The Influence of Centralization and Extent of Cross-Functional Team Usage on Senior Managers’ Risk-Related Perceptions

Abstract

Despite the importance of managing risk taking in organizations, we know relatively little about how organizational design influences the risk-related perceptions of key organizational decision makers such as senior managers. This study examines how two basic organizational design variables—the extent to which organizations use cross-functional teams (CFTs) and the centralization of decision-making authority—interact to influence senior managers’ perceptions of the extent to which risk taking is supported within their organizations, in a new product development context. Regression results from a survey of 102 senior managers of various organizations in the United States and Norway find that extent of CFT use positively influences senior managers’ perceptions of organizational support for risk taking at low to moderate levels of decision-making authority. At high levels of senior manager authority, risk-related perceptions are not only more positive but also largely independent of the extent of CFT use.

Keywords

control systems cross-functional teams centralization senior managers risk-taking perceptions new product development

Introduction

Risk is a core construct in the study of organizations. A vast number of studies have used a wide variety of perspectives to examine why organizations and their key decision makers make risky decisions and the consequences of these decisions for organizational performance (see Bromiley & Rau, 2010, for a review). A subset of these studies examines risk in a new product development (NPD) context (Cabrales, Medina, Lavado, & Cabrera, 2008; He & Mittal, 2007; Salomo, Weise, & Gemunden, 2007; Sethi, Smith, & Park, 2001). A common idea underlying these studies is that risk taking, properly managed, leads to better NPD.

The problem is that though risk taking is critical to NPD, organizations often take too little risk in these settings (Slater & Narver, 1995). In other settings (e.g., in banks where the widespread of use of new—but poorly understood—financial products precipitated the 2006-2008 banking crisis), the opposite may be the dominating concern. Managing an organization’s risk taking has thus long been a topic of interest to mangers, boards, and researchers (Devers, Wiseman, & Holmes, 2007). In particular, a number of studies in this area have examined the determinants of managers’—especially senior managers’—risk taking (see Bromiley & Rau, 2011, for a review), as senior managers are the key decision makers in an organization (Hambrick & Mason, 1984).

Our study adds to this research by examining this issue from a control systems perspective. We examine how two basic organizational design variables—the extent to which organizations use cross-functional teams (CFTs) and the degree to which decision-making authority is centralized at the level of the senior manager—influence senior managers’ perceptions of the extent to which risk taking is supported within their organizations, in an NPD context. We focus on CFTs and centralization because they represent two different kinds of personal control systems, enabling horizontal coordination and vertical control, respectively. In addition, a number of studies on NPD identify managing these two variables as the most effective ways of organizing for NPD (Anderson, De Dreu, & Nijstad, 2004; Barczak, Griffin, & Kahn, 2009). We examine senior managers’ perceptions, because these perceptions will influence the strategic decisions that senior managers actually make for their organizations.

We see our study as contributing to the literature on control systems in the field of organization theory. Organizational control systems, in general, are ways to improve coordination and reduce the number of exceptions and errors (Perrow, 1967; Van de Ven & Delbecq, 1974; Van de Ven, Delbecq, & Koenig, 1976) and increase efficiency (Mintzberg, 1979; Scott & Davis, 2007; Thompson, 1967). The extent to which organizational control mechanisms have effects beyond organizational structure, efficiency, and task uncertainty—for example, on risk perceptions—has not been a major topic of examination by researchers. Identifying these additional effects of control systems can increase our understanding of how to design organizations to achieve desired outcomes.

Our study also contributes to the literature on risk taking, particularly in an NPD context. Despite the recognition of the importance of risk taking for successful NPD, we know relatively little about how organizational design factors influence risk perceptions, because studies in this area have focused on other organizational or micro-level factors that influence risk taking (Mullins, Forlani, & Walker, 1999; Voss, Sirdeshmukh, & Voss, 2008). By examining the effects of CFT use and centralization of decision-making authority, we gain a clearer picture of how organizational control systems influence key decision makers’ perceptions related to risk taking during NPD.

We now draw on the risk and NPD literatures to propose testable hypotheses relating the extent of CFT use and centralization of authority to senior managers’ perceptions of risk taking.

Theory and Hypotheses

We begin by defining the two key constructs of our study, namely, risk taking and organizational control systems.

Risk Taking

Given our focus on senior managers’ perceptions of the extent to which risk taking is encouraged in their organizations, we follow the tradition of defining risk from the perspective of practicing managers (Calantone, Garcia, & Droge, 2003; March & Shapira, 1987). From this perspective, risk is a construct associated with only potential negative outcomes, because most managers would not consider a decision risky if they expected to obtain desirable outcomes (March & Shapira, 1987; Miller & Leiblein, 1996). In addition, “managers offer a very active perception of risk taking. They see risk taking as part of their jobs, but also believe they substantially influence the outcome after the risky choice” (Bromiley & Rau, 2010, p. 309).

“Senior managers” refer to managers who make critical new product or service-related decisions regarding project prioritization, support, or withdrawal of project resources, and development aspects of new projects. Typically, these managers have strategic business unit responsibilities, one of which is the successful commercialization and launch of new products.

Organizational Control Systems

Organizations coordinate and regulate risk taking by implementing control systems (Scott & Davis, 2007). Organizations use many different types of control systems such as personal controls, cultural controls, output controls, and bureaucratic controls (Scott & Davis, 2007). Personal control systems attempt to “shape and influence the behavior of a person in a face-to-face interaction in the pursuit of a company’s goals” (Hill & Jones, 2013, p. 425). As organizations increase in size and complexity, personal control systems tend to give way to impersonal coordination systems and more formalized decision processes (Mintzberg, 1979; Scott & Davis, 2007). A key determinant of the transition between the use of personal and impersonal control systems is task uncertainty: The higher the task uncertainty, the greater the use of personal control systems (Van de Ven & Delbecq, 1974; Van de Ven et al., 1976).

In a product development setting, with its high levels of uncertainty, this implies a relatively frequent use of personal control systems. Personal control systems, however, come in many flavors. A supervisor talking to a subordinate about a task is using a personal control system, as is an employee conversing with a peer about how to complete a joint task (Hill & Jones, 2013). In our study, we examine two fundamentally different types of personal control systems—horizontal and vertical—for their effect on individuals’ risk-taking perceptions.

Horizontal Control Systems: Extent of CFT Use

Horizontal personal control systems reduce communication barriers and enable better coordination among departments. CFTs, which draw together members from different departments to work on a common task, are thus a type of horizontal coordination channel. This view of CFTs is consistent with that of some prior studies. Mintzberg (1979), for example, in his description of an adhocracy, describes CFTs as an essential integration mechanism in a horizontally differentiated structure that is used to integrate knowledge from specialized functional areas to solve complex problems.

