512

The Determinants of Organizational 
Innovation: An Interpretation and 
Implications for Research Libraries

Ronald C. Jantz

Ronald C. Jantz is Digital Library Architect at Rutgers University Libraries; e-mail: rjantz@rutgers.edu. 
© 2015 Ronald C. Jantz, Attribution-NonCommercial (http://creativecommons.org/licenses/by-nc/3.0/) 
CC BY-NC

The research reported here is focused on a specific type of change 
in an organization: an innovation. In an empirical analysis of research 
libraries, it was found that five factors had a significant impact on the in-
novation performance of the library. These factors relate to the strategy, 
organizational structure, and leadership of the research library. The study 
sample consisted of 50 libraries that were members of the Association of 
Research Libraries. This paper will discuss the theoretical model, explain 
the effects of these five variables, highlight certain additional correlations 
that are meaningful, and discuss implications for research libraries.

here is a common perception that an innovation is the result of personal 
creativity, inspiration, and the fortuitous convergence of a variety of exter-
nal factors, none of which can be predicted or managed. Although some 
innovations emerge by accident or serendipity, according to Peter Drucker, 

successful innovations result from a “conscious, purposeful search for innovation op-
portunities which are found in only a few situations.”1 There are certain attributes of 
organizational culture that can foster this more conscious, purposeful search. Some 
50 years ago, Jesse Shera, a preeminent library scholar, understood this cultural chal-
lenge and suggested a librarian mindset for creating a more innovative library: “In a 
world of mingled menace and promise, the winds of change blow as surely through 
the library stacks as they do through the corridors of the United Nations… The librar-
ian, therefore, must be both critic and architect—destroyer of that which is obsolete 
and builder of his own future.”2 

In their study of isomorphic processes, DiMaggio and Powell have suggested that the 
traditions and norms of the institution create an “iron cage” that restricts change.3 Given 
the library’s well-established norms, organizational members are typically conditioned to 
focus on and protect existing practice. To counter this internal resistance, Peter Drucker has 
argued, based on many years of study, that innovation can be systematically managed.4 
Shera’s admonition that librarians become both builders of the future and destroyers of 
that which is obsolete suggests that major changes will be a part of the research library 
future. To accomplish these changes, as Kanter has noted,5 members of the organization 
will need to challenge beliefs and received wisdom to realize major new innovations. 

doi:10.5860/crl.76.4.512 crl14-580



The Determinants of Organizational Innovation  513

The central question for leaders is whether the positive impact of the research library 
on scholarly communication and information access can be substantially improved 
without implementing major innovations and undertaking the requisite organizational 
changes. If major change is warranted, how then is this change brought about? These 
questions suggest that library leaders will need to create a culture in which innovation 
and creativity become accepted norms.

The purpose of this paper is to report and further explicate the findings of a recent 
in-depth study by the author on organizational innovation in research libraries and 
to explore important and more subtle bivariate relationships that emerge from the 
underlying dataset.6 The in-depth study revealed important determinants of innova-
tion in research libraries. These determinants are related to the strategy, structure, and 
leadership of the library and can provide suggestions for how to change the culture to 
become more innovative. Implications for the research library are presented, suggesting 
that library leaders will need to develop a creative and innovative culture to survive 
and prosper in the turbulent and rapidly changing environment of the 21st-century 
research university.

Organizational Innovation and the Diffusion Model
Angle and Van de Ven have succinctly described the generic innovation process as “a 
purposeful, concentrated effort to develop and implement a novel idea that is of substantial 
technical, organizational, and market uncertainty.”7 An innovative capability can be at-
tributed to an individual, a group, an organization, or a nation or state. In this discussion, 
organizational innovation of the research library is examined with specific attention to 
those attributes that can lead to a more innovative library. For the research library and 
similar nonprofit institutions, innovations are primarily of two types: administrative and 
technical. Given this basic typology, innovation can be defined as the introduction into the 
organization of a technical innovation (a new product, a new service, a new technology) 
or a new administrative practice; or a significant improvement to an existing product, 
service, technology, or administrative practice.8 This study focuses exclusively on technical 
and service innovations in the research library that are delivered to the end user.

Innovations are created in a series of steps or stages, a process that innovation 
scholars have defined as the diffusion model.9 An idea will pass through these major 
stages in the organization in a transformative process that ultimately yields a success-
ful innovation. Three major stages of innovation diffusion have been defined and are 
depicted in figure 1: the initiation of the innovation, a decision to adopt, and the imple-
mentation of the innovation. Although figure 1 illustrates an orderly and sequential 
process, there is much feedback, organizational churn, and significant elapsed time 
before an innovation is successfully implemented.

There are multiple substages for each of these major stages.10 When the leadership 
of an organization makes decisions about introducing a change, their choices occur 

FIGURE 1
Stages in Rogers’ Diffusion of Innovation Model11



514  College & Research Libraries May 2015

over time in the various stages of the innovation diffusion process. The research library 
faces challenges in each of the three stages of diffusion. 

In the initiation stage, performance gaps or new opportunities become evident 
through the effects of external enablers. Marcus and Weber cite considerable research 
suggesting that most innovation stimuli originate from external jolts or trigger events.12 
The emergence of new knowledge, environmental turbulence, a new technology, and 
competitive threats are examples of enablers that can create opportunities. If leaders 
have a positive attitude toward change, then initiation of an innovation becomes pos-
sible. The second substage of initiation is that of attitude formation. This is a critical 
stage involving the leader, the leadership team, and the strategy of the organization. 

If there is an alignment of the innovation with the strategy of the organization and basic 
agreement among the leadership of the library, then the diffusion process can proceed to 
the decision stage. In the adoption decision stage, there is much information gathering and 
communication within the leadership team to decide to proceed to the implementation 
stage. If consensus is reached and decision processes are well defined and the decision 
clearly communicated, then the diffusion process proceeds to the implementation stage.

The implementation stage typically has three substages. In the initial implementation, 
the innovation is put on a trial basis and evaluated to determine if it is practical for a 
long-term commitment. The second stage involves a formal commitment in which the 
organization establishes appropriate processes and policies, possibly making structural 
changes to support the innovation. In the third stage, full implementation results when 
a majority of the potential clients have successfully used the innovation. In assessing 
the extent of implementation, many factors should be considered including the time 
for diffusion, marketing efforts, the fit of the innovation to the client group, and a 
possible flawed implementation. It is in the third stage where scholars have identified 
“implementation failure” as a major reason that an organization does not reap the 
full benefits of an innovation.13 An example in the research library world is the insti-
tutional repository (IR). Although most libraries in this study indicated that they had 
implemented an IR, many responded that full implementation had not been achieved.

Literature Review
The literature review discusses the theoretical framework and prior research regarding 
the major constructs used in this research.

Behavioral Integration. In most organizations, decisions do not emanate unilat-
erally from the singular leader at the apex of the organization. Rather, leaders and 
managers—the top management team (TMT)14—are constantly communicating and 
making tradeoffs between meeting current needs and developing capabilities for the 
future. Building on upper echelons theory, Hambrick proposed the concept of behav-
ioral integration in the top management team and linked the concept to organizational 
outcomes.15 The leadership of an organization is called upon to make decisions in a 
complex environment of information overload that is often ambiguous and contradic-
tory and frequently subject to multiple interpretations. According to Hambrick, an 
integrated team can navigate the complexities of decision making more effectively. 
However, for many organizations, the concept of a leadership “team” is a misnomer. 
The group that is nominally the TMT may have little interaction—rarely meeting as a 
team—and group members may have their own agendas. Even in this less integrated 
team, the leaders and managers of the organization, with their respective styles, mo-
tives, biases, and experience, can significantly affect organizational outcomes.

Behavioral integration is a meta-construct that involves group processes such as 
collaboration that were previously represented by separate constructs. Behavioral 
integration should not be confused with social integration or interdependence. The 



The Determinants of Organizational Innovation  515

meta-construct also does not imply consensus; in fact, behaviorally integrated teams 
may have considerable disagreement—a situation that can create more content-rich 
discussions. Hambrick theorized that behavioral integration consisted of three impor-
tant factors: 1) level of collaborative behavior; 2) quantity and quality of information 
exchange; and 3) emphasis on joint decision making.16 

Carmeli and Schaubroeck note a behaviorally integrated TMT is characterized by 
intense interaction that produces open information exchange and collaborative-based 
decisions.17 These decisions are typically more complex than those encountered in 
other work groups. These researchers examined the impact of behavioral integration 
on organizational decline using a sample of 116 TMTs from various industries. It was 
found that behavioral integration was negatively related to organizational decline (that 
is to say, more behavioral integration results in less decline) and was positively related 
to the perceived quality of strategic decisions. In addition, their in-depth case studies 
indicate that it is reasonable to impute causality to these hypothesized relationships. 
In a survey of teams from 96 service organizations, Carmeli found that TMT behav-
ioral integration is positively associated with both human resource performance and 
economic performance.18 A component of behavioral integration is the exchange of 
information in the leadership team. Damanpour and Aravind report that internal com-
munication and “a climate conducive to the dispersion of ideas across the organization” 
were positively related to innovation in two waves of meta-analyses.19 Leadership is 
also seen as a major factor in contributing to organizational creativity—an important 
antecedent of innovation. Uhl-Bien coined the phrase “enabling leadership” where the 
formal leaders of the organization are particularly well-suited to stimulate creativity 
because of their networks, access to resources, and authority.20 A more behaviorally 
integrated leadership team may be able to carry out this enabling leadership, resulting in 
an organization generating more ideas that lead to improved innovation performance. 

