PROJECT DESCRIPTION

Statement of the Problem

The question of why and how organizations innovate has captured the interest of many disciplines and resulted in much research. In an effort to answer this question, researchers have explored how various organizational, individual, and environmental factors foster or inhibit innovation. Unfortunately, although the literature on innovation in the private and public fields is vast, the current innovation models offer little direction to those who want to influence organizational innovation (Meyer & Goes, 1988). In fact, the most consistent theme running through the innovation literature is that the empirical findings are inconsistent (Wolfe, 1994). This warrants attention and detailed re-examination of the factors included in the conceptual models tested.

In seeking to identify the alternative sources of the inconsistency in innovation research results, Fiol (1996) notes that the innovation researchers have focused primarily “on the means to effectively squeeze innovative activity out of organizations, with little regard for the continuous accumulation of knowledge that provide the source of that capability” (p. 1013). Variables most often employed in innovation studies include: formalization, professionalization, specialization, organizational size and resources, slack (the resources an organization has beyond what are minimally required to maintain operations), complexity, centralization, managerial attitude toward change, technical knowledge resources, administrative intensity, external communication, internal communication, and vertical differentiation (Damanpour; 1987; Damanpour, 1996; Kimberly & Evanisco, 1981; Meyer & Goes, 1988; Schin & McClomb, 1998).

It is obvious that the existing innovation models focus predominantly on structural explanations of innovation, failing to acknowledge the role of the human factor in the process of innovation (Claver & Llopis, 1998; Daft, 1982). Based on his review of innovation models and empirical evidence, Daft (1982) suggests that organizational employees are central in the consideration of the accurate modeling of organizational innovation. Their perceptions of organizational values and expectations are especially important in understanding the process of innovation. This process is unstructured and highly uncertain; behaviors necessary to produce innovation are unknown; and the probability of efforts being successful is unclear (Hauser, 1998a; O’Reilly, 1989; Zaltman, Duncan, & Holbek, 1973). Because of this uncertainty, the innovation process cannot be traditionally planned, organized, and guided by formal rules and procedures; it is rather fostered through the creation of an innovative culture (Russel, 1990). In situation of change and flux organizational culture serves as a supplement to structure and as a complement to leadership, and therefore is essential for assessing the organization’s potential to innovate (Hauser, 1998). A number of authors also point to organizational culture as an effective way of motivating and directing the solution of unstructured problems, and argue that culture, not just structure, is central to organizing for innovation (Amabile, 1997; Dougherty, 1996; Morgan, 1996; Pervaiz, 1998; Roberts, Watson, & Oliver 1989; Schein, 1994). Furthermore, organizational culture is seen as fostering individual creativity (Amabile, 1997; Higgins, 1995). Creativity, or the production of novel ideas, is the first step toward innovation, which is the successful implementation of those ideas. When creativity takes place in the right organizational culture, it results in innovation.
Given the complexity of the innovation phenomenon and the inconsistency of innovation research results, it is increasingly evident that the cultural perspective might be useful for understanding innovation (Roberts et al., 1989). Unfortunately, while recognized as important, the influence of organizational culture on organizational innovativeness remains at the level of theory. The existing research is plagued with conceptual and methodological issues that preclude researchers from drawing conclusions regarding the role of culture in fostering or inhibiting innovation. This research will apply innovative theories of cultural models along with methodological tools from the field of cognitive anthropology (Romney, Weller, Batchelder, 1986) to study the effect of organizational culture on organizational innovativeness, and will address some of the issues plaguing the culture and innovation research.

Theoretical Background

A careful examination of the literature on organizational culture and innovation reveals that authors have focused on cultural consensus (the degree to which employees share organizational values) and sought to identify organizational values, norms, beliefs, and assumptions characterizing innovative organizations. Furthermore, some theoretical propositions exist with regard to the structure of organizational culture and its effect on innovation.

Cultural Consensus and Innovation

Numerous theoretical propositions exist regarding the relationship between cultural consensus and innovation. For a long time the tendency was to treat culture as monolithic, unitary, homogeneous, and widely shared (Martin & Meyerson, 1988). It was seen as alleviating anxiety, clarifying the ambiguous, bringing predictability to the uncertain, and controlling behaviors, which disrupt harmony and predictability (Martin, 1992). This concept of a strong and unifying culture became popularized in the 1980s with the publication of Peters and Waterman (1982) book “In Search of Excellence”. Homogeneity of organizational values was seen as essential for becoming an excellent company. The authors emphasized the coherence of culture and the need to integrate it throughout all parts and levels of the organization. Proponents of this uniformist approach argued that strong cultures exert a greater degree of control over employees’ behaviors and beliefs, and have greater potential for implicit coordination (Denison, 1990; Pervaiz, 1998). Furthermore, they argued that by virtue of deeply held assumptions and beliefs, organizations are able to facilitate behaviors in accordance with organizational practices and principles, to promote a more rapid decision process, and to promote employees’ loyalty to the organization (Denison, 1990; Wilkins & Ouchi, 1983).