As CFTs bring together a number of organizational members with different areas of expertise, they are “best practice” for NPDs (Barczak et al., 2009; Brown & Eisenhardt, 1995). The diversity present in CFTs increases the amount and variety of information available to project team members. This in turn increases the understanding of the NPD process among team members and results in better products and process outcomes (see Brown & Eisenhardt, 1995, for a review). The link between the use of CFTs and NPD performance is in fact among “the most important and empirically robust” findings in this field (Brown & Eisenhardt, 1995, p. 367).

At the organizational level, as organizations begin to implement CFTs extensively, the practice of setting up and using CFTs for NPD should develop common elements. That is, the extensive use of CFTs will make the use of CFTs become a type of routine, where routines are defined as “repetitive, recognizable patterns of interdependent actions, carried out by multiple actors” (Feldman & Pentland, 2003, p. 95). Organizational routines are tools to formalize task resolution. Formalization of behavior by routines is one of the most researched relations in organization theory (Donaldson, 2001). As the routinization of CFT use occurs, it helps manage the uncertainty inherent in NPD both by increasing reliability and predictability of CFT work (March, 1994) and by encoding lessons from history (Levitt & March, 1988). As such, consistent with managerial perceptions on risk taking (March & Shapira, 1987), managers may come to perceive risk taking during NPD—when using CFTs—as manageable and controllable.

We suggest that the above control effects associated with the routinization of CFT use, combined with the positive relations between CFT use and NPD performance discussed earlier, should lead senior managers to have more positive perceptions of organizational support for risk taking during NPD. This is because of the “house money effect” (Thaler & Johnson, 1990), which predicts,

if an individual has already won a positive amount of money from the sequence of decisions, he is willing to take fair bets as long as the payoffs are small relative to the amount already won. . . . They are willing to risk some, but not all, of their previous winnings. (Neilson, 1998, pp. 27-28)

In the context of our study, we submit that although senior managers may not actually make more risky decisions, they may be more willing to believe that their organizations would support at least some risk taking during NPD, given the previously observed positive outcomes with NPD and their own perceptions that, with the use of CFTs, risk taking during NPD is controllable.

A few studies provide evidence in support of our argument. The financial literature finds evidence for the “house money effect” at the individual level (Keasey & Moon, 1996; Thaler & Johnson, 1990). At the organizational level, Mishina, Pollock, and Bragaw (2012) find that CEO decision making is subject to the same effect. We therefore hypothesize as follows:

Hypothesis 1 (H1): Extent of CFT use will positively influence senior managers’ perceptions of the extent to which their organizations encourage risk taking during NPD.

We have reasoned above that extent of CFT use acts as a type of horizontal personal control system that influences senior managers’ perceptions related to risk taking. As with other control systems, we note that the exact effects of CFT use on senior managers’ risk perceptions will also depend on other organization specific factors, and specifically, on other control systems in use within the organization. We therefore turn to the second type of personal control system we examine in this study, namely, vertical control systems.

Vertical Personal Control Systems: Centralization of Decision-Making Authority at the Senior Manager Level

Vertical control refers to centralized decision making or the top-down control of tasks and processes in organizations; in other words, control by hierarchy (Bouwman, Van Den Hooff, Van De Wijngaert, & Van Dijk, 2005). In our study, vertical control refers to the extent to which decision-making authority is centralized at the level of the senior manager. We now examine how vertical control will moderate the relations between extensiveness of CFT use and senior managers’ perceptions.

Vertical and Horizontal Control Systems and Risk Perceptions

In organization theory, horizontal control through coordination among teams or units as a means of managing high uncertainty tasks is an alternative to vertical control or centralization (e.g., Van de Ven et al., 1976). Both types of personal control systems correlate and serve the same purpose (Van de Ven et al., 1976). We submit, however, that although centralization of authority and coordination through horizontal channels may lead to equivalent outcomes in terms of organizational control, they will have different implications for managers’ perceptions of organizational support for risk taking when they are used jointly for managing high uncertainty tasks like NPD.

When we examine CFTs and centralization, it is apparent that they exert control through different mechanisms. Organizations use CFTs to delegate decision-making authority to a set of team members, thereby diffusing responsibility among these members. In contrast, centralization, or vesting a senior manager with a high degree of decision-making authority (e.g., Gulick & Urwick, 1937; Mintzberg, 1979), concentrates authority in one point of the organizational hierarchy.

A number of studies on NPD find that effective NPD requires that delegation and centralization work hand in hand: CFTs are most effective when senior managers at least partially regulate their behavior or else exert some form of “subtle control” (Brown & Eisenhardt, 1995; Griffin, 1997; McDonough, 2000). We build on this to suggest that organizations need to balance the effects of two different types of personal control systems—extent of CFT use and senior managers’ decision-making authority. Specifically, the extent of senior managers’ decision-making authority should moderate the relation between extent of CFT use and managers’ perceptions relating to organizational support for risk taking.

To understand how this may work, let us look at the interaction from an individual manager’s perspective. A vast literature on managerial incentives suggests that managers are fundamentally risk averse (Jensen & Meckling, 1976). Indeed, agency theory assumes that in the modern corporation, firm owners would have to compensate their agents (managers) for making risky decisions that might expose managers to employment risk (Jensen & Meckling, 1976). The key here is the extent to which managers have the decision-making authority required to suppress or encourage any deviations in the CFT routine. Authority is the flip side of responsibility (Arrow, 1974). When an NPD manager has a great deal of decision-making authority, he or she will be clearly recognized, within their organization, as being ultimately responsible for NPD performance. Although these managers would probably be sensitive to the value of risk taking in generating new products, given their responsibility for the success of the project, they would also value to an equal (if not greater) extent, controllability through the CFT routine. Indeed, one of the benefits of routines is that they protect decision makers from the consequences of their actions (Scott, 2002). Senior managers with a great deal of authority may therefore be less likely to perceive greater organizational support for risk taking with increased CFT use.

The situation changes when senior managers have lower levels of decision-making authority. In this scenario, there should be a positive relationship between extent of CFT use and risk perceptions. As in the scenario discussed above, managers with less authority (and therefore, less responsibility for NPD performance), should still be risk averse, given their exposure to employment risk. At the same time, however, their lower decision-making authority, and therefore, responsibility, should make them more susceptible to the value of risk taking and using CFTs during NPD. We propose the following hypothesis:

Hypothesis 2 (H2): Decision-making authority negatively moderates the relations between CFT use and senior managers’ perceptions of risk-taking support such that the higher the decision-making authority, the lower the effect of CFT use on senior managers’ perceptions of organizational support for risk taking.