Organizational Structure. Structural contingency theory suggests that the structure of 
the organization should change to adapt to and align with various factors in the external 
environment.21 According to Burns and Stalker,22 the organic structure, as opposed to the 
mechanistic or bureaucratic structure, is appropriate for adapting to changing conditions 
in the environment. Blau proposed a formal theory of structural differentiation in orga-
nizations, and Jansen et al. developed a scale for structural differentiation that character-
izes the extent to which tasks in an organization are distributed across different units.23 

Ettlie, Bridges, and O’Keefe have developed a model that focuses on organizational 
strategy and a structure that leads to both radical and incremental innovations.24 In 
their study of the food processing industry, findings suggest that radical innovation is 
supported by an aggressive technology policy, a centralized decision process, informal 
structures, and a concentration of technical specialists, whereas incremental innova-
tions emerge from large, complex, decentralized organizations. In contrast to many of 
the innovation studies, Ettlie found that large size appears to negatively affect radical 
innovations in that more formal structures are required to manage large organizations. 
Ettlie and colleagues suggest that the more traditional institutions might offset the size 
influence by partitioning structurally for radical and incremental innovation—an early 
hint at an ambidextrous organization. Highlighting bureaucratic effects, Jansen et al. 
found that a more hierarchical structure negatively affects exploratory innovation, 
whereas a more rule-based, formal structure will positively affect incremental inno-
vation.25 The multidimensional complexity of organizational innovation can be seen 
in these studies where there are different views of the types of organizations that are 
required in the initiation and implementation stages. Organizational size, a concept 
that has caused much variation in research results, appears to further complicate the 
understanding of what structures are best to achieve a more flexible organization.26 If 



516  College & Research Libraries May 2015

major change and radical innovations are desired, these studies seem to suggest that a 
more complex, informal, decentralized organization is required in the initiation stage 
and a centralized process for decision making is required in the adoption stage. To 
manage the complexities of a radical implementation, a further shift to a more formal, 
less complex organization would be required in the implementation stage. 

Ambidexterity. The ambidextrous orientation of the organization is characterized by 
the ability to simultaneously conduct exploratory activities while also supporting and 
enhancing current services. The benefits of the library professional framework and 
long periods of success can result in organizational complacency and a limited ability 
to acquire and apply new knowledge to exploratory work. This ability to acquire and 
use new knowledge is an integral part of the learning organization. March identified 
two characteristics of the learning organization, noting “the essence of exploitation 
is the refinement of existing competences, technologies, and paradigms” while “the 
essence of exploration is experimentation with new alternatives.”27 In their literature 
review, Klein and Knight identified a strong learning orientation as a critical factor 
in the innovation implementation stage.28 He and Wong demonstrated empirically 
that exploratory and exploitative strategies have a strong positive impact on product 
innovation.29 Tushman and O’Reilly have extended these organizational learning con-
cepts to characterize the ambidextrous organization as one that can simultaneously 
engage in both exploration and exploitation activities and also cope with the resulting 
organizational contradictions. These contradictions become important for long-term 
survival as suggested by these researchers: “Ambidextrous organizations build in 
contradictions as they operate for today and tomorrow.”30 

The External Environment. Although there is considerable literature on innovation 
and public institutions, Noordegraaf and Stewart suggest that more attention should be 
given to the environment in which the institution is embedded.31 Outside the borders of 
the university, there are external pressures that are creating environmental uncertainty 
that affect the university and the university library. These pressures originate, in part, 
from the political and economic environment and the emergence of for-profit firms that 
offer competing services. For much of the 20th century, research libraries resided in a 
stable, predictable environment. However, the external environment of the 21st century 
appears to be much more unstable, a situation that often favors more radical innovations. 

Dess and Beard’s model defines three components of the external environment: 
munificence, complexity, and dynamism.32 Boyne and Meier clarify the concept of 
dynamism by explaining that the crucial element of dynamism is not the frequency 
or magnitude of change that can usually be anticipated, but the unpredictability or 
uncertainty of the environment.33 In reporting on public service firms, these researchers 
indicated that environmental turbulence can have a negative effect on performance and 
that managers can mitigate these effects from the environment by maintaining structural 
stability.34 Koberg et al. found that environmental dynamism positively affects both 
incremental and radical innovations, although the effect is more positive for radical 
innovations.35 In a study of the financial services sector, Jansen et al. conclude that ex-
ploratory innovation is more effective in dynamic environments.36 Andrews used both 
objective and subjective measures of environmental dynamism and reported that both 
measures were negatively related to achievements in public organizations.37 In studying 
logistics innovation adoption, Germain found that organizational size and environmen-
tal uncertainty predict expensive, radical innovations but not incremental innovation.38 
Damanpour reported that the effect of environmental uncertainty is significant for the 
size-innovation relationship and helps explain variance beyond control variables.39 In 
a related study,40 he also indicated that the relationship between bureaucratic control 
and innovation is less negative under conditions of high environmental uncertainty. 



The Determinants of Organizational Innovation  517

This prior research indicates that the external environment affects innovation and 
performance outcomes in organizations in many ways, interacting with size, structure, 
and bureaucratic processes. Although there are varying results, it appears that organi-
zations in turbulent environments are likely to implement more radical innovations.

Methods
The Population and Sample. In this research, the unit of analysis is the research library 
as represented by the views and perspectives of the leadership or top management 
team of the institution. The population for the current study was the research libraries 
in the United States that are members of the Association of Research Libraries (ARL). A 
sample of 50 libraries was taken from this population. To create the sample, university 
librarians or directors from each ARL institution were contacted to obtain approval 
for their leadership team to participate in this research. 

To obtain data for each institution, the top management team in the library responded 
to survey questions regarding various aspects of the organizational structure, the 
external environment, and demographics of the leadership team. The average size of 
the library team was 3.6 members, and a total of 183 library leaders responded to the 
survey that included 102 questions.41 All teams included the university librarian. For 
the regression analysis, team responses were aggregated to create an organizational 
response for each library. A summary of TMT demographics is provided in table 1. 
The size and geographic distribution of each ARL institution in the sample are shown 
in table 2 and figure 2 respectively. 

TABLE 1
Demographics of the Research Library Top Management Teams (N=50)

Variable Mean Standard Deviation
TMT Age (Years) 55.5 4.13
TMT Size (Number of Members) 3.6 0.80
TMT Tenure in the Profession (Years) 25.1 6.15

TABLE 2
Size and Institutional Characteristics of Research Libraries (N=50)

Size (Total FTE) Percentage
100–199 30.0
200–299 32.0
300–399 16.0
400–499 10.0
>= 500 12.0
Total 100.0
Institutional Characteristics Percentage
Type of Institution (Public/Private) 84.0/16.0
Region—Urban/City 84.0
Region—Suburb 10.0
Region—Town 6.0
Total 100.0



518  College & Research Libraries May 2015

For the sample of 50 libraries, potential nonresponse bias was assessed by deter-
mining if there was any significant difference in either the size of the library or the 
geographical region between participating and nonparticipating libraries in the popula-
tion. For size, an independent samples t test was performed to assess whether the mean 
size of libraries participating in the study differed significantly from the libraries that 
did not participate. The Wilcoxon rank-sum test was used to test that the distribution 
of geographical regions among the library participants was not significantly different 
from that of the nonparticipating libraries. These tests demonstrated that there was no 
significant difference of size or geographical distribution between participating and 
nonparticipating libraries. 

Innovation Characteristics. There are two important innovation characteristics 
that were used in this study. First, a fundamental innovation typology distinguishes 
between administrative and technical innovations. Administrative innovations focus 
on the internal mechanisms of the organization and the social system (such as recruit-
ment of personnel, allocation of resources, and the reward system) whereas a technical 
innovation focuses externally on the client and end user to provide new products and 
services. Primarily because of methodological reasons, this research focuses exclusively 
on technical innovations since they are much easier to identify and describe. Admin-
istrative innovations remain an important area for further research.

