| − | Journal of Organizational Behavior<br>J. Organiz. Behav. 26, 379–408 (2005)<br>Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/job.312<br>Validating the organizational climate<br>measure: links to managerial practices,<br>productivity and innovation<br>MALCOLM G. PATTERSON,1 MICHAEL A. WEST,2*<br>VIV J. SHACKLETON,2 JEREMY F. DAWSON,2 REBECCA LAWTHOM,3<br>SALLY MAITLIS,4 DAVID L. ROBINSON1 AND ALISON M. WALLACE5<br>1Institute of Work Psychology, University of Sheffield, Sheffield, U.K.<br>2Work and Organizational Psychology, Aston Business School, Birmingham, U.K.<br>3Manchester Metropolitan University, Manchester, U.K.<br>4Sauder School of Business, University of British Columbia, Vancouver, British Columbia, Canada<br>5University of Queensland, Brisbane, Queensland, Australia<br>Summary This paper describes the development and validation of a multidimensional measure of<br>organizational climate, the Organizational Climate Measure (OCM), based upon Quinn and<br>Rohrbaugh’s Competing Values model. A sample of 6869 employees across 55 manufacturing<br>organizations completed the questionnaire. The 17 scales contained within the measure had<br>acceptable levels of reliability and were factorially distinct. Concurrent validity was measured<br>by correlating employees’ ratings with managers’ and interviewers’ descriptions of managerial<br>practices and organizational characteristics. Predictive validity was established using measures<br>of productivity and innovation. The OCM also discriminated effectively between organizations,<br>demonstrating good discriminant validity. The measure offers researchers a relatively<br>comprehensive and flexible approach to the assessment of organizational members’ experience<br>and promises applied and theoretical benefits. Copyright # 2005 John Wiley & Sons, Ltd.<br>Introduction<br>Central to most, if not all, models of organizational behavior are perceptions of the work environment,<br>referred to generally as ‘organizational climate’ (Rousseau, 1988). Primarily understood as an intervening<br>variable between the context of an organization and the behavior of its members, and attempting<br>to understand how employees experience their organizations, the concept has inspired many<br>descriptions and operationalizations. Despite the level of interest surrounding organizational climate,<br>however, there are few well-validated measures of the construct. In this paper, we describe the<br>development of a new measure of organizational climate which is both theoretically grounded and<br>empirically validated.<br>Received 9 December 2003<br>Revised 12 April 2004<br>Copyright # 2005 John Wiley & Sons, Ltd. Accepted 30 September 2004<br>* Correspondence to: Michael A. West, Work and Organizational Psychology, Aston Business School, Birmingham, B4 7ET,<br>U.K. E-mail: m.a.west@aston.ac.uk<br>The climate concept<br>While climate has been consistently described as employees’ perceptions of their organizations, the<br>construct has suffered over the years from conflicting definitions and inconsistencies in operationalization.<br>The dominant approach conceptualizes climate as employees’ shared perceptions of organizational<br>events, practices, and procedures. These perceptions are assumed to be primarily descriptive<br>rather than affective or evaluative (Schneider & Reichers, 1983). More recent work contradicts this<br>view, suggesting strong evaluative or affective components (Patterson, Warr, & West, 2004). At the<br>individual level of analysis, referred to as ‘psychological climate’ (James & Jones, 1974), these perceptions<br>represent how work environments are cognitively appraised and represented in terms of their<br>meaning to and significance for individual employees in organizations (James & Jones, 1974; James &<br>Sells, 1981).<br>Most empirical studies have used an aggregate unit of analysis, such as the work group, department,<br>or organization (hence group, departmental, and organizational climate constructs). Such climates<br>have been operationally constructed by aggregating individual scores to the appropriate level and using<br>the mean to represent climate at that level. The rationale behind aggregating individual data to a unit<br>level is the assumption that organizational collectives have their own climate and that these can be<br>identified through the demonstration of significant differences in climate between units and significant<br>agreement in perceptions within units (James, 1982). Perceptual agreement implies a shared assignment<br>of psychological meaning allowing individual perceptions to be aggregated and treated as a<br>higher-level construct. Most research is now focused on aggregate rather than on psychological<br>climate (Schneider, Bowen, Ehrhart, & Holcombe, 2000). This paper therefore describes the development<br>of a measure of organizational-level climate that is intended to support research focusing on<br>organizational climate, given the current emphasis on organizational level climate in both theory<br>and research (Schneider, Smith, & Goldstein, 2000).<br>Climate and culture<br>Consensus is not easily achieved in this area, however, since there are both theoretical differences and<br>disciplinary differences in what climate represents. Many of these differences are revealed in the<br>debate about the distinction between organizational climate and culture. Indeed, the two terms are<br>sometimes used interchangeably. Goodman and Svyantek (1999), for example, used the Organizational<br>Climate Questionnaire (OCQ) to operationally define dimensions of organizational culture.<br>While this seems paradoxical, the OCQ’s authors, Litwin and Stringer (1968), did describe the variables<br>measured by the OCP as assessing the shared beliefs and values of organizational members that<br>constitute the perceived work environment, and shared beliefs and values are often incorporated as<br>central elements in definitions of organizational culture. Hence the problems of conceptual and definitional<br>overlap.<br>There is no doubt that culture and climate are similar concepts since both describe employees’ experiences<br>of their organizations. Organizational climate, according to Schneider (2000), represents the<br>descriptions of the things that happen to employees in an organization. Climate (he suggests) is behaviorally<br>oriented. Climates for safety or service, for example, represent the patterns of behavior that support<br>safety or service. Organizational culture, in contrast, comes to light when employees are asked why<br>these patterns exist. The question is answered in relation to shared values, common assumptions, and<br>patterns of beliefs held by organizational members, and it is these which define organizational culture.<br>Svyantek and Bott (2004) propose the definitions which help distinguish between climate and culture.<br>Organizational culture is defined as a set of shared values and norms held by employees that guide<br>380 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>their interactions with peers, management, and clients. Organizational climate is more behaviorally<br>oriented in that climates for creativity, innovation, safety, or service, for example, may be found in<br>the workplace. These climates represent employees’ perceptions of organizational policies, practices,<br>and procedures, and subsequent patterns of interactions and behaviors that support creativity, innovation,<br>safety, or service in the organization. Thus climate can be understood as a surface manifestation<br>of culture (Schein, 1985; Schneider, 1990). Exploring organizational cultural values and assumptions<br>in relation to, for example, individualism/collectivism can help explain employees’ perceptions of the<br>climate for teamwork in their organizations.<br>The quest to differentiate the concepts has influenced approaches to measurement, with most climate<br>research utilizing quantitatively based questionnaire measures applied comparatively across several<br>organizations, while most culture researchers have advocated the use of qualitative measures and a<br>focus on single organizations. The approach taken in the research described here derives from climate<br>research and involves the development of a quantitatively based questionnaire measure of organizational<br>climate. What dimensions should such a measure seek to tap?<br>The dimensions of climate<br>An initial assumption of theory and research in the area of organizational climate was that social environments<br>could be characterized by a limited number of dimensions. For example, Campbell, Dunnette,<br>Lawler, and Weick (1970) identified four dimensions common to a number of climate studies (individual<br>autonomy; degree of structure imposed on the situation; reward orientation; and consideration,<br>warmth, and support). James and his colleagues (James & James, 1989; James & McIntyre, 1996;<br>James & Sells, 1981) describe four dimensions they identified across a number of different work contexts:<br>(1) role stress and lack of harmony; (2) job challenge and autonomy; (3) leadership facilitation<br>and support; and (4) work group cooperation, friendliness, and warmth. James suggested that individuals<br>developed a global or holistic perception of their work environment (e.g., James & Jones, 1974),<br>which could be applied to any number of contexts and industries.<br>However, over the years the number of climate dimensions identified as targets of assessment<br>has proliferated, leading to confusion and slow theoretical progress. For example, Glick’s (1985)<br>review of the field described an abbreviated list of climate dimensions including leader’s psychological<br>distance (Payne & Mansfield, 1978), managerial trust and consideration (Gavin & Howe, 1975),<br>communication flow (Drexler, 1977), open-mindedness (Payne & Mansfield, 1978), risk orientation<br>(Lawler, Hall, & Oldham, 1974), service quality (Schneider, Parkington, & Buxton, 1980); equity<br>(James, 1982), and centrality (Joyce & Slocum, 1979). Since Glick’s review, the development<br>of new climate scales has continued. For example, the Business Organization Climate Index<br>(Payne & Pheysey, 1971) was revised in 1992 with the addition of scales measuring concern for<br>customer service, the impact of information quality, and ability to manage culture (Payne, Brown,<br>& Gaston, 1992).<br>Schneider (1975, 1990, 2000) eschews the use of general multidimensional measures of climate and<br>argues for a facet-specific climate approach where climate has a focus and is tied to something of interest.<br>Schneider suggests that the dimensions of organizational climate will differ depending on the purpose<br>of the investigation and the criterion of interest, and that general measures of organizational<br>climate will contain dimensions that are not relevant for each specific study. This line of argument<br>has encouraged the development of measures of several dimensions of climate such as service<br>(Schneider, 1990) and innovation (Anderson & West, 1998; West, 1990).<br>Rather than considering the global and domain-specific approaches to organizational climate as<br>opposite sides of one coin, it is worthwhile viewing both as a valid basis for the investigation of work<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 381<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>environment perceptions. Which approach is favored depends largely on the interests of the investigation.<br>The global approach is advantageous in terms of its provision of an overall snapshot of organizational<br>functioning, allowing a view of the ways whole organizations operate (Ashkanasy,Wilderom, &<br>Peterson, 2000). A multidimensional global approach can also highlight subcultures and identify the<br>effects of particular dimensions on specific outcome measures, such as organizational productivity or<br>innovation (Ashkanasy et al., 2000). The domain-specific approach contributes more precise and targeted<br>information for use in areas such as the improvement of customer satisfaction and the improvement<br>of company safety. What has research revealed about the relationship between organizational<br>climate and outcomes of theoretical interest?<br>Links between climate and outcomes<br>Research has suggested that climate perceptions are associated with a variety of important outcomes at<br>the individual, group, and organizational levels. These include leader behavior (Rousseau, 1988;<br>Rentsch, 1990), turnover intentions (Rousseau, 1988; Rentsch, 1990), job satisfaction (Mathieu,<br>Hoffman, & Farr, 1993; James & Tetrick, 1986; James & Jones, 1980), individual job performance<br>(Brown & Leigh, 1996; Pritchard & Karasick, 1973), and organizational performance (Lawler et<br>al., 1974; Patterson et al., 2004).<br>As a global or summary construct, organizational climate has been related to several important work<br>outcomes. Brown and Leigh (1996) demonstrated that perceptions of a motivating and involving organizational<br>climate were positively related to supervisory ratings of performance. Organizational climate<br>has been shown to relate to group process variables across organizational levels (Griffin &<br>Mathieu, 1997). Day and Bedeian (1991) showed that employees performed better (as rated by<br>their supervisors) in organizational climates they perceived as structured (unambiguous) and supportive<br>of risk.