Method

Data and Sample

Previous research on organizations has sometimes been criticized for being too focused on the U.S. context (Boyacigiller & Adler, 1991). As we were cognizant of this criticism, we took advantage of the fact that the authors of this study were based in two countries—the United States and Norway—and examined managers from both these countries. A list of 170 senior managers was generated from a database of the Product Development and Management Association’s member companies, the authors’ contacts, and a list of Norwegian firms. The list was pre-screened to ensure that it contained only senior managers responsible for their division or business unit’s NPD projects. Eighty-seven of these managers were from the United States and 83 from Norway. The senior managers were then contacted by telephone and requested to participate in a 20-min mail survey. Respondents were assured of complete anonymity in their responses. A summary report of results was offered to encourage interest and participation.

A total of 110 questionnaires were returned, with 68 responses from the United States and 42 responses from Norway. Senior managers were well-represented across functional areas, which included marketing (26.4%), research and development (29.1%), production (13.6%), and other areas (30.9%) such as design/engineering (5.8%), finance (4.4%), sales (4.4%), and IT (2.9%). In addition, these senior managers belonged to firms from multiple industries. About half (51%) of the senior managers from the U.S. firms were employed by manufacturing firms, one third (33%) by firms providing business services, one tenth (9.8%) came from information and communication firms, and the remaining came from food processing, distribution, and software development firms. About 69% of the Norwegian responses were from managers in manufacturing firms, and 14% from managers in firms engaged in professional, scientific, and technical activities. The remaining managers belonged to a variety of firms such as those in wholesale, retail, and trading; information and communication; support service activities; electricity, gas, steam, and air conditioning supply; and arts, entertainment, and recreation. The manufacturing firms in this study produce a wide variety of products ranging from games and toys to chemicals, rubber hose, and diagnostic equipment. The service firms provide a variety of business services including, among others, information-based services, services to educational institutions, accounting, advertising, and claims processing. Survey respondents held titles such as director or VP of NPD, Marketing, and Market Development.

As we required information about risk taking, extent of CFT use, and senior manager decision-making authority, we decided to use a structured questionnaire to collect the data. This questionnaire was developed in two stages. We first reviewed the literature to uncover previously used scale items measuring risk taking, decision-making authority, and CFT use by organizations in a NPD context. We then obtained feedback on the questionnaire from academics familiar with the NPD literature before administering it to the respondents.

Measures: Independent Variables

Extent of CFT use

Extent of CFT use was measured with an eight-item scale from Maltz and Kohli (2000). Sample items include “To what extent does your division/Strategic Business Unit (SBU) use teams that include managers from multiple functions to develop new products” and “To what extent does your division/SBU unit use teams that include managers from multiple functions to identify new markets or customers.” The items were measured on a 5-point Likert-type scale anchored by never and very often. The coefficient alpha of this scale was .88, indicating sufficient reliability.

An exploratory factor analysis of the items in this scale (with maximum likelihood extraction and direct oblimin rotation) revealed one factor with an Eigenvalue of 4 that explained 50% of the variance, and a second factor with an Eigenvalue of 0.6 that explained an additional 8% of the variance. While only one of the Eigenvalues exceeds 1, that the analysis extracted two factors suggests that the scale may not be purely uni-dimensional. A closer look at the pattern matrix indicated that while most of the items loaded onto the first factor, two of the items (which also had the lowest item-total correlations) loaded on the second factor. We re-ran the factor analysis after removing these two items. This factor analysis extracted one factor that explained 52% of the variance. The coefficient alpha of this revised scale was .86, indicating sufficient reliability.

Senior manager decision-making authority

Senior manager decision-making authority was measured with a five-item scale from Ayers, Dahlstrom, and Skinner (1997) on a 7-point Likert-type scale anchored by strongly disagree and strongly agree. Sample items include, “In general, when product-related decisions have to be made for which no rules or procedures exist, I have authority to make those decisions” and “In general, very few actions are taken without the help of a superior” (reverse-scored item). The coefficient alpha of this scale was .66, indicating sufficient reliability (Peter, 1981).

An exploratory factor analysis of the items in this scale (with maximum likelihood extraction and direct oblimin rotation) revealed one factor with an Eigenvalue of 1.6 that explained 32% of the variance, and a second factor with an Eigenvalue of 0.9 that explained an additional 19% of the variance. As the Eigenvalue of the second factor is very close to 1, this suggests that the scale may not be uni-dimensional. A closer look at the pattern matrix indicated that the two items (which the lowest item-total correlations) loaded on the second factor. We re-ran the factor analysis after removing these two items. This factor analysis extracted one factor that explained 48% of the variance. The coefficient alpha of this revised scale was .71, indicating sufficient reliability.

Dependent Variable

Perceptions of the extent to which organizations support risk taking

Individual perceptions of the extent to which their organizations support risk taking were measured with a six-item scale from Calantone et al. (2003) on a 7-point Likert-type scale anchored by strongly disagree and strongly agree. Sample items include, “In my business unit, we believe that higher financial risks are worth taking for higher rewards” and “We accept occasional new product failures as being normal.” The coefficient alpha of this scale was .87, indicating sufficient reliability. An exploratory factor analysis of the items in this scale (with maximum likelihood extraction) revealed one factor that explained 53% of the variance, suggesting a uni-dimensional construct of risk taking.

Establishing Construct Validity

We carried out an exploratory factor analysis (extraction of factors with maximum likelihood method and oblimin rotation) to check whether respondents were able to distinguish among risk taking, CFT use, and degree of decision-making authority. The analysis extracted three distinct factors with Eigenvalues greater than 1, corresponding to the three constructs. Table 1 presents the results of the factor analysis along with a list of items in the three scales.

Table 1.

Results of the Exploratory Factor Analysis on Risk Taking, CFT Use, and Decision-Making Authority: Pattern Matrix.