In addition to the technical-administrative distinction, this study deals with the 
extent of change, which varies along a continuum from incremental to radical. Accord-
ing to Zaltman et al,42 a radical innovation is always “to some extent disruptive of the 
status quo” and involves changes in the organization’s subsystems, values, incentives 
and power. Radical innovations involve new knowledge that is used to create totally 
new products and services or to make fundamental changes in an existing product 
or service. In contrast, an incremental innovation uses existing knowledge to create 
minor improvements in a product or service. Research has been quite consistent in 

FIGURE 2
Geographical Distribution across ARL Regions of Research Libraries (N=50)



The Determinants of Organizational Innovation  519

demonstrating that radical innovations are riskier than incremental innovations, and 
they have more profound organizational effects, albeit with the promise of greater 
returns. Although the incremental and radical concepts represent a continuum, we 
view these concepts as endpoints or polar opposites on a continuum.43 

Assessing Innovations. A special panel of librarians and library science scholars was 
convened to assess the radicalness of specific innovations in research libraries.44 The 
online panel survey listed 40 innovations. For each innovation, the panel member was 
asked to assess radicalness based on five criteria. Candidate innovations were technical, 
providing new services to library patrons as opposed to administrative innovations. All 
of the innovations have been implemented in at least one research library in the period 
2007–2011. Using the definitions of radical and incremental, panel members identified 
the extent of agreement or disagreement for each criterion by selecting a number on the 
nine-point Likert scale. After analysis by the author to remove similar innovations, a 
final list of 32 innovations was developed. Library leaders answered two questions for 
each innovation: 1) had a decision to adopt the innovation been made and 2) what was 
the extent of implementation of the innovation in their institution.45 The complete list 
of radical, incremental, and mid-range innovations is shown in table 3, appendix A.46

Major Constructs. The major constructs for this study are defined as those indepen-
dent variables that were expected to have the most impact on innovation performance: 
behavioral integration, structural differentiation, the ambidextrous orientation, and the 
external environment. Innovation researchers, as noted below, have developed scales 
and assessed construct validity for each of these concepts. For each major construct, a 
team score was computed as an average of the scores for each individual respondent. 
Questions for each construct can be viewed at a PDF version of the survey.47 

Simsek and colleagues created a scale and validated the construct for behavioral 
integration, a measure that consists of three concepts: collaborative behavior, informa-
tion exchange, and joint decision making.48 For this independent variable, a single, 
overall measure of behavioral integration was constructed from the resulting nine-item 
measure (survey questions 1–9). Using Cronbach’s alpha,49 the behavioral integration 
scale appeared to have good internal consistency, α = 0.93. Lubatkin et al. extended 
He and Wong’s constructs for exploratory and exploitative orientations, resulting in a 
six-item scale for each concept.50 This scale has been adapted for research libraries as 
an ambidextrous orientation, a perspective of the library leaders that their organization 
is able to conduct both exploratory activities while also supporting current services. 
A nine-point Likert scale was used to capture responses for each question (survey 
questions 21–32). The scales for exploration and exploitation appeared to have good 
internal consistency, α = 0.90 and α = 0.87, respectively. From a scale developed by 
Jansen et al.,51 five questions characterize the structural differentiation construct. The 
resulting scale captures the extent that leaders have partitioned their organization into 
separate units (survey questions 10–14). The scale for structural differentiation had 
acceptable internal consistency, α = 0.66. In accordance with Milliken and Waldman et 
al., environmental uncertainty was captured as a perception of the external environment.52 
A six-item scale, adapted from Khandwalla,53 was used to measure environmental 
uncertainty (survey questions 15–20). The scale for environmental uncertainty had 
acceptable internal consistency, α = 0.71.

The Dependent Variable: Innovation Performance. A unique dependent variable, 
innovation performance, was created to account for three different dimensions that 
contribute to the innovativeness of the organization. These three dimensions capture 
more of the innovation context and address methodological issues with previous 
studies that have focused exclusively on the adoption decision. Innovation magnitude 
was captured by determining the number of innovations for which there had been a 



520  College & Research Libraries May 2015

decision to adopt (figure 1, stage 2). The extent of implementation (figure 1, stage 3) 
of each of the innovations was determined from the leadership team’s view of how 
much of the implementation had been completed. Finally, the balance between the two 
main types of innovation, radical and incremental, was assessed. Regarding balance, 
many researchers have emphasized the need to balance between these two types of 
innovation activity.54 Too much focus on incremental innovation sacrifices the future, 
whereas too much radical innovation can adversely affect current services. Those librar-
ies with larger magnitude, extensive implementation, and a balance between radical 
and incremental received higher innovation performance scores. 

Research Model and Analysis 
Research Model. Innovation scholars have demonstrated that the external environment, 
organizational factors, and leadership characteristics represent the primary antecedents 
of organizational innovation.55 From the innovation literature and the author’s research, 
a model was constructed in which the effects of both the external environment and 
organizational factors on the innovativeness of the research library were taken into 
account. Innovativeness in research libraries was hypothesized to be the consequence 
of several major factors including the leadership team, the organizational structure, 
and the learning strategy of the organization. In addition to the major factors, several 
additional predictor variables were expected to have causal effects within the unique 
environment of the research library.56 From the extant literature, earlier research by 
the author,57 and observations within research libraries, the following research ques-
tions were formulated:

1. What effect does the behavioral integration of the senior team have on the in-
novation performance of the research library?

2. How does a more structurally differentiated organization affect the innovation 
performance of the research library?

3. What is the effect of an ambidextrous orientation on the innovation performance 
of the research library?

4. Does an ambidextrous orientation mediate the relationship between senior 
team integration and innovation performance?

5. How do the factors in the external environment affect the innovation perfor-
mance of the research library?

6. How do contextual factors (demographics and enablers) of the research library 
affect innovation performance?

The research model with research questions is illustrated in figure 3. In the explor-
atory analysis, 17 possible predictor variables of innovation performance were identi-
fied (appendix B, table 4). The exploratory analysis was conducted using a hierarchical 
regression approach. In each step, a predictor variable was entered in an order based 
on theory and evidence from reported research. Predictors adding little or no incre-
ment to variation (R2) were eliminated from the final analysis.58 The direct effects of 
the major constructs are illustrated in figure 3 along with the mediating effect of the 
ambidextrous orientation (RQ4). Beyond these major constructs, the demographics (age 
and level of education) of the senior team and certain enabler variables were expected 
to have an impact on innovation in the research library.59 

Analysis. The hierarchical regression analysis in the current study yielded five 
variables that were significant predictors of innovation in research libraries and that ac-
counted for 49 percent of the total variance in research library innovation performance. 
Four of these five variables relate to organizational mechanisms and the leadership 
team and represent a unifying context for creating a more innovative culture. The fifth 
variable, organizational size (based on total FTE), did have a significant and positive 



The Determinants of Organizational Innovation  521

impact on the innovative performance of the library. After controlling for size and 
type (public or private) of the organization, the factors of primary interest remained 
significant in impacting the innovativeness of the library. The size variable has been 
the focus of many innovation studies; however, size may be a proxy for other effects 
such as slack resources.60 It is generally understood that small organizations are more 
flexible, whereas larger organizations typically have additional resources to dedicate 
to the more innovative projects. In related studies of nonprofit organizations, the size 
of the organization was typically found to have a positive impact on innovation.61 

The overall regression, including the two controls and the four predictor variables, 
was statistically significant: R = .70, R2 = .49, adjusted R2 = .41, F (6,40) = 6.36, p < .001 
(the t-values for the best predictors are shown in appendix C, table 5). The R2 value 
indicates how much variance in the dependent variable can be explained by variation 
in the independent variables. The final step in the regression (model 5) indicates that 
49 percent of the variation in research library innovation performance can be explained 
by the controls and the four predictor variables. 

Interpretation of Results and Implications for Research Libraries
Behavioral Integration. With respect to RQ1, it was found that behavioral integration 
had a positive and significant impact on innovation performance. Why is the integration 
of the senior team an important concept in the management of the research library? 
First, there is a growing body of literature that suggests that the more integrated 
TMT, rather than the singular leader, can have a major impact on organizational per-
formance.62 Tushman et al. report that the strategy emanating from senior leadership 
and the resulting organizational design are powerful levers that enable an organiza-
tion to pursue dramatically different types of activities.63 To successfully innovate, 
TMT collaboration and active engagement is needed to create supportive structures, 
strategies, and an innovative organizational culture.64 A behaviorally integrated team 

FIGURE 3
Research Model with Research Questions



522  College & Research Libraries May 2015

has the confidence and trust to have open debates on unpopular issues and risky 
proposals, resulting in a greater likelihood of achieving agreement and concerted 
action on controversial projects. Through increased quality and quantity of informa-
tion, more ideas are available for discussion and joint agreement. The collaborative 
aspect of the behavioral construct suggests that team members will be more willing to 
discuss unique, nontraditional ideas. The flow of new ideas, the willingness to discuss 
controversial projects, and the collaborative aspects of the team will lead to more radi-
cal innovation proposals flowing into the decision process. By contrast, members of 
the less behaviorally integrated team are likely to withdraw into their own domains 
where they have undisputed control. This behavior is typical of the classical manager 
in which rules and processes become dominant and that lead to support of the status 
quo and, at best, more incremental innovations.

