9.2.25 .2 .. (in. ‘ .v: A 3 r5! .- VI. \ \Y. Mfiéwh ,. . $9 2... 5 621—? ..r 31.. 3.1.3.. KI» ~¢ w 0.3, 5 i 3.32.? .111} .45. v. 3 ’ e 1 , .3. . Nunez? 5%... ‘3 if». ouar 5 Z... . ._ E: 39$ 5... a .2521: .3! 3 . 2 . an 1 THESlS & (1&7 LIBRARY l Michigan State 2 University J This is to certify that the dissertation entitled TWO ESSAYS ON TOP- AND BOTTOM-LINE EFFECTS OF TEAM USE IN NEW PRODUCT DEVELOPMENT presented by SERDAR SALIH DURMUSOGLU has been accepted towards fulfillment of the requirements for the Ph.D. degree in Marketing and Supply Chain Management Mefibr Ptofessor’s Signature / 8 W 300?“ U U Date MSU is an affirmative-action, equal-opportunity employer .-n-l-l-I-O-I-l-Q-l-I-I-‘-o-a-t-o--. PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 6/07 p:/ClRC/Date0ue.indd-p.1 TWO ESSAYS ON TOP- AND BOTTOM-LINE EFFECTS OF TEAM USE IN NEW PRODUCT DEVELOPMENT By Serdar Salih Durmusoglu A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Marketing and Supply Chain Management 2007 ABSTRACT TWO ESSAYS ON TOP- AND BOTTOM-LINE EFFECTS OF TEAM USE IN NEW PRODUCT DEVELOPMENT By Serdar Salih Durmusoglu The primary objective of this dissertation was to investigate and subsequently to advance the understanding of new product development (NPD) team use in organizations. To achieve this objective, two studies related to NPD team use were conducted. While the first study provided a meta-analytic review of the extant literature, the second took a more in depth look into how NPD team use and other organizational factors such as a cross-functionality of the team, team leader’s existence and accountability in all NPD phases, and inherent degree of product innovativeness can contribute to the success of the implementation of a growth strategy with new product introductions. Two methods were used; a meta analysis followed by path analysis, and longitudinal survey study. The data collected in the longitudinal study was analyzed by regression and multivariate analysis of variance. Many insights are gleaned from the meta-analytic review, including the gaps in the current body of knowledge as well as what the accumulated knowledge reveals on the effects of several antecedents of NPD team performance. In the second study, the differential effects of proficient NPD task execution on market-based and process-based success measures were identified. Moreover, the interaction of these task execution proficiencies with NPD team cross-functionality, team leader existence and accountability, and inherent degree of product innovativeness are related to outcomes. Finally, the changes in the proficiency of NPD phase task execution over time is observed in an effort to shed some light whether learning by doing in successive executions of tasks improve NPD outcomes. Copyright by SERDAR SALIH DURMUSOGLU 2007 ACKNOWLEDGEMENTS I would like to start with thanking Dr. Roger J. Calantone, my dissertation committee chair and advisor. I am very appreciative to him as his guidance ’ assisted me considerably all throughout the Ph.D. program. I am also thankful that he put me on a research project with Dr. Gina C. McNally a few years ago. We worked on various research projects and this dissertation surely benefited much from her input. I am glad that she was on my dissertation committee. I would like to express special thanks also to my other committee members, Dr. Cheri Speier, for her constructive suggestions on my dissertation, and to Dr. V. Sambamurthy, for inspiring me to maintain a high standard in all my research endeavors, including in this dissertation. I would also like to send my gratitude to Steve. Some of the data in this dissertation would not be complete if he were to avoid my numerous calls and e- mails last few years. He would answer my calls even when he was on his way to the airport or recovering from a cold at home. The administrative assistants in the Department of Marketing and Supply Chain Management have always provided outstanding support whenever I needed something. I extend my thanks to all of them. I would also like to thank two of my fellow doctoral students; Mark, for being a person I could talk to and Wesley, for being an understanding office mate. Dr. Clay Voorhees motivated me to work harder and keep on being great. His advice was also invaluable during my job search process. I am grateful to him. My dear friends Anne-Laure, Kurt, and Erica have proven, over and over again, their sincere interest in my aspirations of becoming a scholar. Friends like you were all I needed. Finally, I would like to thank Gokce, my wife. She and l, two doctoral students, have been through a lot in the past couple of years. Thank you for being by my side and making the graduate student life tolerable. I couldn’t ask for more. Among other wonderful things, you are resilient and smart, understanding and caring, calm and hard working. Thank you for all you have done for me. I am happy to be sharing my life with you. vi TABLE OF CONTENTS LIST OF TABLES ............................................................................................... ixi LIST OF FIGURES .............................................................................................. xi CHAPTER 1 INTRODUCTION ................................................................................................... 1 1.1. BACKGROUND: CHARACTERISTICS OF NPD TEAMS ....................... 1 1.2. OVERVIEW OF THE REMAINING CHAPTERS ..................................... 8 CHAPTER 2 ANTECEDENTS OF NPD TEAM PERFORMANCE: A META-ANALYTIC REVIEW .............................................................................................................. 10 2.1 INTRODUCTION ................................................................................... 10 2.2 A FRAMEWORK FOR ANALYZING NPD TEAM PERFORMANCE ....... 12 2.3 METHOD: META ANALYSIS .................................................................. 20 2.3.1 Database Development ................................................................... 21 2.3.2 Study Context and Data Characteristics ......................................... 24 2.3.3 Study Design Characteristics .......................................................... 30 2.3.4 Dependent Variable: NPD Team Performance ............................... 32 2.4 ANALYIS AND RESULTS ...................................................................... 33 2.4.1 Meta-Analytic Analysis Procedures ................................................ 33 2.4.2 Correlation Effect Sizes and Their Interpretations .......................... 34 2.4.3 The Effect of Stage Setters and Cohesion on NPD Team Performance: A Path Analysis ................................................................. 42 2.5 CONCLUSIONS ...................................................................................... 44 CHAPTER 3 A LONGITUDINAL STUDY OF ORGANIZATIONAL CHANGE BY IMPLEMENTATION OF NEW PRODUCT-DRIVEN GROWTH STRATEGY: PROFICIENCY OF PERFORMING NPD PROCESS AND THE MODERATING EFFECT OF CROSS-FUNCTIONAL TEAM USE ............. 48 3.1 INTRODUCTION .................................................................................... 48 3.2 THEORETICAL BACKGROUND ........................................................... 53 3.2.1 Organization Development and Change .......................................... 53 3.2.2 Strategic Planning and Goal Setting: An Organizational Change Theory in Teleological School of Thought .................................. 54 3.2.3 New Product Strategy Implementation ............................................ 55 3.2.4 NPD Team Performance ................................................................. 56 vii 3.3 HYPOTHESES ...................................................................................... 58 3.3.1 Proficiency of NPD Phase Activities on NPD Team Performance ............................................................................................. 58 3.3.2 The Moderating Role of Perceived Cross-Functionality in Organizing for NPD .................................................................................. 61 3.3.3 The Moderating Role of the Project Leader ..................................... 64 3.3.4 The Moderating Role of Inherent Product lnnovativeness ............... 65 3.3.5 Improvement in NPD Task Execution Proficiencies over Time: Team Learning by Doing .......................................................................... 66 3.4 METHOD ............................................................................................... 68 3.4.1 Research Site: Organizational Change at Alpha Corporation .......... 68 3.4.2 Sample and Data Collection Procedures ......................................... 71 3.4.3 Respondents and Demographics .................................................... 73 3.4.4 Measures ......................................................................................... 77 3.4.4.1 NPD Team Performance ................................................... 77 3.4.4.2 Antecedents of NPD Team Performance ........................... 79 3.4.4.3 Moderators ......................................................................... 80 3.5 ANALYSIS AND RESULTS ................................................................... 81 3.5.1 Seemingly Unrelated Regression (SURE) ....................................... 81 3.5.2 One-Way Mutlivariate Analysis of Variance (MANOVA) .................. 87 3.6 CONCLUSIONS ..................................................................................... 93 CHAPTER 4 CONTRIBUTIONS AND FUTURE RESEARCH DIRECTIONS .......................... 96 4.1 INTRODUCTION .................................................................................... 96 4.2 STUDY #1: ANTECEDENTS OF NPD TEAM PERFORMANCE: A META-ANALYTIC REVIEW ...................................................................... 97 4.2.1 Theoretical Contributions ................................................................. 97 4.2.2 Managerial Contributions ................................................................. 99 4.2.3 Future Research Directions ........................................................... 100 4.3 STUDY #2: A LONGITUDINAL STUDY OF ORGANIZATIONAL CHANGE BY IMPLEMENTATION OF NEW PRODUCT-DRIVEN GROWTH STRATEGY: PROFICIENCY OF PERFORMING NPD PROCESS AND THE MODERATING EFFECT OF CROSS- FUNCTIONAL TEAM USE .......................................................................... 100 4.3.1 Theoretical Contributions ............................................................... 100 4.3.2 Managerial Contributions ............................................................... 101 4.3.3 Future Research Directions ........................................................... 102 APPENDIX ........................................................................................................ 103 CODING SHEET FOR STUDIES ................................................................ 104 REFERENCES .................................................................................................. 106 viii LIST OF TABLES TABLE 1-1 OVERVIEW OF CHAPTERS ............................................................... 2 TABLE 1-2 CHARACTERISTICS OF NPD TEAMS BASED ON CONTEXTUAL DIMENSIONS ....................................................................... 6 TABLE 2-1 STUDIES INCLUDED IN THE META-ANALYTIC REVIEW ............... 25 TABLE 2-2 STUDY CONTEXT AND DATA CHARACTERISTICS OF EMPIRICAL STUDIES ................................................................................. 29 TABLE 2-3 DESIGN CHARACTERISTICS OF EMPIRICAL STUDIES ................ 31 TABLE 2-4 TOTAL NUMBER OF CORRELATIONS REPORTED BETWEEN NPD TEAM PERFORMANCE AND ITS PREDICTORS ............ 35 TABLE 2-5 SUMMARY OF EFFECT SIZES BETWEEN ANTECEDENTS AND OVERALL NPD TEAM PERFORMANCE ............................................ 38 TABLE 2-6 SUMMARY OF EFFECT SIZES BETWEEN ANTECEDENTS AND MARKET-BASED NPD TEAM PERFORMANCE (EFFECTIVENESS) ................................................................................... 39 TABLE 2-7 SUMMARY OF EFFECT SIZES BETWEEN ANTECEDENTS AND PROCESS-BASED NPD TEAM PERFORMANCE (EFFICIENCY) ............................................................................................. 41 TABLE 2-8 MEANS, STANDARD DEVIATIONS, AND CORRELATIONS FOR THE VARIABLES IN THE PATH MODEL ............................................. 43 TABLE 2-9 SUMMARY OF CORRECTED CORRELATION EFFECT SIZES ON NPD TEAM PERFORMANCE DIMENSIONS ............................. 46 TABLE 3-1 RESPONDENT PARTICIPATION IN DATA COLLECTION ROUNDS ..................................................................................................... 74 TABLE 32 DESCRIPTIVE STATISTICS FOR RESPONDENT TENURE AND EXPERIENCE (IN YEARS) .................................................................. 75 TABLE 3-3 FUNCTIONAL REPRESENTATION OF THE RESPONDENTS ........ 77 TABLE 3-4 DESCRIPTIVE STATISTICS AND CORRELATIONS BETWEEN VARIABLES .............................................................................. 82 TABLE 3-5 SURE REGRESSION RESULTS ...................................................... 84 TABLE 3-6 SUMMARY OF HYPOTHESES TESTING ........................................ 88 TABLE 3-7 MEAN RATINGS OF NPD PHASE EXECUTION PROFICIENCIES ACROSS ROUNDS ......................................................... 90 TABLE 3-8 BOX’S TEST FOR NPD PHASE EXECUTION PROFICIENCIES ACROSS ROUNDS ......................................................... 92 TABLE 3-9 MANOVA RESULTS FOR NPD PHASE EXECUTION PROFICIENCIES ACROSS ROUNDS ......................................................... 92 LIST OF FIGURES FIGURE 2-1 THE MODEL FOR ANALYZING TEAM EFFECTIVENESS SUGGESTED BY COHEN AND BAILEY (1997) .......................................... 15 FIGURE 2-2 THE MODEL FOR ANALYZING NPD TEAM EFFECTIVENESS SUGGESTED BY MCDONOUGH (2000) ....................... 17 FIGURE 2-3 GUIDING THEORETICAL MODEL FOR META-ANALYSIS ON NPD TEAM PERFORMANCE ................................................................ 18 FIGURE 2-4 ANTECEDENTS OF NPD TEAM PERFORMANCE: PATH MODEL ........................................................................................................ 43 FIGURE 2-5 SIGNIFICANT PATH COEFFICIENTS BETWEEN THE ANTECEDENTS, COHESION, AND NPD TEAM PERFORMANCE ............ 45 FIGURE 3-1 THE EFFECT OF NPD TASK EXECUTION PROFICIENCIES AND MODERATORS ON NPD TEAM EFFECTIVENESS AND EFFICIENCY ................................................................................................ 59 FIGURE 3-2 MEAN PROFICIENCY OF TASK EXECUTION RATING FOR EACH NPD PHASE ............................................................................. 91 xi CHAPTER 1 INTRODUCTION This dissertation consists of four chapters. Chapter 1 is presents an overview of the dissertation chapters, provides a brief background of the main topic of the dissertation new product development (NPD) teams and the research objectives. Chapter 2 and Chapter 3 are two studies that deal with a unique aspect of NPD team performance. The last chapter, Chapter 4, delineates on the theoretical and managerial contributions of the study and provides suggestions for future research. Table 1-1 provides an overview of the chapters, their topics, and research objectives. 