This is changing as researchers increasingly recognize that organizational culture is full of negations, instabilities, and struggles (Martin, 1992). The proponents of this view argue that intra-cultural diversity must be recognized, and they see organizational cultures as systems of meanings that are characterized by multiple meanings and partial consistency and consensus, which in turn are negotiated among multiple, complementary, and competing perspectives (Batteau, 2001; Geertz, 1983; Martin & Meyerson, 1988; Trice, 1993). They see organizations not as homogeneous worlds but rather as pluralistic worlds with multiple realities. Glynn, Barr, and Dacin (2000) note “that by shifting our view from a focus on sources of competitive and cognitive similarity to a focus including sources of cognitive plurality, we can enrich our understanding of long-term competitive advantage and better serve organizational adaptability”       (p.729).

Authors who recognize the heterogeneous nature of organizational culture argue that strong cultures are problematic for innovations. They maintain that organizations with strong cultures might have difficulties in implementing new ways of functioning, in responding to changes in the external environment, and in generating new solutions to the problems that arise (Nemeth, 1997; Trice, 1993).  Many organizations, including successful and enduring ones, use strong organizational cultures for social control. While creating uniformity, loyalty, and commitment to the organization, the cult-like strong cultures inhibit innovation and an organization’s ability to respond to change (Nemeth, 1997). The result of such an organization-wide sharing of normative prescriptions can be rigidity, lack of tolerance, and bitterness toward new ideas. When employees develop an emotional attachment to a set of cultural values, they might reject and dislike any changes that introduce different values (Trice 1993). Furthermore, loyalty, cohesion, and uniformity can result in a lack of reflection, which will inhibit the organization’s capacity to anticipate and successfully respond to new circumstances.
Collins and Porras (1994) note that innovation requires an organizational culture that is entirely opposite to a strong predictable culture, which emphasizes adherence to organizational expectations. One must be able to feel free to deviate from organizational expectations, to “think outside the box”, and challenge the existing ways of doing things in the organization. Organizational plurality adds richness and variation, and it creates complex and varied understandings, which increase breadth, flexibility, and innovation (Glynn et al., 2000). These authors recognize that an organization may not necessarily have one culture, but rather multiple subcultures. Thus, in assessing the relationship between innovation and organizational culture, it is critical to consider not only the strength of the organizational culture, but also its structure.

The Structure of Culture and Innovation

For the purpose of this proposal, the structure of organizational culture is defined as the existence of differentiated subcultures and the nature of their relationship to each other and to an overall unifying culture. Scholars are increasingly recognizing that organizations have multiple subcultures, and they argue that subcultures should receive more attention in future research (Foster-Fishman & Keys, 1997; Glynn et al., 2000). Since many organizations retain a structure that is based on informal social relationships, commonality of role, position, task, or physical proximity, such informal and formal groupings might develop their own sub-cultures. Additionally, demographic subcultures can form based on such group identities as gender, race, ethnicity, and age.

It has been suggested that the existence of subcultures with diverse ideologies can have an effect on organizational innovation (Hauser, 1998a; Hauser, 1998b; Trice & Beyer, 1993). Hauser (1998b) argues that conflicting subcultures can have severe negative impacts on innovation. Deep and longstanding divisions can lead to limited learning and distrust between subcultures.  Neutral subcultures do not have systematic effects on innovation. Complementary subcultures, on the other hand, have different but not conflicting values and thus provide a basis for numerous diverse ideas. The conflict between such subcultures can have positive outcomes because the consensus that is reached is based on values that are not in conflict.
Another important aspect to consider in assessing the relationship between the structure of culture and innovation is the relation between the subcultures and the overall culture. The subcultures might have their own set of cultural values, which do not necessarily conform to the prescribed organizational values (Sackmann, 1991). These subcultures can have different relations to overall culture (Gregory, 1983; Smircich, 1983). They have been described as enhancing, countercultural, and orthogonal. In the enhancing subculture, top managements’ view is reproduced in an exaggerated form. Countercultures, on the other hand, can be described as resisting top management’s views. Finally, orthogonal cultures are neither positively nor negatively oriented toward top management. These subcultures are usually occupational groupings.