	Factor 1	Factor 2	Factor 3
Risk taking
1. In my business unit, we believe that higher financial risks are worth taking for higher rewards.	.543	.227	.115
2. We accept occasional new product failures as being normal.	.713	.074	−.199
3. We are willing to take big financial risks in product development.	.818	.004	.076
4. We encourage the development of innovative marketing strategies knowing well that some will fail.	.714	−.076	.034
5. As senior managers, we prefer to “play it safe.”^a	.702	.066	.209
6. As senior managers, we like to implement plans only if we are very certain that they will work.^a	.785	−.152	.034
CFT use
To what extent does your division/SBU unit use teams that include managers from multiple functions to:
1. Visit external customers	−.066	.578	−.053
2. Develop new products	−.241	.894	.073
3. Develop product performance goals	−.046	.727	.114
4. Identify new markets or customers	.130	.600	−.153
5. Determine new product funding priorities	.088	.697	.056
6. Review the performance of a product or product line	.244	.690	−.048
Decision-making authority
Please indicate your level of support with the following statements:
1. In general, when product-related decisions have to be made for which no rules or procedures exist, I have authority to make those decisions	−.012	.047	.732
2. In general, very few actions are taken without the help of a superior.^a	.023	−.065	.876
3. In general, even small matters have to be referred to someone higher in the organization for a final answer.^a	.059	.003	.445

Note. Extraction method: Maximum likelihood; rotation method: Oblimin with Kaiser normalization. χ² = 122.430, df = 63, p < .01. CFT = cross-functional team.

Reverse scored.

As an additional check, we re-ran the factor analysis on all the items for risk taking, CFT use, and decision-making authority (including the items that we had deleted earlier). The results of this analysis revealed four factors, with the problematic items cross-loading on two factors. We therefore conducted subsequent analyses without these items.

To assess convergent and discriminant validity, we used Partial Least Squares–Structural Equation Modeling (PLS-SEM; instead of a regular structural equation model) to conduct a confirmatory factor analysis, as we have a small sample size. For convergent validity, the average variance extracted (AVE) should be 0.50 or higher, meaning that “the latent variable explains more than half of its indicators’ variance” (Hair, Ringle, & Sarstedt, 2011). The AVEs for CFT use, decision-making authority, and risk taking were 0.54, 0.65, and 0.60, respectively, indicating acceptable convergent validity.

Discriminant validity is indicated by two criteria: the AVE of each latent construct should be greater than the latent construct’s highest squared correlation with any other latent construct, and an indicator’s loading with its associated latent construct should be higher than its loadings with all the remaining constructs (Hair et al., 2011). Both these criteria were met, indicating acceptable discriminant validity. Table 2 presents the final cross-loadings of the items on the constructs of decision-making authority, CFT use, and risk taking.

Table 2.

Establishing Convergent Validity: Results of the Confirmatory Factor Analysis for Risk Taking, CFT Use, and Decision-Making Authority—Cross-Loadings From PLS-SEM Analysis.

	Factor 1	Factor 2	Factor 3
Risk taking
1. In my business unit, we believe that higher financial risks are worth taking for higher rewards.	.749	.308	.192
2. We accept occasional new product failures as being normal.	.666	.249	−.005
3. We are willing to take big financial risks in product development.	.866	.235	.249
4. We encourage the development of innovative marketing strategies knowing well that some will fail.	.738	.171	.222
5. As senior managers, we prefer to “play it safe.”^a	.855	.256	.313
6. As senior managers, we like to implement plans only if we are very certain that they will work.^a	.763	.122	.215
CFT use
To what extent does your division/SBU unit use teams that include managers from multiple functions to:
1. Visit external customers	.082	.527	−.099
2. Develop new products	−.004	.684	.021
3. Develop product performance goals	.134	.680	.060
4. Identify new markets or customers	.190	.761	−.094
5. Determine new product funding priorities	.206	.806	.074
6. Review the performance of a product or product line	.342	.907	.002
Decision-making authority
Please indicate your level of support with the following statements:
1. In general, when product-related decisions have to be made for which no rules or procedures exist, I have authority to make those decisions.	.217	.038	.846
2. In general, very few actions are taken without the help of a superior.^a	.265	−.033	.882
3. In general, even small matters have to be referred to someone higher in the organization for a final answer.^a	.178	−.006	.677

Note. CFT = cross-functional teams; PLS-SEM = Partial Least Squares–Structural Equation Modeling.

Reverse scored.

Controls

Following Baird and Thomas (1985), we controlled for external environmental, industry, organizational, and senior manager characteristics that might influence risk-taking attitudes in a NPD context. At the external environmental level, we controlled for country of operation (to capture any effects attributable to differences in national culture). We coded the country of operation as an indicator variable (1 = Norway, 0 = United States).

At the industry level, we controlled for industry turbulence, as organizations in stable industries may take fewer risks than organizations in turbulent industries (Wezel & Saka-Helmhout, 2006). We measured industry turbulence with an eight-item scale from Sarin and Mahajan (2001). The coefficient alpha of this scale was .75, indicating sufficient reliability.

We controlled for organization size as larger organizations may have more slack, allowing them to conduct more (or more risky) NPD. We measured organization size by asking respondents to indicate the total number of employees in their organization (<50 employees, 50-250 employees, and so on). We coded organization size with a dummy variable (1 = organizations with 250 employees or less, 0 = organizations with more than 250 employees). We chose this coding because about half the organizations in our sample employed about 250 employees or less, whereas the other half of the sample employed more than 250 employees.

We also controlled for senior manager characteristics, as senior managers have a critical influence on organizational decisions (Hambrick & Mason, 1984). Specifically, we controlled for senior managers’ age and functional background. We focused on age because some research on risk identifies age as a key socio-economic factor influencing risk taking (Kogan & Wallach, 1964). We controlled for functional background as it may determine managers’ past experiences with risk taking; these experiences may, in turn, influence managers’ perceptions related to risk taking in the future (Sitkin & Weingart, 1995).

Common Method Bias Considerations

Using key informants to provide data on both the dependent and independent variables raises concerns about common method bias. We therefore employed two procedures to assess the extent to which this bias may be present in the data used in this study. The first procedure used Harmon’s single-factor test (Podsakoff & Organ, 1986). We subjected the items used to construct the scaled measures (risk taking, CFT use, senior managers’ decision-making authority) to an exploratory factor analysis (principal components and varimax rotation). The factor solution extracted three factors, with the three scales emerging as they should, and the first factor accounting for 30% of the covariance among the items. In addition, the first factor extracted did not appear to be a general factor. This type of result suggests that the data set is not subject to a significant common method bias (Podsakoff & Organ, 1986). Similar results were obtained in a factor analysis of all the scaled items plus the control variables.