Decision Awareness. The enabler variable, decision awareness, emerged from the 
dataset of library leaders’ responses to survey questions. This variable was constructed 
by determining when there was a consensus in the leadership team regarding the 
decision to adopt an innovation. From these responses, it became evident that there 
was considerable diversity within a leadership team regarding the knowledge or 
awareness that a decision had been made to proceed with a specific innovation. This 
uncertainty regarding a decision raised obvious questions about the resultant impact 
on innovation performance in the library. How were decisions being communicated 
and documented? If decisions to commit to an innovation were unclear, how would 
this lack of clarity affect the innovativeness of the library? To capture the decision 
awareness factor, a score was computed for each team based on the response to the 
following question for each innovation: “Has the library made a decision to adopt 
the innovation?”65 Decision awareness had a significant and positive impact on the 
innovation performance of the library.

When senior leaders and managers communicate accurately with each other, the 
innovativeness of the organization is likely to improve. This effect is highly intuitive 
and perhaps most apparent in the implementation stages of an innovation. During the 
implementation stage, it may be necessary to move resources between units to support 
the development of the innovation. If a decision hasn’t been properly communicated 
or if there is confusion about the status of a decision, it becomes difficult for the leader-
ship team to agree to reallocate resources to what might be considered a controversial 
project. In addition to the significant regression results, decision awareness is positively 
correlated with behavioral integration with (r = 0.38, p < .01) (see appendix B, table 4). 
The correlation of decision awareness with behavioral integration suggests that the 
organization has a communication process leading to more clarity and less confusion 
regarding major decisions. The implications for the organization are that decisions 
must be clearly communicated, not only within the leadership team, but also to the 
members of the organization.

Structural Differentiation. Organizational structure is important for innovation 
insofar as it is designed to facilitate the generation of new ideas and the successful 
implementation of these ideas. Duncan has identified the persistent dilemma for in-
novative organizations—different organizational structures appear to be appropriate 
for the initiation and implementation stages of the innovation process.66 

For the structural differentiation variable, library leaders responded to statements such 
as 1) in our library, exploratory and production activities are structurally separated; 
and 2) our library has units that are specialized in specific functions. With respect to 
RQ2, a more structurally differentiated library had a significant and negative impact 
on innovation performance, suggesting that a more structurally differentiated library 
will be less innovative. There are several explanations for this finding. First, an orga-



The Determinants of Organizational Innovation  523

nizational structure is inherently a stabilizing force that perpetuates the status quo. 
More structure in an organization can create barriers to communication and the free 
flow of ideas. This effect is evident in the correlation of the structure variable, structural 
differentiation, and another independent variable, decision awareness, which captures the 
senior team’s view of whether a decision to adopt has been made. These two variables 
are negatively correlated (r = –0.33, p < .05) (see appendix B, table 4), suggesting that 
a more structurally differentiated library results in less awareness of decisions and 
restricts the communication of these decisions. Second, one might speculate that library 
leaders have not dedicated sufficient attention to integrating more diverse units into 
the larger library system. In particular, new units for doing exploratory work will 
require extra effort from leaders to integrate these units into the mainstream of the 
library. This integration is essential for ideas, especially the more controversial ideas, 
to be communicated, shared effectively, and transformed into innovations.

The negative impact of structural differentiation appears to conflict with much of the 
theory; however, there are several possible explanations for this effect. In fact, there is 
still quite a bit of variation in the empirical studies regarding organizational structure. 
In contrast to the negative effects of structure, there is a steady stream of research 
demonstrating that different structures can significantly improve the organization’s 
ability to innovate.67 Earlier studies and empirical analyses have indicated that differ-
ent structures are needed in the various stages of innovation. Informal, decentralized 
organizations work best in the early diffusion stages to create more ideas and initiate 
potential innovations. As the organization moves through the decision process and into 
the implementation stage, well-defined processes and an efficiency-oriented culture 
are needed to meet schedules and to develop a high-quality product or service. These 
shifting structural requirements highlight the importance of flexibility in the organi-
zational structure and the inherent challenge to library leadership to create optimal 
organizational structures. Conducting effective exploratory work presents one of the 
most significant challenges for flexible structures.

Exploratory activities are candidates to be located in separate units. The benefits of 
exploratory work—the generation and use of new knowledge—cannot be overstated. 
In a classic paper, Cohen and Levinthal observe that an increased R&D investment 
creates a capacity to assimilate and exploit new knowledge that, in turn, enhances 
the innovativeness of the organization.68 As Tushman and O’Reilly have pointed out, 
incremental innovation can be managed within a traditional functional organization.69 
For the library to produce more radical innovations, increased exploratory activities and 
cross-functional teams provide opportunities for increasing innovation performance. 
Tushman and O’Reilly describe the entrepreneurial unit that is needed to generate 
discontinuous innovations: “these units are relatively small, have loose, decentralized 
product structures, experimental cultures, loose work processes, strong entrepreneurial 
and technical competencies, and relatively young and heterogeneous employees.”70 

In a recent ARL SPEC kit, German and Namachchivaya investigated how research 
libraries are supporting R&D activities. For their sample, 69 percent of the libraries 
were investing in R&D activities; however, many of these libraries indicated that R&D 
is diffused throughout the organization.71 This diffuse approach makes it very difficult 
to manage and track R&D activities and is likely to produce incremental innovations 
that reflect the immediate needs of the specific unit. Research into R&D processes has 
found that the more autonomous exploratory teams are more effective.72 A relatively 
small exploratory unit can evaluate external knowledge, transfer technology into the 
library, prototype, and demonstrate possible new products and services, thus provid-
ing the leadership team with alternatives for making innovation adoption decisions. 
Members who engage in these exploratory activities will use different vocabulary and 



524  College & Research Libraries May 2015

different processes and perhaps be viewed as inefficient or unproductive—a signifi-
cant challenge in managing R&D. Leaders will need to consider how to structure the 
organization of the library to support these entrepreneurial members and the related 
R&D activities. 

Ambidextrous Orientation. With respect to RQ3, the ambidextrous orientation was 
shown to have a positive and significant impact on innovation performance (model 
5, appendix C, table 5). The ambidexterity of the organization is characterized by the 
ability to simultaneously conduct exploratory activities while also supporting and 
enhancing current services. There are two very important results regarding the am-
bidexterity of the library. First, in reference to RQ3, the ambidextrous orientation has 
a significant and positive impact on innovation performance. Second, as illustrated 
in figure 3, the ambidextrous orientation mediates the relationship between behavioral 
integration and innovation performance. In the study, it was shown that two-thirds of 
the impact of behavioral integration and the ambidextrous orientation is achieved 
through this mediation effect. Thus, library leadership teams that are highly integrated 
and support both exploratory efforts and existing services are more likely to have 
innovative organizations. In the ambidextrous organization, the challenge of manag-
ing contradictions will become more evident as technology provides opportunities 
for many new services. For example, while the library continues to conduct in-class 
bibliographic instruction, an exploratory team might be working to eliminate this 
type of instruction by providing sophisticated online tutorials that are delivered to 
the student at the time of need.

The effect of an ambidextrous orientation is most relevant for the mission, vision, and 
strategy of the organization. Elenkov et al. report that leaders who want to change the 
status quo “should formulate and articulate a shared and inspiring vision of the fu-
ture.”73 Similarly, Sarros et al. found that the behavior attribute “articulates vision” was 
the factor most strongly related to an organizational climate that fosters an innovative 
culture.74 An ambidextrous library operates in multiple modes simultaneously—focus-
ing on existing services while also experimenting, prototyping, and incurring risk in 
undertaking these activities. These quite different activities can be restricted by the 
natural barriers in the organization—structural, social, and psychological. Thus, an 
ambidextrous orientation imposes a requirement on leaders to articulate and support 
the resulting inconsistent organizational architectures and processes.75 Organizational 
contradictions will emanate from many dimensions: focus and flexibility, autonomy 
and democracy, individual and the group. Managing in this more innovative environ-
ment is based on processes in which leaders recognize and embrace contradiction.76 
These contradictions have practical consequences in the way the library supports 
traditional services such as bibliographic instruction, reference desk queries, and print 
book collections.77 

Unexpected Results
This section highlights the results regarding leadership demographics that were un-
expected and appear to be in conflict with much of reported theory. There is extensive 
research that examines the demographics of both the singular leader and the leadership 
team; the research reported here focuses exclusively on the leadership team.

Demographics of the Senior Team. Leadership demographics can have a major 
impact on the creation of an innovative library. Overreliance on leaders who have 
spent their professional lives in the library will likely result in a dependence on 
traditional management approaches and a focus on current products. Within the 
framework of upper echelons theory, Hambrick and Mason have proposed that the 
demographic characteristics of a leadership team such as age and tenure can be used 



The Determinants of Organizational Innovation  525

to predict organizational outcomes such as performance and innovativeness.78 In this 
research, the hypotheses related to the demographics of the library leadership team 
were not supported. This section provides further insight and interpretation of these 
unexpected results.