Hauser (1998a) further suggests that the relationship between organizational culture and subcultures can influence innovation. He discusses four types of organizational cultures. A pluralistic culture is characterized by consensus within existing subcultures and absence of an overall unifying culture. The relationships between these subcultures will have varying and diametrically opposed effects on innovation. An integrated pluralistic culture, characterized by consensus within existing subcultures and an overall unifying culture, will result in an effective innovation processes. Absence of consensus within existing subcultures and absence an overall unifying culture are characteristic of a chaos culture, which has no consistent effect on innovation. Lastly, consensus within existing subcultures combined with an overall unifying culture lead to a monoculture, which does not foster creativity and diversity. According to Hauser (1998b), an overall unifying culture with strong innovation-oriented values combined with complementary subcultures is the most effective cultural pattern for organizational innovativeness. It guarantees collaboration and mutual understanding due to the overall organizational values and at the same time has the potential for diverse knowledge and functional conflicts spurred by subcultures.

While it is important to consider cultural consensus and structure of organizational culture, it is equally important to consider what the content of that consensus is as well as what values characterize the subcultures. Overall, what makes a subculture different is the distinctive profile of values it exhibits.

Organizational Values and Innovation

While not systematic, the most extensive research evidence exists regarding the organizational values, values, norms, and beliefs characterizing an innovative organization. Among those are: working in collaboration with others, team-orientation, creativity and innovation, risk taking, experimentation, flexibility, respect for individuals, tolerance, being supportive, taking initiative, sharing information freely, external-orientation, future-orientation, tolerance of mistakes, and people-orientation (Amabile, 1997; Dellana & Hauser, 2000; Hurley & Hult, 1998; Kitchell, 1995; Russell, 1990). Validity, interpretability, and generalizability of those results, though, are constrained by the empirical weaknesses of the studies. The major problem is the lack of systematic and valid measurement of these values at the cultural level. Most researchers used newly developed measures, rather than reliable and valid instruments. Furthermore, the authors have generated many culture dimensions, which make comparisons of the results difficult. In order to advance the theory of culture and innovation, researchers should seek to develop and rely on a more robust set of culture attributes, which would allow meaningful comparisons across organizations, as well as valid investigation of cultural differences and their relative impact on other organizational factors (Cook & Rousseau, 1988; O’Reilly, Chatman, & Caldwell, 1991).

Alternate Explanatory Variables

To build on existing theory and empirical evidence, this study will include a range of variables that have been suggested to be related to innovation. A number of theorists and researchers argue that leadership is the most important factor affecting innovation (King, 1990; Osborne, 1998; Schin & McClomb, 1998). Leaders can create and manage an organizational culture promoting innovation, can be product champions or hero innovators who support innovation throughout the process of its implementation, and can create the organizational structure needed to support innovativeness (Peters & Waterman, 1982; Van de Ven, 1986). Furthermore, leaders can enhance organizational capacity to innovate by directing resources and energy toward implementing new programs and by lending power and legitimation to innovative activities (Hasenfeld, 1983).

Many authors identify transformational leadership as an ideal leadership style for promoting innovation (Bass, 1985; Howell & Higgins, 1990). Transformational leaders use charisma, individualized consideration, inspiration, and intellectual stimulation to encourage creativity and enhance employees’ capacity to innovate. Despite numerous theoretical propositions, the empirical investigations of the relationship between leadership and innovation have seldom been conducted (Schin & McClomb, 1998; Waldman & Bass, 1991). King and Anderson (1995) noted that academics and managers accept theoretical prescriptions regarding the leadership styles needed to foster innovation despite very limited empirical evidence.

Among other variables shown to be related to organizational innovativeness are size (Kaluzny, Veney, & Gentry, 1974; Kimberly and Evanisco, 1981; Mohr, 1969), centralization, complexity, formalization, inter-organizational relations, and funding intensity (Damanpour; 1987; Damanpour, 1996; Kimberly & Evanisco, 1981; Meyer & Goes, 1988; Shin & McClomb, 1998). The study will control for these variables, while examining the relationship between organizational culture and innovation.

Prior Research and Long-term Goals of the Principal Investigator’s Project

The proposed research builds on the principal investigator’s prior research on innovations in nonprofit human service organizations. In her previous study titled “Organizational Culture and Innovation in Nonprofit Human Service Organizations” the author tested innovation model where cultural consensus and organizational values were major independent variables and leadership and size served as control variables (Jaskyte, 2000).

Twenty local chapters of the Association of Retarded Citizens (ARC) in a Southern state participated in that study. Regression diagnostic tests, employed to assess small sample data, detected one influential case, which was deleted from the data set, leaving the author with a sample of 19 organizations. Since conceptual and operational definitions of the major variables of the study pertained to an organization, the organization was used as the unit of analysis. Individual employee responses (all employees of participating organizations were asked to fill out a survey) were aggregated to obtain estimates of organizational culture. These aggregated estimates, in turn, were used in data analysis. A key feature of this study was using cultural consensus analysis (Romney, et al. 1986) to examine agreement within organizations on organizational values. Cultural consensus analysis is the only model currently available that provides an explicit method for moving from the individual level of analysis to the aggregate, such as an organization, in the analysis of organizational culture (Jaskyte & Dressler, in press).

