Details of these results are available onrequest from the first author.
Hypothesis 1 predicted that employeeswould have generally negative reactions todense work environments. Table I showsthat all the correlations involving spatialdensity and the reaction measures were sta-tistically significant and in the expected di-rection. Employees in dense work areas withlittle space available tended to be late towork more often and were more interestedin transferring than were individuals whoworked in less dense areas. Individuals indense areas also reported higher levels ofcrowding and lower levels of work area sat-isfaction than those in low-density areas.Thus, Hypothesis 1 was supported.
To examine whether crowding mediatedrelations between spatial density and theother affective and behavioral reactions (Hy-pothesis 2), a series of analyses were con-ducted that followed recommendations byBaron and Kenny (1986). Each step musthold in order to support a conclusion of me-diation. First, the independent variable mustsignificantly influence the mediator. Asshown in Table I, spatial density was signifi-cantly related to crowding (r = .46, p < .01).Second, the independent variable must sig-nificantly influence the dependent variable.As noted above, spatial density was signifi-cantly related to each of the employee reac-tionswork area satisfaction (r = .30, p