<br>Domain-specific climate has also been linked with several important work outcomes. Using their<br>model of service climate, Schneider and colleagues demonstrated that service climate is related to customer<br>perceptions of service quality (Schneider, 1980; Schneider et al., 1980; Schneider, White, &<br>Paul, 1998). Safety climate has been significantly linked with safety behaviors in accidents teams<br>(Hofmann & Stetzer, 1996), and safety compliance in the health sector (Murphy, Gershon, & DeJoy,<br>1996). Research in the area of innovation suggests that group climate factors influence levels of<br>innovative behavior in health care and top management teams (West & Wallace, 1991; West &<br>Anderson, 1996).<br>While progress in understanding that dimensions of climate predict outcomes in a variety of studies,<br>knowledge develops haphazardly in this field in a way that appears not to be synergistic or to lead to<br>theory development. This is partly because virtually every study referred to above uses a different measure<br>of climate, each assessing rather different dimensions. The accruing knowledge is not cumulative,<br>hence the study we describe here which seeks to develop an inclusive, robust and theoretically based<br>approach to the measurement of climate. Moreover, many instruments are not validated, are poorly<br>designed, and fail to specify the level of analysis.<br>Existing measures of climate<br>The lack of a theoretical basis for many climate instruments has resulted in much of the variation in<br>climate dimensions employed in different measures. For example, Wilderom, Glunk, and Maslowski<br>382 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>(2000) located and summarized 10 studies relating climate to organizational performance. They<br>reported that different aspects of climate emerged as important in different studies. This diffuse pattern<br>of results is likely to be due, in part, to the variety of methods of assessment of climate employed in<br>these studies.<br>The inability to draw clear research conclusions through a lack of theory and subsequent<br>inconsistent operationalization of climate is compounded by the fact that most climate<br>instruments have not been validated. With the exception of some domain-specific climates such as<br>Schneider’s service climate (Schneider et al., 1998), there are few measures with demonstrated reliability<br>and validity.<br>One of the best-known general measures of organizational climate is the Organizational Climate<br>Questionnaire (OCQ) by Litwin and Stringer (1968). It comprises 50 items that assess nine dimensions<br>of climate. A number of studies (e.g., Sims & LaFollette, 1975; Muchinsky, 1976) have suggested that<br>a six-factor structure is more appropriate and pointed out that the existing nine scales showed poor<br>split-half reliabilities. A review by Rogers, Miles, and Biggs (1980) showed that most studies had<br>found six factors and that there was virtually no agreement among researchers regarding which items<br>loaded best on the different factors. They concluded that the OCQ lacked validity and was not a consistent<br>measurement device. Such measurement problems are not unusual in this area of research and<br>prompted the development of the measure described here.<br>A number of culture questionnaires have been published over the last 25 years, but they can also be<br>seen as measures of climate as they tap the surface manifestations of underlying cultural assumptions<br>(Schein, 2000). Again, these instruments suffer from a number of problems including a lack of a theoretical<br>basis, little validity information (Ashkanasy et al., 2000), little or no confirmatory studies and/<br>or small sample sizes used for their development.<br>A further methodological weakness of climate research is the vague or poorly specified descriptive<br>level of items in many climate measures. Each climate questionnaire item should clearly focus on the<br>specific collective unit which corresponds to the climate being studied (team, department, or organization).<br>Unfortunately, in many studies respondents have not been instructed to focus on a specific organizational<br>unit, but rather to provide descriptions relating to their ‘work environments’ (Howe, 1977;<br>Schneider & Reichers, 1983). This ambiguity in the frame of reference of climate items can lead to<br>individuals describing perceptions of different parts of the organization, some assuming the questionnaire<br>asks them to describe their department and others assuming the referent is the organization<br>(Rousseau, 1988).<br>A related issue concerns the type of respondents included in studies of organizational climate. Organizational<br>climate is a characteristic of an entire organization and, as Wilderom et al. (2000) argue, ‘it<br>seems crucial that researchers investigate all sorts of organizational members, representative of all the<br>various hierarchical, departmental, divisional and/or professional entities’ (p. 207). However, investigations<br>often focus only on managerial employees (Gordon & DiTomaso, 1992; Kotter & Heskett,<br>1992; Sheridan, 1992; Denison, 2001; Denison & Mishra, 1995; Weber, 1996). Clearly, for inclusiveness,<br>we need measures of organizational climate that assess the experiences of employees throughout<br>the workforce. The content and wording of such measures should therefore be relevant and comprehensible<br>to all organizational members.<br>This paper describes the development of a global multidimensional measure of organizational<br>climate intended to address the conceptual and methodological issues outlined above. The<br>measure is designed to be theoretically grounded, to explicitly and consistently specify the appropriate<br>frame of reference, and to be applicable across a range of work settings and to target all employee<br>levels (lower level as well as managerial employees). We report data, from a large sample of<br>employees and organizations, describing its factor structure and internal reliability, and assessing the<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 383<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>measure’s discriminant and consensual validity and (using separate source data) concurrent and predictive<br>validity. First, we describe the Competing Values framework, which underpins our measure.<br>Competing Values model<br>We tackle the problem of what dimensions to include in a measure of organizational climate by adopting<br>a clear theoretical framework to drive this choice. In choosing a framework we were mindful of the<br>need to clearly anchor the measure at the organizational level. The measure had to assess a broad class<br>of organizational, rather than psychological variables that constitute the organizational context for<br>individual actions and that therefore enable comparative studies of organizational climate.We propose<br>that one meta-theoretical model, the Competing Values model (Quinn & Rohrbaugh, 1983; Quinn &<br>McGrath, 1985; Gifford, Zammuto, & Goodman, 2002), provides a framework of values that underlie<br>organizational climates. The model represents broad managerial ideologies that have emerged over<br>time (the content of these ideologies are discussed below). Building on Beyer’s (1981) work on ideologies<br>in organizations, Zammuto, Gifford, and Goodman (2000) argue that as managerial ideologies<br>become institutionalized they form a common basis for the ideologies of individual organizations,<br>imported via education, consultants, training, and management books. As a result, Western managers<br>‘draw from a common, limited set of managerial ideologies that are the foundation for the assumptions,<br>values, and beliefs on which individual organizations’ cultures are based. Differences in which<br>parts of these broad managerial ideologies are imported into organizations’ ideologies lead to differences<br>in organizational cultures’ (Zammuto et al., 2000, p. 264). The essential point is that an organization’s<br>ideology and culture will shape decision-making and action in that organization (Beyer,<br>1981; Zammuto et al., 2000), and consequently will play an important role in the development of<br>its climate.<br>The Competing Values model, developed in a series of articles and studies by Quinn and colleagues<br>(e.g., Quinn & Rohrbaugh, 1981, 1983; Quinn & McGrath, 1985), proposes that organizational effectiveness<br>criteria in the literature can best be understood when organized along fundamental dimensions—<br>flexibility versus control and internal versus external orientation. The framework’s four<br>quadrants describe four broad domains of valued outcomes and associated managerial ideologies about<br>the means through which these outcomes may be achieved. It calls attention to how opposing values<br>exist in organizations and how ‘individual organizations are likely to embrace different mixtures of<br>values that are reflected in their desired ends and in the means to attain them, such as their structural<br>designs and mechanisms of co-ordination and control’ (Zammuto & O’Connor, 1992, p. 711). The<br>intention of the model is to encapsulate into one framework the major approaches to organizational<br>values and effectiveness over the last 100 years.<br>A major strength of this model is its derivation from four major schools of study of organizational<br>effectiveness, reflecting long traditions in management and organizational psychology:<br>� The human relations approach (internal focus and flexibility in relation to the environment) reflects<br>the tradition derived from the socio-technical (Emery & Trist, 1965) and human relations schools<br>(e.g., McGregor, 1960). This approach emphasizes the well-being, growth and commitment of the<br>community of workers within an organization.<br>� The internal process approach (internal focus and tight control within the organization) reflects a<br>Tayloristic concern with formalization and internal control of the system in order that resources<br>are efficiently used.<br>� The open systems approach (external focus and flexible relationships with the environment)<br>emphasizes the interaction and adaptation of the organization in its environment, with managers<br>384 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>seeking resources and innovating in response to environmental (or market) demands (Shipper &<br>White, 1983).<br>� The rational goal approach (external focus but with tight control within the organization) reflects a<br>rational economic model of organizational functioning in which the emphasis is upon productivity<br>and goal achievement (Hall, 1980; Clinebell, 1984).<br>By combining these orientations into one model, Quinn and colleagues aimed to provide a broad<br>conceptual map of the domains of theory in the field over the last 60 years. Such a map is useful in<br>identifying the required topography of a climate measure, applicable to a wide range of organizations.<br>The model is also useful in reflecting the means for implementing those values in terms of managerial<br>practices, and the ends or outcomes which are emphasized or which compete in each domain.<br>It is important, however, to recognize that the model does not propose that organizations can be<br>located predominantly in one quadrant but, reflecting the rich mix of competing views and perspectives<br>in organizations, proposes that organizations will be active in, and give emphasis to, each domain,<br>but with differing strengths. Quinn (1988) argued that a balance of competing organizational values is<br>required for organizational effectiveness. Similarly, we suggest that organizations will not have strong<br>climates in relation to each of the dimensions we identify within any one quadrant but that the strength<br>of emphasis will vary both across quadrants and across dimensions within quadrants. The value of the<br>model for the development of a climate measure is that it aids inclusiveness and ensures that dimensions<br>tap the core values of organizations.<br>Identifying climate dimensions<br>Dimensions were generated within domains to ensure that there was an adequate sample of dimensions<br>identified within each of the four broadly conceptualized domains. In order to develop subscales that<br>reflected each of these four domains, we searched the literatures referred to above to develop such an<br>inclusive set of dimensions. We selected those dimensions that were most frequently utilized in<br>research studies from 1960 to 2000 on climate and that fitted into the competing values model with<br>a relatively unambiguous location in one of the four quadrants. This process was iterative, involving a<br>large number of meetings to achieve consensus by members of the research team. In addition, where no<br>suitable dimension could be found in the literature we identified an appropriate construct (e.g., the<br>concept of reflexivity as a dimension within the open systems quadrant).<br>Debates and conceptual analyses among the authors and ultimately psychometric analyses served to<br>sharpen our conceptualization of the nature and structure of organizational climate, located in the<br>Competing Values model, as well as to identify deficiencies in the dimension pool. The key question<br>addressed was ‘Does the pool adequately reflect the four domains of the Competing Values model?’