1.1 BACKGROUND: CHARACTERISTICS OF NPD TEAMS Until the mid 19703, the three key generic approaches to coordination in organizations were standardization or rules, plans and schedules, and mutual adjustment (March and Simon 1958; Thompson 1967). The more recent approach is the use of teams in organizations (Van de Ven, Delbecq, and Koenig 1976). Similar to Hackman’s (1987) definition, and building on early work of Alderfer (1977), Cohen and Bailey (1997, p. 241) define organizational teams as “a collection of individuals who are interdependent in their tasks, who share responsibility for outcomes, who see themselves and who are seen by others as an intact social entity embedded in one or more larger social systems such as the business unit or corporation, and who manage their relationships across organizational boundaries.” Table 1-1 Overview of Chapters Chapter Topics Research Objectives 1 Introduction 2 A meta-analytic review of Develop a theoretical framework for literature on NPD teams 3 A longitudinal study of formalized NPD process implementation for product- driven growth 4 Contributions and future research directions investigating NPD team performance; conduct a meta- analytic review of the literature to identify knowledge gaps and resolve conficlting extant knowledge Investigate the effect of use of cross-functional teams and proficient execution of NPD process tasks on NPD success; identify the order of improvement in NPD task execufion Provide a summary of both theoretical and managerial contributions of the dissertation; suggest future research topics Literature provides numerous typologies of organizational teams. For example, Sundstrom, DeMeuse, and Futrell (1990) classify four types of teams: advice and involvement teams, production and service teams, project and development teams, and action and negotiation teams. On the other hand, Cohen and Bailey (1997) differentiate between work teams, parallel teams, project teams, and management teams. Despite different labels, Sundstrom and colleagues’ (1990) and Cohen and Bailey’s (1997) typologies mostly overlap. Based on the descriptions that these authors provide, it can be concluded that advice and involvement teams correspond to parallel teams; production and service teams correspond to work teams; and project and development teams correspond to project teams. While Sundstrom, DeMeuse, and Futrell (1990) include a category for action and negotiation teams; Cohen and Bailey (1997) ignore these teams, but create a category for management teams. Although there are large overlaps between different typologies of organizational teams, many of them either lack some types of categories or bundle several similar team types into the same category. A common example is classifying NPD teams and other project teams in the same category. Although project teams and NPD teams share certain characteristics such as limited duration and production of one-time outputs, the type of Skills required from an enterprise resource planning (ERP) software implementation team would differ considerably from what is needed from NPD teams. An ERP software attempts to integrate all or several departments and functions across a company onto a single computer system that can serve all those departments’ particular needs. On the other hand, the goal of an NPD team is to create a new product that has the highest potential to succeed in the market. As such, ERP software implementation project team members would need to understand the firm’s processes and routines in great detail, but do not need to know much about the firm’s new product strategy. To address the need for a widely accepted taxonomy of organizational teams, Devine (2002) provides a review of existing classification systems and derives a taxonomy consisting of fourteen team types based on seven underlying contextual dimensions. These contextual dimensions are fundamental work cycle, physical and cognitive ability requirement, temporal duration, task structure, active resistance, hardware dependence, and finally, health risk. Fundamental work cycle refers to the smallest meaningful collective activity for the team. This smallest collective activity is determined by its output, which corresponds to the most micro unit by which team performance can be evaluated. The second dimension, ability requirement, is the level of physical activity required to accomplish goals as opposed to cognitive ability requirements such as information acquisition and processing. This dimension makes the distinction between teams that produce informational outputs such as plans, decisions and choices, and the teams where members pull, push, and carry objects around in order to get the job done. Temporal duration, the third contextual dimension of the team taxonomy, refers to the length of time for which the team exists. The main distinction is between the teams that work for pre-determined amount of time and teams that disband when they accomplish their goals. An example of a team that disbands is a union contract negotiation team that ceases to exist when a contract is signed. While both types of teams exist for a finite amount of time, the former work during the pre-determined period called shifts, which can usually be measured by hours. On the other hand, the latter type of teams will exist for longer periods of time, which can range from hours to years. The fourth dimension, task structure, is the level of agreement among team members about what the group is supposed to accomplish. For tasks that have low task structure, members might have Unclear and conflicting ideas of what actions to take in order to achieve their goals. Conversely, for teams with unstructured tasks, ambiguity exists as to which and how the inputs, such as resources and knowledge, should be combined to generate the outcomes. The fifth dimension, active resistance, refers to the existence of a competing team, or an individual directly trying to prevent the team from accomplishing its goals. Teams facing resistance must overcome the opposition in order to be effective. Hardware dependence, the sixth contextual dimension, is the reliance of a team on technology to perform the activities. This dimension has to do with how much the team is constrained by communication systems, machines, tools or other specialized equipment when working to achieve the desired collective outcome. Finally, health risk is the likelihood of human injury when the team is in action. Devine (2002) explicates three levels of health risk: low, moderate and high. At low health risk levels, there is no likelihood of death; at moderate levels, the likelihood of death is confined to team members only; and at high health risk levels, the loss of human life is possible for both team members and other individuals such as customers or citizens. In a broad sense, NPD teams are a type of project team due to their limited duration and task of producing one-time outputs (Cohen and Baliey 1997; Mankin, Cohen, and Bikson 1996). Moreover, based on Devine’s (2002) recent and detailed taxonomy, these teams can be classified as design teams. Table 1-2 presents the characteristics of NPD teams along the seven contextual dimensions of team typology developed by Devine (2002). Table 1-2 Characteristics of NPD Teams based on Contextual Dimensions Contextual Dimensions Fundamental work cycle Ability requirement (physical and cognitive) Temporal duration Task structure Active resistance Hardware dependence Health risk Concept, model, prototype, product Low on physical, high on cognitive abilities Mission; definite end with product launch Varies depending on the inherent degree of innovativeness of the new product Competition for senior manager attention and firm’s limited resources; might also face resistance from senior managers or other departments Moderate: design tools, communication technologies to facilitate activities; increases as the team gets more geographically dispersed Low for team members; fatality may be caused by some new products such as drugs Concepts, prototypes and products are the smallest unit of outputs that an NPD team produces and can be evaluated against. Hence, these outputs determine the fundamental work cycle of NPD teams. In terms of abilities, NPD teams do not require physical abilities, such as pushing or carrying large and heavy objects, but their tasks are demanding for cognitive abilities due to coordination, information processing and technology use requirements. NPD teams cease to exist when the project is over, that is the new product is launched or, in some occasions, when the project is terminated. Hence, their temporal duration is dependent on the accomplishment of their task of developing and subsequently launching a new product. The degree of task structure ranges from low to high in NPD teams. Really new products require more creativity, which in turn might result in following a different and less structured process for developing them. On the other hand, incrementally new products are handled by following more routine and clearly identified steps in each phase of NPD. Although there is rarely another competing team that competes for developing a similar product in a firm, NPD teams still would compete for a firm’s limited resources. Therefore, a concept that has higher potential to generate value and better satisfy customers could facilitate an NPD team’s activities since senior managers would give more importance to this team’s needs as compared to other teams working on other NPD projects. Consequently, there is some competition for senior manager support and the firm’s scarce resources such as human resources or capital. Moreover, NPD teams can encounter active resistance from a senior manager due to incongruence in perceptions or goals. For example, vice president of operations might prefer incremental new products because existing product platforms can be used to produce them. However, a really new product would highly likely necessitate new production processes, which would require time to get the desired operating efficiencies. NPD team members are moderately dependent on hardware. Prototyping and project management tools are among the tools they use to perform their tasks successfully. Moreover, information and communication tools such as video conferencing equipment or shared workspaces would be a necessity for coordinating activities if team members are not co-located. Finally, while the risk of any NPD team member getting injured during development is quite low, the loss of human life is a definite possibility for certain new products such as a pharmaceutical drug that demonstrates unexpected adverse side effects after its release in the market; or a new model car catching fire because of a design failure of placing an easily flammable type of wiring near the engine. Most of these taxonomies have ‘task type’ as the common denominator when dividing the teams into groups. Differences and similarities in task types utilized seem to generate the differences and overlaps between the taxonomies. Moreover, no matter which taxonomy is adopted in classifying NPD teams, in essence, these teams are unique in many ways. Take McGrath’s (1984) typology, a commonly referred grouping, is developed around four general task categories: generating of ideas and plans, selecting between alternatives, negotiating conflicts of interest, and executing work. NPD teams perform all of these tasks and therefore are different than many other organizational teams. And since task type moderates the relationship between organizational teams and their performance (Stewart and Barrick 2000), separate studies of NPD team performance are frequently conducted. 1.2 OVERVIEW OF THE REMAINING CHAPTERS To contribute to the advancement of knowledge on NPD teams, this dissertation addresses broad theoretical concepts related to NPD. The first study, presented in the next chapter, is a meta-analytic review of NPD team literature. Although NPD teams have risen in prominence since the 19803 and the majority of firms use them for NPD, there is limited consensus in the literature as to what constitutes the nomological network of NPD team performance. Therefore, the main objective of this first study is to integrate literature on work group effectiveness and NPD teams and develop a theoretical framework for investigating NPD teams. Furthermore, this study presents a review of the literature and subsequently identifies some patterns and knowledge gaps on NPD team performance. By synthesizing the current literature on NPD teams, the first study contributes to managerial practice by enabling managers to get a more comprehensive understanding of the relevant and important factors for improving team productivity, subsequently NPD success. The second study, the topic of Chapter 3, examines the relationship between proficiency of discovery, development and commercialization phase activities on new product sales, ROI, new product quality and NPD team efficiency outcomes of budget and schedule adherence. Moreover, the moderating role of cross functional team use, team leader accountability and inherent level of new product innovativeness between the phase execution proficiencies and NPD team performance outcomes are investigated. The final chapter, Chapter 4, summarizes the both theoretical and managerial contributions and concludes with suggestions for future research. CHAPTER 2 ANTECEDENTS OF NPD TEAM PERFORMANCE: A META-ANALYTIC REVIEW 2.1 INTRODUCTION New product development (NPD) teams transform a set of inputs such as customer needs and technical and manufacturing capacity of the firm into a set of outputs such as product design, product platform, and promotions. NPD teams are mostly cross-functional, which refers to a team that consists of core members from different functions and that these core members remain in place, are stable and actively participate from the early stages of product development through product introduction (Griffin 1997a). Scholars have been investigating antecedents of NPD team performance for almost two decades. However, these studies find many conflicting results on the effect of antecedents on NPD team performance. For example, formalized NPD process use is a variable that is examined in many studies on NPD teams. In studies investigating the effect of formal NPD processes on market-based performance figures such as new product sales, the impact is found to be either very low and insignificant (e.g., Faraj and Sproull 2000; Atuahene-Gima and Evangelista 2000) or significant and positive (e.g., Moenaert et al. 1994; Lynn, Skov, and Abel 1999). When NPD team performance is measured by process- related variables such as budget or schedule adherence, the results still vary considerably from very low and insignificant (e.g., Faraj and Sproull 2000) to medium size effects (e.g., Sarin and McDermott 2001 ). As such, a quantitative 10 synthesis of the literature would therefore provide a big step in clarifying the nature of the relationships. Another issue that needs resolution is whether cross-functional team use enhances NPD outcomes. Use of cross-functional teams in NPD