It was hypothesized that cultural consensus (or the strength of agreement within the organization) and organizational values (the content of the consensus), would be related to organizational innovativeness. Furthermore, it was hypothesized that the interaction of cultural consensus and organizational values would be significantly related to organizational innovativeness after controlling for the effects of organizational size and leadership.

The results of the study showed that organizational innovativeness was inversely related to cultural consensus (r = -.570, p<.05) and positively related to two value dimensions: innovation (r = .437, p<.05), and aggressiveness    (r = .430, p<.10). Organizational innovation was significantly and inversely correlated with the stability value dimension (r = -.554, p<.05). Further correlational analyses provided a better understanding of the combined effect of cultural consensus and organizational values on organizational innovativeness. Cultural consensus was positively correlated with two value dimensions: team orientation (r = .577, p<.01), and stability (r = .531, p<.05), and inversely correlated with organizational innovativeness (r = -.570, p<.05). Thus, high cultural consensus in this sample of organizations was characterized by an emphasis on stability and teamwork that inhibits innovation.

Unlike the cultural uniformist perspective, the study showed that a strongly shared culture might not be appropriate for fostering innovation, especially considering its content. The higher the cultural consensus on such values as stability, security, low level of conflict, predictability, rule-orientation, team-orientation, working in collaboration with others, the less innovative the organization will be. Innovative organizations had weak cultural consensus and possessed such values as being innovative, willing to experiment, quick to take advantage of opportunities, and risk taking. The results of a multiple regression analysis, in which organizational innovativeness was regressed on cultural consensus (due to the multicollinearity between cultural consensus and organizational values, the interaction terms of the two variables were not included in the equation), leadership and size showed that the only significant predictor of organizational innovativeness was cultural consensus, accounting for 36.5% of the variance.

The results of this study call for further exploration of organizational culture. The wide variation in the degree to which values were shared within organizations indicated that organizational culture cannot be dichotomized as only strong or weak, but should rather be treated as a more variable phenomenon. Further, future research should seek a better understanding of the content of consensus that is conducive to innovation. It is not known what the relationship between organizational culture and innovativeness would have been had strong cultural consensus been characterized by organizational values supporting innovation. While the results of this study showed that organizational culture characterized by weak cultural consensus and innovativeness and aggressiveness value dimensions was conducive to innovation, future research needs to investigate whether different combinations of cultural consensus and organizational values can produce innovation.

The proposed project will contribute to further examination of the model. Having a larger sample of organizations will allow testing of a more complex model and will permit examination of structure of culture along with cultural consensus and organizational values. The long-term goals of the project involve further building and refinement of the innovation model locally as well as internationally in large samples of different types of nonprofit organizations, as well as in business and public organizations.

Research Aims, Questions, and Hypotheses

The aim of this proposal is twofold: 1) examine the relationships between organizational culture and innovation while controlling for alternate explanatory variables; and 2) improve conceptualization and operationalization of organizational culture and innovation constructs.

(1)  Drawing from the literature on innovation and culture, a number of research questions and hypotheses were formulated:

• A general research question is: How does organizational culture affect organizational innovativeness?
More specific research questions pertaining to three culture variables are:
1. Is cultural consensus (the degree to which employees agree on a set of organizational values) associated with organizational innovativeness?
2. Are organizational values associated with organizational innovativeness?
3. Is the interaction of cultural consensus and organizational values associated with organizational    innovativeness?
4. Is the structure of organizational culture associated with organizational innovativeness?
5. Will organizational culture be related to organizational innovativeness after controlling for the effects of alternate explanatory variables?
6. Will the effect of organizational culture differ depending on the type of innovation, its nature, and the stage of its implementation?
• It is predicted that cultural consensus, organizational values, interaction of cultural consensus and organizational values, and structure of culture will be associated with organizational innovativeness after controlling for the effects of alternate explanatory variables.

(2) Conceptual/Methodological Development:

• An in-depth understanding of innovations, their type, stage of implementation, and nature will be obtained, as it has been suggested that different types of innovations might be differentially explained, and that the effect of various factors might differ depending on the stage of innovation process.
• Building on the intra-cultural diversity perspective, and recognizing that organizational culture is not monolithic and unitary but rather heterogeneous, this study will assess three domains of organizational culture: cultural consensus (the degree to which organizational values are shared), the structure of organizational culture, and organizational values (the content of consensus).
• In addition to the issue of cultural heterogeneity, this study will also address the issue of the unit of analysis. As distinct from previous studies where a leader or a sample of upper management employees was used as a proxy for the organization to report on the culture of the organization, this study will include all employees and will use the organization as the unit of analysis. Individual responses will be aggregated to obtain estimates of organizational culture. These aggregated estimates, in turn, will be used in data analysis.
• The study will employ the cultural consensus model to obtain the culture estimates (Romney, Weller, & Batchelder, 1986). The cultural consensus model makes it possible to evaluate the degree of diversity and sharing among individuals within the organization and hence to characterize different organizations as sharing more-or-less strongly an organizational culture.  Similarly, using cultural consensus analysis, we can estimate the structure of culture, and the content of values that are being shared.
• The study will control for many variables that were shown to be related to innovation in previous studies.