<br>Below we present the Competing Values model and the climate scales which we attributed to each of<br>the quadrants represented in the model. We did not attempt to create an equal number of dimensions<br>and scales in each quadrant since some dimensions are inherently more complex and multifaceted<br>(e.g., human relations) than others (internal process).<br>The Human Relations Model (internal focus, flexible orientations) has norms and values associated<br>with belonging, trust, and cohesion, achieved through means such as training and human resource<br>development. Coordination and control are accomplished through empowerment and participation,<br>and interpersonal relations are supportive, cooperative, and trusting in nature. Climate dimensions<br>which we identified as representing this quadrant are:<br>� employee welfare—the extent to which the organization values and cares for employees (e.g.,<br>Robinson & Rousseau, 1994; Guest, 1998);<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 385<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>� autonomy—designing jobs in ways which give employees wide scope to enact work (e.g., Cherns,<br>1976; Klein, 1991);<br>� participation—employees have considerable influence over decision-making (e.g., Miller &<br>Monge, 1986; Hollander & Offerman, 1990; Heller, Pusi, Strauss, & Wilpert, 1998);<br>� communication—the free sharing of information throughout the organization (e.g., Callan, 1993;<br>Hargie & Tourish, 2000);<br>� emphasis on training—a concern with developing employee skills (e.g., Gattiker, 1995; Morrow,<br>Jarrett, & Rupinski, 1997);<br>� integration—the extent of interdepartmental trust and cooperation (e.g., Lawrence & Lorsch, 1967;<br>Nauta & Sanders, 2000);<br>� supervisory support—the extent to which employees experience support and understanding from<br>their immediate supervisor (e.g., Cummins, 1990; Eisenberger et al., 2002).<br>In the Internal Process Model (internal focus, control orientation) the emphasis is on stability, where<br>the effects of environmental uncertainty are ignored or minimized. Coordination and control are<br>achieved by adherence to formal rules and procedures. The Internal Process Model represents the classic<br>bureaucracy. Scales which reflect this model are:<br>� formalization—a concern with formal rules and procedures (e.g., Pugh, Hickson, Hinings, &<br>Turner, 1968; Hall, 1991);<br>� tradition—the extent to which established ways of doing things are valued (e.g., Coch & French,<br>1948).<br>The emphasis of the Open Systems Model (external focus and flexible orientation) is on readiness,<br>change and innovation, where norms and values are associated with growth, resource acquisition, creativity<br>and adaptation. Climate dimensions which are likely to reflect this orientation are:<br>� flexibility—an orientation toward change (e.g., Garrahan & Stewart, 1992; King & Anderson,<br>1995);<br>� innovation—the extent of encouragement and support for new ideas and innovative approaches<br>(e.g., West & Farr, 1990);<br>� outward focus—the extent to which the organization is responsive to the needs of the customer and<br>the marketplace in general (Kiesler & Sproull, 1982; West & Farr, 1990);<br>� reflexivity—a concern with reviewing and reflecting upon objectives, strategies, and work processes,<br>in order to adapt to the wider environment (West, 1996, 2000).<br>The primary emphasis in the Rational Goal Model (external focus and control orientation) is on the<br>pursuit and attainment of well-defined objectives, where norms and values are associated with productivity,<br>efficiency, goal fulfillment, and performance feedback. Climate dimensions which might reflect<br>this model are:<br>� clarity of organizational goals—a concern with clearly defining the goals of the organization (e.g.,<br>Locke, 1991);<br>� effort—how hard people in organizations work towards achieving goals (e.g., McCaol, Hinsz, &<br>McCaol, 1987);<br>� efficiency—the degree of importance placed on employee efficiency and productivity at work (e.g.,<br>Ostroff & Schmitt, 1993);<br>� quality—the emphasis given to quality procedures (e.g., Deming, 1986; Hackman &Wageman, 1995);<br>� pressure to produce—the extent of pressure for employees to meet targets (e.g., Taira, 1996);<br>� performance feedback—the measurement and feedback of job performance (e.g., Annett, 1969;<br>Kopelmann, 1986).<br>386 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Organizational Context<br>Method<br>Pilot study and development of the main item pool<br>Approximately 10 items per scale were generated to tap into each of 19 proposed dimensions of climate.<br>We used broad and comprehensive sets of items to tap the organizational climate dimensions.<br>Subsequently, each item set was reduced and items modified during the piloting. Weaker items (i.e.,<br>those that respondents found ambiguous or confusing), and the more important items (those which<br>loaded highly on the constructs) were returned identified. Scale items were also selected on the basis<br>of item differentiation and measurement width within scales by examining both inter-item correlations<br>and semantic context. After all these processes, 95 items remained. Items from across all the scales<br>were randomly arranged, in order that effects due to response set were minimized. The response format<br>used was a 4-point Likert scale of definitely false, mostly false, mostly true, and definitely true.<br>Because the questionnaire was designed for use within all levels of the workforce, care was taken to<br>use straightforward wording. The first page contained instructions explaining the purpose and format<br>of the questionnaire, and informed respondents that it would take about 25 minutes to complete. They<br>were asked to answer all questions, as far as possible.<br>Administering the questionnaire<br>The Organizational Climate Measure (OCM)# was administered to employees from 55 organizations<br>within the U.K. manufacturing sector.1 The size of the companies ranged from 60 to 1929 employees,<br>The Companies<br>The companies were spread throughout the United Kingdom, and were drawn fromthe manufacture of<br>metal goods and mechanical engineering sector, plastics and rubber sector, and a third miscellaneous<br>category.Theywere predominantly single site and averaged approximately 260 employees.The climate<br>data were collected between 1994 and 1996.At this time the companieswere facing testing trading conditions,<br>with a greater than usual probability of bankruptcy and an increased threat to jobs. In response<br>many firms set about trying to improve productivity through introducing new management practices,<br>reorganizing production methods and so on. Almost all the firms in our sample reported undergoing<br>significant change in structure, human resource management policies, and work design. In short, the<br>study took place at a time of considerable uncertainty and change for manufacturing organizations.<br>Employees’ Climate Perceptions<br>The organizational climate surveys across the companies reinforced the researchers’ views that<br>organizational changes were implemented with limited effectiveness. Employees reported low<br>levels of participation, lack of open communication, and little decentralization. Pressure on employees<br>to produce and formalization (the use of rules and formal procedures to control activities) dominated<br>employees’ climate perceptions of their organizations.<br>1Researchers interested in using the OCM for research can do so without charge on application to the second author.<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 387<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>with an average of 256 employees. Some 12 051 questionnaires were distributed and in 49 organizations<br>all employees were asked to complete the OCM. In the six organizations with more than 500 employees,<br>a random sample of 60 per cent of the workforce was selected to complete the questionnaire.<br>Two methods were used to administer the questionnaire. In 37 companies a postal survey was carried<br>out. Here, climate questionnaires were sent out to employees at their place of work in personally<br>addressed envelopes which also contained a pre-paid return envelope. Seventeen companies chose to<br>have the climate survey administered by researchers on site.<br>Results<br>Overall, completed questionnaires were received from 6869 employees, constituting a 57.0 per cent<br>response rate across the sample. There was no evidence that response rates varied by organizational<br>size or sector.<br>There were three stages in the analysis. First, using the data provided by 50 per cent of the respondents,<br>we employed confirmatory factor analysis to examine how closely the data derived from<br>employees’ responses to the 95 items were consistent with a 19-latent factor model of organizational<br>climate. Based on these findings, the model was refined to improve the fit between the model and the<br>data in terms of internal consistency of scale items and discrimination between latent constructs. The<br>adequacy of the refined model was then formally assessed (cross-validated) using the remaining<br>50 per cent of the sample, again employing confirmatory analysis.<br>The final part of the analysis centered on demonstrating the usefulness and adequacy of our climate<br>measures, in terms of consensual, discriminant, and external (concurrent, predictive) validity and measurement<br>invariance (i.e., does the instrument measure the same constructs across different samples?).<br>A common problem with data collected from lengthy questionnaires is missing data. Within our<br>6869 returned questionnaires we found that, on average, 8 per cent of questions were left unanswered<br>by each respondent. A list-wise deletion of cases would result in the loss of around 80 per cent of the<br>sample. Far more efficient methods are now available to researchers, including use of full information<br>maximum likelihood estimation (FIML) (Arbuckle, 1996), which uses all available data to produce<br>estimates, and the expectation–maximization (EM) algorithm, which uses the information contained<br>in cases with and without missing data to compute a maximum likelihood covariance matrix and so<br>makes more efficient use of the available data than standard missing data techniques. Graham, Hofer,<br>and MacKinnon (1996) demonstrated that parameter estimates based on the analysis of the EM covariance<br>matrix are excellent in that they are unbiased and efficient.We made use of both these procedures<br>during the analysis.<br>Throughout the analysis we made the assumption that all missing data points may be regarded as<br>missing at random, since the percentage missing for each question varied from only 1 per cent to<br>2.5 per cent (i.e., there appeared to be no ‘favored’ or ‘unfavored’ items). These figures suggested that<br>piloting had been successful in eliminating ‘unfriendly’ or ‘difficult’ questions and that missing<br>responses could be regarded as reasonably random in occurrence throughout the sample.<br>One further assumption was also made throughout the analysis. The covariances observed between<br>item responses were assumed to be reasonably homogeneous across different categories of job type,<br>age, sex, department, etc. In other words, the covariations (and therefore factor structure) among<br>responses for persons who are (say) management staff and those who are non-management staff were<br>considered to be reasonably similar. A more detailed discussion of this assumption is given in the<br>Results section.<br>388 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Model exploration and refinement<br>A common analytical technique used for scale development is to split the data at random into two<br>subsamples, and then to use the first subsample for exploratory analysis of the factor structure and<br>the second subsample to confirm this structure. We broadly followed this approach, although rather<br>than using exploratory factor analysis (EFA) followed by confirmatory factor analysis (CFA), we used<br>CFA on both parts of the data. This was due to the problems of missing data mentioned above: while<br>full information maximum likelihood estimation (FIML) is an appropriate technique for dealing with<br>missing data for CFA, no such equivalent technique is available for EFA.<br>Since we hypothesized the existence of the 19 different climate constructs a priori, and intentionally<br>wrote items to tap into each of these domains, we started by assessing how well a 19-factor structural<br>equation model described the covariation between items. For this, data provided by the first 50 per cent of<br>the sample were used. Each of the sets of five items was specified to load onto a single factor only, representing<br>the underlying construct in mind. Correlation among the latent factors was permitted, based on<br>the assumption that the 19 constructs reflect both opposing/similar domains within the two-dimensional<br>Competing Values framework (internal versus external focus, and control versus flexible orientation).<br>The covariance matrix was computed using EM (expectation maximization) by Schafer’s (1997)<br>NORM program. The EMcovariance matrix was then entered into theAMOS structural equation modeling<br>program (Arbuckle, 1996) for fitting of the 19-factor confirmatory factor analysis model.<br>Detailed results of this analysis are not repeated here, but are available on request from the authors.