activities has been consistently increasing in many industries since mid seventies (Henke, Krachenberg, and Lyons 1993; Page 1993). However, as McDonough (2000) states, firms are still struggling to reap the benefits expected from cross- functional teams. This is clearly reflected in the studies that examined the effect of cross-functional team use in NPD teams. For example, while Keller (2001) finds a significant positive impact, many other studies (e.g., Sarin and McDermott 2001, Lovelace, Shapiro, and Weingart 2001, Carbonell and Rodriguez 2006, and Ancona and Caldwell 1992c) find that use of NPD teams hinder performance outcomes such as adherence to budgets and adherence to schedules. This discrepancy is observed in studies that utilize a composite, overall measure of NPD team performance. Ancona and Caldwell’s (1992b) and Pelled, Eisenhardt, and Xin’s (1999) studies show a negative influence of cross-functionality on performance. On the other hand, Stock (2006) and Lee and Chen (2007) both find a significant positive effect. In conclusion, an empirical synthesis of the current literature on NPD teams would serve as a significant step in determining the state of knowledge on this phenomenon. This meta-analysis of studies on NPD teams contributes to literature by facilitating future research in two ways. It facilitates future research by identifying the structural variables that are prominent in extant studies and by identifying 11 their effect sizes in predicting NPD team performance. The results foster not only both theoretical development and suggest future research routes, but also provide guidance to managers that aspire to improve NPD team effectiveness and performance by providing a more comprehensive understanding of the relevant and important factors. The remainder of this essay is organized as follows. First, an overview of recent conceptual models that have been proposed for analyzing organization and NPD team performance is provided. Next, these models are integrated into a comprehensive framework for analyzing NPD team effectiveness. Then, the nature, purpose and steps followed in this meta-analysis are described. After presenting collection of relevant studies in building the database, the criteria for inclusion of studies in the analysis are explained. The chapter concludes with intended contributions to theory and managerial practice. 2.2 A FRAMEWORK FOR ANALYZING NPD TEAM PERFORMANCE Consistent with frameworks analyzing team performance based on the dominant “input-process-output” (I-P-O) approaches (cf. McGrath 1984; Gladstein 1984; Hackman 1987), Denison, Hart, and Kahn (1996) provide a framework for studying and assessing the effectiveness of cross-functional teams in firms. Their three-domain model encompasses organizational context, internal process and outcome measures. Factors such as coordination with other teams and functional departments, autonomy and power, problems and resolutions with other departments, resources, mission and direction, and rewards for performance are factors in the organizational context domain. On the other 12 hand, norms, importance of team’s work, effort given by team members, creativity, and democracy on member viewpoints are among the factors to consider in the team processes domain. Finally, outcomes can be grouped into two subdomains of organizational outcomes and member outcomes. While success of the product is an organizational outcome, factors such as learning, satisfaction with personal growth are among the individual team member outcomes. While earlier frameworks analyzed team performance based on IPO approaches, recent conceptualizations of analyzing organization teams have voiced concerns that I-P-O framework “constrains thinking about teams” (llgen et al. 2005, p. 520). Recent conceptualizations, therefore, augmented the l-P-O models of teams by adding mediator variables and cyclical feedback loops in the relationships between team-related variables (of. Keller 2001). Further, Ilgen et al. (2005) propose an “input—mediator-output-input” (IMOI) framework for analyzing teams in organizations. In this framework, they explain their reasoning for replacing “process” with “mediator” by arguing that not all explanatory variables identified in the literature can be processes and that some constructs used by researchers, who are trying to invoke the l-P-O framework, as process are actually emergent affective or cognitive states. Marks, Mathieu, and Zaccaro (2001) present ‘cohesion’, an emergent state, as an example that researchers use in l-P-O models. They note that teams with low cohesion may be less willing to manage existing conflict, which is a process. The distinction between emergent states and processes is important because emergent states do not 13 represent team actions that lead toward outcomes (Marks, Mathieu, and Zaccaro 2001). Next, the “I” at the end of the IMOI framework highlights the cyclical feedback inherent between the variables in the nomological net of organization teams. And finally, elimination of the hyphen between the letters acknowledges the fact that some of the relationships between the variables may be nonlinear or conditional, thus encouraging scholars to consider moderating variables when examining organization team effectiveness. Based on a qualitative review of the literature, Cohen and Bailey (1997) develop a detailed framework, with mediating and nonrecursive relationships between sets of variables. Cohen and Bailey’s (1997) qualitative review includes four types of organization teams: work, parallel, project and management. Their findings suggest that team effectiveness is influenced by environment, design factors, internal and external processes and psychological traits. Based on their findings, they develop a heuristic model for investigating team effectiveness, which is depicted in Figure 2-1. Cohen and Bailey’s (1997) integrative review of the literature concludes that determinants of team effectiveness varies depending on the type of team. These results have justified the earlier concerns brought about by Goodman Ravlin, and (1987), who argued that a model for explaining one type of teams (e.g., sales teams) would only be generalizable to all other types of teams (e.g., management teams or sports teams) at some level of abstraction, but a more granular examination is needed to advance theory on team effectiveness. 14 Figure 2-1 The Model for Analyzing Team Effectiveness Suggested by Cohen and Bailey (1997) Task Design e.g., autonomy, interdependence Group Composition e.g., size, tenure Organizational Context e.g., rewards, supervision Internal Processes e.g., conflict, communication External Processes e.g., conflict, communication Effectiveness A Environmental Factors e.g., turbulence, industry characteristics Group Psychosocial Traits e.g., norms, shared mental models, beliefs Performance Outcomes e.g., quality, productivity Attitudinal Outcomes e.g., job satisfaction, trust Behavioral Outcomes e.g., turnover absenteeism Following Cohen and Bailey’s (1997) work, Stewart (2006) conducts a meta-analysis of the literature on a specific set of organization team variables, namely, design factors. The three sets of team design factors reviewed are composition, task design, and context and leadership. Compared to Cohen and Bailey ( 1997), Stewart’s (2006) review is more limited in scope because it focuses on studies that examine only project and management teams. In his meta-analysis, Stewart (2006) finds that aggregated measures of individual abilities of team members and dispositions correlate positively with team performance. On the other hand, team member heterogeneity and performance correlate near zero, but this effect is slightly positive for project 15 teams. Moreover, performance slightly increases when project teams include more members. Further, the relationship between team autonomy and performance is positive. However, this relationship is weaker for teams that do knowledge work (e.g., NPD teams) than for teams that do physical work. In addition, intra-team coordination leads to higher performance in teams that require high cognitive skills than in teams that require physical abilities. Finally, the relationship between the leadership that a team receives from its designated leader and performance is found to be significantly positive. In the NPD teams’ context, McDonough (2000) takes a step in developing a framework for analyzing team effectiveness, and based on interviews with professionals and a review of literature, proposes a general framework for investigating cross-functional team success (See Figure 2-2.). In this model, stage setter variables such as project goals, empowerment and human resources influence cross-functional team performance through the mediation of team behaviors such as cooperation and commitment. Moreover, enabling variables such as team leaders, top management teams and champions are hypothesized to partially moderate the relationship between the stage setter variables and team behaviors. A closer look at Cohen and Bailey’s (1997) model on analyzing organization teams and McDonough’s (2000) model on NPD team performance suggests that these two models are complementary and an integration of the two could provide a comprehensive framework for understanding NPD team effectiveness. Figure 2-3 depicts this integrated model. 16 Figure 2-2 The Model for Analyzing NPD Team Effectiveness Suggested by McDonough (2000) Stage Setters Enablers Team Behaviors Performance - Team Leaders - Senior Managers - Champions - Project Goals - Cooperation Cross- - Empowerment - Commitment Functional - Human - Ownership Team Resources - Respect/T rust Success - Climate T Contextual Effects T e.g., industry In the integrated model, first, the set of variables such as design, group composition and organizational context that Cohen and Bailey (1997) includes in the same group corresponds to what McDonough (2000) refers to as stage setting elements. McDonough (2000) refers to these variables as elements that reflect management actions, which are set at the outset of the projects and create the foundation of the team activities. For example, group composition variables include McDonough’s (2000) conceptualization of human resources since McDonough (2000) explains that team outcomes will be enhanced with increasing variety of perspectives team members bring to the team. 17 80.00.0080 ..0>00.3 :00 000.0050 .0.o_>000m - .03.. 00.00.0000 00_ :00 000.0050 30.03:: - 000.0050 3.0000003...— 0000.0000 0.300000 0000.00.00 .0003 :00 308.2th 000.0050 0000000000.“. 00.00 00:00.0 2,00 $0.03 .8005 300 :00 300002.00th 00.00050 00.000.005.35. 3.0; 3.000000%... 0.00... 0.0.00.5 .000 6.2.3 - 0.0000. \\ .3000. 00.05 - 00002 - 0000026000 0900.000 - 0000.80.00 6.00000E .2000 0.0000. 0.00. :00. 00.05.0030 - 000000.030 0.000000. .0.000 0.0000. 0.00. :00. 00.03.0020 - Our—05.33500— EGOF Dn—z 0.0200. .0 00>... - 0000000. .0 00>... - 000.00. 000. 2,0. .0> 000.-.... - 00.30000. 0.0.. -230. .0> 0.05m, - 00.30006 02.00530 .m> 02.00.30 - ‘ 00:0..200.000 0.90% .w wCOEOu-ammfl: 00.00000 4000.08.08 000900000 :00 0.20.002 .005 00.000.030.080 00.0000 :00 0000000.“. .0535 00.000.025.00 40.0000 :00 00000005 .0520. 40.0 0000.392 30.00000. 0000.393 0003005000000 00:0..200300 . .5030:— - 0.300". 3000.00.35 0.20.00: / £59.... 80 .0000 .000 00.030. :00 00000 3002050090 0.300.. .000 :00 005000030 .000... 000000000920. $000050 :00 09000 0.00... 0.033 0W0“ 0000000000 000... 002 00 0_0>.00<-0.0.2 .0. .0005. 0000.000... 00.0.30 m-m 0.30.“. 18 Second, Cohen and Bailey (1997) postulates that environmental factors such as turbulence and industry characteristics influence the stage setting variables. McDonough (2000), on the other hand, hypothesizes that contextual effects have an impact on all antecedent variables. The only “contextual effect” identified in McDonough’s (2000) work is industry, which indicates that these variables are the same as what Cohen and Bailey (1997) refers to as environmental variables. Since the industry effect could only be argued to influence the stage setting variables, environmental factors are expected to have an effect on these antecedent variables. ' Third is the inclusion of enablers in the framework. Enablers are individuals that are located within different hierarchical levels of the organization and that could facilitate NPD activities and could be team leaders, champions or senior management. Team leaders could be considered enablers because they can challenge the team members to perform their best, instill encouragement and resolve conflicts effectively. Champions are individuals who take an inordinate interest in getting a particular product developed (Rosenau et al. 1996). As such, they can play a crucial role especially when they help the NPD team to overcome resistance from senior management. Finally, senior management can support the NPD team by demonstrating commitment to the project, thus helping to ease resistance to the members from other departments. The framework suggests that these enabling individuals act as moderators. The fourth set of variables is the mediating variables. These variables could be in the form of internal and external processes as Cohen and Bailey 19 (1997) postulate, or they could be team behavior variables such as cooperation within the team, commitment of the team members or respect of the members to each other. Team processes are defined as members’ interdependent acts that convert inputs to outcomes through cognitive, verbal, and behavioral activities directed toward organizing task work to achieve collective goals” (Marks, Mathieu, and Zaccaro 2001). Team process involves interactions within the team, i.e., internal processes, as well as outside the team, i.e. external processes. In addition to directly affecting team effectiveness outcomes, these variables have nonrecursive relationships with team psychosocial traits such as norms and beliefs. Finally, this guiding framework distinguishes between four types of outcomes: team effectiveness outcomes such as new product quality or new product sales; team efficiency outcomes such as budget adherence, team attitudinal outcomes such as job satisfaction of team members or trust between the members; and team behavioral outcomes such as employee turnover or absenteeism. 2.3 METHOD: META ANALYSIS Meta-analysis is the application of statistical procedures to a sample of empirical studies for the purpose of integrating and synthesizing the knowledge on a specific phenomenon (Wolf 1986). Hunter and Schmidt (2004) note that integration of studies reveals the simpler patterns of relationships thereby providing a basis for theory development. Lipsey and Wilson (2001) note that meta-analysis is a form of survey research, in which research reports, rather than 20 people are surveyed. Therefore, a coding form is developed as the instrument, a sample of population of research reports is gathered, and each research is ‘interviewed’ by a coder who reads it carefully and codes the appropriate information about its characteristics and quantitative findings (See Appendix for the coding sheet used for coding the extant literature). The meta-analysis in this dissertation consisted of the following steps, which follows the procedure proposed by Glass, McGaw, and Smith (1981): 1. Development of a framework listing factors that contribute to explaining NPD team effectiveness (this has been discussed in the previous section); 2. Selection of studies to be included in the review; 3. Coding of the characteristics of studies included in the review; 4. Analysis of the study characteristics and identification of variables influencing NPD team effectiveness; 5. Presentation and discussion of findings; 6. Delineation of directions for future research. 