The second procedure we used was partial correlation controlling for the unrotated first factor score from the factor analysis including all of the variables in the study (Podsakoff & Organ, 1986). A comparison of the correlation coefficients before and after controlling for the first unrotated factor score indicates that some of the correlation coefficients were lowered by partialing out the first unrotated factor score. This is expected because this procedure may remove any of the true variance as well as any variance due to common method used in the study (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). As the first factor extracted corresponded to risk taking, significant changes were observed in the correlations of risk taking with other variables. The results thus appear to indicate that though some common method variance may exist, it may not be a significant problem with these data.

Establishing the Cross-National Invariance of Instruments

An issue in cross-national research is that of establishing whether the instruments used to measure constructs are cross-nationally invariant (Steenkamp & Baumgartner, 1998). To check for cross-national invariance, we first carried out an exploratory factor analysis for each of the instruments for each country. The factor analyses indicated a similar pattern of factor loadings for risk taking, CFT use, and senior managers’ degree of decision-making authority for each country. The factor loading patterns for the scales for turbulence, however, were not similar across the two countries. An analysis of the factor loading patterns suggested that three items in the turbulence scale were possibly interpreted differently by respondents in the two countries. Results from re-running the factor analyses after removing these problematic items indicated similar factor loading patterns across the two countries. The coefficient alpha for the revised scale for assessing turbulence was .73.

To confirm measurement equivalence across countries, particularly for our constructs of interest (CFT use, decision-making authority, and risk taking), we then ran a PLS-SEM procedure for these constructs in the two groups (United States and Norway) separately. A comparison of the outer weights or loadings in these two groups revealed similar patterns of loadings, except for one item relating to centralization of decision-making authority. We re-ran the PLS-SEM procedure after deleting this item. The results showed similar patterns of outer weights or loadings across the two groups, suggesting configural invariance, and therefore, a basic level of measurement equivalence (Sarstedt & Ringle, 2010; Steenkamp & Baumgartner, 1998). Table 3 presents the results of this analysis.

Table 3.

Establishing Measurement Equivalence of the Scales for Risk Taking, CFT Use, and Decision-Making Authority Across the United States and Norwegian Samples: PLS-SEM Results.

Outer loadings: U.S. sample
Risk taking
1. In my business unit, we believe that higher financial risks are worth taking for higher rewards.			.770
2. We accept occasional new product failures as being normal.			.652
3. We are willing to take big financial risks in product development.			.857
4. We encourage the development of innovative marketing strategies knowing well that some will fail.			.765
5. As senior managers, we prefer to “play it safe.”^a			.850
6. As senior managers, we like to implement plans only if we are very certain that they will work.^a			.740
CFT use
To what extent does your division/SBU unit use teams that include managers from multiple functions to:
1. Visit external customers		.573
2. Develop new products		.772
3. Develop product performance goals		.680
4. Identify new markets or customers		.770
5. Determine new product funding priorities		.821
6. Review the performance of a product or product line		.883
Decision-making authority
Please indicate your level of support with the following statements:
1. In general, when product-related decisions have to be made for which no rules or procedures exist, I have authority to make those decisions.	.864
2. In general, very few actions are taken without the help of a superior.^a	.961

Outer loadings: Norwegian sample
Risk taking
1. In my business unit, we believe that higher financial risks are worth taking for higher rewards.			.462
2. We accept occasional new product failures as being normal.			.678
3. We are willing to take big financial risks in product development.			.780
4. We encourage the development of innovative marketing strategies knowing well that some will fail.			.762
5. As senior managers, we prefer to “play it safe.”^a			.782
6. As senior managers, we like to implement plans only if we are very certain that they will work.^a			.838
CFT use
To what extent does your division/SBU unit use teams that include managers from multiple functions to:
1. Visit external customers		.454
2. Develop new products		.956
3. Develop product performance goals		.856
4. Identify new markets or customers		.556
5. Determine new product funding priorities		.623
6. Review the performance of a product or product line		.554
Decision-making authority
Please indicate your level of support with the following statements:
1. In general, when product-related decisions have to be made for which no rules or procedures exist, I have authority to make those decisions.	.931
2. In general, very few actions are taken without the help of a superior.^a	.831

Note. CFT = cross-functional teams; PLS-SEM = Partial Least Squares–Structural Equation Modeling.

Reverse scored.

We also ran the PLS-SEM analysis for the entire sample (both U.S. and Norway data, but with the deleted item) again, and compared the results with those obtained when we tested for convergent validity. The two sets of results showed similar patterns. In addition, the composite reliability measure of decision-making authority scale was .87, indicating acceptable reliability.

Establishing Model Invariance Across Countries

We next conducted a Chow test for model invariance. That is, we tested whether the samples from the two countries should be pooled or analyzed separately. This is important for both theoretical and empirical reasons. From a theoretical standpoint, the extent of similarity between Norway and the United States is debatable. Although the United States and Norway are similar on some economic and social indicators (e.g., life expectancy, GDP per capita, and literacy rate) and share similar profiles on two of Hofstede’s (1980) four dimensions of national culture, they differ on two other dimensions of national culture (Sornes, Stephens, Saetre, & Browning, 2004) and in other aspects as well (Esping-Andersen, 1990). Esping-Andersen (1990), in his highly influential book, The Three Worlds of Welfare Capitalism, identifies three major types of welfare states: liberal, corporatist-statist, and social democratic. Social democratic states such as Norway fuse welfare and work, whereas liberal states like the United States use means-tested approaches and modest universal transfers or social insurance plans.

It is possible that these differences may lead to different approaches to business; the empirical evidence regarding this, however, is mixed. A qualitative study of women members of boards of directors in both countries, for example, finds not only substantial similarities but also some differences across the two countries (Bilimoria & Huse, 1997). Similarly, Sornes et al. (2004) compares how workers in the United States and Norway use information and communication technologies, and finds both similarities and differences across the two groups, reflecting their similarities and differences in national culture. In our sample, an exploratory analysis comparing responses from senior managers from the United States and Norway found similar results, except for one variable. Norwegian and U.S. managers differed significantly in their responses to the dependent variable—their perceptions of the extent to which their organizations’ support risk taking (F = 6.889, p < .05)—but not in any of the independent variables of interest.

Given this difference in means for the variable perceived organizational support for risk taking, we wanted to test whether the model explaining the variance in the dependent variable would differ across the two samples, that is, whether there would be differences in the coefficients of the independent variables. This is also a test of whether the two samples should be analyzed separately or pooled together for further analysis. Results of the Chow test, F = .873 for a critical F(10, 79) of 1.95, indicate that the null hypothesis that coefficients are the same for both countries (i.e., the independent variables have the same effect on the dependent variable across the two samples) cannot be rejected at an alpha of 5%. We therefore pooled the two samples for further analyses.