In addition to Hambrick’s original conceptualization,79 there is considerable em-
pirical research on the impact of TMT characteristics such as education and age on 
innovativeness, R&D intensity, and strategic choices.80 For example, Bantel and Jackson 
have shown that innovative banks are managed by more educated teams who have 
diverse functional areas of expertise.81 Wiersema and Bantel report that firms most 
likely to undergo changes in corporate strategy had TMTs characterized by lower 
average age and shorter organizational tenure.82 Earlier theoretical propositions by 
this author suggested that leadership age and tenure would negatively impact the 
more radical innovations in research libraries.83 In the current research, there were 
no significant correlations between the leadership team’s average age or tenure in the 
profession and innovation performance (AGE and TEN variables, appendix B, table 
4). Much of the innovation literature suggests that older leaders are risk-averse and 
tend to preserve the status quo, characteristics that can restrict major change in the 
organization. However, the negative aspects of age appear not to have had an impact 
on innovation performance in the sample of 50 ARL libraries. An interesting correlation 
analysis provides additional insight for this finding. By dichotomizing the average age 
of the leadership team and innovation performance, the effect of age on innovation is 
partitioned into four quadrants as illustrated in the following table:

TABLE 6
The Number of Innovative Libraries Associated with Young and Mature 

TMTs (N=50)
Innovative Libraries Non-Innovative Libraries Total

Young TMTs 15 11 26
Mature TMTs 12 12 24
Total 27 23 50

There is almost an equal distribution across the four cells of table 6, suggesting that 
the mature teams are as innovative as the young teams. This phenomenon is somewhat 
counterintuitive and deserves further explanation in light of the previous discussion of 
ambidexterity and team integration. A team will need to work together for several years 
before it can become more integrated. Further, several more years of collaboration would 
be required for the team to support the more controversial and conflict-laden ambidextrous 
orientation. These effects tend to positively affect the mature team’s impact on innovation 
and reduce the impact of the young team. There are, however, some remaining caveats. 
First, it is likely that an inverse U-shaped relationship actually exists between age of 
the leadership team and innovation performance, suggesting that the middle years are 
perhaps the most innovative. Second, from a methodological perspective, there was little 
variation in the age of the library leaders (Mean = 55.52, SD = 4.13), making it more dif-
ficult to discern any significant impact of age on the innovativeness of the library. Finally, 
in proposing a situational approach to the study of leadership, researchers point out that 
demographic characteristics are imperfect proxies for psychological constructs and do not 
take into account the dynamics of the decision process in complex social organizations.84 
Many leaders are quite flexible and able to adapt their management styles to the situa-
tion, suggesting that a fixed construct does not adequately model leadership behavior.



526  College & Research Libraries May 2015

Leadership Education. Bantel and Jackson have observed that TMTs with higher 
levels of knowledge and significant cognitive diversity will perform better.85 In a 
study of banks, these researchers found that the more innovative banks are managed 
by more educated teams with diverse areas of functional expertise. One of the inter-
esting and unexpected results from the current study was the finding regarding the 
senior leadership’s level and diversity of education. Correlations of educational level 
(r = –.10) and educational diversity (r = .08) with innovation performance were small 
and not significant. These variables model the impact of knowledge accumulation and 
the low correlations raise interesting questions. One possible explanation for the low 
correlations relate to the type of innovation. The research reported here focused on 
technical innovations in contrast to administrative innovations. Leaders and managers, 
especially in large organizations, are primarily involved with the administrative aspects 
of the organization and have relatively little time to dedicate to technical aspects, thus 
reducing their influence on technical innovations. In a further exploratory analysis, 
it was expected that those TMTs in which members had science backgrounds might 
have a positive impact on innovation performance. In their study of Fortune 500 com-
panies, Wiersema and Bantel found that TMTs with science backgrounds were more 
positively oriented toward change in their organizations.86 Given that innovations 
frequently require organizational change, leaders with science backgrounds might be 
more favorably disposed toward major technology-based innovations in the library. 
In the current study, a dummy variable was created in which 1 indicated the presence 
of science degrees within the TMT and 0 indicated no science degrees. The correlation 
of this variable with innovation performance was positive (r = .11) but not significant 
(EDS variable, appendix B, table 4).

Environmental Uncertainty. In the research model of figure 3 (RQ5), the uncer-
tainty of the external environment was the only major factor that was not significantly 
related to innovation performance. The mean for team responses on environmental 
uncertainty was 4.47 on a nine-point Likert scale (SD = 0.72, N=50), suggesting that 
most TMTs did not view their environment as all that uncertain. From these scores, 
one can speculate that library leaders felt reasonably confident that they can predict 
the effects of budget allocations, government control, and political attitudes on their 
library (survey questions 15–20). 

There are some rather subtle explanations for this effect. The general definition of 
environmental uncertainty is “the individual’s perceived inability to predict something 
accurately.”87 However, as Milliken notes,88 even though the environment might be 
changing dramatically, leaders may be confident that they can predict the volatility or 
variability in the environment. It is the turbulent and unexpected changes that can lead 
to a perception of environmental uncertainty. However, related to innovations, there 
is one remaining important observation. For whatever reason, if leaders do not view 
their environment as uncertain, there will be less motivation to make major changes 
and the institution is more likely to continue on a track of incremental innovation.

Limitations
Because of the small population (99 research libraries) and the sample of 50, structural 
equation modeling (SEM) could not be used in this study. However, SEM provides 
a powerful tool to examine multiple latent factors and to more effectively model the 
causes and effects of mediators. A future study could use SEM and the larger popula-
tion of academic libraries to undertake a more comprehensive path analysis. 

Respondent bias is always an issue in a survey-based study. In this study, it was 
very important to obtain responses from the leadership team as opposed to the singu-
lar leader. By considering the responses of all of the TMT members, respondent bias 



The Determinants of Organizational Innovation  527

has been minimized. This aspect was especially important when the library leaders 
were asked to assess the extent of implementation for each innovation. There is little 
convincing research that either contradicts or supports the generally accepted belief 
that top administrators can provide reliable information about their organizations.89 
Although self-assessment measures can be prone to bias, they are the most commonly 
used approach, given that other measures are difficult to obtain and can also be biased. 

Regarding the underpinnings for upper echelons theory, demographics are, at 
best, imperfect proxies for psychological constructs, the major problem being that 
demographic characteristics do not co-vary perfectly with psychological attributes. 
This stability and lack of precision in demographic characteristics can lead to conflict-
ing empirical results. This limitation may, in part, help explain the unexpected results 
regarding the age and educational level of the library leadership team. It should be 
noted that effective leaders are able to modulate their styles based on the situation at 
hand. TMT effectiveness can vary from situation to situation, whereas demographics 
remain relatively constant for extended periods of time. 

In a cross-sectional study, it is difficult to account for how innovative capability 
increases over time. For example, research libraries may be achieving ambidexterity 
by partially overlapping exploratory and traditional activities over time rather than 
conducting these activities simultaneously. As Piao points out,90 too much or too little 
overlap in these activities tends to reduce firm longevity, whereas moderate overlap 
will extend firm longevity. Longitudinal studies will provide a more complete picture 
of how research libraries create both incremental and radical innovations.

Further Research 
Assessing innovation performance in complex social organizations presents many 
subtle complexities. The 17 indicators used in this research covered diverse aspects 
of the research library, some revealing significant effects on innovation performance 
while others lacked the expected correlation with each other or with the dependent 
variable. Additional research is warranted, not only to better understand the unsup-
ported hypotheses but to also explore other dimensions of the research library.

The predictors of management or administrative innovations can be quite different 
from those affecting technical innovations and are more specific to the organization in 
which they are created.91 Many innovations are initiated because there is leadership 
dissatisfaction with some condition in the organization (such as loss of customers or 
clients) or leaders perceive an external threat such as loss of budget. How do adminis-
trative innovations originate in the research library and do they follow or lead a major 
new technical innovation? Is overall innovative performance related to having a balance 
of both technological and administrative innovations? Understanding these questions 
will help leadership create and sustain an innovative capability in their organizations.

It must be acknowledged that the research library is embedded in a larger institu-
tional framework that is also undergoing change. Bass summarizes the condition in 
the modern university as one embedded in a state bureaucracy, complicated by union 
contracts, faculty norms, and traditions.92 The parent institution can restrict or promote 
the innovativeness of the library. The policies of the university can either impede or 
facilitate the research library’s ability to innovate. What does it mean for a university 
to be innovative and how does the innovative university contribute to the success of 
library innovations and vice versa? 