<br>Refinement of factor structure<br>There were very strong correlations between the latent factors labeled Innovation and Flexibility (0.94)<br>and between Participation and Communication (0.98). This indicated that the shared variance in items<br>designed to measure Innovation corresponded almost precisely to the shared variation in items<br>designed to reflect Flexibility. The same applied to items representing Participation and Communication.<br>An examination of item correlations showed that Innovation items correlated just as highly with<br>Flexibility items as they did with items within their own subscale (again this was also true for the items<br>representing Participation and Communication).<br>In view of the fact that we wanted subscales to measure reasonably distinct dimensions of climate,<br>the first modification to the model was therefore to allow the two subsets of Innovation and Flexibility<br>items to load onto a new single factor which we called Innovation & Flexibility. Similarly, items originally<br>designed to tap aspects of Participation and Communication were allowed to load onto a single<br>construct, provisionally labeled Involvement. When the revised model was refitted to the first<br>50 per cent sample of data, there were no correlations over 0.80.<br>Scale refinement<br>The process of subscale refinement was based on several criteria. We examined regression weights for<br>small loadings in order to identify items that had insufficient covariation with other items in a subscale.<br>Unidimensionality and internal consistency were also assessed by calculation of reliability coefficients<br>(Cronbach’s alpha) and inter-item correlations. We also removed those items loading highly on other<br>factors. This was assessed by examination of modification indices.<br>In the new 17-factor model, all but one of the regression weights of the latent variables on their<br>respective items were highly significant. This one item was removed. Also, modification indices<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 389<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>suggested that two items loaded highly on several other factors. Reliability analysis showed that there<br>would be no change in the value of Cronbach’s alpha if the items were removed, and inspection of the<br>remaining items in the scales showed that measurement width would be unaffected. Therefore these<br>items were also removed. We also reduced the number of items in some scales on the basis of<br>low reliability/regression weights and overlap into other domains of climate. Where two scales had<br>been combined (Participation and Communication, Flexibility and Innovation), the new composite<br>scales were reduced to six items, in order that the number of items for each should not be too overrepresentative,<br>but allowing the scale to retain its breadth of content. In total, 13 of the 95 items were<br>removed.<br>Assessment of proposed dimension structure and scale measures<br>To assess the adequacy of our refined subscales and factor structure, a confirmatory factor analysis of<br>the proposed model was performed on the remaining 50 per cent of the data. To overcome the problem<br>of not getting fit indices with FIML, a commonly used (but undocumented) approach to assessing<br>goodness of fit in missing data situations was used (J. Graham, personal communication, 1998).<br>The first step involves assessing the overall goodness of fit of the specified model and the second step<br>requires estimating parameters and obtaining accurate standard errors for more exact critical ratio<br>tests. We computed EM estimates of item covariances on the same remaining approximately<br>50 per cent of the dataset using the same technique described earlier and fitted the proposed model<br>to the estimated covariance matrix. The fit indices produced are based on the assumption that there<br>are no missing data in the N¼3428 cases and therefore are too conservative. The correct value of<br>N to be used in the calculation of fit indices should be somewhat smaller than 3428 (nominal N) to<br>reflect the fact that many of these cases did not contain answers to all items. A very crude approximation<br>would be to use a value of N0 ¼3428�proportion of the data points actually observed. This value<br>of N0 should then give fit indices that are much closer to the true values. In fact, since the proportion of<br>missing data points in our sample was relatively small, the sample sizes N and N0 were very similar and<br>gave the same values (to within 0.003) for the fit indices of interest. In light of the above, we can reasonably<br>claim the true values of the indices should be very similar to those obtained by using EM.<br>Internal homogeneity<br>Internal consistency estimates (using Cronbach’s alpha) of the final 17 scales are presented in Table 1.<br>With the exception of the Autonomy scale, all alpha values are at or above 0.73. Also shown are the<br>means, standard deviations, and correlations for each scale. The items comprising each scale can be<br>found in the Appendix, along with which scales map onto each of the four quadrants of the Competing<br>Values model.<br>The standardized estimates of regression weights produced by FIML can also be found in the<br>Appendix. All critical ratios were significant, indicating that the latent factors explained significant<br>proportions of variance in the scale items.<br>Confirmatory factor analysis of dimensional structure<br>Four commonly used fit indices obtained by fitting the model to the EM covariance matrix are<br>presented in Table 2. In view of the large sample size, the usual chi-square statistic was considered<br>390 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Table 1. Cronbach’s alpha, means, SDs, and correlations for the climate scales<br>Scales Cronbach Mean SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.<br>alpha<br>1. Autonomy 0.67 2.28 0.52<br>2. Integration 0.86 2.52 0.68 0.23<br>3. Involvement 0.87 2.29 0.69 0.37 0.61<br>4. Supervisory 0.88 2.64 0.67 0.22 0.49 0.52<br>Support<br>5. Training 0.83 2.53 0.76 0.21 0.52 0.64 0.51<br>6. Welfare 0.91 2.63 0.79 0.33 0.55 0.70 0.60 0.64<br>7. Formalization 0.77 2.81 0.60 �0.26 0.30 0.26 0.26 0.37 0.32<br>8. Tradition 0.73 2.42 0.64 �0.26 �0.44 �0.53 �0.32 �0.48 �0.49 �0.28<br>9. Innov. & 0.86 2.49 0.64 0.21 0.53 0.61 0.54 0.58 0.63 0.31 �0.56<br>Flexibility<br>10. Outward 0.83 3.20 0.63 0.06 0.38 0.36 0.38 0.38 0.47 0.37 �0.43 0.48<br>Focus<br>11. Reflexivity 0.76 2.64 0.54 0.16 0.54 0.56 0.45 0.52 0.55 0.33 �0.51 0.65 0.53<br>12. Clarity of 0.87 2.53 0.73 0.14 0.48 0.58 0.43 0.53 0.54 0.32 �0.44 0.61 0.42 0.58<br>Org. Goals<br>13. Efficiency 0.80 2.03 0.70 0.18 0.42 0.53 0.27 0.46 0.41 0.27 �0.47 0.38 0.21 0.35 0.39<br>14. Effort 0.79 2.76 0.58 0.12 0.44 0.45 0.45 0.45 0.47 0.35 �0.39 0.48 0.39 0.38 0.43 0.34<br>15. Perform. 0.78 2.44 0.66 0.15 0.48 0.59 0.42 0.56 0.50 0.35 �0.49 0.52 0.35 0.54 0.56 0.48 0.44<br>Feedback<br>16. Pressure to 0.79 2.90 0.63 �0.22 �0.12 �0.22 �0.12 �0.14 �0.24 0.11 0.04 �0.11 0.05 �0.04 �0.08 �0.18 0.14 �0.04<br>Produce<br>17. Quality 0.80 3.28 0.62 0.02 0.36 0.33 0.40 0.42 0.46 0.43 �0.35 0.47 0.60 0.44 0.45 0.20 0.51 0.38 0.04<br>N¼6869.<br>All correlations are statistically significant with p<0.001, except for those between Autonomy and Quality ( p¼0.101), Reflexivity and Pressure to Produce ( p¼0.032), and<br>Outward Focus and Pressure to Produce ( p¼0.001).<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 391<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>inappropriate for model assessment (although quoted here for completeness). Instead we used the<br>Bentler–Bonnett normal fit index (NFI), the non-normal fit index (NNFI) and the comparative fit index<br>(CFI), all of which are thought to reflect fit relatively well regardless of the sample size. In addition we<br>also examined the RMSR (root mean square [standardized] residual), which represents the average<br>absolute value by which the observed sample variances and covariances differ from those predicted<br>by the model.<br>A value of RMSR below 0.05 is regarded as evidence that the model predicts the observed covariances<br>among items very well. Our results therefore suggest that the proposed model provides an adequate<br>explanation of the relationships in the data. Values of NFI, NNFI, and CFI above 0.9 are usually<br>considered indicative of a good fit. Our statistics fall short of this recognized level. However, it is widely<br>accepted that achieving high levels of fit with large numbers of items is difficult or even impossible (e.g.,<br>Floyd &Widaman, 1995). In our case, limitations on the overall level of fit are due not only to the large<br>number of items but also to the specification of the factor model. Even though questions were worded<br>and subsequently selected to be unidimensional, facets of climate are non-independent by nature, so<br>success in confirming factor structure will always be limited by restricting items to load on just one<br>factor (when there are 16 possible others). Moreover, the items were randomly arranged in the questionnaire,<br>which may have contributed to more response overlap. One final, but important, point relates<br>to this observation. Researchers are unlikely to wish to use the whole OCM in a single study, especially<br>where their use of the measure is focused by the appropriate use of theory. They will be more likely to<br>focus on the use of scales from one quadrant of the Competing Values model. For example, researchers<br>examining innovation are more likely to focus on scales from the Open Systems quadrant, whereas<br>those interested in people management issues may focus only on scales in the Human Relations quadrant.<br>Where the focus is thus defined, the fit of items to subscales within quadrants is more convincing.<br>Climate scales which are likely to map onto the Competing Values quadrants are:<br>� Human Relations—comprising, Involvement, Autonomy, Supervisory Support, Integration, Welfare,<br>Training, and Effort;<br>� Internal Process—Formalization and Tradition;<br>� Open Systems—Reflexivity, Innovation & Flexibility, and Outward Focus;<br>� Rational Goal—Clarity of Organizational Goals, Pressure to Produce, Quality, Performance Feedback,<br>and Efficiency.<br>Confirmatory factor analyses for each of these four proposed models on the second 50 per cent of<br>cases yielded the sets of fit indices shown in Table 3.<br>We do not predict or expect there necessarily to be a high level of intercorrelation between scales<br>within quadrants, thus producing a second-order factor structure mapped on to the four quadrants.<br>When we incorporated all scales into one CFA and tested for four underlying constructs, there was<br>Table 2. Goodness of fit indices for the 17-factor CFA model<br>Index Specified sample size<br>NominalN N0<br>NFI 0.83 0.83<br>NNFI 0.84 0.84<br>CFI 0.85 0.85<br>RMSR 0.042<br>Chi-squared (d.f.) 25195.9 (3103)<br>392 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>no neat second-order factor structure mapping onto the four quadrants. The quadrants describe the organizational<br>terrain; the model does not propose that organizations will have a set of practices that map on<br>to all or most of these climate dimensions within one domain. In some organizations, a human relations<br>orientation will be evidenced by supervisory support and an emphasis on employee welfare. In other<br>organizations, a human relations orientation will be evidenced by an emphasis on employee involvement<br>and autonomy. The one represents a ‘patriarchal’ approach to human relations; the other a more<br>participative, democratic approach. There is no reason why we would necessarily expect them to be<br>consistent. The same applies to the relationships between dimensions within the other three domains.<br>Generalizability<br>Throughout the analysis we have assumed that the same factor structure holds for each of several populations<br>(e.g., management/non-management, shop floor/non-shop floor) and that all parameter values<br>(i.e., correlations and regression weights) are roughly equal to those estimated across these subsamples.<br>Since we intend the scales to be used at various levels within organizations, this assumption was<br>tested for management/non-management and shop floor/non-shop floor. The results gave almost identical<br>fit indices for the 17-factor model across all job types (these are shown in Table 4). There was no<br>reason from these analyses to doubt that parameter estimates, or factor structures, would be any different<br>between the two subsamples.<br>Consensual and discriminant validity<br>The scales have high face validity, because items were selected specifically to represent the constructs<br>of interest, and because the analyses presented suggest factorially distinct scales relating to these<br>Table 3. Fit indices for the four competing values domains<br>Domains Indices using nominal N or N0<br>NFI NNFI CFI Chi-squared (d.f.)<br>Human relations 0.88 0.88 0.89 7372.4 (506)<br>Internal process 0.91 0.88 0.92 645.2 (26)<br>Open systems 0.93 0.92 0.93 1583.6 (101)<br>Rational goal 0.90 0.89 0.91 3053.8 (220)<br>Table 4. Fit indices for the different subgroups<br>Indices using nominal N or N0<br>NFI NNFI CFI Chi-squared (d.f.)<br>Management 0.81 0.83 0.84 14885.7 (3103)<br>Non-management 0.85 0.85 0.86 32205.7 (3103)<br>Shopfloor 0.82 0.83 0.84 22525.7 (3103)<br>Non-shopfloor 0.81 0.84 0.84 20354.3 (3103)<br>Sample included 1656 managers and 5067 non-managers, and 3012 shopfloor workers and 2257 non-shopfloor workers.