Results

Table 4 presents the means, standard deviations, and intercorrelations among the variables used in this study.

Table 4.

Means, Standard Deviations, and Intercorrelations.

	M (SD)	Country^a	Organization size	Turbulence	Marketing	R&D	Manufacturing	Age	CFT use	Decision-making authority
1. Organization size	0.5364 (0.5010)	.355**
2. Turbulence	2.1389 (0.4857)	−.101	−.079
3. Marketing	0.3048 (0.4625)	.034	.122	−.175
4. R&D	0.3429 (0.4769)	.094	−.088	−.024	−.478**
5. Manufacturing	0.0286 (0.1674)	.101	−.183	−.293**	−.114	−.124
6. Age	44.03 (9.41)	.065	−.030	.127	−.534**	.215*	−.074
7. CFT use	3.5455 (0.7820)	−.011	−.302*	.107	−.118	.080	.102	−.158
8. Decision-making authority	2.0131 (0.5138)	.141	.065	.215*	−.113	.134	−.179	.027	−.001
6. Risk taking	4.0970 (1.1203)	.245**	.086	.148	−.277**	.200*	−.077	.244*	.193*	.243*

Note. CFT = cross-functional teams.

Norway = 1; United States = 0.

p < .05. **p < .01.

Table 4 indicates that risk-taking perceptions correlate with the country code (r = .245, p < .01), extent of CFT use (r = .193, p < .05), and senior managers’ decision-making authority (r = .243, p < .05).

H1 states that a positive relationship will exist between extent of CFT use and senior managers’ perceptions of the extent to which their organizations support risk taking. H2 states that senior managers’ decision-making authority will negatively moderate this relation. We examined both these hypotheses using multiple regression analyses. We regressed extent of CFT use, decision-making authority, and the interaction of these two factors on risk-taking perceptions. Moderation effects are indicated by significant interactions between degree of authority and CFT use (Baron & Kenny, 1986). Because the regression includes interaction terms, extent of CFT use and degree of authority were centered before carrying out the regression to reduce multicollinearity (Neter, Kutner, Nachtsheim, & Wasserman, 1996). The regression results indicated that the effect of industry turbulence on risk taking was both non-significant and miniscule (b = .010, p = .967). Given the size of the sample and the number of variables in the analysis, this variable was dropped from further analysis to improve the statistical power of the results. Dropping this variable did not significantly change the magnitude or direction of the results. Table 5 presents the results of the final regression analysis. The first column reports regression results with only the controls, the second column reports the results with the main effects added, and the third column reports the results with all variables and interaction terms. We will mainly discuss the results with the interactions.

Table 5.

Regression Results: Dependent Variable—Senior Manager Perceptions of Organizational Support for Risk Taking.

	I	II	III
Constant	3.562** (.683)	2.683** (.688)	2.585** (.645)
Controls
Country	0.594* (.239), .255	0.392^† (.230), .169	0.227 (.220), .098
Organiza tion size	0.067 (.231), .029	0.238 (.229), .105	0.328 (.215), .144
Functional background: Marketing	−0.594^† (.310), −.243	−0.225 (.309), −.092	−0.286 (.290), −.117
Functional background: R&D	0.001 (.265), .001	−0.024 (.250), −.010	−0.109 (.235), −.046
Functional background: Manufacturing	−1.447^† (.800), −.177	−1.156 (.765), −.143	−1.755* (.733), −.217
Age	0.011 (.013), .090	0.027* (.013), .228	0.030* (.013), .252
Independent variables
Extent of CFT use		0.441** (.154), .279	0.748** (.166), .473
Degree of authority		0.201** (.073), .256	0.228** (.069), .291
Interactions
CFT use × Degree of authority			−0.390** (.104), −.370
R ²	.184	.286	.382
n	102	101	101
F	3.560**	4.613**	6.252**

Note. The figures in the table represent regression coefficient, followed by the standard error (in parentheses), and then the standardized coefficient. CFT = cross-functional teams.

†

p < .1. *p < .05. **p < .01.

Consistent with H1, extent of CFT use is positively related to senior managers’ perceptions of the extent to which their organizations support risk taking (b = .748, p < .01).

The regression results in Table 5 also indicate that the interaction term between extent of CFT use and senior managers’ decision-making authority is negative and significant (b = −.390, p < .01). The effect size, a measure of the magnitude of the phenomenon (namely, the influence of the independent variables of interest on risk taking), is high (f² = .44), as is the power of the test (>.99 at α = .01; Cohen, 1992).

To examine whether decision-making authority negatively moderates the relation between extent of CFT use and risk-taking perceptions (H2), we plotted the interactions in Figure 1 (Aiken & West, 1991).

Figure 1.

The moderating effect of senior manager decision-making authority on the relation between extent of CFT use and perceptions of organizational support for risk taking.

The figure indicates that when senior managers have a high degree of decision-making authority, extent of CFT use has an insignificant and negligible relation with risk perceptions (slope = .037, p > .10). Extent of CFT use has a positive effect on risk perceptions when senior managers’ decision-making authority is moderate (slope = .6, p < .01) or low (slope = 1.163, p < .01). At these low to moderate levels of authority, risk perceptions become more positive as CFT use increases. These results are consistent with H2.

Discussion

Our study examines how the use of vertical and horizontal personal control systems—in particular, extent of CFT use and senior managers’ decision-making authority—interact to influence senior managers’ perceptions of the extent to which their organizations support risk taking during NPD. The results of our study find a positive relation between extent of CFT use and senior managers perceived organizational support for risk taking. It is interesting to note that this effect depends on the level of senior managers’ decision making; the extensive use of CFTs leads higher perceived organizational support for risk-taking perceptions only at low to moderate degrees of authority.

From an organization theory perspective, these findings are consistent with bureaucratic contingency theory (Donaldson, 2001), which suggests that organizations may use both horizontal and vertical control as complementary control systems to enhance performance. Our study extends this prediction to risk perceptions. Interestingly, it appears that there is a ceiling effect of risk taking for senior managers with high authority; beyond that ceiling, the extensive use of CFTs does not increase senior managers’ perceptions organizational support for risk taking. Moreover, the ceiling effect is exerted by a vertical control system—at high degrees of senior manager authority, risk-related perceptions are not only more positive but also largely independent of the extent of CFT use.

A large body of literature on upper echelons finds that senior managers’ perceptions are critical in determining strategic choices, and, therefore, organizational performance (Hambrick & Mason, 1984). Combined with studies that find that while the involvement of senior managers is critical for NPD, it is also an area sorely in need of improvement by organizations (Barczak et al., 2009), the results of our study suggest the importance of designing and balancing organizational control systems to influence managerial perceptions in ways that lead to desired organizational outcomes.