Research sites

This research will be conducted in partnership with a nonprofit organization Communities in Schools (CIS). Community in Schools (CIS) is the nation’s leading community based organization that focuses on bringing community resources into schools to help young people successfully learn, stay in school, and prepare for life. For the second year in a row, Communities in Schools has been selected by Worth magazine as one of the top 100 nonprofits in the U. S., based on innovation, strategic insight, and effectiveness. Communities in Schools (CIS) works closely with local business leaders and school superintendents and takes a community development approach to supporting education by unifying the existing resources of communities around children, families, and schools as a support system to help young people realize their full potential and take responsibility for their future. Since 1997, CIS has been lowering dropout rates and the incidence of school disruption, and has increased graduation rates.

Methods

Sample. 48 local sites of CIS Georgia will participate in the first phase of the study (first year of the project). The project will be extended to include the CIS in the States of South Carolina, North Carolina, and Florida (a total of 74 research sites) in the second phase of the study (second and third year of the project).

Procedures. To collect data on organizational innovativeness, the research team, comprised of faculty and graduate students from the School of Social Work and from the Institute for Nonprofit Organizations, and information and evaluation manager of CIS will carry out personal interviews with the executive directors. Prior to the interview, each director will receive an interview guide. This guide includes 14 questions regarding different types of innovations, which were developed on the basis of the most widely used typologies of innovation (Perri, 1993, Damanpour, 1987). Directors will be asked to report whether or not one or more innovations had been attempted in a particular area and describe them. They will also be asked to identify the stage of innovation adoption, the degree to which innovation is radical or incremental, and whether it is original - developed within an organization, or adapted – borrowed from other organizations. Having the guide before the interview will give the executive directors ample time to think about their responses. Additionally, the executive directors will be asked about the number of departments and programs associated with an organization, number of job classifications of personnel, number of organizations their organizations work with, and funding intensity. Organizational documents, memos, informational booklets, annual reports, and financial documents will also be collected and analyzed to get a better sense of organizational functioning and culture. Additionally, demographic information for the counties where the CIS sites are located will be used to place each of the organizations in the larger context and to account for the differences in population and environmental characteristics.

While on the site, the team members will distribute surveys to all employees, which will include the measures of organizational culture, leadership, formalization, centralization, and demographic questions, and will wait until the surveys are filled out. Additional surveys and pre-paid envelopes will be left for those employees and volunteers who will not be on site at the time of the researchers’ visit.

Questionnaire data will provide an in-depth understanding of intra-organizational diversity, as it will be analyzed using cultural consensus model - an innovative analytic tool from the field of cognitive anthropology. This model will enable the researchers to determine cultural consensus (the extent to which employees share organizational values), the content of that cultural consensus, and the structure of organizational culture. A more detailed description of the model should help the readers understand how when analyzed using adequate analytic techniques questionnaire data can provide in-depth understanding of organizational culture.

Cultural Consensus Model

 Consensus analysis is a technique developed in the field of cognitive anthropology that allows for the assessment of the amount of agreement among individuals about some domain of cultural knowledge. Further it estimates the content of that agreement (Romney et al., 1986).  It takes into account the sources of cultural variability – individuals not being equally knowledgeable about the organizational culture, and differences in the models of various organizational groupings. Rather than treating culture as stable set of values held by all employees, this technique makes possible the evaluation of whether there is a culturally central or more diversified understanding of cultural values within each organization (Caulkins & Hyatt, 1999).

 Cultural consensus model is based on three major assumptions (Romney et al., 1986). First, the degree of sharing among individuals is a result of their joint agreement with cultural consensus. Second, the probability of an individual answering a question correctly (i.e. in accordance with the prevailing model) is a result of that individual’s competence with respect to that domain of knowledge. Lastly, the model assumes that the cultural reality is the same for all individuals and therefore there is a culturally correct answer for every question.

 In cultural consensus analysis the data consists of the responses of each individual on each question. A computer program ANTHROPAC (Borgatti, 1992) analyzes the matrix of informant intercorrelations and produces information on: (1) cultural consensus, or the degree of agreement among respondents about a set of organizational values, (2) how knowledgeable each respondent is about the cultural values, (3) the “cultural knowledge key,” or culturally correct information about the content of consensus according to the aggregated individual answers.