<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 393<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>constructs. The instrument should also demonstrate significant differences in employee perceptions<br>across organizations if it is to be useful in discriminating between organizations. It should also produce<br>data indicating reasonable consensus between employees within organizations. To determine the discriminant<br>validity of the OCM, analyses of variance were performed using all the scales, with organizations<br>as the independent variables. There were significant between-organization differences on all<br>scales. F-values ranged from 5.21 (d.f.¼53) for Integration to 27.25 (d.f.¼53) for Pressure to Produce.<br>In order to assess the inter-rater reliability for each scale, the results of the one-way analysis of variance<br>were converted to intra-class correlations (ICC) (Shrout & Fleiss, 1979) with the two versions<br>commonly known as ICC(1) and ICC(2) (Bliese, 2000). These coefficients assess the ratio of variation<br>within organizations to variation among organizations. Thus, high coefficients are related to small<br>within-organization variance. Whereas ICC(1) assesses the reliability of a single rating (i.e., the reliability<br>of one respondent’s answers), ICC(2) assesses the reliability of a mean rating (or an aggregated<br>score). The ICC(1) values for the dimensions in this study are between 0.04 and 0.22, with an average<br>of 0.16, and are within the range reported by previous studies.<br>Within the literature there are few guidelines for appropriate values of ICC(1). However, a review by<br>James (1982) found values ranging from 0 to 0.50, with 0.12 as the median, and Bliese (2000) wrote<br>that, in his experience, values typically range between 0.05 and 0.20. Also he reported that he had<br>never encountered values higher than 0.30 and would be surprised to find values greater than this in<br>most applied field research. Compared with these, the values in this study appear reasonable. Bliese<br>also suggests that values of ICC(2) above 0.70 should be considered acceptable—all the values here<br>are above 0.75, with all but two at least 0.89, demonstrating excellent inter-rater reliability in these<br>scales. The scales demonstrating highest coefficients are Pressure to Produce, Quality, and Efficiency,<br>and those with lowest coefficients are Integration and Reflexivity, with p<0.001 for all scales. These<br>are shown in Table 5.<br>As critics have argued that the intra-class coefficient is a measure of consistency and not agreement<br>(Shrout & Fleiss, 1979; James, 1982) a further agreement index was utilized. The within-group<br>agreement index of multiple item scales developed by James, Demaree, and Wolf (1984, 1993) is a<br>Table 5. Reliability and agreement of scales<br>Scale Internal consistency ICC(1) ICC(2) Average agreement<br>(Cronbach’s alpha) (rWG(j))<br>Autonomy 0.67 0.08 0.89 0.84<br>Integration 0.86 0.08 0.79 0.82<br>Involvement 0.87 0.12 0.93 0.80<br>Supervisory Support 0.88 0.10 0.92 0.83<br>Training 0.83 0.10 0.91 0.69<br>Welfare 0.91 0.17 0.95 0.77<br>Formalization 0.77 0.12 0.93 0.81<br>Tradition 0.73 0.15 0.94 0.77<br>Innov. & Flexibility 0.86 0.12 0.93 0.83<br>Outward Focus 0.83 0.17 0.90 0.85<br>Reflexivity 0.76 0.06 0.75 0.85<br>Clarity of Org. goals 0.87 0.12 0.93 0.76<br>Efficiency 0.80 0.23 0.97 0.79<br>Effort 0.79 0.12 0.93 0.85<br>Perform. Feedback 0.78 0.10 0.91 0.75<br>Pressure to Produce 0.79 0.22 0.96 0.83<br>Quality 0.80 0.22 0.96 0.84<br>394 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>technique for assessing agreement among the judgments made on a single variable in regard to a single<br>target. It is therefore an appropriate measure to use when testing climate perceptions within an organizational<br>unit. Agreement ranged from 0.69 to 0.85. Scales which yielded high indices included Outward<br>Focus, Reflexivity, Effort, Quality, and Autonomy, and the scale which produced the lowest index<br>was Training (a mean index of 0.69). Avalue of 0.7 or above has been suggested as the cut-off point to<br>indicate within-group inter-rater agreement (James, 1982). This is consistent with the values suggested<br>by Nunnally (1978) for acceptable internal consistency reliability estimates in this type of research.<br>Organizational climate agreement indices produced a median of 0.87 in a study by Ostroff (1993). All<br>agreement indices suggested a sufficient level of agreement between raters within organizations,<br>except for training, and this was very close to the recommended level. Full results can be seen in<br>Table 5.<br>Concurrent validity<br>The development of the OCM was undertaken as part of a wider longitudinal study examining relationships<br>between management practices, organizational climate, employee attitudes, and company economic<br>performance (West, Patterson, Pillinger, & Nickell, 1998). As indicated above, in addition to<br>conducting climate surveys, researchers spent a minimum of one day in each company conducting<br>semi-structured interviews with senior management about a range of company practices such as competitive<br>strategies, organizational structure, quality practices, and human resource management issues.<br>The interviews were conducted with the person primarily responsible for the practice in question (e.g.,<br>the chief executive for competitive strategy, the HRM director for training). Evidence from each interviewee<br>was cross-checked with additional information from company documents and a tour of the<br>manufacturing facility.<br>The interviewers’ ratings (which were made before the climate surveys were carried out) provided<br>rich data for determining the concurrent validity of the climate dimensions. As the data were not collected<br>from the respondents who completed the climate questionnaires, the correlations were not<br>inflated by common method variance. The following section describes the correlations between the<br>climate scales and aspects of the interview data relating to associated areas of organizational functioning.<br>All correlations are shown in Table 6.<br>Human Relations quadrant<br>Strong support for the concurrent validity of the Training scale was provided by a significant association<br>with interviewer ratings of the sophistication of company training practices (r¼0.52, p<0.01).<br>Also, the Autonomy dimension was significantly correlated with the level of responsibility inherent in<br>people’s jobs, as rated by interviewers (r¼0.46, p<0.01). The Integration scale was positively correlated<br>with the use of problem solving and cross-functional teams (r¼0.54, p<0.01) and interviewers’<br>ratings of the extent to which a total quality culture had been implemented (r¼0.40,<br>p<0.01). Both of these practices can help break down functional barriers.<br>Involvement was positively related to interviewers’ ratings of the coherence and coverage of the<br>appraisal system (r¼0.34, p<0.05). Such a relationship would be expected, since appraisal offers<br>a formal opportunity for discussion between management and staff of objectives, strategies, and processes<br>in the individual’s work. Similarly, company scores on the Welfare scale were related to interviewers’<br>ratings of the extent of harmonized terms and conditions (r¼0.34, p<0.05) and the level of<br>benefits offered to employees (r¼0.55, p<0.01).<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 395<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Table 6. Correlations of climate scales with similar variables from management interviews<br>Training Operator Use of cross- Total Appraisal Harmonization Level of Comm’n to Age Eff’ness of<br>responsibility functional quality benefits workforce of org. market<br>teams culture research<br>Autonomy 0.28 0.46 0.16 0.12 0.16 0.19 �0.02 0.14 �0.01 �0.18<br>Integration 0.27 0.15 0.54 0.40 0.08 �0.04 0.55 0.51 �0.18 0.11<br>Involvement 0.42 0.24 0.11 0.11 0.34 0.24 0.16 �0.02 �0.06 0.06<br>Supervisory 0.06 �0.08 0.40 0.21 0.15 �0.01 0.26 0.26 �0.09 0.03<br>Support<br>Training 0.52 0.20 0.48 0.44 0.19 0.29 0.47 0.58 �0.21 0.41<br>Welfare 0.45 0.07 0.38 0.22 0.28 0.34 0.55 0.38 �0.18 0.16<br>Formalization 0.23 �0.34 0.39 0.41 0.22 0.12 0.33 0.35 �0.07 0.31<br>Tradition �0.42 �0.12 �0.45 �0.36 �0.30 �0.15 �0.64 �0.51 0.08 �0.21<br>Innov. & 0.28 0.20 0.44 0.31 0.20 0.19 0.54 0.50 �0.34 0.05<br>Flexibility<br>Outward Focus 0.29 �0.25 0.40 0.37 0.13 0.03 0.62 0.45 0.03 0.46<br>Reflexivity 0.43 0.22 0.47 0.37 0.35 0.15 0.57 0.56 0.03 0.20<br>Clarity of 0.49 0.10 0.41 0.41 0.43 0.20 0.55 0.61 0.13 0.21<br>Org. goals<br>Efficiency 0.16 0.06 0.54 0.28 0.18 0.22 0.44 0.44 �0.14 0.07<br>Effort 0.12 �0.23 0.46 0.25 0.06 �0.01 0.25 0.33 0.01 0.18<br>Perform. 0.43 0.00 0.56 0.42 0.48 0.19 0.49 0.58 0.20 0.18<br>Feedback<br>Pressure �0.12 �0.25 0.03 0.02 �0.07 �0.18 �0.22 �0.01 0.25 0.05<br>to Produce<br>Quality 0.24 �0.35 0.27 0.17 0.11 �0.06 0.49 0.35 �0.04 0.33<br>p<0.05 for correlations of 0.28 or more; p<0.01 for correlations of 0.38 for more.<br>396 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Internal Process quadrant<br>The Formalization scale, assessing the degree to which employees perceive the company as relying on<br>formal rules and procedures, was negatively associated with interviewers’ ratings of employee responsibility<br>(r¼�0.34, p<0.05). The Tradition scale was negatively related to adoption of a number of<br>management practices associated with the ‘new manufacturing paradigm’ (Dean & Snell, 1991);<br>sophisticated training practices (r¼�0.42, p<0.01); use of appraisals (r¼�0.30, p<0.05); a total<br>quality culture (r¼�0.36, p<0.05), use of cross-functional teams (r¼�0.45, p<0.01), and frequent<br>and comprehensive communication to the workforce (r¼�0.51, p<0.01).<br>Open Systems quadrant<br>In older companies systems and patterns of behavior tend to become institutionalized, acting as a<br>source of organizational inertia. This inertia sets limitations on the ability of organizations to adapt<br>(Hannan & Freeman, 1977). Therefore the negative relationship between company age and the Innovation<br>& Flexibility scale provided some validation for this scale (r¼�0.34, p<0.05). The Outward<br>Focus dimension correlated positively with managers’ rated effectiveness of their company’s market<br>research activities (r¼0.46, p<0.01).<br>Rational Goal quadrant<br>The goal clarity and the feedback scales showed associations with closely related management practices.<br>The Clarity of Organizational Goals scale was positively related to interviewers’ ratings of the<br>frequency and comprehensiveness of communication to the workforce (r¼0.61, p<0.01), and Performance<br>Feedback was associated with the coherence and coverage of the appraisal system (r¼0.48,<br>p<0.01), and effective communication (r¼�0.58, p<0.01).<br>The Quality dimension was not significantly associated with interviewers’ ratings of the effectiveness<br>of company quality procedures.With this notable exception, overall, the associations between the<br>climate dimensions and closely related management practices assessed by interviews with senior management<br>indicate good concurrent validity.<br>Predictive validity<br>Approximately one year after the collection of climate data, the managing director or head of production<br>operations of each company completed a survey relating to organizational innovation in the areas<br>of products, production technology, and work organization. Researchers rated responses on a 5-point<br>scale, from 1 ‘not at all innovative’ to 5 ‘very innovative.’ These ratings were based on the extent of<br>innovation in relation to new products, adaptations of existing products, effects on production processes,<br>and numbers of staff affected by the innovations. Questionnaires were rated by three researchers<br>to check the reliability of the ratings. We examined whether the Innovation & Flexibility and<br>Reflexivity climate scores provided by employees were associated with researcher ratings of organizational<br>innovation one year later. The Innovation & Flexibility and Reflexivity scales were significantly<br>associated with researchers’ ratings of innovativeness in products (r¼0.53, p<0.01; r¼0.41,<br>p<0.01 respectively), technology (r¼0.44, p<0.01; r¼0.31, p<0.01), and work organization<br>(r¼0.36, p<0.01; r¼0.39, p<0.01).<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 397<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>We were also able to investigate the predictive validity of the measure of organizational climate in<br>relation to organizational productivity. Productivity data (logarithm of financial value of net sale per<br>employee) for 42 companies from our sample were collected from company accounts lodged in Companies<br>House (the central repository for the compulsory submission of annual accounts made by every<br>limited company in the UK). Subsequent productivity (collected one year after the climate survey) was<br>significantly correlated in controlled analyses with eight aspects of organizational climate (training,<br>welfare, supervisory support, effort, innovation and flexibility, quality, performance feedback, and formalization).<br>For a detailed report of these findings see Patterson et al. (2004).