The results of this study have some practical implications. The literature on NPD indicates that risk taking is a critical driver of innovative NPD, at least in turbulent environments (Calantone et al., 2003). This study suggests that the use of CFTs alone does not guarantee greater risk taking; rather, the extent of CFT use matters. Put another way, the use of CFTs comes with a cost—that of more positive perceptions of organizational support for risk taking, at least in the minds of senior managers. The question for organizations is, does CFT use generate a sufficiently high return to compensate for these changed perceptions? Organizations may be able to use CFTs as a true “best practice” only when they are able to recognize and manage the more positive risk-related perceptions that accompany extensive CFT use by vesting senior managers with an appropriate degree of authority.

As with all research, this study has some limitations. The managers in this study represent an available sample. Like other survey research, the results of this study present a cross-sectional portrait of the relations between CFT use, degree of senior manager authority, and risk-taking perceptions, and therefore can only capture correlations, not the dynamic or causal nature of these relationships. Furthermore, we used single respondents from multiple organizations. This approach allowed us to gather data from a larger sample of firms, but may have limited the depth to which we were able to examine the relations among the variables of interest in this study. Based on the characteristics of our sample, however, we believe that our informants were able to provide relevant and accurate responses to our research questions about CFT use, degree of authority, and risk perceptions (see Huber & Power, 1985).

This study represents a first step toward developing a more detailed understanding about how horizontal and vertical control systems influence the perceptions of individuals within an organization regarding its internal climate. One avenue for future research would be to examine the mechanisms through which control systems such as CFT use influence risk perceptions at an organizational level. We also note here that though our results indicate that similar models explain managerial perceptions related to risk taking across organizations in both the countries, our data also find a difference in means of senior managers’ perceptions of organizational support for risk taking—but not in the independent variables of interest. We find this contrast interesting and point to it as an avenue for future research. Future studies in this area may also find country-specific differences relating to the extent of CFT use, centralization of decision-making authority, and risk in countries that vary more sharply in their cultural and business environments.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported in part by a faculty research grant from Northern Illinois University in 2004.

Associate Editor: William L. Gardner

Author Biographies

Devaki Rau is an Associate Professor of Management at Northern Illinois University. Her research interests include strategic decision making, top management teams, and organizational learning. Her research has been published in journals such as the Strategic Management Journal, Journal of Management, Journal of Applied Psychology, Small Group Research, and Journal of Business Research. She earned her PhD from the University of Minnesota.

Thorvald Haerem (thorvald.haerem@bi.no) is an Associate Professor at Norwegian Business School. His research interests include organizational and individual routines, decision making, and information processing. He has published research in journals such as the Journal of Applied Psychology, Organizational Studies, Organization Science, Journal of Behavioral Decision Making, and Academy of Management Review. He earned his PhD from Copenhagen Business School.

Elisa Fredericks has published on the development of new products in Industrial Marketing Management, Journal of Product Innovation Management, Journal of Qualitative Research and Journal of Nonprofit and Public Sector Marketing as well as being an active conference participant. She is an Associate Professor of Marketing at Northern Illinois University. Her research and teaching includes product development and management and cross functional integration. She earned her PhD at the University of Illinois at Chicago and has a BS and MBA from New York University.

References

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: SAGE.

Anderson

De Dreu

C. K. W.

Nijstad

B. A.

(2004). The routinization of innovation research: A constructively critical review of state-of-the-science. Journal of Organizational Behavior, 25, 147-173.

Arrow

K. J.

(1974). The limits of organization. New York, NY: W.W. Norton.

Ayers

Dahlstrom

Skinner

S. J.

(1997). An exploratory investigation of organizational antecedents to new product success. Journal of Marketing Research, 34, 107-116.

Baird

I. S.

Thomas

(1985). Toward a contingency model of strategic risk taking. Academy of Management Review, 10, 230-243.

Barczak

Griffin

Kahn

K. B.

(2009). Perspective: Trends and drivers of success in NPD practices: Results of the 2003 PDMA Best Practices Study. Journal of Product Innovation Management, 26, 3-23.

Baron

R. M.

Kenny

D. A.

(1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Bilimoria

Huse

(1997). A qualitative comparison of the boardroom experiences of U.S. and Norwegian women corporate directors. International Review of Women and Leadership, 3(2), 63-76.

Bouwman

Van Den Hooff

Van De Wijngaert

Van Dijk

(2005). Information and communication technology in organizations. London, England: SAGE.

10.

Boyacigiller

N. A.

Adler

N. J.

(1991). The parochial dinosaur: Organizational science in a global context. Academy of Management Review, 16, 262-290.

11.

Bromiley

Rau

(2010). Risk taking and strategic decision making. In Nutt

Wilson

(Eds.), Handbook of decision making (pp. 307-326). Chichester, UK: John Wiley.

12.

Bromiley

Rau

(2011). Strategic decision making. In Zedeck

(Ed.), APA handbook of industrial and organizational psychology (pp. 161-182). Washington, DC: American Psychological Association.

13.

Brown

S. L.

Eisenhardt

K. M.

(1995). Product development: Past research, present findings, and future directions. Academy of Management Review, 20, 343-378.

14.

Cabrales

A. L.

Medina

C. C.

Lavado

A. C.

Cabrera

R. V.

(2008). Managing functional diversity, risk taking, and incentives for teams to achieve radical innovations. R&D Management, 38, 35-50.

15.

Calantone

R. J.

Garcia

Droge

(2003). The effects of environmental turbulence on new product development strategy planning. Journal of Product Innovation Management, 20, 90-103.

16.

Cohen

(1992). A power primer. Psychological Bulletin, 112, 155-159.

17.

Devers

C. E.

Wiseman

R. M.

Holmes

R. M.

Jr. (2007). The effects of endowment and loss aversion in managerial stock option valuation. Academy of Management Journal, 50, 191-208.

18.

Donaldson

(2001). The contingency theory of organizations. Thousand Oaks, CA: SAGE.

19.

Esping-Andersen

(1990). The three worlds of welfare capitalism. Cambridge, UK: Polity Press.

20.

Feldman

M. S.

Pentland

B. T.

(2003). Reconceptualizing organizational routines as a source of flexibility and change. Administrative Science Quarterly, 48, 94-118.

21.

Griffin

(1997). PDMA research on new product development practices: Updating trends and benchmarking best practices. Journal of Product Innovation Management, 14, 429-458.

22.