2.3.1 Database Development Potentially relevant studies are identified in two stages. In the first stage, the initial step was to perform keyword searches of the electronic databases such as ABI/lnform (ProQuest) and JSTOR using words as ‘NPD teams’, ‘product teams’, ‘development teams’, ‘innovation teams’, ‘product development teams’, ‘cross-functional’, and so forth. Next, references of the papers identified through these keyword searches of electronic databases were examined. Third, the 21 Social Sciences Citation Index was investigated for articles that cited several seminal studies such as Ancona and Caldwell (1992a; 1992b). In order to detect studies that did not come up with the keyword searches, reference examination and citation tracking, leading marketing, new product development, management, and engineering management journals where manually searched for articles on NPD teams. This manual search included articles published since 1980 or the inception of the journal, whichever is more recent. Journals to searched manually are Journal of Marketing, Journal of Marketing Research, Journal of Product Innovation Management, Management Science, Organization Science, Academy of Management Journal, Administrative Science Quarterly, Journal of the Academy of the Marketing Science, lEEE Transactions on Engineering Management, Research Technology Management, Research Policy, R&D Management, Industrial Marketing Management, Journal of Business and Industrial Marketing, Journal of Management, Information Systems Research, MIS Quarterly, Decision Sciences and European Journal of Innovation Management. As can be seen, this list of journals include both top-tier and second-tier publications across a breadth of disciplines. In the second step, file drawer problem is addressed (Wolf 1986; Rosenthal 1995) and rejected and working papers are sought. Inclusion of work that did not get published is especially important to reduce bias in the meta analysis results, since they may be rejected due to insignificant results. Therefore, a request for working papers and unpublished articles, etc. was posted on American Marketing Association’s electronic listserver ELMAR. To 22 motivate participation, anyone who provided a study was promised to be sent a draft of the meta-analysis once it is completed. Further, more than a dozen researchers, identified as prominent researchers on NPD teams based on the studies identified in the first stage of study identification, were contacted via e- mail and were requested to send any working papers or unpublished articles. These researchers were also promised a draft of the meta-analysis if they shared their studies. As a result of these five different ways of search conducted in two stages, more than 180 studies were gathered when the search process was concluded in March 2007. These papers are published in the years between 1982 and 2007. This time frame is appropriate for analysis of NPD team literature because, as noted earlier, use of teams in NPD activities started in the early 19803. Once studies were gathered, again, a two-staged process was executed to determine inclusion of a study in the meta-analysis. First, as noted earlier, Cohen and Bailey’s (1997) and McDonough’s (2000) frameworks were integrated to generate a comprehensive model of examining NPD team performance. Consequently, the integrated model in Figure 2-3 provided the initial basis for inclusion. Only studies that investigate the relation between NPD team performance and at least one variable from the sets of variables in this comprehensive model of NPD team performance were selected. This yielded a list of 78 studies to be included in the database. Next, this list was further reduced to determine a subset of the 78 studies to be used in calculating effect sizes by selecting articles that reported the correlation coefficients or its variants 23 among these studies. In conclusion, three of the 78 studies were analytic, two were conceptual, and 73 of them were empirical studies. The study database included 74 articles that were published in either journals or conference proceedings; one study that was never published, and three were work-in-progress papers. Fifty-four of the published papers (73.0%) were in first-tier journals, corresponding almost three fourths of the database. Table 2-1 presents a list of the studies in the current database. Procedures similar to other meta-analyses in marketing literature were followed by first designing a coding sheet that specified the information to be extracted from each study (of. Montoya-Weiss and Calantone 1994; Henard and Szymanski 2001). This was done in an effort to reduce coding error (Lipsey and Wilson 2001). After the coding sheet was prepared, one marketing scholar who is knowledgeable and has extensive experience in both new product literature and meta-analysis procedures provided comments and suggestions for its improvement. This scholar was also the second coder of the studies. Once all 78 studies were coded by the first investigator, the second coder coded a random sample of the studies and a high level of agreement on coding was obtained. 2.3.2 Study Context and Data Characteristics Information on nine research context variables was coded for each empirical study: 1) organization size (small, large, or both) where small is defined as if a firm’s number of employees is less than 500 or its annual sales is less than $1 billion; 2) team type (cross-functional, other, or both); 3) team proximity 24 Table 2-1 Studies Included in the Meta-Analytic Review* Authors of Studies Status Year Journal Type Paper Type Akgun and Lynn 1 2002a 2 empirical Akgun and Lynn 1 2002b 2 empirical Akgun et al. 1 2006 1 empirical Akgun, Lynn, and Byrne 1 2006 1 empirical Akgun, Lynn, and Reilly 1 2002 2 empirical Akgun, Lynn, and Yilmaz 1 2006 2 empirical Ancona and Caldwell 1 1992a 1 empirical Ancona and Caldwell 1 1992b 1 empirical Ancona and Caldwell 1 1990 2 empirical Ang and Slaughter 1 2001 1 empirical Atuahene-Gima and Evangelista 1 2000 1 empirical Barczak and Wilemon 1 1989 1 empirical Barczak and McDonough 3 2006 NA empirical Barczak and McDonough 2 2003 NA empirical Carbonell and Rodriguez 1 2006 2 empirical Carmel 1 1 995 1 empirical Chen 1 2006a 2 empirical Chen 1 2006b 2 empirical Chen, Reilly, and Lynn 1 2005 2 empirical Cohen, Eliashberg, and Ho 1 1997 1 analytic Cohen, Eliashberg, and Ho 1 2000 2 analytic Cohen, Eliashberg, and Ho 1 1996 1 analytic Cordero, Farris, and DiTomaso 1 1998 1 empirical Denison, Hart, and Kahn 1 1996 1 empirical Faraj and Sambamurthy 1 2006 1 empirical Faraj and Sproull 1 2000 1 empirical Faraj, Guinan, and Sambamurthy 3 2003 NA empirical Fedor et al. 1 2003 2 empirical Gobeli, Koenig, and Bechinger 1 1998 1 empirical Griffin 1 1997a 1 empirical Guinan, Cooprider, and Faraj 1 1998 1 empirical 25 Table 2-1 (cont’d). Authors of Studies Status Year Journal Type Paper Type Hauptman and Hirji 1 1996 1 empirical Hirst and Mann 1 2004 1 empirical Hoegl and Gemuenden 1 2001 1 empirical Hoegl and Parboteeah 1 2006 1 empirical Hoegl and Proserpio 1 2004 1 empirical Hoegl, Ernst, and Proserpio 1 2007 1 empirical Hoegl, Parboteeah, and Gemuenden 1 2003 2 empirical Hoegl, Parboteeah, and Munson 1 2003 2 empirical Keller 1 2001 1 empirical Keller 1 1 986 1 empirical Keller, Julian, and Kedia 1 1996 1 empirical Kratzer, Leenders, and van Engelen 1 2004 2 empirical Lee and Chen 1 2007 2 empirical Leenders, van Engelen, and Kratzer 1 2007 1 empirical Leenders, van Engelen, and Kratzer 1 2003 2 empircial Lovelace, Shapiro, and Weingart 1 2001 1 empirical Lynn, Akgun, and Keskin 1 2003 2 conceptual Lynn, Reilly, and Akgun 1 2000 1 empirical Lynn, Skov, and Abel 1 1999 1 empirical Madhavan and Grover 1 1998 1 empirical McDonough 1 1 993 1 empirical McDonough and Barczak 1 1992 1 empirical McDonough and Barczak 1 1991 1 empirical McDonough, Kahn, and Griffin 1 1999 1 empirical Moenaert et al. 1 1994 1 empirical Moenaert et al. 1 2000 1 empirical Montoya-Weiss, Massey, and Song 1 2001 1 empirical Pelled and Adler 1 1994 1 empirical Pelled, Eisenhardt, and Xin 1 1999 1 empirical Pinto and Pinto 1 1990 1 empirical 26 Table 2-1 (cont’d). Authors of Studies Status Year Journal Type Paper Type Pinto, Pinto, and Presscott 1 1993 1 empirical Qiu et al. 3 NA NA empirical Sarin and Mahajan 1 2001 1 empirical Sarin and McDermott 1 2003 2 empirical Sethi 1 2000a 1 empirical Sethi 1 2000b 1 empirical Sethi and Nicholson 1 2001 1 empirical Sethi, Pant, and Sethi 1 2003 1 conceptual Sethi, Smith, and Park 1 2001 1 empirical Shim and Lee 1 2001 1 empirical Stock 1 2006 1 empirical Susman and Ray 1 1999 2 empirical Thamhain 1 2003 1 empirical Troy, Szymanski, and Varadarajan 1 2001 1 empirical Valle and Avella 1 2003 2 empirical van den Bulte and Moenaert 1 1998 1 empirical Voss 1 1 985 1 empirical *Status: 1=published article, 2=unpublished article, 3=working paper Journal Type: 1: first tier (e. 9., Journal of Marketing, Journal of Marketing Research, Journal of Product Innovation Management, Management Science, Organization Science, Academy of Management Journal, Administrative Science Quarterly, Journal of the Academy of Marketing Science, IEEE Transactions on Engineering Management, Information Systems Research, and MIS Quarterly) 2=second tier (e. 9., Research Technology Management, PDMA Proceedings, Research Policy, R&D Management, Technova tion, Industrial Marketing Management, Journal of Business and Industrial Marketing, Journal of Management, European Journal of Innovation Management, Decision Sciences, and Creativity and lnnova tion Management), NA=Not Applicable 27 (co-located, partially dispersed, virtual or more than one type); 4) innovation type (product, process and other, or both); 5) type of product developed (goods, services, and/or software); 6) product innovativeness degree (incremental, radical, or both); 7) data collection countries (US, other countries, or both US and other countries), 8) served market technological innovation degree (high technology, low technology, or both), and 8) industry type (manufacturing, non- manufacturing, or both). Table 22 presents the counts and percentages of the study characteristics in the final set of studies included in the meta-analysis. Similar to previous meta-analyses in marketing literature (cf. Montoya-Weiss and Calantone 1994), study characteristics were coded only when the authors reported explicit information. As a result, percentages of studies reporting study characteristics are also provided in the table. An important organizational characteristic, firm size is not reported in more than half of the studies. However, small and large firms would differ considerably that could affect how an NPD team operates. For example, resource constraints would be different in small firms than in large firms. Moreover, smaller firms would be closer to hallway companies and the team members might know each other well even though team longevity might be low. Similar high percentage of non reporting is obsen/ed for other study context and data characteristics such as data collection countries (38.4%), team type (56.3%), product innovativeness degree (81.9%). Consequently, this review of the literature highlights the necessity for more detailed reporting of study context and data characteristics 28 Table 22 Study Context and Data Characteristics of Empirical Studies Classifier Variable Number of Count and Percentage across Applicable Studies Empirical Studies Organizational Characteristics Organization size 713 Small only: 11; 15.5% Large only: 9; 12.7% Both: 14; 19.7% Not explicitly reported: 37; 52.1% Data collection countries 73 US only: 18; 24.7% Other only: 23; 31.5% Both: 4; 5.5% Not explicitly reported: 28; 38.4% Industry type 71 a Manufacturing only: 40; 56.3% Non-manufacturing only: 5; 7.0% Both: 18; 25.4% Not explicitly reported: 8; 11.3% Served-market technological 71a Hi-tech markets only: 37; 52.1% innovation degree Low-tech markets only: 2; 2.8% Both: 8; 11.3% Not explicitly reported: 24; 33.8% Team Composition Characteristlc Team type 71 a Cross-Functional only: 29; 40.8% Other only: none; 0.0% Both: 2; 2.8% Not explicitly reported: 40; 56.3% Task/Product Characteristlcs Innovation type 73 Product only: 66; 90.4% Process or other only: none Both: 5; 6.8% Not explicitly reported: 2; 2.7% Type of product developed 73 Goods only: 19; 26.0% Services only: 1; 1.4% Software only: 16; 21.9% More than one type: 10; 13.7% Not explicitly reported: 27; 37.0% Product innovativeness degree 72 Incremental only: none; 0.0% Really new only: 1; 1.4% Both: 12; 16.7% Not explicitly reported: 59; 81.9% a This is not applicable to two studies since the data is collected from students. b This is not applicable to one study where innova tiveness degree is the dependent variable. 29 in future studies. One promising observation is that innovation types that the teams are dealing with (e.g., goods, services, or software) are explicitly expressed in almost al studies (90.4%). 