<br>Discussion<br>The OCM was developed and tested with a large sample of employees drawn from 55 manufacturing<br>organizations. The results revealed that the instrument has sound psychometric properties and provides<br>researchers with a robust means for assessing 17 dimensions of employee perceptions of their work<br>environments. Moreover, the internal reliability of the scales was acceptable in almost all cases and<br>good in many.<br>Uniquely, we were able to assess the concurrent and predictive validity of many of the scales drawing<br>on intensive interviews conducted with managers in the 55 organizations; upon interviewer ratings<br>of dimensions of company functioning such as innovative orientation, sophistication and extent of<br>human resource management practices, richness and responsibility of front line jobs; and on productivity<br>and innovation data.<br>The measure also discriminated effectively between organizations, demonstrating good discriminant<br>validity. Agreement between raters (employees) within organizations was high, showing that<br>the measure satisfied the conditions for aggregation of perceptions of employees. The fact that the<br>measure does effectively discriminate between organizations is important, since it suggests that variations<br>in climate can reliably be reflected in the data gathered using this instrument. Studies of organizational-<br>level factors and relationships with climate can therefore be embarked upon with a high<br>degree of confidence, especially where researchers have identified, a priori, organizations which are<br>likely to differ in characteristics under study, and which can be related to climate (e.g., economic performance,<br>innovation, employee commitment).<br>We did not predict nor find a second-order factor structure mapping on to the four quadrants of the<br>Competing Values model. The Competing Values model proposes that organizations span all four<br>quadrants, having a greater or lesser degree of emphasis across all four, rather than fitting into one<br>or other. The model specifically rejects a typological approach, arguing that understanding of organizations<br>should be based on the relative emphasis they give to each of the competing values. Forcing<br>them into one or other quadrant would misrepresent their inherent span of activity and experience and<br>of those who work within them. Employees experience control, flexibility, and internal and external<br>focus in their organizations. It is the degree of emphasis and associated types of managerial practices<br>(effectively the application of managerial ideologies) that is of interest to climate researchers.<br>Similarly, we would not expect that organizations would have a high degree of emphasis on every or<br>even most dimensions of climate within a quadrant. For two organizations with strong human relations<br>orientations we might find that one emphasizes supervisory support but not autonomy, while another<br>might emphasize training but not involvement. The Competing Values model provides a topography of<br>organizational climate—it does not propose that organizations characterized by a strong human relations<br>orientation will have high scores on every or most dimensions within that quadrant.<br>398 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>The OCM was developed from a theoretical base, the Competing Values model, itself reflecting considerable<br>prior theoretical development. By basing the measure on underlying conceptual bases in the<br>organizational sciences, we aimed to provide a measure which would be useful for a broad range of<br>research interests, enabling researchers to test central theoretical propositions about relationships, such<br>as those between climate and organizational effectiveness. The measure consequently offers a number<br>of unique advantages.<br>First it is relatively comprehensive, offering researchers the advantage of being able to assess<br>employees’ experience over many fundamental dimensions of climate. It is unlikely that most<br>researchers will wish to apply all 17 scales of the OCM in their research since the logistics of administering<br>the questionnaires, and analyzing the resultant data sets, are likely to prove too cumbersome.<br>Moreover, it might suggest a lack of theoretical focus. Indeed, the instrument can be used in a more<br>refined way by selecting scales most applicable to the research questions being posed. This might<br>require the use of only one or two scales, such as those tapping innovation orientation or traditionalism.<br>Alternatively, whole domains might be assessed by using all scales in one quadrant (e.g., all seven<br>scales in the Human Relations domain). The wider applicability of the scales in this regard has been<br>demonstrated by our application of them in a service setting, where we have demonstrated factorial<br>invariance, reliability, and validity of 12 of the scales in their use amongst 5000 employees in 27 hospitals<br>in the United Kingdom (Hill, West, Patterson, & Borrill, 1997).<br>The measure can be used to study mergers and acquisitions since it will provide an indication of the<br>similarity or difference between organizational climates or value emphasis before and after merger<br>(are they similar in terms of emphases and in which quadrants of the model?). Similarly, it can be<br>applied to studies of organizational change and specifically culture change to determine the effects<br>on employees’ experience of change. It is possible, for example, to determine whether most change<br>occurs in scales in the open systems or internal process quadrants. Are the intended consequences of<br>the change (a more human relations-oriented organization for example) borne out by the results of the<br>application of the measure? The measure can also be used to assess differences between subcultures in<br>large organizations. Moreover, researchers can examine organizational change processes generally to<br>determine trends over time—are most organizational changes focused on internal process, open systems,<br>rational goal or human relations values?<br>The measure’s development also opens up possibilities for testing theory. Quinn proposed that<br>strong emphasis across all four domains (climate strength) would be associated with organizational<br>effectiveness and the measure can be employed to test this prediction. Moreover, it enables testing<br>of other ideas about climate strength (where all or most climate dimensions within a quadrant intercorrelate<br>highly) as a predictor of organizational outcomes. The alternative conception of strong climates<br>(where organizational members agree in their climate ratings) and how this relates to<br>organizational outcomes can also be tested. Furthermore, because the measure is based on the four<br>value domains, researchers can both develop and test theory about the relationship between specific<br>climate dimensions in or across quadrants and a broad range of outcomes (e.g., turnover, productivity,<br>accidents, citizenship, and innovation).<br>The research also suggested some gaps in both the model and the measure. The body of work on<br>organizational ethics (e.g., Trevino & Weaver, 1994; Victor & Cullen, 1988) and governance<br>(e.g.,Wood, 1994) is not reflected in the Competing Values model. Values that emphasize ethics apply<br>to all four quadrants. Specifically, we believe that the Competing Values model neglects dimensions<br>that are common to or span all four areas of values or emphases. Ethics is an issue in all areas. The<br>model, we propose, should incorporate a central (Venn diagram-like) component that integrates common<br>elements such as an emphasis on ethical practice and policy. We see this as an area for future<br>theoretical and psychometric development. Finally, it is important that the measure is applied across<br>a variety of organizational types and in a variety of national contexts.<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 399<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>In conclusion, we hope researchers find this measure useful in their research and that we are<br>able to more easily accumulate understanding about the role of climate in organizational<br>functioning.<br>Author biographies<br>Malcolm G. Patterson is a Research Fellow in the Institute of Work Psychology, University of Sheffield,<br>from where he received his MSc in Occupational Psychology. His current research interests<br>include human resource management and performance, stress management interventions, and evidence-<br>based practice.<br>Michael A. West is Professor of Organizational Psychology and Director of Research at Aston Business<br>School. He has also been a member of the Centre for Economic Performance at the London<br>School of Economics since 1991. He graduated from the University of Wales in 1973 and received<br>his PhD in 1977. He then spent a year working in the coal mines of South Wales before beginning<br>his academic career.<br>Viv J. Shackleton is Senior Lecturer in Work and Organizational Psychology at Aston University<br>Business School. Previously he has worked at Ashridge Management College, UK, and Rochester University,<br>NY. His research interests are in the areas of assessment, recruitment and selection, and in<br>cross-cultural management.<br>Jeremy F. Dawson is a statistician in the Work and Organizational Psychology group at Aston Business<br>School, Aston University, UK. Much of his work involves analysis of large data sets collected by<br>the group for research into human resource management, organizational climate, innovation, and team<br>working. His own research interests focus on the methodology of research in small groups, in particular<br>the accuracy of incomplete data, and measures of diversity.<br>Rebecca Lawthom is a Principal Lecturer in Psychology at Manchester Metropolitan University. She<br>contributes to courses in social, community and counselling psychology. She has research interests in<br>gender and organizations, and marginalized groups. She is a co-author of Researching life stories: theory<br>and method in a biographical age (2004, Routledge Falmer), which explores gender and narrative<br>methodology.<br>Sally Maitlis is an Assistant Professor of Organizational Behavior at the Sauder School of<br>Business, University of British Columbia. She received her PhD from the University of<br>Sheffield. Her research interests include the social and political aspects of organizational sensemaking<br>and decision-making processes, and narrative and discursive approaches to the study of emotion<br>in organizations. Her work has been published in the Academy of Management Journal, Organization<br>Science, Organization Studies, Journal of Management Studies, and American Behavioral<br>Scientist.<br>David L. Robinson was project statistician at the Institute of Work Psychology, University of Sheffield,<br>during the time the analysis in this work was undertaken. He is now Statistician/Research Assistant<br>in the Department of Oral Health and Development at the University’s Dental School. He was<br>awarded an MSc in Statistics with distinction from the University of Sheffield in 1997 and is currently<br>in the final stages of a PhD on statistical shape analysis.<br>Alison M. Wallace is currently completing her postgraduate studies in Organizational Psychology at<br>the University of Queensland, Australia. Her research interests include affective experiences of work<br>and climate and culture, and the application of research methods and insights in the field of public<br>policy.<br>400 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>References<br>Anderson, N., & West, M. A. (1998). Measuring climate for work group innovation: development and validation<br>of the team climate inventory. Journal of Organizational Behavior, 19, 235–258.<br>Annett, J. (1969). Feedback and human behaviour. Harmondsworth, U.K.: Penguin.<br>Arbuckle, J. L. (1996). Full information estimation in the presence of incomplete data. In G. A. Marcoulides,<br>& R. E. Schumacker (Eds.), Advanced structural equation modeling (pp. 243–277). Mahwah, NJ:<br>Erlbaum.<br>Ashkanasy, N. M.,Wilderom, C. P. M., & Peterson, M. F. (2000). Handbook of organizational culture and climate.<br>London: Sage.<br>Beyer, J. M. (1981). Ideologies, values, and decision making in organizations. In P. C. Nystrom, & W. H.<br>Starbuck (Eds.), Handbook of organizational design (Vol. 2, pp. 166–202). New York: Oxford University<br>Press.<br>Bliese, P. D. (2000). Within-group agreement, non-independence and reliability. In K. J. Klein, & S. W. J.<br>Kozlowski (Eds.), Multilevel theory, research and methods in organizations: Foundations, extensions and new<br>directions (pp. 349–381). San Francisco, CA: Jossey-Bass.<br>Brown, S., & Leigh, T. W. (1996). A new look at psychological climate and its relationship to job involvement,<br>effort, and performance. Journal of Applied Psychology, 81, 358–368.<br>Callan, V. J. (1993). Subordinate–manager communication in different sex dyads: consequences for job<br>satisfaction. Journal of Occupational and Organisational Psychology, 66, 13–27.<br>Campbell, J. P., Dunnette, M. D., Lawler, E. E., & Weick, K. E. (1970). Managerial behavior, performance, and<br>effectiveness. New York: McGraw Hill.<br>Cherns, A. (1976). The principles of socio-technical systems design. Human Relations, 29, 783–792.<br>Clinebell, S. (1984). Organizational effectiveness: an examination of recent empirical studies and the development<br>of a contingency view. In W. D. Terpening, & K. R. Thompson (Eds.), Proceedings of the 27th annual<br>conference of the Midwest Academy of Management (pp. 92–102), Department of Management, University of<br>Notre Dame.