Conclusion
Organizational innovation is a complex process that involves creativity, risk taking, 
trial and error, coping with failure, and dealing with an increasingly turbulent exter-



528  College & Research Libraries May 2015

nal environment. Most would agree that leaders and managers can have a significant 
impact on the culture and innovativeness of the organization. This study has found 
that there are important determinants related to the leadership team, decision making, 
and organizational structure that can affect innovation. These determinants suggest 
that successful innovation will require a mix of collaborative leaders who can put in 
place a strategy and organizational structure for conducting exploratory work while 
also supporting current services. Articulating vision, mission, and strategy to inspire 
and motivate organizational members requires a lot of time and energy from leaders. 
A starting point is a simple, emotionally engaging vision statement that characterizes 
the library’s commitment to innovate. Such a mission statement will help create the 
climate in which major innovations can be initiated and passed through the diffusion 
stages to successful implementation.

Innovation almost always involves an encounter with the unknown. Many scholars 
have reported that innovation can be successfully managed and sustained over time. 
Innovation is not dependent on individual creativity and can be systematized any-
where “because it has everything to do with organization and attitude and very little 
to do with nurturing solitary genius.”93 The most important lesson from this research 
is that library leaders can affect positive and major change in their institutions, but this 
change will require them to embrace the inherent conflict and paradox of an innovative 
strategy. To become more innovative, library leaders will need to acknowledge this 
paradox, articulate its relevance, and promulgate a vision throughout the organization 
that will enable the creation of new services for the 21st century information society.



The Determinants of Organizational Innovation  529

Appendix A. 
TABLE 3

The Final List of Innovations Used in the Survey (N = 32)
Innovation 
Number

Radical Incremental Midrange

1. Provision of a service to 
publish e-journals

The sharing of a 
technology platform 
(such as, an OPAC or 
institutional repository) 
with another library

A collaboration 
with another library 
to share collection 
development

2. The provision of a 
GIS (Geographical 
Information System) 
service to students and 
faculty, including access 
to GIS software, training, 
and consultation

A device and associated 
service to allow students 
and faculty to check out 
their own books

A service for the 
submission, access 
and preservation of 
ETDs (electronic 
theses and 
dissertations)

3. Provision of a service to 
faculty and students for 
multimedia production 
including instruction, 
software and equipment 
platforms to support 
multimedia creation and 
publication

The use of live chat and 
instant messaging for 
reference service

The offering of a 
Wi-Fi service to the 
local community (that 
is to say, not members 
of the university 
community)

4. Provision of a science 
data service including 
archiving, preservation, 
and access to research 
data and liaison support 
to researchers

Installation of a coffee 
bar/restaurant/café in the 
library

The provision of 
digital exhibits for 
special collections or 
other unique materials 
that are owned by the 
library

5. Replacement of stack 
book storage and 
preservation with digital 
book storage and digital 
preservation

Provision of a service 
to inexpensively print, 
bind and trim bookstore-
quality paperbacks from 
digital book files that are 
out of copyright

The embedding of 
library liaisons with 
students and the 
instructor in course 
management systems

6. Made the transition to 
a bookless (that is, no 
print books) library for 
certain disciplines (for 
example, engineering)

Provision of a mobile 
device lending service 
(laptop, netbook, iPad 
and so on)

Provision of federated 
searching across 
the library OPAC, 
the institutional 
repository, and other 
open repositories



530  College & Research Libraries May 2015

7. Implemented a liaison 
service to provide 
assistance to faculty 
researchers for managing 
their copyrights (for 
instance, to fulfill article 
deposit requirements 
from the National 
Institute of Health and 
other institutions)

Use of RSS feeds (or 
similar technology) to 
provide library news and 
event descriptions to 
library patrons

Use of digital object 
identifiers (such as, 
DOIs or Handles) 
to create long-
term, stable links 
to digital resources 
that are locally 
owned or created 
(digitized resources, 
dissertations, special 
collections, and so on)

8. The creation of an 
institutional repository 
to contain the research 
output of the university 
(such as, faculty 
research or dissertations)

Delivery of 
bibliographic instruction 
using online tutorials

Provision of mobile 
access to the library 
website and online 
catalog

9. Provided a dynamic 
mapping application for 
the OPAC to provide 
patrons directional 
information to find a 
shelved item

Reconfiguration of 
physical space and 
redesigned services to 
provide information 
or learning commons 
(a central location 
for workstations, 
information resources, 
and librarian assistance)

Creation of a website 
or portal for faculty 
that provides services 
and assistance for 
article deposit into the 
institution’s repository

10. A service to digitize and 
provide online access 
to historic, print course 
catalogs

The development of a 
flexible bibliographic 
instruction course 
structure that gives 
students the option 
of attending sessions 
in the classroom, 
participating online, 
or doing both

11. Outsourced reference 
service to another 
organization (for 
example, another 
library, nonprofit, or a 
commercial organization)

Implementation of 
faceted browsing for 
the library OPAC

12.
Provided for the 
digitization and access 
to historic university 
yearbooks



The Determinants of Organizational Innovation  531

Appendix B

a Significant (2-tailed) at .01 level, b Significant at .05 level 

Key to Variables: IP=Innovation performance, BI= Behavioral integration, 
DA=Decision awareness, LDR=Singular leader attitude, AO=Ambidextrous orientation, 
AOB=Ambidexterity (balance), SZ=Organizational size, STR=Structural differentiation, 
STF=Staff/professional profile, AFL=Affluence, ENM=Environmental munificence, 
ENU=Environmental uncertainty, AGE=TMT average age, TEN= TMT average professional 
tenure, TNP=TMT average tenure in position, EDL=TMT level of education, EDD=TMT 
educational diversity, EDS=TMT science background.

TABLE 4
Bivariate Correlations with Innovation Performance for all Predictor 

Variables (N=50)



532  College & Research Libraries May 2015

Appendix C 

TABLE 5
t-Values for the Five Best Individual Predictors (N=47)94

Independent 
Variable

Unstandardized
Regression 
Coefficient

Standardized 
Regression 
Coefficient

t-Value p-Value R2increment R
2

Organizational 
Size (model 1)

25.67 .42 t(44) = 
3.11

.003** .184 .184

Behavioral 
Integration 
(model 2)

4.59 .32 t(43) = 
2.26

.029** .087 .271

Structural 
Differentiation 
(model 3)

–4.88 –.29 t(42) = 
–2.29

.027** .081 .351

Decision 
Awareness 
(model 4)

0.37 .33 t(41) = 
2.20

.033** .069 .420

Ambidextrous 
Orientation 
(model 5)

2.62 .35 t(40) = 
2.30

.027** .068 .488

*p < .10, **p<.05, ***p< .01

Notes

 1. Peter F. Drucker, “The Discipline of Innovation,” in Managing Innovation, eds. Jane Henry 
and David Walker (London: Sage Publications, 1991), 9.

 2. Jesse H. Shera, Documentation and the Organization of Knowledge (Hamden, Conn.: Archon 
Books, 1966), 95.

 3. Paul J. DiMaggio and Walter W. Powell, “The Iron Cage Revisited: Institutional Isomor-
phism and Collective Rationality in Organizational Fields,” American Sociological Review 48 (1983): 
150–54.

 4. Peter Drucker, “The Discipline of Innovation,” Harvard Business Review (May/June 1985): 
67–72.

 5. Rosabeth M. Kanter, “Change-Master Skills: What It Takes to Be Creative,” in Managing 
Innovation, eds. Jane Henry and David Walker (London: Sage Publications, 1991), 54–61.

 6. Ronald C. Jantz, “Incremental and Radical Innovations in Research Libraries: An Explor-
atory Examination Regarding the Effects of Ambidexterity, Organizational Structure, Leadership, 
and Contextual Factors” (PhD diss., Rutgers University, 2013). Last modified: April 2, 2013.

 7. Harold L. Angle and Andrew H. Van de Ven, “Suggestions for Managing the Innovation 
Journey,” in Research on the Management of Innovation: The Minnesota Studies, eds. Andrew H. Van 
de Ven, Harold L. Angle, and Marshall Scott Poole (New York: Harper and Row, 1989), 664.

 8. Fariborz Damanpour and Shanthi Gopalakrishnan, “Theories of Organizational Structure 
Innovation Adoption: The Role of Environmental Change,” Journal of Engineering and Technology 
Management 15 (1998): 5–10.

 9. Everett M. Rogers, Diffusion of Innovations, 5th ed. (New York: Free Press, 2003), 5–6.
 10. Robert B. Duncan, “The Ambidextrous Organization: Designing Dual Structures for In-

novation,” in The Management of Organization Design: Strategies and Implementation, eds. Ralph H. 
Killman, Louis R. Pondy, and Dennis P. Slevin ( New York: North-Holland, 1976), 168–70.

11. Figure 1 adapted from Rogers, Diffusion of Innovations, 170.
12. Alfred A. Marcus and Mark J. Weber, “Externally Induced Innovation,” in Research on the 



The Determinants of Organizational Innovation  533

Management of Innovation: The Minnesota Studies, eds. Andrew H. Van de Ven, Harold L. Angle, 
and Marshall Scott Poole (New York: Harper and Row, 1989), 537–39.