2.3.3 Study Design Characteristics Counts and percentages of study design characteristics in the final set of 78 studies are presented in Table 2-3. As in study context and data characteristics, study design characteristics were coded only when the authors explicitly reported them in their papers. Although majority of the studies have collected their data using a single method (70.4% survey and 7.0% case study only), about one fifth (22.5%) glean insights based on data collected with multiple methods: survey and a form of case study. The results also show that more than three quarter of the studies used cross-sectional data as opposed to 13.7% using longitudinal data. All but two studies use primary data; one study uses secondary data sources while another uses both primary and secondary data. A variety of organizational members are used as key informants. While team leaders are the respondents of 14.1% of the studies, team members are key informants for 12.7 % of the studies. Senior managers are used as key informants in 8.5% of the studies. Most importantly, the compilation of studies reveals the methodological rigor in the literature: more than half of the studies collect responses from multiple types of informants for investigating NPD teams. This is highly appropriate as an NPD team’s work affects its members, their departments and the firm as a whole. Therefore, obtaining perceptions from informants at multiple levels of the organization would represent the intricacies 30 Table 2-3 Design Characteristics of Empirical Studies Number of Count and Percentage across Empirical Classrfier Variable Applicable Studies Studies Data Characteristics Data collection methods 71" Survey only: 50; 70.4% Case study only”: 5; 7.0% Both: 16; 22.5% Data temporality 73 Cross-sectional: 58; 79.5% Longitudinal: 10; 13.7% Not explicitly reported: 5; 6.8% Data type 73 Primary only: 71; 97.3% Secondary only: 1; 1.4% Both: 1; 1.4% Data source 71c Managers/senior managers only: 6; 8.5% (type of informants) Team/project leaders only: 10; 14.1% Team members only: 9; 12.7% More than one type of informant: 41; 57.7% Not explicitly reported: 5; 7.0% Performance Measurement Item Characterisitcs Item plurality 69d Single only: 6; 8.7% Mulitple only: 54; 78.3% Both: 8; 11.6% Not explicitly reported: 1; 1.4% Item subjectivity 68° Subjective only: 63; 92.6% Objective only: 3; 4.4% Both: none; 0.0% More than one type: 1; 1.5% Not explicitly reported: 1; 1.5% a Two out of seventy-four have not explicitly reported data collection method. b Case studies include semi-structured interviews, company records, etc. ° Not applicable two studies: the study where data is obtained from secondary sources; and the study where informants are students in an executive MBA course. d Not applicable to four studies that have only qualitative data. ‘9 Excludes one study with secondary data only and not applicable to four studies that have only qualitative data. 31 involved in an NPD team’s performance. Literature shows high methodological rigor for measuring NPD team performance. Performance is measured with multiple items in 78.3% of the studies, while 11.6% use both single and multiple items for measuring dimensions of performance. Finally, subjective measures are used for operationalizing NPD team performance in all except four of the studies. 2.3.4 Dependent Variable: NPD Team Performance NPD team performance is defined and operationalized in several different ways in the literature. In effect, team performance construct includes variables that describe an NPD team’s effectiveness and efficiency (Madhavan and Grover 1998). While effectiveness refers to the degree to which expectations regarding the quality of the outcome (e.g., a new product’s quality, a new software’s robustness, etc.) are met by the team, efficiency refers to the adherence to schedule and budget constraints set for the team at the beginning of the project. Put differently, Hoegl and Gemuenden (2001) notes that effectiveness reflects a comparison of actual versus intended outcomes, while efficiency reflects a comparison of actual versus intended inputs. In this dissertation, effectiveness and market-based performance labels are used interchangeably as well as efficiency and process-based performance labels. This multidimensional construct resulted in many authors measuring only team effectiveness, only team efficiency or a combination of the two, thereby measuring an overall team performance. In this dissertation the relationships 32 between antecedent variables and all three NPD team performance measures (i.e., overall, effectiveness, and efficiency) are all considered. 2.4 ANALYIS AND RESULTS 2.4.1 Meta-Analytic Analysis Procedures As noted earlier, the correlation effect sizes are analyzed in this study. Analysis is on the model-level correlations as opposed to study-level correlations. This is consistent with other meta-analyses in marketing literature (cf. Henard and Szymanski 2001) and is suggested by Glass, McGaw, and Smith (1981 ). Model-level analysis is done by analyzing correlations across all models and all studies. Caution is exerted in coding of the correlations to ensure that conceptual dissimilar, but similarly labeled are not combined inappropriately as well as conceptually similar but dissimilarly labeled constructs are combined. To do this, construct operationalizations are thoroughly investigated. In addition to calculating simple means by averaging raw correlations and then sample size- adjusted means of correlation effect sizes obtained from each study, effects were corrected for measurement error by dividing the correlation coefficient by the product of the square root of the reliabilities of the two constructs (Hunter and Schmidt 2004). As noted by Hunter and Schmidt (2004), low reliability deflates correlations. For the studies that did not report the reliability of the construct or used a single item where reliability was not applicable, the simple mean of the reliabilities for that construct across all studies was used (cf. Kirca, Jayachandran, and Bearden 2005). 33 2.4.2 Correlation Effect Sizes and Their Interpretations Table 2-4 presents the total number of correlations reported between NPD team performance dimensions and their predictors in the studies. As can be seen, the literature provides correlations for all relationships, but the number of studies reporting a particular correlation is limited to only one study for quite a number of relationships. For example, correlation between team effectiveness and the antecedent variables of task interdependence, behavior control, interfunctional climate and existence of norms have only been examined by one study. Table 2-4 therefore presents the gaps in the literature, and subsequently, provides guidance as to which variables should be included in future studies. For example, only one study examines the role of a particular team psychosocial trait, namely norms, on NPD team overall performance and no study investigates the effect of norms on either NPD team effectiveness or efficiency. Hence, future studies should include norms in their set of antecedent variables and shed some light on their influence on NPD team performance. Next, corrected correlation effect sizes between predictors and overall NPD team performance, effectiveness, and efficiency are calculated. These correlations are interpreted by following the guidelines suggested by Cohen (1977; 1986). These guidelines suggest that correlations less than 0.10 indicate small, correlations equal to 0.25 indicate medium, and correlations greater than 0.40 indicate a large effect of the predictor on the outcome variable. 34 6063.2... 00.0. 0.0. 20.00.0260 6:0. b..E..xo.l 0 N. n - 0 00002600300 .0000. 600. 0 N 0 m 00006060000009.6000 .0000. 600. 0.0.000m - 0.06.6 .000.60:t0.0. N 000000 600 0.60.0 000 0 >606 .000 0 00.030. 00009606606860 . 6.600 .0.>0000 00.00_.0> 0.0.000 6003030090 N F F - 0.....0 606000006 60.0.0 5 0 - m b..000.60200065.296 600.602 - 00000.0.0 60.900 9. 0.. [x ('3th 01me!- I cm 0.. MP 0 0N.0 600. 0 3300020500. 600. 00.n0..0> 00.030600 030.0 m - - m 6.0000 x00. - 000000000006. x00. 00.65.06.200: 00000.0 005.066. F >6000.00\.006.030060 600. 00.00..0> 00.000 0.00... 0.3.0.0 008m 000')” v f\ 1- V 1-[\1- (D .GuOh. .tmn— Damn-OOOOOLQ .tOQ Emma-00x50: .tfln— =9—0>° 0.0.0.090 0.. 000 00006600... 600... 002 0003.00 000000... 000006000 .0 .00602 .06... TN 0.00... 35 0 0 0 N 2.00.0600 6000.00.00. 0F 0 m 0 0000026500. 6 6000.0 00.000 0.060“. 6.0060965 F - - F 06.00 .0 00006.08 0:9... 0200000050 E00 ._. 0F FF 0 0 00.60.0000 30.: .00606 600. 0.20.005. .050 N F - F 00.60.0500 .06060 n 0 F - 00.60.006600 00.000 00000090 .090Fxm 0 F F 0 6.0000 00000006600. N F - F 2.2.0.0. 600. 0F 0 .V F 0:000 0003600563600. 0 F F F m N 000002000000000000 NF 0 0 F 00.00.0500 .0606. NF 0 0 F 00000606600 .0606. n m - .V 00000.0 .0606. 00000090 .0980. 0.0.0.005. .000... €00 00000000090 €00 00000-0...02 €00 =0.0>0 ..0....8. #N 0.03 36 Table 2-5 summarizes total number, minimum, maximum, and range of correlations as well as cumulative sample size of the studies, simple, sample size-adjusted and reliability-adjusted correlations for the bivariate correlations between antecedent variables and overall team performance. Four of the fifteen simple means, namely, formalized process use, team size, functional diversity, degree of product innovativeness indicate a low effect size. Five of the effects indicate low to medium sizes: task conflict, team tenure, goal support, team leader power and team interpersonal conflict. Two correlations, internal process and team leader effectiveness indicate medium effects. Finally, four effect sizes, outcome-based rewards, goal clarity (assuming 0.37 is close to 0.40, which is the recommended value for interpreting an effect size as large), cohesion, and team member satisfaction all indicate large effect sizes. Sample size adjustments on these correlations do not present a different picture in terms of categorization of effect sizes of the variables except that cohesion’s effect size goes down from 0.56 to 0.37, which can be assumed marginally large. Reliability adjustments on the correlations do not change the categorization of the effect sizes either, but cohesion’s effect increases to 0.42, which can be now interpreted as large as opposed to marginally large. Next, Table 2-6 summarizes total number and range of correlations as well as cumulative sample size of the studies, simple, sample size-adjusted and reliability-adjusted correlations for the bivariate correlations between antecedent variables and market-based NPD team performance. As seen in the table, there are six low effect sizes based on simple correlation means: physical distance, 37 0.00000 .00 0000 0000.00.09 0000 00005002 00 80350.0 00.0: 00.0.8.8 000 0000:. 00.00.00 00.0000m 0 0.00000 A050 8 800 00..QE00 80 00.00.0000 0.08.0 0880 0000.008 00. 0.. 000.: 0.90.0 a 00.0 00.0 3.0 0 —0 00.0 00.0 000 00.0 0 0 0000000800. .0 006000008030 00>. 9.800“. 302009.300 0.0.0 0.0 00.0 000 00.0 00.0 000 00.0 0 .- 000000000 .00.. .3202 500. 9.80.00: .005 00.0- 00.0- 00.0- K0 00.0 00.0 000 00.0- v 0 8.008 080.090.000.00. 00.0 00.0 00.0 000 00.0 00.0 000 00.0 N 0 0.000020008020008 00.0 00.0 00.0 00. 00.0 00.0 000 00.0- .V 0 0008.0 .0520. 000082.. .0520. 0030.00: 3.0 . .0 0.0 000 00.0 00.0 000 00.0- 0 0 808000.028 .008. 502 00.0 00.0 00.0 000 3.0 3.0 000 00.0 0 0 00008060000009.0800 .0000. E00. 000.0000 00.0 0 .0 m 00 000 00.0 00.0 000 00.0- 0 0 000000 .000 00.0 00.0 00.0 000 00.0 00.0 000 00.0 v 0 300.0 .000 00.0 00.0 00.0 000 00.0 00.0 000 00.0 0 0 00.0.0. 0008-80.90.05850 5.80 0803.590 0 0.0 0 0.0 00.0 000 00.0 00.0 000 00.0. 0 0 .0.08082002030820 .0000000. 0 0.0 3.0 00.0 000 00.0 00.0 000 00.0 v 0 33000000500. E00. 000 00.0 00.0 000 00.0 00.0 000 00.0. v 0 0.00 E000 8.002.800 .000.- 0 0.0 0 0.0 00.0 000 00.0 00.0 000 000- 0 0 8.008 x00. 00.0 00.0 00.0 000 00.0 00.0 000 00.0- 0 0 8000.00.02.00: 00088 000.088. 00.08 0.00» 20000 0080 0000.2 000.2 0000.2 0~.0 00:_0> 00:.0> 0800.008 00.030 0000.03.00“. E00... 002 =0..0>O 00000.0( 00.0:..0< 0.0.0.0 0.0.000 0000.008 0000.008 .0 000050 .0 000050 000 00.00.0000 2:50:00. .000 0.9000 02.0.0an 00 0000.”. x05. 000 0.2 .08.? .08... 0000002000 .000... 002 _.0.0>O 000 0.0000095. 0002000 0000 8000 .6 008030 0-0 0.00... 38 000.08. 0.0 000000.00 000 0.00.80. .0... 08.00.0000 0 0.00008 .00 0.00 00.0.3000. 00...: 00.0.3000. .0 00008.0... 00.0: 00.0.8.8 0.0 0000... 00.09.00 00.3000... 0 0.00.0.0 000.0 .0 .000 0.0.0.000 .0. 00.00.0000 0.0000. 000.00 0000.008 0... 0.. 000... 0.0.0.0 0 0 .0 0 .0 0 ..0 0.0 00.0 00.0 000 00.0- 0 0 00.00.08 8000.00.00. 00.0 .00 .00 000 00.0 00.0 000 .00- 0 .- 00000>00>0§ .0 00.00.008.90 .0 00... 0.200“. 300500.500 .00 00.0 00.0 000 00.0 00.0 00... .00 0 0 808.0000 .00.. .0020... 000. 0.0.0.8... .200 00.0 00.0 .00 000 00.0 00.0 0.0 .00 0 0 0.000 0020.00.20.58... .00 00.0 00.0 00.. 00.0 00.0 0:0 0.0 0 . 000000000800028 000 .00 .00 .00 0.0 00.0 000 00.0 v 0 000050.80 050... 00.0 00.0 00.0 00.. .00 .00 0.0 00.0 0 . 00000005500 00.0.... 00000020 .0530. 0.82.82 00.0 .00 00.0 .00 00.0 00.0 .20 00.0 0 0 08000 .80 00.0 00.0 3.0 .00 0.0 00.0 0:0 00.0 0 0 0.50.0 000 00.0 3.0 00.0 000 3.0 0.0 0.0 00.0 0 0 0.0.0 .80 0 .0 0 .0 .. ..0 0 .0 00.0 00.0 0.0 .00 0 0 00.020. 0008.00.90.88? 08.50 32.00.0005 .00 00.0 00.0 000 00.0 00.0 0:0 00.0 0 0 0000800500. :00. 00.0 00.0 00.0 000. .00 00.0 000 .00- 0 0 00.0 500. 020000.000 .0000 00.0- 00.0- .00 .0. 0 . .0 0.0 0:0 00.0- 0 0 0800.00 00.00... 00.0 .00 00.0 00. 00.0 00.0 000 00.0- 0 0 02.00.202.000 0008.0 000.0000. 00.000 0.00» 0.2000 0080 0000.). 000.2 0000.2 00.0 00:.0> 00:.0> 00000.008 00.020 8000.02.80 .0000 002 00000-00000: 00.00.00. 00.00.00. 0.0.0.0 0.0.000 0000.008 0000.008 .0 .0005: .0 .0005: .0... 00.00.0000 0.5000 -000 00200 00.03500 .0 00:00 x02 0.0 .02 0.0. 0.0. 0000028000. 0000000000 .000 .- 0.02 000000000... 000 0.000000.0< 0002000 000.0 .0000 .0 .000.an 0-0 0.00... 9 3 team size, team tenure, outcome-based rewards, degree of product innovativeness, and task complexity. The effect size of formalized process use is 0.20, indicating a low to medium effect size. The correlation effect sizes of all four internal processes, i.e., internal communication, internal coordination, cohesion, and teamwork are medium. The effect size of goal support is 0.33, and hence can be considered as medium to large. Finally, three of the fifteen simple correlation means are large, indicating large effect sizes. These variables are goal clarity, goal stability, and team member job satisfaction. The effect sizes remain the same when they are adjusted for sample sizes, except the effect of goal support increases to 0.37, close to being considered as large. In the third step, when the effect sizes are corrected for both sample size and reliability of the predictor and outcome variables, all effect sizes either remain the same or increase. The effect of goal support exceeds the cutoff of 0.40 and is now 0.43, clearly a large effect size. Finally, Table 2-7 summarizes total number and range of correlations as well as cumulative sample size of the studies, simple, sample size-adjusted and reliability-adjusted correlations for the bivariate correlations between antecedent variables and NPD team efficiency. Simple mean correlations reveal that three team composition variables, namely, team size, team tenure and cross- functionality as well as team leader power have small effect size on NPD team, efficiency. Next, two team design variables of formalized process use and physical distance have effect sizes somewhere between low and medium. On the other hand, outcome-based rewards and internal coordination have medium 40 600200 02 020 20.00.002.00 020 5.22.2020 22.: 0200020200 0 0.000000 002 020.: 000.22.002.02 202.... 00.22.002.02 20 20.30.5020 02.0: 00003000 020 0200E 00000.80 52.50.2000 0 00000.20 x080 20 2020 0.203200 200 000020.802: 02000.2 000200 20.00.0200 020 0.. 200E 03.25 0 00.0 0 .0 00.0 000 00.0 00.0 000 00.9 0 0 00000....0500. .0 0900000290 .0 00>. 8200". 320529.35 00.0 00.0 00.0 000. :0 00.0 000 00.0 3 0 00.80.0000 300 .0020... E00. 280.00: .050 00.0 00.0 00.0 000 00.0 00.0 000 00.0 0 0 £000 {025020003503 .00 00.0 00.0 000 00.0 00.0 0:0 00.0- : 0 00000200000200.0008 00.0 00.0 00.0 000 00.0 00.0 0:0 0 0.0 0 0 000050.08 .052... 00.0 0 <0 0 .0 000 00.0 00.0 0:0 00.0- 0 0 00.000.55.08 .052... 00330.5 .0522. 880.00: 9.0 00.0 :0 000 00.0 00.0 0:0 0 00- 0 0 8000.20.03... .0000. 0.00. 0.0305 00.0 00.0 00.0 000 00.0 00.0 0:0 00.0 0 0 000000 .000 00.0 00.0 00.0 :0 2.0 00.0 0:0 00.0 0 0 E500 .000 3.0 00.0 00.0 0000 00.0 3.0 000 00.0 0 0 200.0 .000 _0.0 00.0 00.0 000 00.0 3.0 000 $0 0 0 00.0.50. 00.000-60.05050050 .0280 3:25.505 .00. 00.0- 00.0 000 00.0 00.0 000 00.0- 0 0 a.000.800.000.020.0% .0000000. 000 00.0 8.0 000 00.0 00.0 000 00.0- 0 0 230000.058. 500. 00.0- 00.0- 00.0- 000. 