<br>Coch, L., & French, J. R. (1948). Overcoming resistance to change. Human Relations, 1, 512–532.<br>Cummins, R. C. (1990). Job stress and the buffering effect of supervisory support. Group and Organization<br>Studies, 15, 92–104.<br>Day, D. V., & Bedeian, A. G. (1991). Predicting job performance across organizations: the interaction of work<br>orientation and psychological climate. Journal of Management, 17, 589–600.<br>Dean, J. W., & Snell, S. A. (1991). Integrated manufacturing and job design: moderating effects of organizational<br>inertia. Academy of Management Journal, 34, 774–804.<br>Deming, W. E. (1986). Out of the crisis. Cambridge, MA: MIT Press.<br>Denison, D. R. (2001). Organizational culture: can it be a key lever for driving organizational change? In<br>C. L. Cooper, S. Cartwright, & P. C. Earley (Eds.), International handbook of organizational culture and climate<br>(pp. 347–372). Chichester: Wiley.<br>Denison, D. R., & Mishra, A. K. (1995). Toward a theory of organizational culture and effectiveness. Organization<br>Science, 6, 204–223.<br>Drexler, J. (1977). Organizational climate: its homogeneity within organizations. Journal of Applied Psychology,<br>62, 38–42.<br>Eisenberger, R., Stinglhamber, F., Vandenberghe, C., Sucharski, I., & Rhoades, L. (2002). Perceived supervisor<br>support: contributions to perceived organizational support and employee retention. Journal of Applied<br>Psychology, 87, 565–573.<br>Emery, F. E.,& Trist, E. L. (1965). The causal texture of organizational environments. Human Relations, 18, 21–32.<br>Floyd, F. J., & Widaman, K. F. (1995). Factor analysis in the development and refinement of clinical assessment<br>instruments. Psychological Assessment, 7, 286–295.<br>Garrahan, P., & Stewart, P. (1992). The Nissan enigma: Flexibility at work in a local economy. London: Mansell.<br>Gattiker, U. E. (1995). Firm and taxpayer returns from training of semiskilled employees. Academy of<br>Management Journal, 38, 1152–1173.<br>Gavin, J. R., & Howe, J. G. (1975). Psychological climate: some theoretical and empirical considerations.<br>Behavioral Science, 20, 228–240.<br>Gifford, B. D., Zammuto, R. F., & Goodman, E. A. (2002). The relationship between hospital unit culture and<br>nurses quality of life. Journal of Healthcare Management, 47, 13–26.<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 401<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Glick, W. H. (1985). Conceptualizing and measuring organizational and psychological climate: pitfalls in<br>multilevel research. Academy of Management Review, 10, 601–616.<br>Goodman, S. A., & Svyantek, D. J. (1999). Person–organization fit and contextual performance: do shared values<br>matter? Journal of Vocational Behavior, 55, 254–275.<br>Gordon, G. G., & DiTomaso, N. (1992). Predicting corporate performance from organizational culture. Journal of<br>Management Studies, 29, 783–798.<br>Graham, J. W., Hofer, S. M., & MacKinnon, D. P. (1996). Maximizing the usefulness of data obtained with<br>planned missing value patterns: an application of maximum likelihood procedures. Multivariate Behavioral<br>Research, 31, 197–218.<br>Griffin, M. A., & Mathieu, J. E. (1997). Modeling organizational processes across hierarchical levels: climate,<br>leadership, and group process in work groups. Journal of Organizational Behavior, 18, 731–744.<br>Guest, D. (1998). Is the psychological contract worth taking seriously? Journal of Organizational Behavior, 19,<br>649–664.<br>Hackman, J. R., & Wageman, R. (1995). Total quality management: empirical, conceptual, and practical issues.<br>Administrative Science Quarterly, 40, 309–342.<br>Hall, R. H. (1980). Effectiveness theory and organizational effectiveness. Journal of Applied Behavioral Science,<br>16, 536–545.<br>Hall, R. H. (1991). Organizations: Structures, processes and outcomes. Englewood Cliffs, NJ: Prentice-Hall.<br>Hannan, M. T., & Freeman, J. (1977). The population ecology of organizations. American Journal of Sociology,<br>82, 929–964.<br>Hargie, O., & Tourish, P. (2000). Handbook of communication audits. London: Routledge.<br>Heller, F., Pusic´, E., Strauss, G., & Wilpert, B. (1998). Organizational participation: Myth and reality. Oxford:<br>Oxford University Press.<br>Hill, F., West, M. A., Patterson, M. G., & Borrill, C. (1997). Is organizational climate a meaningful construct?<br>Working paper, Institute of Work Psychology, University of Sheffield, IWP memo no. 102.<br>Hofmann, D. A., & Stetzer, A. (1996). A cross-level investigation of factors influencing unsafe behaviors and<br>accidents. Personnel Psychology, 49, 307–339.<br>Hollander, E. P., & Offerman, L. R. (1990). Power and leadership in organizations: relationships in transition.<br>American Psychologist, 45, 179–189.<br>Howe, J. G. (1977). Group climate: an exploratory analysis of construct validity. Organizational Behavior and<br>Human Performance, 19, 106–125.<br>James, L. R. (1982). Aggregation bias in estimates of perceptual agreement. Journal of Applied Psychology, 67,<br>219–229.<br>James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-group interrater reliability with and without<br>response bias. Journal of Applied Psychology, 69, 85–98.<br>James, L. R., Demaree, R. G., & Wolf, G. (1993). An assessment of within-group interrater agreement. Journal of<br>Applied Psychology, 78, 306–309.<br>Jones, A. P., & James, L. R. (1979). Psychological climate: dimensions and relationships of individual and<br>aggregated work environment perceptions. Organizational Behavior and Human Performance, 23, 201–250.<br>James, L. A., & James, L. R. (1989). Integrating work environment perceptions: explorations into the<br>measurement of meaning. Journal of Applied Psychology, 74, 739–751.<br>James, L. R., James, L. A., & Ashe, D. K. (1990). The meaning of organizations: the role of cognition and<br>values. In B. Schneider (Ed.), Organizational climate and culture (pp. 40–129). San Francisco, CA:<br>Jossey-Bass.<br>James, L. R., & Jones, A. P. (1974). Organizational climate: a review of theory and research. Psychological<br>Bulletin, 81, 1096–1112.<br>James, L. R., & Jones, A. P. (1980). Perceived job characteristics and job satisfaction: an examination of reciprocal<br>causation. Personnel Psychology, 33, 97–135.<br>James, L. R., & McIntyre, M. D. (1996). Perceptions of organizational climate. In K. R. Murphy (Ed.), Individual<br>differences and behavior in organizations (pp. 416–450). San Francisco, CA: Jossey-Bass.<br>James, L. R., & Sells, S. B. (1981). Psychological climate: theoretical perspectives and empirical research. In<br>D. Magnusson (Ed.), Toward a psychology of situations: An interactional perspective (pp. 275–292). Hillsdale,<br>NJ: Erlbaum.<br>James, L. R., & Tetrick, L. E. (1986). Confirmatory analytic tests of three causal models relating job perceptions to<br>job satisfaction. Journal of Applied Psychology, 71, 77–82.<br>Joyce, W. F., & Slocum, J. W., Jr. (1979). Climates in organizations. In S. Kerr (Ed.), Organizational behavior<br>(pp. 317–333). Columbus, OH: Grid.<br>402 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Kiesler, S., & Sproull, L. (1982). Managerial responses to changing environments: perspectives on problem<br>sensing from social cognition. Administrative Science Quarterly, 27, 548–570.<br>King, N., & Anderson, N. (1995). Innovation and change in organizations. London: Routledge.<br>Klein, J. A. (1991). A reexamination of autonomy in light of new manufacturing processes. Human Relations, 44,<br>21–38.<br>Kopelman, R. E. (1986). Objective feedback. In E. A. Locke (Ed.), Generalizing from laboratory to field settings<br>(pp. 119–145). Lexington, MA: Lexington Books.<br>Kotter, J. P., & Heskett, J. L. (1992). Corporate culture and performance. New York: Free Press.<br>Lawler, E. E., Hall, D. T., & Oldham, G. R. (1974). Organizational climate: relationship to organizational<br>structure, process and performance. Organizational Behavior and Performance, 11, 139–155.<br>Lawrence, P. R., & Lorsch, J.W. (1967). Differentiation and integration in complex organizations. Administrative<br>Science Quarterly, 10, 21–38.<br>Litwin, G. H., & Stringer, R. A. (1968). Motivation and organizational climate. Boston, MA: Harvard University<br>Press.<br>Locke, E. A. (1991). The motivation sequence, the motivation hub, and the motivation core. Organizational<br>Behavior and Human Decision Processes, 50, 288–299.<br>Mathieu, J. E., Hoffman, D. A., & Farr, J. L. (1993). Job perceptions–job satisfaction relations: an empirical<br>comparison of three competing theories. Organizational Behavior and Human Decision Processes, 56,<br>370–387.<br>McCaol, K. D., Hinsz, V. B., & McCaol, H. S. (1987). The effects of commitment to performance goal on effort.<br>Journal of Applied Social Psychology, 17, 437–452.<br>McGregor, D. (1960). The human side of enterprise. New York: McGraw-Hill.<br>Miller, K. I., & Monge, P. R. (1986). Participation, satisfaction and productivity: a meta-analytic review. Academy<br>of Management Journal, 29, 727–793.<br>Morrow, C. C., Jarrett, M. Q., & Rupinski, M. T. (1997). An investigation of the effect and economic utility of<br>corporate-wide training. Personnel Psychology, 50, 91–119.<br>Muchinsky, P. M. (1976). An assessment of the Litwin and Stringer Organizational Climate Questionnaire: an<br>empirical and theoretical extension of the Sims and LaFollette study. Personnel Psychology, 29, 371–392.<br>Murphy, L. R., Gershon, R. M., & DeJoy, D. (1996). Stress and occupational exposure to HIV/AIDS. In C. L.<br>Cooper (Ed.), Handbook of stress, medicine, and health (pp. 176–190). Boca Raton, FL: CRC Press.<br>Nauta, A., & Sanders, K. (2000). Interdepartmental negotiation behavior in manufacturing organizations.<br>International Journal of Conflict Management, 11, 135–161.<br>Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill.<br>Ostroff, C. (1993). Comparing correlations based on individual-level and aggregated data. Journal of Applied<br>Psychology, 78, 569–582.<br>Ostroff, C., & Schmitt, N. (1993). Configurations of organizational effectiveness and efficiency. Academy of<br>Management Journal, 36, 1345–1362.<br>Patterson, M. G.,Warr, P. B., &West, M. A. (2004). Organizational climate and company performance: the role of<br>employee affect and employee level. Journal of Occupational and Organizational Psychology, 77, 193–216.<br>Patterson, M., West, M. A., Lawthom, R., & Nickell, S. (1997). Impact of people management practices on<br>business performance. London: Institute of Personnel and Development.<br>Payne, R. L., Brown, A. D., & Gaston, K. (1992). Reliability and validity of an updated version of the Business<br>Organisation Climate Index (BOCI). (Discussion paper). Sheffield, England: Sheffield University Management<br>School.<br>Payne, R., & Mansfield, R. (1978). Correlates of individuals’ perceptions of organization climates. Journal of<br>Occupational Psychology, 51, 209–218.<br>Payne, R. L., & Pheysey, D. C. (1971). G.G. Stern’s organizational climate index: a reconceptualization<br>and application to business organizations. Organizational Behavior and Human Performance, 6, 77–78.<br>Pritchard, R. D., & Karasick, B. W. (1973). The effects of organizational climate on managerial job performance<br>and satisfaction. Organizational Behavior and Human Performance, 9, 126–146.<br>Pugh, D. S., Hickson, D. J., Hinings, C. R., & Turner, C. (1968). Dimensions of organizational structure.<br>Administrative Science Quarterly, 13, 65–91.<br>Quinn, R. E. (1988). Beyond rational management: Mastering the paradoxes and competing demands of high<br>performance. San Francisco, CA: Jossey-Bass.<br>Quinn, R. E., & McGrath, M. R. (1985). The transformation of organizational culture: a competing values<br>perspective. In P. J. Frost, L. F. Moore, M. R. Louis, C. C. Lundberg, & J. Martin (Eds.), Organizational culture<br>(pp. 315–344). Beverly Hills, CA: Sage.<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 403<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Quinn, R. E., & Rohrbaugh, J. (1981). A Competing Values approach to organizational effectiveness. Public<br>Productivity Review, 5, 122–140.<br>Quinn, R. E., & Rohrbaugh, J. (1983). A spatial model of effectiveness criteria: toward a competing values<br>approach to organizational analysis. Management Science, 29, 363–377.<br>Rentsch, J. (1990). Climate and culture: interaction and qualitative differences in organizational meanings.<br>Journal of Applied Psychology, 75, 668–681.<br>Robinson, S. L., & Rousseau, D. M. (1994). Violating the psychological contract: not the exception but the norm.<br>Journal of Organizational Behavior, 15, 245–259.<br>Rogers, E. D., Miles, W. G., & Biggs, W. D. (1980). The factor replicability of the Litwin and Stringer<br>Organizational Climate Questionnaire: an inter- and intra-organizational assessment. Journal of Management,<br>6, 65–78.<br>Rousseau, D. M. (1988). The construction of climate in organizational research. In C. L. Cooper, & I. T. Robertson<br>(Eds.), International review of industrial and organizational psychology (Vol. 3, pp. 139–158). New York:<br>Wiley.<br>Schafer, J. L. (1997). NORM [Computer software]. Philadelphia: Pennsylvania State University.<br>Schein, E. (2000). Sense and nonsense about culture and climate. In N. M. Ashkanasy, C. P. M. Wilderom, &<br>M. F. Peterson (Eds.), Handbook of organizational culture & climate (pp. xxiii–xxx). Thousand Oaks, CA:<br>Sage.