Gulick

Urwick

(Eds.). (1937). Papers on the science of administration. New York, NY: Institute of Public Administration.

23.

Hair

J. F.

Ringle

C. M.

Sarstedt

(2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing Theory and Practice, 19, 139-151.

24.

Hambrick

D. C.

Mason

P. A.

(1984). Upper echelons: The organization as a reflection of its top managers. Academy of Management Review, 9, 193-206.

25.

Mittal

(2007). The effect of decision risk and project stage on escalation of commitment. Organizational Behavior and Human Decision Processes, 103, 225-237.

26.

Hill

C. W. L.

Jones

G. R.

(2013). Strategic management: An integrated approach. Mason, OH: South-Western Cengage Learning.

27.

Hofstede

(1980). Culture’s consequences: International differences in work related values. Newbury Park, CA: SAGE.

28.

Huber

Power

(1985). Retrospective reports of strategic level managers: Guidelines for increasing accuracy. Strategic Management Journal, 6, 171-180.

29.

Jensen

Meckling

(1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. The Journal of Financial Economics, 3, 305-360.

30.

Keasey

Moon

(1996). Gambling with the house money in capital expenditure decisions: An experimental analysis. Economics Letters, 50, 105-110.

31.

Kogan

Wallach

M. A.

(1964). Risk taking: A study in cognition and personality. New York, NY: Holt, Rinehart, and Winston.

32.

Levitt

March

J. G.

(1988). Organizational learning. Annual Review of Sociology, 14, 319-340.

33.

Maltz

Kohli

A. K.

(2000). Reducing marketing’s conflict with other functions: The differential effects of integrating mechanisms. Journal of the Academy of Marketing Science, 28, 479-492.

34.

March

J. G.

(1994). A primer on decision making: How decisions happen. New York, NY: Free Press.

35.

March

J. G.

Shapira

(1987). Managerial perspectives on risk and risk taking. Management Science, 33, 1404-1418.

36.

McDonough

E. F.

III. (2000). Investigation of the factors contributing to the success of cross-functional teams. Journal of Product Innovation Management, 17, 221-235.

37.

Miller

K. D.

Leiblein

M. J.

(1996). Corporate risk-return relations: Returns variability versus downside risk. Academy of Management Journal, 39, 91-122.

38.

Mintzberg

(1979). The structuring of organizations. Englewood Cliffs, NJ: Prentice Hall.

39.

Mishina

Pollock

Bragaw

N. A.

(2012, January). Know when to hold ’em, know when to fold ’em? Differentiating between loss aversion and the house money effect in CEO reactions to firm performance. Academy of Management Proceedings (Vol. 2012, No. 1, pp. 1-1). doi:10.5465/AMBPP.2012.78

40.

Mullins

J. W.

Forlani

Walker

O. C.

(1999). Effects of organizational and decision-maker factors on new product risk taking. Journal of Product Innovation Management, 16, 282-294.

41.

Neilson

W. S.

(1998). Reference wealth effects in sequential choice. Journal of Risk and Uncertainty, 17, 27-47.

42.

Neter

Kutner

M. H.

Nachtsheim

C. J.

Wasserman

(1996). Applied linear statistical models (4th ed.). Chicago, IL: Richard D Irwin.

43.

Perrow

(1967). A framework for the comparative analysis of organizations. American Sociological Review, 32, 194-208.

44.

Peter

J. P.

(1981). Construct validity: A review of basic issues and marketing practices. Journal of Marketing Research, 18, 133-145.

45.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J. Y.

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903.

46.

Podsakoff

P. M.

Organ

D. W.

(1986). Self-reports in organizational research: Problems and prospects. Journal of Management, 12, 531-544.

47.

Salomo

Weise

Gemunden

H. G.

(2007). NPD planning activities and innovation performance: The mediating role of process management and the moderating effect of product innovativeness. Journal of Product Innovation Management, 24, 285-302.

48.

Sarin

Mahajan

(2001). The effect of reward structure on the performance of cross-functional product development teams. Journal of Marketing, 65, 35-53.

49.

Sarstedt

Ringle

C. M.

(2010). Treating unobserved heterogeneity in PLS path modeling: A comparison of FIMIX-PLS with different data analysis strategies. Journal of Applied Statistics, 37, 1299-1318.

50.

Scott

R. W.

(2002). Organizations: Rational, natural, and open systems (5th ed.). Upper Saddle River, NJ: Prentice Hall.

51.

Scott

R. W.

Davis

G. F.

(2007). Organizations and organizing: Rational, natural, and open system perspectives. Upper Saddle River, NJ: Pearson-Prentice Hall.

52.

Sethi

Smith

D. C.

Park

C. W.

(2001). Cross-functional product development teams, creativity, and the innovativeness of new consumer products. Journal of Marketing Research, 38, 73-85.

53.

Sitkin

S. B.

Weingart

L. R.

(1995). Determinants of risky decision-making behavior: A test of the mediating role of risk perceptions and propensity. Academy of Management Journal, 38, 1573-1592.

54.

Slater

S. F.

Narver

J. C.

(1995). Market orientation and the learning organization. Journal of Marketing, 59, 63-74.

55.

Sornes

J.-O.

Stephens

K. K.

Saetre

A. S.

Browning

L. D.

(2004). The reflexivity between ICTs and business culture: Apply Hofstede’s theory to compare Norway and the United States. Informing Science Journal, 8, 123-142.

56.

Steenkamp

J.-B. E. M.

Baumgartner

(1998). Assessing measurement invariance in cross-national consumer research. Journal of Consumer Research, 25, 78-90.

57.

Thaler

R. H.

Johnson

E. J.

(1990). Gambling with the house money and trying to break even: The effects of prior outcomes on risky choice. Management Science, 36, 643-660.

58.

Thompson

(1967). Organizations in action. New York, NY: McGraw-Hill.

59.

Van de Ven

A. H.

Delbecq

A. L.

(1974). A task contingent model of work-unit structure. Administrative Science Quarterly, 19, 183-197.

60.

Van de Ven

A. H.

Delbecq

A. L.

Koenig

(1976). Determinants of coordination modes within organizations. American Sociological Review, 41, 322-338.

61.

Voss

G. B.

Sirdeshmukh

Voss

Z. G.

(2008). The effects of slack resources and environmental threat on product exploration and exploitation. Academy of Management Journal, 51, 147-164.

62.

Wezel

F. C.

Saka-Helmhout

(2006). Antecedents and consequences of organizational change: “Institutionalizing” the behavioral theory of the firm. Organization Studies, 27, 265-286.