13. Katherine J. Klein and Joann Speer Sorra, “The Challenge of Innovation Implementation,” 
Academy of Management Review 21 (1996): 1055.

14. The top management team (TMT) is the relatively small group of influential leaders and 
managers at the top of the organization. In the research library, the TMT is defined as the uni-
versity librarian and direct reports. In this study, the phrases “library leadership” and “senior 
team” are considered synonymous with the TMT concept.

15. Donald C. Hambrick and Phyllis A. Mason, “Upper Echelons: The Organization as a Reflec-
tion of Its Top Managers,” Academy of Management Review 9 (1984): 198–202; Donald C. Hambrick, 
“Top Management Groups: A Conceptual Integration and Reconsideration of the Team Label,” in 
Research in Organizational Behavior, Vol. 16, eds. Barry M. Staw and L. L. Cummings (Greenwich, 
Conn.: JAI Press, 1994): 188–200.

16. Hambrick, “Top Management Groups,” 189.
17. Abraham Carmeli and John Schaubroeck, “Top Management Team Behavioral Integration, 

Decision Quality, and Organizational Decline,” Leadership Quarterly 17 (2006): 442.
18. Abraham Carmeli, “Top Management Team Behavioral Integration and the Performance 

of Service Organizations,” Group & Organization Management 33 (2008): 725–26.
19. Fariborz Damanpour and Deepa Aravind, “Organizational Structure and Innovation 

Revisited: 
From Organic to Ambidextrous Structure” in Handbook of Organizational Creativity, ed. Michael 

D. Mumford (London; Waltham, MA: Elsevier/Academic Press, 2012), 503. 
20. Eunice M. L. Soriano de Alencar, “Creativity in Organizations: Facilitators and Inhibitors” 

in Handbook of Organizational Creativity, ed. Michael D. Mumford (London; Waltham, MA: Else-
vier/Academic Press, 2012), 100; Mary Uhl-Bien, Russ Marion, and Bill McKelvey, “Complexity 
Leadership Theory: Shifting Leadership from the Industrial Age to the Knowledge Era” (2007), 
Leadership Institute Faculty Publications Paper 18, available online at http://digitalcommons.unl.
edu/leadershipfacpub/18 [accessed 23 January 2014].

21. D. Charles Galunic and Kathleen M. Eisenhardt, “Reviewing the Strategy-Structure-
Performance Paradigm” in Research in Organizational Behavior, Vol. 16, eds. Barry M. Staw and L. 
L. Cummings (Greenwich, Conn.: JAI Press, 1994), 216. 

22. Tom Burns and George Macpherson Stalker, The Management of Innovation (London: Ta-
vistock, 1961), 105.

23. Peter M. Blau, “A Formal Theory of Differentiation in Organizations,” American Sociologi-
cal Review 35 (1970): 201–18; Justin J. P. Jansen, Michiel P. Tempelaar, Frans A. J. van den Bosch, 
and Henk W. Volberda, “Structural Differentiation and Ambidexterity: The Mediating Role of 
Integration Mechanisms,” Organization Science 20 (2009): 809.

24. John E. Ettlie, William P. Bridges, and Robert D. O’Keefe, “Organization Strategy and 
Structural Differences for Radical versus Incremental Innovation,” Management Science 30 (1984): 
682– 95.

25. Justin J. P. Jansen, Frans A. J. Van Den Bosch, and Henk W. Volberda, “Exploratory In-
novation, Exploitative Innovation, and Performance: Effects of Organizational Antecedents and 
Environmental Moderators. Management Science 52, no. 11 (2006): 1667–79.

26. Fariborz Damanpour, “Organizational Size and Innovation,” Organization Studies 13 (1992): 
395; Richard Germain, “The Role of Context and Structure in Radical and Incremental Logistics 
Innovation Adoption,” Journal of Business Research 35 (1996): 122–24.

27. James G. March, “Exploration and Exploitation in Organizational Learning,” Organization 
Science 2 (1991): 85.
28. Katherine J. Klein and Andrew P. Knight, “Innovation Implementation: Overcoming the 

Challenge,” Current Directions in Psychological Science 14 (2005): 245.
29. Zi-Lin He and Poh-Kam Wong, “Exploration and Exploitation: An Empirical Test of the 

Ambidexterity Hypothesis,” Organization Science 15 (2004): 487–90.
30. Michael L. Tushman and Charles O’Reilly, Winning through Innovation: A Practical Guide to 

Leading Organizational Change and Renewal (Boston: Harvard Business School Press, 1997): 35–37, 
170.

31. Mirko Noordegraaf and Rosemary Stewart, “Managerial Behavior Research in Private and 
Public Sectors: Distinctiveness, Disputes, and Directions, Journal of Management Studies 37, 

no. 3 (2000): 427–43.
32. G. Dess and D. Beard, “Dimensions of Organizational Task Environments,” Administrative 

Science Quarterly 29 (1984): 54–57.
33. George A. Boyne and Kenneth J. Meier, “Environmental Turbulence, Organizational Stabil-

ity, and Public Service,” Administration & Society 40, no. 8 (2009): 803.
34. Boyne and Meier, “Environmental Turbulence,” 805–06.



534  College & Research Libraries May 2015

35. Christine S. Koberg, Dawn R. Detienne, and Kurt A. Heppard, “An Empirical Test of Envi-
ronmental, Organizational, and Process Factors Affecting Incremental and Radical Innovation,” 
Journal of High Technology Management Research 14 (2003): 37–38.

36. Jansen, Van Den Bosch, and Volberda, “Exploratory Innovation,” 1667.
37. Rhys Andrews, “Organizational Task Environments and Performance: An Empirical 

Analysis,” International Public Management Journal 12, no. 1 (2009): 12–15.
38. Richard Germain, “The Role of Context and Structure in Radical and Incremental Logistics 

Innovation Adoption,” Journal of Business Research 35, no. 2) (1996): 123–24.
39. Fariborz Damanpour, “Organizational Complexity and Innovation: Developing and 
Testing Multiple Contingency Models,” Management Science 42, no. 5 (1996): 705.
40. Fariborz Damanpour, “Bureaucracy and Innovation Revisited: Effects of Contingency 
Factors, Industrial Sectors, and Innovation Characteristics,” Journal of High Technology Manage-

ment Research 7, no. 2 (1996): 159.
41. See survey questions at http://dx.doi.org/doi:10.7282/T37D2S88. 
42. Gerald Zaltman, Robert Duncan, and Jonny Holbeck, Innovations & Organizations. (Malabar, 

Fla.: Robert E. Krieger Publishing, 1984), 24.
43. Although the term “radical” has some pejorative connotations, it will be used in lieu of 

perhaps more appropriate but more awkward phrases such as “nonincremental” and “discon-
tinuous” innovations.

44. Questions have been adapted from Hubert Gatignon, Michael L. Tushman, Wendy Smith, 
and Philip Anderson, “A Structural Approach to Assessing Innovation: Construct Development 
of Innovation Locus, Type, and Characteristics,” Management Science 48, no. 9 (2002): 1112. 

45. See survey questions 33–96 for each innovation: http://dx.doi.org/doi:10.7282/T37D2S88 .
46. Midrange innovations are those rated by the special panel as falling between the radical 

and incremental endpoints.
47. See the following questions on the innovation survey for each major construct: behavioral 

integration (1–9), an ambidextrous orientation (21–32), structural differentiation (10–14), and the 
external environment (15–20): http://dx.doi.org/doi:10.7282/T37D2S88.

48. Zeki Simsek, John F. Veiga, Michael H. Lubatkin, and Richard N. Dino, “Modeling the 
Multilevel Determinants of Top Management Team Behavioral Integration,” Academy of Manage-
ment Journal 48, no. 1 (2005): 73–78.

49. Cronbach’s alpha provides a means of assessing internal consistency reliability of a 
multiple-item scale. See Rebecca M. Warner, Applied Statistics: From Bivariate through Multivariate 
Techniques (Los Angeles: Sage, 2013), 931.

50. Michael H. Lubatkin, Zeki Simsek, Yan Ling, and John F. Veiga, “Ambidexterity and 
Performance in Small- to Medium-sized Firms: The Pivotal Role of Top Management Team Be-
havioral Integration,” Journal of Management 32 (2006): 665–66; He and Wong, “Exploration and 
Exploitation,” 492. 

51. Jansen, Tempelaar, van den Bosch, and Volberda, “Structural Differentiation and Ambi-
dexterity,” 809.

52. Francis Milliken, “Three Types of Perceived Uncertainty about the Environment: State, 
Effect and Response Uncertainty,” Academy of Management Review 12, no. 1 (1987): 134–35; David 

A. Waldman, Gabriel G. Ramirez, and Robert J. House, “Does Leadership Matter? CEO Leadership 
Attributes and Profitability under Conditions of Perceived Environmental Uncertainty,” Academy 
of Management Journal 44, no. 1 (2001): 136.