<br>Schein, E. H. (1985). Organizational culture and leadership: A dynamic view. San Francisco, CA: Jossey-Bass.<br>Schneider, B. (1975). Organizational climates: an essay. Personnel Psychology, 36, 19–36.<br>Schneider, B. (1980). The service organization: climate is crucial. Organizational Dynamics, 9, 52–65.<br>Schneider, B. (1983). An interactionist perspective on organizational effectiveness. In L. L. Cummings, & B. M.<br>Staw (Eds.), Research in organizational behavior (Vol. 5, pp. 1–31). Greenwich, CT: JAI Press.<br>Schneider, B. (1990). The climate for service: an application of the climate construct. In B. Schneider (Ed.),<br>Organizational climate and culture (pp. 383–412). San Francisco, CA: Jossey-Bass.<br>Schneider, B. (2000). The psychological life of organizations. In N. M. Ashkanasy, C. P. M. Wilderon, &<br>M. F. Peterson (Eds.), Handbook of organizational culture and climate (pp. xvii–xxi). Thousand Oaks, CA:<br>Sage.<br>Schneider, B., Bowen, D. E., Ehrhart, M. G., & Holcombe, K. M. (2000). The climate for service: evolution of a<br>construct. In N. M. Ashkanasy, C. P.M.Wilderom, &M. F. Peterson (Eds.), Handbook of organizational culture<br>and climate (pp. 21–36). Thousand Oaks, CA: Sage.<br>Schneider, B., Parkington, J. J., & Buxton, V. M. (1980). Employee and customer perceptions of service in banks.<br>Administrative Science Quarterly, 25, 252–267.<br>Schneider, B., & Reichers, A. E. (1983). On the etiology of climates. Personnel Psychology, 36, 19–39.<br>Schneider, B., Smith, D. B., & Goldstein, H. W. (2000). Attraction–selection–attrition: toward a person–<br>environment psychology of organizations. In W. B. Walsh, K. H. Craik, & R. H. Price (Eds.), Person–<br>Environment Psychology: New Directions and Perspectives, 2nd edn. (pp. 61–85). Mahwah, NJ: Erlbaum.<br>Schneider, B., White, S. S., & Paul, M. C. (1998). Linking service climate and customer perceptions of service<br>quality: tests of a causal model. Journal of Applied Psychology, 83, 150–163.<br>Sheridan, J. E. (1992). Organizational culture and retention. Academy of Management Journal, 35, 1036–1056.<br>Shipper, F., & White, C. S. (1983). Linking organizational effectiveness and environmental change. Long Range<br>Planning, 16(3), 99–106.<br>Shrout, P. E., & Fleiss, J. L. (1979). Interclass correlations: use in assessing rater reliability. Psychological<br>Bulletin, 86, 420–428.<br>Sims, H. P., & LaFollette, W. (1975). An assessment of the Litwin and Stringer Organizational Climate<br>Questionnaire. Personnel Psychology, 28, 19–38.<br>Svyantek, D. J., & Bott, J. P. (2004). Organizational culture and organizational climate measures: an integrative<br>review. In J. C. Thomas (Ed.), Comprehensive handbook of psychological assessment: Industrial and<br>organizational assessment (Vol. 4, pp. 507–524). Hoboken, NJ: Wiley.<br>Taira, K. (1996). Compatibility of human resource management, industrial relations and engineering under mass<br>production and lean production: an exploration. Applied Psychology: An International Review, 45, 97–117.<br>Trevino, L. K., & Weaver, G. (1994). Business ETHICS/BUSINESS Ethics: one field or two? Business Ethics<br>Quarterly, 4, 113–128.<br>Victor, B., & Cullen, J. B. (1988). The organizational bases for ethical work climates. Administrative Science<br>Quarterly, 333, 101–125.<br>Weber, Y. (1996). Corporate culture fit and performance in mergers and acquisitions. Human Relations, 49, 1181–<br>1202.<br>404 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>West, M. A. (1990). The social psychology of innovation in groups. In M. A.West, & J. L. Farr (Eds.), Innovation<br>and creativity at work: Psychological and organizational strategies (pp. 81–100). Chichester: Wiley.<br>West, M. A. (1996). Reflexivity and work group effectiveness: a conceptual integration. In M. A. West (Ed.),<br>Handbook of work group psychology (pp. 555–579). Chichester: Wiley.<br>West, M. A. (2000). Reflexivity, revolution, and innovation in work teams. In M. M. Beyerlein, D. A. Johnson, &<br>S. T. Beyerlein (Eds.), Product development teams (pp. 1–29). Stamford, CT: JAI Press.<br>West, M. A., & Anderson, N. (1996). Innovation in top management teams. Journal of Applied Psychology, 81,<br>680–693.<br>West, M. A., & Farr, J. L. (1990). Innovation and creativity at work: Psychological and organizational strategies.<br>Chichester: Wiley.<br>West, M. A., Patterson, M., Pillinger, T., & Nickell, S. (1998). Innovation and change in manufacturing. Institute<br>of Work Psychology, University of Sheffield.<br>West, M. A., & Wallace, M. (1991). Innovation in health care teams. European Journal of Social Psychology, 21,<br>303–315.<br>Wilderom, C. P. M., Glunk, U., & Maslowski, R. (2000). Organizational culture as a predictor of organizational<br>performance. In N. M. Ashkanasy, C. P. M. Wilderom, & M. F. Peterson (Eds.), Handbook of organizational<br>culture and climate (pp. 193–209). Thousand Oaks, CA: Sage.<br>Wood, D. J. (1994). Business and society, 2nd edn. New York: Harper Collins.<br>Zammuto, R. F., Gifford, B., & Goodman, E. A. (2000). Managerial ideologies, organization culture, and the<br>outcomes of innovation: a competing values perspective. In N. M. Ashkanasy, C. P. M. Wilderon, & M. F.<br>Peterson (Eds.), Handbook of organizational culture and climate (pp. 261–278). Thousand Oaks, CA: Sage.<br>Zammuto, R. F., & O’Connor, E. J. (1992). Gaining advanced manufacturing technologies’ benefits: the roles of<br>organizational design and culture. Academy of Management Review, 17, 701–729.<br>Appendix<br>Organizational Climate Measure#<br>The Organizational Climate Measure (OCM) consists of 17 scales, divided in to four quadrants: human<br>relations, internal process, open systems, and rational goal. Items marked with an asterisk (*) are<br>reversed before the scale is calculated. Regression weights quoted are from the confirmatory factor<br>analysis on the second 50 per cent of the data; all are highly statistically significant.<br>The response scale is: 1¼‘Definitely false’, 2¼‘Mostly false’, 3¼‘Mostly true’, 4¼‘Definitely<br>true’.<br>Item Regression weight<br>Autonomy<br>Management let people make their own decisions much of the time 0.585<br>Management trust people to take work-related decisions without getting permission first 0.556<br>People at the top tightly control the work of those below them* 0.528<br>Management keep too tight a reign on the way things are done around here* 0.516<br>It’s important to check things first with the boss before taking a decision* 0.513<br>Integration<br>People are suspicious of other departments* 0.699<br>There is very little conflict between departments here 0.719<br>People in different departments are prepared to share information 0.719<br>Collaboration between departments is very effective 0.728<br>There is very little respect between some of the departments here* 0.766<br>Continues<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 405<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Appendix. Continued<br>Item Regression weight<br>Involvement<br>Management involve people when decisions are made that affect them 0.633<br>Changes are made without talking to the people involved in them* 0.752<br>People don’t have any say in decisions which affect their work* 0.739<br>People feel decisions are frequently made over their heads* 0.738<br>Information is widely shared 0.695<br>There are often breakdowns in communication here* 0.744<br>Supervisory Support<br>Supervisors here are really good at understanding peoples’ problems 0.749<br>Supervisors show that they have confidence in those they manage 0.712<br>Supervisors here are friendly and easy to approach 0.715<br>Supervisors can be relied upon to give good guidance to people 0.807<br>Supervisors show an understanding of the people who work for them 0.851<br>Training<br>People are not properly trained when there is a new machine or bit of equipment* 0.765<br>People receive enough training when it comes to using new equipment 0.804<br>The company only gives people the minimum amount of training they need to do their job* 0.741<br>People are strongly encouraged to develop their skills 0.649<br>Welfare<br>This company pays little attention to the interests of employees* 0.729<br>This company tries to look after its employees 0.907<br>This company cares about its employees 0.918<br>This company tries to be fair in its actions towards employees 0.823<br>Formalization<br>It is considered extremely important here to follow the rules 0.676<br>People can ignore formal procedures and rules if it helps get the job done* 0.640<br>Everything has to be done by the book 0.591<br>Its not necessary to follow procedures to the letter around here* 0.650<br>Nobody gets too upset if people break the rules around here* 0.581<br>Tradition<br>Senior management like to keep to established, traditional ways of doing things 0.688<br>The way this organization does things has never changed very much 0.469<br>Management are not interested in trying out new ideas 0.645<br>Changes in the way things are done here happen very slowly 0.714<br>Innovation & Flexibility<br>New ideas are readily accepted here 0.715<br>This company is quick to respond when changes need to be made 0.702<br>Management here are quick to spot the need to do things differently 0.701<br>This organization is very flexible; it can quickly change procedures to meet new 0.702<br>conditions and solve problems as they arise<br>Assistance in developing new ideas is readily available 0.682<br>People in this organization are always searching for new ways of looking at problems 0.716<br>Outward Focus<br>This organization is quite inward looking; it does not concern itself with what is 0.774<br>happening in the market place*<br>Ways of improving service to the customer are not given much thought* 0.677<br>Customer needs are not considered top priority here* 0.767<br>This company is slow to respond to the needs of the customer* 0.652<br>This organization is continually looking for new opportunities in the market place 0.588<br>Continues<br>406 M. G. PATTERSON ET AL.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Appendix. Continued<br>Item Regression weight<br>Reflexivity<br>In this organization, the way people work together is readily changed in 0.551<br>order to improve performance<br>The methods used by this organization to get the job done are often discussed 0.747<br>There are regular discussions as to whether people in the organization are working 0.664<br>effectively together<br>In this organization, objectives are modified in light of changing circumstances 0.482<br>In this organization, time is taken to review organizational objectives 0.707<br>Clarity of Organizational Goals<br>People have a good understanding of what the organization is trying to do 0.664<br>The future direction of the company is clearly communicated to everyone 0.795<br>People aren’t clear about the aims of the company* 0.659<br>Everyone who works here is well aware of the long-term plans and direction of this company 0.842<br>There is a strong sense of where the company is going 0.839<br>Efficiency<br>Time and money could be saved if work were better organized* 0.768<br>Things could be done much more efficiently, if people stopped to think* 0.605<br>Poor scheduling and planning often result in targets not being met* 0.657<br>Productivity could be improved if jobs were organized and planned better* 0.829<br>Effort<br>People here always want to perform to the best of their ability 0.660<br>People are enthusiastic about their work 0.692<br>People here get by with doing as little as possible* 0.604<br>People are prepared to make a special effort to do a good job 0.688<br>People here don’t put more effort into their work than they have to* 0.637<br>Performance Feedback<br>People usually receive feedback on the quality of work they have done 0.639<br>People don’t have any idea how well they are doing their job* 0.629<br>In general, it is hard for someone to measure the quality of their performance* 0.635<br>People’s performance is measured on a regular basis 0.629<br>The way people do their jobs is rarely assessed* 0.702<br>Pressure to Produce<br>People are expected to do too much in a day 0.583<br>In general, peoples’ workloads are not particularly demanding* 0.647<br>Management require people to work extremely hard 0.722<br>People here are under pressure to meet targets 0.585<br>The pace of work here is pretty relaxed* 0.672<br>Quality<br>This company is always looking to achieve the highest standards of quality 0.791<br>Quality is taken very seriously here 0.811<br>People believe the company’s success depends on high-quality work 0.649<br>This company does not have much of a reputation for top-quality products* 0.592<br>Quadrant Scales<br>Human Relations Autonomy<br>Integration<br>Involvement<br>Supervisory Support<br>Training<br>Continues<br>VALIDATING THE ORGANIZATIONAL CLIMATE MEASURE 407<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>Appendix. Continued<br>Quadrant Scales<br>Welfare<br>Internal Process Formalization<br>Tradition<br>Open Systems Innovation & Flexibility<br>Outward Focus<br>Reflexivity<br>Rational Goal Clarity of Organizational Goals<br>Efficiency<br>Effort<br>Performance Feedback<br>Pressure to Produce<br>Quality<br># The Organizational Climate Measure (OCM) is copyright Aston Organisation Development, 2003.<br>Copyright # 2005 John Wiley & Sons, Ltd. J. Organiz. Behav. 26, 379–408 (2005)<br>408 M. G. PATTERSON ET AL.<br>
| + | <br> |