 There are several ways of estimating the degree of sharing using cultural consensus analysis.  The technique extracts the eigenvalues or latent roots of the correlation matrix of respondents.  If there is a single eigenvalue, or if the first eigenvalue is large relative to others, this is indicative of substantial sharing of responses (Romney, et al. (1986) suggest a ratio of 3:1 as a useful rule of thumb for determining if the first eigenvalue is large relative to others).
 
In addition to providing an estimate of the overall consensus level, consensus analysis also calculates “cultural competence” coefficients for each informant. These coefficients are the factor loadings of each informant on the first factor, and indicate how strongly that individual’s knowledge is correlated with the composite group knowledge. The average of the cultural competence coefficients can thus be used as another measure of consensus—the larger the average competence, the greater the sharing within the group. Finally, cultural consensus analysis produces the knowledge key, or the consensus answers, for each question. The consensus answers are weighted by the cultural competence individual responses, where more weight is put on the more knowledgeable individuals than less knowledgeable ones.

Measures

Organizational Innovativeness. For the purpose of this study, organizational innovativeness is defined as the number of innovations an organization adopted within the last two years (Gopalakrishnan & Damanpour, 2000). Innovation is defined as the implementation of an idea, service, process, procedure, system, structure, or product that is new to the prevailing organizational practice. Organizational innovativeness is operationalized with fourteen items developed on the basis of Perri (1993) and Damanpour’s (1987) typologies of innovation. Numerous studies relied on these typologies in operationalizing organizational innovativeness (Damanpour, 1990; Damanpour, 1996; Gopalakrishnan & Damanpour, 2000; Russell, 1990; Roberts et al., 1989; Schin & McClomb, 1998). Of the fourteen items the following five indicate administrative innovation: the creation of a new performance evaluation system, the introduction of a new training topic for employees or volunteers, the creation of a new employee/volunteer incentive/reward system, the creation of a new recruitment system, the creation of a new performance evaluation system, and the creation of a new organizational structure or shape. The following items indicate technological product innovation: the introduction of new services/programs, significant change in already existing services/programs, the extension of the services to new groups of clients previously not served by the organization, the production of a new product, the introduction of a new activity/event for clients/employees/volunteers, and the redesign of a product already being produced into something new and significantly different.  Finally, technological process innovation is indicated with the following items: the creation of a new way of service delivery and the significant conversion of an existing way of service delivery. The last item will ask the respondent to identify any other innovations that the organization implemented which were not mentioned during the interview.
The executives directors will be asked to identify whether innovation is original (developed within an organization) or adapted (borrowed from other organizations), and to indicate its stage of implementation and the degree to which it is radical or incremental.

Organizational Culture. Organizational culture in this study is defined as a set of shared values that help organizational members understand organizational functioning and thus guide their thinking and behavior (Desphande & Webster, 1989). It will be measured using the Organizational Culture Profile (OCP) developed by O’Reilly, Chatman, and Caldwell (1991). The instrument contains a set of 54 value statements, 21 of which factored substantially alike in numerous studies forming seven value dimensions:  attention to detail, innovation, outcome orientation, aggressiveness, team orientation, stability, and people orientation (Chatman & Jehn, 1994; O’Reilly et. al., 1991; Sheridan, 1992). Those twenty-one value statements will be used for the purpose of this study. The OCP shows reasonable reliability and convergent validity. The instrument has demonstrated moderate test-retest reliability (median r = .74, range = .65 -.87). The convergent validity of the instrument was established through the significant positive correlation (r = .28, p<.05) between person-organization fit assessed with the OCP and normative commitment, defined as attachment to an organization based on value congruence (O’Reilly et al., 1991). Social desirability bias was addressed by casting the items in neutral terms.

A balanced incomplete block design method will be used to obtain the rankings of each value (Romney & Weller, 1988). Twenty one organizational values will be arranged in the subsets of five and employees will be asked to order each set of values from most characteristic of their organization to the least characteristic. A complete rank-order will be obtained by summing ranks assigned to each value (see Romney & Weller, 1988 for a more detailed discussion of this method).

The Alternate Explanatory Variables

Transformational Leadership. Transformational leadership is defined as a set of practices employed for developing relationships between leaders and employees. It will be measured by the Leadership Practices Inventory (LPI) (Kouzes & Posner, 1993). All employees will be asked to rate a set of 30 behaviorally based statements regarding five leadership practices: Challenging the process, Inspiring the shared vision, Enabling others to act, Modeling the way, and Encouraging the heart. This instrument was selected because it measures leadership behaviors that are consistent with the transformational leadership style, which has been identified as ideal for promoting innovation. Furthermore, it has been shown to have adequate psychometric properties, with internal reliabilities ranging from .81 to .90 and test-retest reliabilities averaging nearly .94 (Posner & Kouzes, 1992).