53. Pradip N. Khandwalla, The Design of Organizations (New York: Harcourt Brace Jovanovich, 
1976), 237–38.

54. March, “Exploration and Exploitation,” 71; Charles A. O’Reilly III and Michael L. Tush-
man, “The Ambidextrous Organization,” Harvard Business Review (Apr. 2004): 76; Michael L. 
Tushman and Charles A. O’Reilly III, “Ambidextrous Organizations: Managing Evolutionary 
and Revolutionary Change,” California Management Review 38 (1996): 11.

55. Fariborz Damanpour (1991), “Organizational Innovation: A Meta-analysis of Effects of 
Determinants and Moderators,” Academy of Management Journal 34 (1991): 558–59; Fariborz 

Damanpour and Marguerite Schneider, “Phases of the Adoption of Innovations in Organizations: 
Effects of Environment, Organization, and Top Managers,” British Journal of Management 17 (2006): 
230–31; Fariborz Damanpour and Deepa Aravind, “Organizational Structure and Innovation 
Revisited: From Organic to Ambidextrous Structure” in Handbook of Organizational Creativity, ed. 
Michael D. Mumford (London; Waltham, MA: Elsevier/Academic Press, 2012): 502–09; Detelin 
Elenkov, William Judge, and Peter Wright, “Strategic Leadership and Executive Innovation Influ-
ence: An International Multi-cluster Comparative Study,” Strategic Management Journal 26 (2005): 
676–79; Richard A. Wolfe, “Organizational Innovativeness: Review, Critique, and Suggestions for 
Research Directions,” Journal of Management Studies 31 (1994): 415–18.

56. The phrases “predictor variable” and “independent variable” will be used interchangeably 



The Determinants of Organizational Innovation  535

in this study.
57. Ronald C. Jantz, “Innovation in Academic Libraries: An Analysis of University Librarians’ 

Perspectives,” Library & Information Science Research 34, no. 1 (2012): 3–12; Ronald C. Jantz, “A 
Framework for Studying Organizational Innovation in Research Libraries,” College & Research 
Libraries 73 (2012): 525–41. 

58. Warner, Applied Statistics, 560.
59. The term “enabler” is taken from an in-depth study of innovation in high schools; Richard 

L. Daft and Selwyn W. Becker, Innovation in Organizations (New York: Elsevier, 1978), 54.
60. Damanpour, “Organizational Innovation,” 559; Fariborz Damanpour, “Organizational 

Size and Innovation,” Organization Studies 13 (1992): 395.
61. Fariborz Damanpour and Thomas Childers, “The Adoption of Innovation in Public Librar-

ies,” Library & Information Science Research 7 (1985): 242; John R. Kimberly and Michael J. Evanisko, 
“Organizational Innovation: The Influence of Individual, Organizational, and Contextual Factors 
on Hospital Adoption of Technological and Administrative Innovations,” Academy of Management 
Journal 24 (1981): 699–708; Richard M. Walker, “An Empirical Evaluation of Innovation Types and 
Organizational and Environmental Characteristics: Towards a Configuration Approach,” Journal 
of Public Administration Research and Theory 18 (2008): 597, 607.

62. Abraham Carmeli and Meyrav Yitzack Halevi, “How Top Management Team Behavioral 
Integration and Behavioral Complexity Enable Organizational Ambidexterity: The Moderat-
ing Role of Contextual Ambidexterity,” Leadership Quarterly 20 (2009): 208; Abraham Carmeli 
and John Schaubroeck, “Top Management Team Behavioral Integration, Decision Quality, and 
Organizational Decline,” Leadership Quarterly 17 (2006): 441; Lubatkin, Simsek, Ling, and Veiga, 
“Ambidexterity and Performance in Small- to Medium-sized Firms.”

63. Michael Tushman, Wendy Smith, Robert Wood, George Westerman, and Charles O’Reilly, 
“Innovation Streams and Ambidextrous Organizational Designs: On Building Dynamic Capa-
bilities” (2002): 2, available online at http://web.mit.edu/sloan/osg-seminar/f02_docs/Tushman-
EtAl_2002.pdf [accessed 23 January 2014]. 

64. Constantine Andriopoulos and Marianne W. Lewis, “Exploitation-Exploration Tensions 
and Organizational Ambidexterity: Managing Paradoxes of Innovation,” Organization Science 20 
(2009): 701–08.

65. The innovation survey posed this question for 32 innovations, all of which had been 
implemented in at least one research library.

66. Duncan, “The Ambidextrous Organization,” 172–73.
67. Damanpour and Aravind, “Organizational Structure and Innovation Revisited,” 502–03. 
68. Wesley M. Cohen and Daniel A. Levinthal, “Absorptive Capacity: A New Perspective on 

Learning and Innovation,” Administrative Science Quarterly, 35 (1990): 138–42, 148.
69. Tushman and O’Reilly, Winning Through Innovation, 156–57.
70. Ibid, 169.
71. Lisa German and Beth Sandore Namachchivaya, SPEC Kit 339: Innovation and R&D (Wash-

ington, DC: Association of Research Libraries, 2013), 16.
72. Marja Alestalo, “Patterns of Influence in Research Work: The Problems of Autonomy and 
Democracy,” R&D Management (Special Issue on the Management of Research Groups) 9 (1979): 221.
73. Elenkov, Judge, and Wright, “Strategic Leadership and Executive Innovation Influence,” 

668.
74. James C. Sarros, Brian K. Cooper, and Joseph C. Santora, “Building a Climate for Innova-

tion through Transformational Leadership and Organizational Culture,” Journal of Leadership & 
Organizational Studies 15 (2008): 153.

75. Andriopoulos and Lewis, “Exploitation-Exploration Tensions and Organizational Am-
bidexterity”; Wendy K. Smith and Michael L. Tushman, “Managing Strategic Contradictions: A 
Top Management Model for Managing Innovation Streams. Organization Science 16 (2005): 533.

76. Smith and Tushman, “Managing Strategic Contradictions,” 523.
77. David W. Lewis, “From Stacks to the Web: The Transformation of Academic Library Col-

lecting,” College & Research Libraries 74 (2013): 167–71.
78. Hambrick and Mason, “Upper Echelons,” 198–202.
79. Hambrick, “Top Management Groups,” 177.
80. Karen A. Bantel and Susan E. Jackson, “Top Management and Innovations in Banking: 

Does the Composition of the Top Team Make a Difference,” Strategic Management Journal 10 
(1989): 107–24; Urs Daellenbach, Anne McCarthy, and Timothy Schoenecker, “Commitment to 
Innovation: The Impact of Top Management Team Characteristics,” R&D Management 29 (1999): 
199–208; Katrin Talke, Soren Salomo, and Katja Rost, “How Top Management Team Diversity 
Affects Innovativeness and Performance Via the Strategic Choice to Focus on Innovation Fields,” 
Research Policy 39 (2010): 907–18.

81. Bantel and Jackson, “Top Management and Innovations in Banking,” 107.



536  College & Research Libraries May 2015

82. Margarethe F. Wiersema and Karen A. Bantel, “TMT Demography and Corporate Strategic 
Change,” Academy of Management Journal 35 (1992): 111.

83. Jantz, “A Framework for Studying Organizational Innovation,” 536.
84. Amy C. Edmondson, Michael A. Roberto, and Michael D. Watkins, “A Dynamic Model of 

Top Management Team Effectiveness: Managing Unstructured Task Streams,” Leadership Quar-
terly 14 (2003): 300; Paul Hersey and Kenneth H. Blanchard, “Life Cycle Theory of Leadership,” 
Training and Development Journal 23, no. 2 (1969): 26–34.

85. Bantel and Jackson, “Top Management and Innovations in Banking,” 118–20.
86. Wiersema and Bantel, “TMT Demography and Corporate Strategic Change,” 110–12.
87. Francis Milliken, “Three Types of Perceived Uncertainty,” 136.
88. Ibid, 138.
89. Koberg, Detienne, and Heppard, “An Empirical Test,” 31.
90. Ming Piao, “Thriving in the New: Implication of Exploration on Organizational Longev-

ity,” Journal of Management 36, no. 6 (2010): 1529.
91. Kristina Jaskyte, “Predictors of Administrative and Technological Innovations in Nonprofit 

Organizations,” Public Administration Review 71, no. 1 (2011): 83–84; Julian Birkinshaw, Gary Hamel, 
and Michael J. Mol, “Management Innovation,” Academy of Management Review 33 (2008): 832.

92. Bernard M. Bass, Leadership and Performance Beyond Expectations (New York: The 
Free Press, 1985): 159–60.
93. Andrew Hargadon and Robert I. Sutton, “Building an Innovation Factory,” Harvard Busi-

ness Review 78, (2000): 157.
94. For the regression analysis, note that three outliers were removed.