00.0 00.0 000 00.0- t 00 00.0 E00. 202009200 Eco... 0.0- 0 . .0- 0 .0- 000 0 F0 :0- 000 00.0- 0 0 08:00.0 00.0.30 000 00.0 0.0 000 00.0 00.0 000 00.0 0 0 00.000.05.800: 0008.0 000.0000. 00.80 .300 2053 008m 0:002 000.2 0:005. 00% 00:_0> 00:_0> 0200.00.80 00620 3:083:00. 508. 9.2 woman-000090. 0000204. 0000060. 295m 0_QE0w c0z0_0:00 5:30:00 .0 .0005: .0 2006:: 5.2... 20.00.0200 00.50.00. 00.0 0.0500 02.0.0500 .0 00:00 x02 000 0.2 .08» .06.. 902225: 00:08.0...00 E00 ._. omz 00000-000090 020 00000000020. 000.300 005m 0005 00 2:08an 0-0.. 0.00... 41 effect sizes. Degree of project innovativeness, with a simple correlation mean of 0.23 can also be considered to have medium effect size. Goal clarity, goal support and teamwork have medium to large effects, with simple means of 0.33 0.32, and 0.34, respectively. Job satisfaction has a marginally large effect size with a simple mean of 0.39. Finally, goal stability demonstrates large effect size. Sample size adjustments reflect no change in the effect size interpretations except for goal stability, teamwork, and job satisfaction. The effect of goal stability goes down from 0.40 to 0.39, therefore is now marginally large. On the other hand, the effect size of teamwork increases from 0.34 to 0.38, which can be considered as marginally large. The effect size of job satisfaction goes down from being marginally large (0.39) to medium to large with an absolute value of 0.37. In the final step, when reliability adjusted correlations are calculated; the effect sizes paint a different picture only for goal clarity, goal stability, goal support, teamwork and satisfaction. All of these effects increase to become large effect size. 2.4.3 The Effect of Stage Setters and Cohesion on NPD Team Performance: A Path Analysis In the final stage of analysis, all 79 studies in the study database were used to generate a correlation matrix of the NPD team-related constructs. The correlation matrix had numerous empty cells because no study examined or reported the relationship between certain constructs. Based on the criteria that there must be multiple correlations that relates a construct to every other 42 00.. 0.. 00. 00. 00. 00.. 00.0 00.090.20.000. E000 0 00.. 0.. 0..- 00.- 0.0 00.0 00.0 E000 0 00.. 00.- 0..- 00. 00.0 080.00.000.00 0 00.. .0. 00. 00... 00.00000 0 00.. 00. .0... $0000.00. .000 E00. 002 0000000000.“. 0 00.. 00... 0000020000. .000 E00. 002 00000.0...02 . 0 0 0 0 0 .00 0.0 000.2 .0005. £00. 0.... :. 00.0.0..0> 0... .0. 00000.0..00 0:0 00000300 0.00.05 0:00.). 0-0 0.0.0..- 0000008000 00000-000090. 000.06.00.00. 00000-0005. 00:05.23; Euoh Dn—Z b30020. 0.2.0.. E00... "'"fl""'"7L"""""""" 00.00000 00.0 E00 0 02.0.0.0 00.0.0... 3 0. ...................... 0 .00.... 0.00.0.0 0005 000.2 500. ”0000.50.00. E00 ... n.n.2 .0 0.0000020... v-m 0..-.0.”. 43 construct (of. Brown and Peterson (1993) and Kirca, Jayachandran, and Bearden (2005)), a path model of antecedents of NPD team performance is examined. This constraint resulted in testing the effects depicted in the model in Figure 2-4. The means, standard deviations and correlations used in the analysis is presented in Table 2-8. Path coefficients are estimated by maximum likelihood estimation using EQS 6.1 (Bentler 2004). Although, the chi-square statistic is significant (x2=19.17, df=4, p<0.01), other fit indices suggest adequate model fit (e.g., MFl=.93, GFI=.94). The path coefficients and their significance are shown in Figure 2-5. The results suggest that a cohesion significantly improves both market-based (B=.47, p<0.01) and process-based NPD team performance ((3:23, p<0.10). Moreover, physical distance (B=-.30, p<0.01) hinders cohesion significantly, while team tenure has a significant positive effect (B=.12, p<0.05). Finally, the direct effects of physical distance and team tenure on NPD team performance dimension are both insignificant, providing support for full mediation by cohesion. 2.5 CONCLUSIONS This study revealed that two factors have a consistent effect size on different NPD team performance dimensions: goal clarity and team member satisfaction. The corrected correlation effect sizes for these variables are both large and positive on overall NPD team performance, NPD team effectiveness and efficiency. Team size also has a consistent, small effect size across 44 Figure 2-5 Significant Path Coefficients between the Antecedents, Cohesion, and NPD Team Performance NPD Team Stage Setters Mediator Performance . E Market-based Physrcal : Performance Team tenure/ longevity . \ N fit It Process-based Performance 5 distance "30*" 5 Cohesion --------------------- *p<. 10, ** p<.05, *** p<.01 performance dimensions, but this effect is positive in sign for overall and market- based performance, while negative for efficiency. Goal stability also has large and positive effect sizes for effectiveness and efficiency, but the study is inconclusive on its effect on overall performance due to inadequate number of extant research. Similarly, the correlation between team leader power/influence and overall performance as well as efficiency demonstrates a positive and small effect size, but no result on its effect on NPD team effectiveness. Finally, internal coordination has a positive and medium effect size on NPD team effectiveness and efficiency. lts effect on overall NPD team performance is indeterminate due to lack of empirical studies. It is also worthwhile to note that outcome-based rewards construct has a positive, but different size effect on each type of performance: large on overall performance, small on market-based performance 45 and medium on process-based performance. Table 2-9 summarizes all corrected correlation effect sizes between the antecedents and three types of NPD team performance. Table 2-9 Summary of Corrected Correlation Effect Sizes on NPD Team Performance Dimensions Sample Size- and Reliability-Adjusted Overall NPD Market-based Process-based Correlation with”, Team Perf. NPD Team Perf. NPD Team Perf. Stage Setters Task Design formalized process use/formalization 0.10 0.24 0.28 physical distance - -0.02 -0.19 task conflict 0.16 - - Team Composition team size 0.04 0.07 005 team tenure/longevity 0.16 0.04 0.07 functional diversity/cross-functionality 0.17 - -0.04 Organizational Context outcome/project-based rewards 0.51 0.16 0.31 goal clarity 0.46 0.48 0.41 goal stability - 0.48 0.44 goal support 0.23 0.43 0.43 Enablers team leader effectiveness/competence 0.38 - - team leader power/influence 0.14 — 0.10 Mediators Internal Processes internal process 0.34 - - internal communication 0.23 0.30 0.22 internal coordination - 0.32 0.29 cohesion/cohesiveness 0.42 0.27 0.24 teamwork/teamwork quality - 0.32 0.42 team/Interpersonal conflict -0.26 - - Other Mediators team member (job) satisfaction 0.76 0.47 0.44 Environmental Factors type of project/degree of innovativeness 0.06 0.02 0.19 task/product complexity - 0.12 - 46 Multivariate path analysis confirms the positive medium-size effect of cohesion on NPD team effectiveness and efficiency as demonstrated by the significant path coefficients. Moreover, it reveals that the effects of physical distance and team tenure on performance are not direct, but rather mediated by cohesion. The results also confirm that physical distance hinders performance through its detrimental effect on an NPD team’s cohesion. 47 CHAPTER 3 A LONGITUDINAL STUDY OF ORGANIZATIONAL CHANGE BY IMPLEMENTATION OF NEW PRODUCT-DRIVEN GROWTH STRATEGY: PROFICIENCY OF PERFORMING NPD PROCESS AND THE MODERATING EFFECT OF CROSS-FUNCTIONAL TEAM USE 3.1 INTRODUCTION Strategy implementation is the crucial link between the marketing strategy formulated and its performance effects (Bonoma 1984; Noble and Mokwa 1999, Atuahene-Gima and Murray 2004). Since the main focus of a marketing strategy is the product (Atuahene-Gima and Murray 2004), a particular marketing strategy can be directed toward existing products or new products. The focus of this essay is on implementation of a marketing strategy that is geared mainly toward introducing new products. Successful implementation of a marketing strategy Is as important as formulating the best strategy for the firm because poor execution can hinder confirmation of the rightness of the new product strategy and new processes associated with the new product strategy, thereby can provoke unnecessary change and lead to a dismissal of an appropriate strategy. Proficient execution of new product strategies necessitates generation of brilliant ideas that are aligned with the firm’s strategy, but it is merely sufficient as these ideas should be transformed into products. However, conversion of these ideas into commercially successful new products is very difficult as demonstrated by a survey of (Product Development Management Association) PDMA members 48 about NPD best practices, which reveals that the top obstacle to successful NPD are seen as the execution of NPD process activities (Page 1993). Firms can be managed successfully using a dominant logic (Prahalad and Bettis 1986) that emphasizes development of core competencies (Prahalad and Hamel 1990). However, as Leonard-Barton (1992) observes, this can result in a problem if the competencies become ‘rigid’ at the expense of innovativeness. In such conditions, transforming a firm from its current competencies of serving existing markets via maintaining the quality of current products to a firm that recognizes new product introductions as the source of its sustainable success is an extraordinary task for top management. The difficulty lies in the need for two levels of change: not only must top management institutionalize successful NPD via modified organization charts, processes, and reward structures, but also they must convince employees of the need for change and ensure adoption and adaptation to the new processes. Indeed, drawing from a survey of PDMA members about NPD best practices, Griffin (1997b) notes that NPD change is evolutionary and moves fonlvard on multiple fronts. Studies documenting successful product innovation practices in established firms exhibit a positive association of project, market, and financial outcomes with organizational characteristics and NPD process proficiency. The data collection methods employed include case studies and extensive interviews (of. Dougherty and Heller 1994; Dwyer and Mellor 1993) or cross-sectional surveys in the US (cf. Calantone, di Benedetto, and Divine 1993) and other countries (of. Song and Parry 1996). Although these studies have 49 served in ascertaining the association between the organizational factors, processes and outcomes, the causality of the effects has not been fully established due to inadeqUacies of these data collection methods in establishing causality. Bollen (1989) notes that the definition of causality has three components: isolation, association and direction of influence. Isolation can be accomplished by assuming that the disturbance, which is the composite of all omitted determinants, is uncorrelated with the exogenous variables. When a cause and its effect are isolated from other influences, then we should be able to show the bivariate association between these variables. Both isolation and association can be shown with cross-sectional data. However, direction of influence can best be supported by utilizing longitudinal data, which enables the researcher to show the temporal priority of the causes (Hume [1739] 1977). Another benefit of using longitudinal data for analyzing the phenomenon of interests in this essay is to capture the lag in performance effects of change programs and identification of ordering of areas that improve in the NPD process. Change programs typically exhibit a lag in profit performance improvement because first employees’ attitudes toward the organization and change must evolve (Mirvis and Lawler 1977). Consequently, analyzing longitudinal data of employees’ assessments of change is useful because the data provide a potential leading indicator of future profit performance while also identifying which areas improve first. 50 The detailed intricacies of how formal project team use is institutionalized are still elusive. Therefore, in an effort to provide evidence on the effect of organizational factors and processes on NPD outcomes, one objective of this essay is to observe and examine an organizational change towards innovation in a traditional manufacturing firm by utilizing longitudinal data. Child (2005) classifies organizational change in three dimensions: radicalness, planned vs. emergent nature, and the focus of change. Consequently, one objective of this study is to observe how a radical, planned change or the part of an organization takes place. Specifically, it examines a firm execute its newly formulated product strategy through implementation of a new NPD process. The second objective of the study relates to the organization of NPD activities. Recent empirical research shows that NPD efforts are predominantly performed by multi-disciplinary, cross-functional teams (Henke, Krachenberg, and Lyons 1993; Page, 1993; McDonough 2000). Nonetheless, there is inconsistency in the extant literature about the effect of use of cross-functional NPD teams on project and market outcomes. While some studies show a negative effect, others suggest a positive one. As such, the second objective of this study is to show that use of cross-functional NPD teams leads to better NPD project, and subsequently, market outcomes. Moreover, this study extends the existing literature. Song and Montoya- Weiss (1998) investigate the moderating role of the degree of product innovativeness on the relative effect of NPD activities on NPD performance. By taking a more comprehensive view and it examines the contingencies in the 51 organizational structure (functional development effort vs. multi-functional structure) and leadership (existence of a dedicated leader throughout all NPD phases) concurrent with the degree of product innovativeness when relating NPD task proficiencies to NPD outcomes. In conclusion, the two research questions investigated are summarized as follows: Research Question 1: When implementing a new product strategy via a formalized product development process, which of the NPD process phases improve before the others? Research Question 2: What is the effect of proficiency in NPD phases and their interactions with the use of cross-functional teams, team leader existence throughout the project and inherent level of product innovativeness on NPD team performance? Moreover, the effect of antecedents on several NPD team performance outcomes such as sales, ROI, and NP quality measured by objectively. Also, methodologically, use of objective outcomes measures as well as longitudinal temporality of data are improvements. The remainder of this chapter is organized as follows. First, the academic study of organization development and change is described. Against this background, the research hypotheses are presented. Next, the method for the study is presented, followed by data analysis and results. 