Size will be operationalized as a total count of full-time, part-time employees, and active volunteers. To be consistent with previous research practice, the log of the number of employees and volunteers will be used as a measure of size (Kimberly & Evanisco, 1981).

Complexity will be operationalized as a number of departments and programs associated with an organization.

Specialization will be measured by the number of job classifications of personnel.

Interorganizational links will be operationalized as the number of organizations with which the organization partners/collaborates/cooperates.

Centralization is defined as the degree to which employees participated in decision making concerning important organizational policies and procedures. The four-item measure developed by Hage and Dewar (1973) will be used to measure centralization. The participants will be provided with a five-step scale to indicate how frequently they participate in decision making regarding the adoption of new programs, adoption of new policies, promotion of professional staff, and hiring of new staff.

Formalization is defined as the extent to which emphasis is placed on organizational rules and procedures (Hage & Dewar, 1973). The four-item measure, also adopted from Hage and Dewar (1973), will be used to measure formalization.

The remaining survey questions will focus on employees’ and executive directors’ gender, race, age, length of employment in an organization, professional background, education, and job title.

Data Analysis

First, data from the face-to-face interviews with the executive directors will be analyzed to obtain the counts of different types of innovations, along with their stage of implementation, and nature (original vs. adopted; radical vs. incremental).

The cultural consensus analysis routine in ANTHROPAC (Borgatti, 1992) will be used to analyze the data on organizational culture. Twenty one value statements will serve as an input to cultural consensus analysis. ANTHROPAC will use ratings of these value statements as units of analysis and the employees as variables in a factor analysis.

• Cultural Consensus. This factor analysis of employees-by-value rankings will indicate the degree of agreement among employees about the cultural values (cultural consensus). If one factor is dominant with the eigen-value at least three times larger than the second factor, it will mean that there is strong cultural consensus. If the analysis reveals several factors in the data, with one factor having an eigen-value at least three times larger than the other, we will still have strong consensus. The smaller the eigen-value is for the first factor, the weaker the consensus.

• Organizational Values (Content of Consensus). In addition to providing the estimate of cultural consensus, ANTHROPAC produces the knowledge key, or the consensus answers, for each of the 21 value statements. These weighted consensus ratings will be averaged to obtain the scores for each value dimensions: innovation, outcome orientation, attention to detail, aggressiveness, team orientation, stability, and people orientation.
• The Structure of Culture. Cultural consensus model allows for identification of the subcultures, as well as for assessing the relationships between them. First, ANTHROPAC creates a person-to-person agreement matrix that can be displayed using multidimensional scaling procedure (MDS). The MDS plots will show each employee in a two dimensional space, where the employees agreeing on a set of values will cluster close to each other in the space. Demographic variables – race, age and gender, as well as organizational variables – job title, and professional background - will be used to see if subcultures exist based on these variables. Once the clusters are identified, cultural consensus analysis will be rerun for each of the clusters to see whether consensus exists among the employees comprising that cluster. If consensus is reached, the subculture will be said to exist. To assess the relationships among the subcultures the cultural answer keys of subcultures will be correlated. If the correlation is negative, the subcultures will be said to have conflicting values. Finally, to assess the relationship between subcultures and the overall culture the answer key of the overall culture will be correlated with the answer key of each of the subcultures.

Since conceptual and operational definitions of the major variables of the study pertain to an organization, the organization will be used as a unit of analysis. Individual employee responses will be aggregated to obtain estimates of organizational culture. These aggregated estimates, in turn, will be used in data analysis. Upon completion of the separate analyses for each organization, a new data set will be created, where the rows will represent each of the organization, and the columns will consist of the measures of cultural consensus, seven dimensions of organizational values, structure of culture, organizational innovativeness, and other explanatory variables. After creating a data file the researchers will proceed to testing the hypotheses.

Hierarchical multiple regression model will be used for testing hypotheses, where independent variables will be entered in a series of equations. The first equation will regress organizational innovativeness on cultural variables. Alternate explanatory variables will then be entered sequentially. The order in which they will be entered will depend on the magnitude of their association with culture variables.

Dissemination of Results and Future Partnership Goals

Discussions of the results will be disseminated to wider audiences through presentations at professional conferences, such as the Association for Research on Nonprofit Organizations and Voluntary Action (ARNOVA) conference, National CIS Conference, and through publications in academic journals, such as Nonprofit and Voluntary Sector Quarterly, Administration in Social Work, Journal of Social Services Research, Nonprofit Management and Leadership, Journal of Community Practice, and others.

The information/findings of NSF Research Project will be discussed at semi-annual meetings with CIS Executive Directors (Spring & Fall), one-on-one meetings between CIS Executive Director and CISGA Community Development Specialist, Annual Conference in fall of 2004, board meetings, and CIS State Directors meetings. The results will also be disseminated through press releases, CIS national multi-track training workshops, vendor booths at conferences (displays, printed materials, etc.), and information updates for legislators distributed during General Assembly 2004 statewide newsletter.

Report of the results of the study will be published and distributed to all participating sites. It will also be available to all interested parties through the CIS Georgia State Office in Atlanta, through the University of Georgia, as well as through a project web-site. To assure higher visibility of the project, the links to this web-site will appear on the Communities in Schools and the UGA homepages. Researchers, nonprofit managers, school representatives will be able to learn more about each participating organization and the partnership between the University of Georgia and CIS. The web-site will not only provide useful information about the partnership and the research project, but will also serve as a learning medium. To facilitate an on-going dialog among the participating organizations, other CIS sites operating across 26 states, as well as their partners – schools and other community organizations, a “web-community” will be created on the project’s web-site. Participants will be able to post their questions, information about the new activities within their organizations, as well as in other organizations providing similar services. They will be encouraged to submit descriptions of the new projects that worked well in their organizations, as well as stories of “how to” and “how not to”, which will reflect their successes as well as failures.

Based on the results of the study, a workshop will be offered to the organizations participating in the first phase of the study. The workshop will consist of two sessions. The first session will involve detailed discussion of the project results. Participants will be encouraged to provide feedback and their interpretation of results. In addition to the discussion of results, this session will provide participants with a number of tools and suggestions for increasing the organization’s capacity to innovate. Participants will leave the session having an understanding of a process for developing innovation and with knowledge of management techniques that can be used to bring innovativeness in an organization. The second session will be designed to facilitate networking among the executive directors. It will provide them with an opportunity to learn about the variety of innovations implemented across different CIS sites and to share their ideas and expertise.

The information about the project and results should help practitioners in their efforts to improve organizational innovativeness. Having knowledge of how organizational culture affects innovativeness of an organization will help not only organizational managers, but also school administrators identify the ways in which culture needs to change in order to foster innovation, and to design managerial practices for influencing and modifying organizational culture.

Both parties – UGA and CIS hope and believe that this project would be only the beginning of a long lasting partnership. Among some of the long-term goals that it might work toward are 1) to design, implement, and assess the series of workshops on improving organizational capacity to innovate; 2) to seek funding to extend the project to include schools and other community organizations, 3) to serve as a role model of effective partnership between research university and the user community by encouraging linkages between other CIS sites and local universities, and by demonstrating the usefulness of research to practitioners, and 4) initiate recognition practice, through CISGA, for local sites demonstrating innovation as recognized in the research findings.

Significance of the Proposed Research

The results of this study will be significant from theoretical, methodological, and practical standpoints.
Building on the existing innovation literature, this study adds an important variable to the innovation models – organizational culture - and explores how cultural consensus, organizational values, and the structure of culture will affect organizational innovativeness while controlling for alternate variables.

The study addresses some of the shortcomings of current innovation and organizational culture studies.
While previous research treated organizational culture as homogeneous, this study builds on the intra-cultural heterogeneity perspective of organizational culture, which views individuals as culture-bearing actors and sources of variation. In departure from previous research, which relied solely on the organizational leaders’ and upper managements’ perceptions of culture, this study measures cultural perceptions of employees from all organizational levels. Since conceptual and operational definitions of the major variables of the study pertain to an organization, the organization will be used as a unit of analysis. Individual employee responses will be aggregated to obtain estimates of organizational culture. These aggregated estimates, in turn, will be used in data analysis.

The study will demonstrate the appropriateness of the cultural consensus model for assessing organizational culture. This is especially important considering the fact that the field of organizational culture has been hindered by the lack of adequate measurement models. Having an appropriate tool will allow researchers to be consistent in their conceptualization and operationalization of organizational. The cultural consensus model makes it possible to evaluate the degree of diversity and sharing among individuals within the organization, the content of consensus, as well as structure of the culture.

The study will further the in-depth understanding of the nature of innovative activities and will account for the type of innovation and stage of its implementation, since it has been suggested that different types of innovations might be differentially explained, and that the effect of various factors might differ depending on the stage of innovation process. Finally, in assessing the relationship between organizational culture and innovation, the research will control for other factors shown to be related to innovation in previous studies.

While the findings of this study will have implications for all types of organizations – nonprofit, business, and public - they will be especially relevant for nonprofit organizations partnering with school systems (CIS network works with more than 2000 schools across the country). Having knowledge of how organizational culture affects innovativeness of an organization will help not only organizational managers, but also school administrators to identify the ways in which culture needs to change in order to foster innovation, and to design managerial practices for influencing and modifying organizational culture. It will lay the groundwork for further research on innovation in other types of organizations.