52 3.2 THEORETICAL BACKGROUND 3.2.1 Organization Development and Change Organization development and change (OD) is a field of interdisciplinary academic study that seeks to understand “planned change processes and to assess the effects of efforts to promote social change” (Alderfer 1977). It is broadly defined as a program of planned interventions that aims to improve the internal operations of the organization by opening up communication, by decreasing internal destructiveness, such as win-lose conflicts, and by increasing creativity in problem solving (Berry and Houston 1993). Early works conceptualized OD as either change in structure and processes or change in individual employees (of. Dalton 1965; Campbell and Dunnette 1968). Recent research adopts a multiple perspective of changing the structure, processes, technology and people as well as their interactions because the early works demonstrated the inadequacy of changing only one of them (Friedlander and Brown 1974). OD encompasses several key activities and assumptions. One is planned interventions, which includes diagnosis of the need for change, design of a plan of action and validation of outcomes. Another is the implementation of organization-wide programs, although aspects of the OD will focus on particular departments or groups. Consequently, King and Anderson (2002) note that OD generates multiple outcomes such as at the organizational, group, and individual levels. As noted earlier, this research focuses on a specific OD effort in a manufacturing firm: cultivate a firm that embraces introduction of new products 53 as its means to grow its sales. At the heart of the analysis lie the changes to the firm’s NPD processes and their effects on the firm’s market and financial outcomes. In the marketing literature, Day (1994) describes six elements in a comprehensive market-oriented change program. First is the diagnosis of current capabilities. Next is the anticipation of future needs for capabilities in light of the strategy for creating customer value. Third is the bottom-up redesign, which is based on the formation of teams responsible for continuous improvement or radical re-design of underlying processes. Fourth is a top-down re-direction from senior managers, who demonstrate a clear, continuing commitment to putting customers first. Fifth is the use of information technology to enable the organization to do things it could not do before. The final element is the monitoring of progress toward improvement targets. Day (1994, p. 49) also suggests that “this change program must be undertaken in conjunction with other actions aimed at enhancing a market orientation. Indeed, the market sensing, customer linking, and channel bonding capabilities cannot be nurtured or productively utilized without concurrent attention to the values, beliefs and behaviors of the members of the organization and being supported by changes in the organization structure, system, control, incentives, and decision processes.” 3.2.2 Strategic Planning and Goal Setting: An Organizational Change Theory in Teleological School of Thought Based on an interdisciplinary literature review, Van de Ven and Poole (1995) classify organizational development and change theories into four families: life-cycle, evolution, dialectic and teleology. Strategic planning and goal 54 setting theories (Chakravarthy and Lorange 1991) are among the examples of process theories founded in the teleology school of thought. In this school of thought, development of an organization proceeds toward a goal or an end date. The assumptions in teleological theories are that the organization going through changes is adaptive and purposeful; it constructs an envisioned end state, takes action to accomplish the task to arrive at the end state and monitors the progress. Another assumption in teleological theory is that organization’s environment and resources limits what can be accomplished. ln teleological theories, development is viewed as a “repetitive sequence of goal formulation, implementation, evaluation, and modification of goals based on what was learned” (Van de Ven and Poole 1995). One important aspect of OD via strategic planning and goal setting is that this theory does not prescribe a necessary trajectory of development the organization must follow to reach the end state. 3.2.3 New Product Strategy Implementation Although the necessary condition for the success of a strategy, strategy formulation has been thoroughly investigated for years and still is an active research area, there is limited investigation on the sufficient condition to achieve the desired ends: implementation. However, marketing strategies can be imitable by competition making implementation proficiency a crucial determinant of success. Firms that can out-execute its competitors in the implementation of a common strategy may be able to reap the expected profits (Cespedes 1991). 55 Implementation of marketing strategy can be defined as the “how-to-do-it” aspects like organizational issues and implementation of specific programs. To date, scholars have argued that broadly two sets of variables improve or hinder marketing strategy implementation. These are structural variables such as a firm’s control systems and policies (e.g., Bonoma 1984) and behavioral factors of organization members responsible for executing a strategy. Unfortunately, there is very limited research that investigates the effect of structural and behavioral variables on implementation outcomes. 3.2.4 NPD Team Performance Noble and Mokwa (1999) argue that implementation success can be both at the individual or firm level. However, for marketing strategy implementations where the results are neither an utter failure nor a great success, the picture might be more complex. For example, in the case of NPD strategy implementation with the use of formalized processes, cross-functional teams and designated team leaders, success can be determined in more granular form by measuring the NPD team performance project by project. This is also helps understand issues more accurately and in more detail. For example, success perception of a strategy implementation may take a long time to be formed in manager or employee cognitions, or may be confounded by the success or failure of a specific project. By adopting a project level perspective and investigating the NPD team performance outcomes, varying success levels of a particular strategy implementation can be observed and the reasons for achieving different performance outcomes can be identified. As such, in this 56 dissertation, implementation success is judged by the success of NPD teams in several dimensions such as sales obtained from the new product developed, return on investment (ROI), product quality, adherence to budgets and adherence to schedules. Another important outcome by which an NPD team’s performance can be measured is the quality of the product developed. Product technical quality refers to the technical performance and reliability of the product relative to specifications (Tatikonda and Montoya-Weiss 2001). Although this definition of product quality portrays more of an engineering perspective on quality, it does represent a major portion of other definitions proposed in the literature. For example, Urban and Hauser (1993) note that quality is the extent that each feature of the new product is designed at satisfying customers. Assuming that the specifications of the new product are correctly specified to fulfill customer needs, the Tatikonda and Montoya-Weiss’ (2001) technical definition of product quality will be synonymous as Urban and Hauser’s (1993) more customer oriented definition. Proficient execution of technical activities such as laboratory testing to assure that the new product under development meets the performance expectations and is at the desired quality level. Moreover, superior marketing will lead to better understanding of customer expectations, which would lead to accurately defining the specifications of the new product. Alpha Corporation has established a brand that is associated with quality product. in fact, the firm stands behind its product by providing warranties. As a result, although speed-to-market is perceived important, maintaining the quality 57 image is perceived as a higher priority by the top management. In integrative model of the NPD process (of. Calantone and Di Benedetto 1988; Calantone, Schmidt, and Song 1996), product quality also plays an integral part. A product development process that adequately incorporates the voice of the customer (Griffin and Hauser 1993) can enhance new product success by generating word of mouth for consumer products or a reputation for industrial products. Studies provide strong evidence that relative quality of new products is a major determinant in new product success, and subsequently, firm performance (Calantone and Cooper 1981; Cooper 1992; Jacobson and Aaker 1987). 3.3 HYPOTHESES The theoretical model that depicts the hypotheses that are elucidated below are shown in Figure 3-1. 3.3.1 Proficiency of NPD Phase Activities on NPD Team Performance In NPD literature, the success of an NPD project is attributed to the proficiency of conducting NPD activities. Using the above reasoning, there are numerous studies that investigate the adequate implementation of these activities on new product performance. Song and Parry (1994; 1996) provide evidence that technical activities are among the important NPD success factors both in Japan and China. Calantone and Di Benedetto (1988) demonstrate the role of marketing, technical and launch activities on the success of the new product as well as the effect of technical activities on the quality of the product developed. Moreover, Calantone, Schmidt, and Song (1996) use a similar model 58 .02.... .000 000.92.. 020.3000 200. 00020 20200202202200 «0 020 2202.20.02.00 3 50000020 «0 .0 0020.020... 02 0 "0.2.00: .00-.022 .0.-.0022 2.02 02. .0. 00200200 0300200 020 00200200 .0003 020 2000022 2.02 02. .202 002.030 2.2000 80.0020. 20... 00.00 02. 00022.2. 52.20.20.800 .0000. E00. 012 000.0202 020 000022 20.200202020022200 «0 020 220.220.0000 «2 50000020 «0 2.. 00.020.02.022 20.3090 200. 02. 2000202 020.200.20.22 00.2.00: 200.022 .0000... 2.02 02. .0. 00200200 0300200 020 00200200 .0003 020 20200.2 2.02 02. 2.02 002.030 2.2020 00:00.2 .2002 00.00 02. 02.0.2.2. 00020200022. .0000... .0 .000. 202022.. 020 000022 20.200220202022200 «0 020 220.220.0000 3 50000020 «0 2.. 00.020.020.02 20.22000 .00. 02. 2000202 020200202. “0.2.0.0: .0200. 9.2.0 00200200 0300200 020 00200200 .0003 020 2000022 0002 02. .202 00220.20 5.200... .0200... .200. 00.00 02. 000.22... 5202920297000... .200. Q22 00200202 02. 020 000022 20200202020228 «0 020 220.220.0000 3 500000.... «0 2.. 00202062022 20.20000 .00. 02. 2000202 020200202. 66.0%. .0200. QQZ .0 00200200 0300200 020 00200200 200002 020 20000.2 2.02 02. 2202 002.030 5.2000 .0200... .200. 00.00 02. 000222... 00022 20200202202200 «0 020 020.220.0000 «2 50500.0 0 0 .0 0020202022 20.20093 00.2.00: 0x00.- 0002.. 22.020.002.600 000202.052... .0000... 00.00020w 0. 0020.020< - 2.02 .0 .0>0.. 0.0.022. 0.0003 0. 0020.020... - 00.00..- 00020. .20E00_0>0o 2.000 .0000... 0.02 - .0.. .0000... 2.02 - 02000 00020. 50500.0 00.0w .0000... 2.02 - 52.20.20.000... 0000.. 2.00.. 00.2 00200.00. 32020202000005 2.00.- 002 00200.00. 520.305 020 0.2000. 00000.0. 000203.023 E00 .- oaz 0.0.000. 020.000.2090 00 20.5003 020.200.... >020.0..2m_ 020 000202.003. 2.00 .- 002 20 0.0.0.0005. 020 00.020620... 20.50006 0.00... 00.2 .o .002”. 02... 0-0 0..-.0.”. 59 and demonstrate the different effect sizes of proficient marketing and technical NPD activities on new product success in the United States and China. In sum, extant literature groups NPD activities based on functional dimensions of marketing and R&D. However, when firms implement team-based NPD processes, each team will be comprised of core members with expertise in domains other than R&D, engineering and marketing, but also others such as finance and purchasing. In line with this reasoning, in this study, the tasks are not grouped into marketing and technical activities, but rather grouped into discovery, development and commercialization phase tasks. Grouping NPD activities into three phases is also consistent with PDMA’s conceptualization of NPD process. For example, studies find that poor initial definition of the new product, an activity in the discovery phase is the number one reason for development delays (Gupta and Wilemon 1990). Formalization of activities help generate more ideas (Troy Szymanski, and Varadarajan 2001), leading to higher product quality. Commercialization phase activities include both the technical launch activities such as initial runs and production scale-up as well as marketing launch activities such as creation of the launch plan, promotion materials, pricing decisions and training of the sales force. Hence, proficient execution of initial runs would enable detection of any production-oriented quality issues, which would also result in containment of project budgets. Moreover, proficient execution of marketing programs would enhance a new product’s sales. Further, proficient promotion formulations would lead to attainment of expected sales 60 prices, limiting selling price-oriented promotions yielding higher returns on the project. Finally, overall, proficient execution of these activities would reduce the time lost in fixing errors. More formally: H1 3, b, c: Execution proficiency of a) discovery, b) development, and c) commercialization phase improve the sales, FiOI, product quality obtained from the new product, and budget adherence and schedule adherence of NPD teams. 3.3.2 The Moderating Role of Perceived Cross-Functionality in Organizing for NPD Until the mid 19703 the three key generic approaches to coordination in organizations were standardization or rules, plans and schedules, and mutual adjustment (March and Simon 1958; Thompson 1967). The more recent approach is the use of teams in organizations (Van de Ven, Delbecq, and Koenig 1976). Historically, innovative groups were separated from the organization (Beer and Walton 1987). This practice started to change with Kanter (1983) who suggested that this separation may be counterproductive to innovation. This idea was further supported by Rubenstein and Woodman (1984) who noted that boundary-spanners such as salespeople should participate in the innovation process by bringing in customer information to the firm. Tushman and Katz (1980) expanded this argument by arguing that everybody involved in innovation activities should act like boundary spanners. Cross-functional teams consist of members from different functional areas including research disciplines such as chemistry, biology, electronics or 61 metallurgy, engineering, manufacturing, marketing and purchasing. Among the advantages of utilizing cross-functional teams are bringing multiple sources of knowledge, information and perspectives to the project; wider variety of contacts outside the project group; and consideration of downstream concerns (e.g., manufacturability) in upstream design (Keller 2001). Cross-functional teams include members that are both permanent and temporary. The permanent members form the ‘core team’, which is defined as a limited group of people responsible for the interfunctional management and coordination of management activities (Moenaert et al. 2000). Although earlier studies refer to cross-functional team use as implementing a core team with members from different functions and this team remains in place, stable and actively participates through product introduction (e.g., Griffin 1997a), this study assumes that an NPD team is cross-functional even when a core team might be formed at early stages of the NPD process such as concept generation and preliminary evaluation, but the active participation of some of the members might commence once the fuzzy front end activities are accomplished. Henceforth, the degree of cross-functionality refers to the extent of the different functional representations in an NPD team, whose members are involved in attending regular team meetings, evaluating opportunities, and determining the most efficient way of creating a new product or refining existing product (cf. Brown and Eisenhardt 1995, Ulrich and Eppinger 1995, and Lee and Chen 2007 for similar definitions). 62 Although using cross functional teams has been increasing in the last decades (Larson and Gobeli 1988; Henke, Krachenberg, and Lyons 1993; Page 1993; McDonough 2000) there is inconsistency in the literature on the influence of employing cross-functional teams in facilitating performance. For example, the quality of cross-functional communication is positively correlated with new product success (Rothwell et al. 1974; Cooper 1979; Zirger and Maidique 1990; Griffin and Hauser 1992). integration at the R&D/marketing interface is important to new product success (Gupta, Raj, and Wilemon 1986; Souder 1987). Further, McDonough’s (2000) study suggest a positive impact of use of cross-functional teams on performance, but Ancona and Caldwell (1992a) find that use of NPD teams hinder performance. In her study of firms in the telecommunications industry, Barczak (1995) finds that firms with high new product performance use cross-functional teams in higher proportion of their projects than firms with low new product performance. More importantly, in their meta-analysis on new product success Henard and Szymanski (2001) demonstrate that cross- functional integration, defined in their study as the degree of multiple-department participation in a new product initiative, is not among the significant drivers of performance. To resolve these inconsistencies in results, a moderating effect of cross- functionality on the relationship between marketing and technical NPD activities and performance are hypothesized. As the level of uncertainty increases, information theory suggests that the need for information increases (Gupta, Raj, and Wilemon 1986; Olson, Walker, and Ruekert 1995). So, a multifunctional 63 team will have access to more and novel information (Granovetter 1973), thereby influencing outcomes at the same skill level. Thus: Hza,b,c: Interactions between the task execution proficiencies in a) discovery, b) development, and c) commercialization phases and the perceived NPD team cross-functionality improve the sales, ROI, product quality obtained from the new product, and budget adherence and schedule adherence of NPD teams. 3.3.3 The Moderating Role of the Project Leader Groups of individuals act differently under the presence and close monitoring of a leader. For example, Lewin, Lippit, and White (1939) find that subordinates under democratic leadership styles act normal even when the leader is not present, whereas they start fighting for power when autocratic leader is not present during a decision making process. As such, an NPD team leader being with the group during decision making from discovery phase through commercialization should alter the influence of task execution proficiency on outcomes. Moreover, NPD team leaders significantly influence NPD success, especially through maintaining project continuity and integrating functions as well as external parties involved in the project (Moenaert et al. 2000). Moreover, team leaders that define task boundaries within which team members should operate improve the likelihood of their NPD project’s success (McDonough and Leifer 1986). Past research suggests that NPD process itself is shaped by the degree of uncertainty managers perceive (Song and Parry 1994; 1997). So, existence of a well-defined manager that stays throughout the whole project should alter the influence of activities on outcomes. Therefore: H3a,b, c: Interactions between the task execution proficiencies in a) discovery, b) development, and c) commercialization phases and perceived NPD team leader accountability improve the sales, ROI, product quality obtained from the new product, and budget adherence and schedule adherence of NPD teams. 3.3.4 The Moderating Role of Inherent Product lnnovativeness Product innovativeness taxonomies include the dimensions which reflect either perspective of the developing firm of the consumers. From the perspective of the firm, product innovativeness represents the level of discontinuity that product generates in the marketing and/or technological process (Garcia and Calantone 2002). Therefore, at a more micro level, within the firm, product newness can be seen as a reflection of the amount of the relevant experience NPD team members can draw on when working on a new product project (Olson, Walker, and Ruekert 1995). In the same vein, inherent level of product innovativeness represents the expected pressure that a new product project will exert on the firm’s existing marketing and technological resources and skills, knowledge and even strategy in all phases of NPD. When inherent product innovativeness is high, then NPD teams will have to cope with more uncertainties from start to finish than when inherent innovativeness is low. Ettlie, Bridges, and O’Keefe (1984) argue that “strategy-structure causal sequence is differentiated by radical versus incremental innovation”. 65 Consequently, critical development activities differ for really new versus incremental products (Song and Montoya-Weiss 1998). For incremental products, the NPD process steps are more clearly and rigidly defined. Improvement in the proficiency of task execution is derived more from past experience and existing knowledge base than in the case of highly innovative products. For really new products, there is more ambiguity in the process steps. Execution of activities will still improve as team will have more freedom to form novel combinations of ideas, leading to novel insights with subsequent execution of these activities, but the strength of the relationship between task execution proficiency to performance will be altered. Exploitation generates clearer, earlier and closer feedback than exploration (Levinthal 1997), which is done during really new NPD projects. So, the effect of proficiency of execution of activities for incremental innovations should be less strong than for radical new product projects. Hence the hypotheses below: H4a, b, c: Interactions between the NPD task execution proficiencies in a) discovery, b) development, and c) commercialization phases and inherent level of product innovativeness improve the sales, ROI, product quality obtained from the new product, and budget adherence and schedule adherence of NPD teams. 3.3.5 Improvement in NPD Task Execution Proficiencies over Time: Team Learning by Doing Proficiencies of carrying out certain tasks improve as the performer acquires more experience executing them. The underlying notion that explains 66 how experience enables to do things better is learning. Individuals learn, and so do groups of individuals in firms. These groups of individuals might be the entire employee base from top management to worker bees. As such, similar to theories that explain organizational learning (OL) borrowing and building from theories on individual learning, theories on team learning borrow from both individual and firm level conceptualizations. New product “teams learn as their members work toward common goals” (Meyers and Wilemon 1989, p. 81). Team learning can be defined as members acquiring and sharing new skills and knowledge, followed by examination of these skills and knowledge, and finally generating solutions to what is helping or hurting team performance (Druskat and Kayes 2000). One of the virtues of OL is the tendency for organizations to become more proficient at their current activities with experience (Alchian 1959; Leonard-Barton 1995). 0L is a three stage process that includes information acquisition, information dissemination and shared interpretation (Sinkula 1994). One of the ways of information acquisition, which is the initial stage of OL, is through direct experience (Slater and Narver 1995). Applying these organization level concepts to team level results in the expectation that the more NPD teams perform NPD process tasks, the more they should get proficient at performing them. In fact, tasks will be carried out “at a higher level of sophistication or with a higher level of understanding” each time team members return to them (Hitchcock and Willard 1995, p. 52). That is, once certain tasks are completed, team members 67 will learn from their mistakes or figure out ways of doing those actions faster and better when they are at the same phase of a future project. In stable environments, mature markets such as the construction materials manufacturing industry, codification of past experiences is called wisdom (Levinthal 1997, p. 167), and the more past experiences are codified the better same or similar tasks will be performed in subsequent attempts. This is due to being aware of the associations and conclusions about the effectiveness of past actions when these actions are codified. Arrow (1962) argues that a major determinant of increasing productivity can be explained with one variable: time. Therefore, NPD teams would get more proficient at their tasks as time goes by due to increasing learning that occurs during NPD phases. Consequently: H5a,b,c: The proficiency of a) discovery, b) development, and c) commercialization phase task executions improve over time. 3.4 METHOD 3.4.1 Research Site: Organizational Change at Alpha Corporation The CD examined in this study takes place at a strategic business unit (SBU) of an international manufacturing conglomerate. To maintain the confidentiality of the firm’s identity, it will be referred as “Alpha Corporation” hereafter. The conglomerate recently abandoned its strategy of growing via acquisitions and mandated its SBUs to adopt organic growth strategies. The observation of Alpha Corporation’s transformation from being an ad hoc incremental innovator to a firm that is committed to growing via new products after they completed the first three stages of a market-oriented OD: diagnosis of 68 current capabilities, anticipation of future needs for capabilities, and re-design of its organization structure and NPD process. Upon receiving the corporate mandate of top line growth, the Alpha Corporation executives first undertook the diagnosis activity. The examination of strengths and weaknesses revealed two important facts. Currently, Alpha Corporation is the market leader in many of its target market segments and has a reputation of high quality products. However, competition from Asian firms is threatening their dominant market share and has the potential to erode profit margins in the future as price is the dominant attribute of customer choice. Alpha Corporation executives decided to focus on the segments where they have leading market share and maintain margins by designing new products for which they could charge premiums. Next, an innovation strategy was formulated based on the market future and Alpha Corporation’s reputation for quality. Since price is the basis of competition faced from Asian firms, the executives determined that a successful NPD program should not only yield acceptable solutions to expressed customer needs, but also result in offerings that serve customers’ latent needs. In this way, new products could be sold at higher price points, thus maintaining the current margins of their existing products. Two steps were taken for identifying latent needs of customers. First, a ‘customer is king’ culture was encouraged. To support this perspective, an organizational chart change was implemented and a new job title has been created to improve conveyance of accurate and richer customer and market 69 information from the end user to the marketing department. Individuals in these positions facilitate the transfer of customer information from sales personnel to the corporate marketing employees who are engaged in NPD activities. Second, in order to turn this more enhanced market information into high quality products, Alpha Corporation’s NPD process was amended by adding stages to the fuzzy front end. These stages not only improve product definition activities, but also ensure the fit of the new product in the firm’s innovation strategy. The NPD process1 implemented at Alpha Corporation is now a staged and gated process (Cooper 1993). Moreover, this stage gate system resembles what Cooper (1994) refers to as a ‘third-generation’ NPD process. Third generation stage-gate processes are identified as encompassing fluid stages, fuzzy and conditional gates focused on project fit in the broader project portfolio and flexible in that each project has its own routing through the process. The new stage gate process at Alpha Corporation encompasses all of these characteristics except flexibility. Currently, a rigid routing of the stages with fuzzy and conditional gates is being enforced. Another aspect of the new NPD process is the use of cross-functional teams. Cross-functional team use is defined as implementing a core team with members from different functions where this core team remains in place and actively participates through product introduction (Griffin 1997b). At Alpha Corporation, the core team generally consists of product managers and engineering personnel. To facilitate communication between marketing and ‘ It is not an innovation process as idea generation is partly determined by the “product roads’. 7O engineering personnel in these NPD teams, the marketing and product development departments are integrated and co-located on the same floor at the corporate office. Past research suggests a positive influence of co-location on product development activities (Rafii 1995). The use of information technologies enables firms to do things they could not do before (Day 1994). The next set of changes executives at Manufacturing Corporation plan to implement are organization-wide information systems and collaborative design tools to further support NPD activities. In sum, Alpha Corporation has gone though all the steps identified by OD researchers to reach the desired end. Moreover, day-to-day senior management involvement in implementing the new process and emphasizing customer needs has helped the firm make great strides toward becoming a market-driven organization (Day 1994). In addition, the firm is improving its NPD processes on the three fronts of discovery, development, and launch activities. As a result, one would expect that employee perceptions of the changes will show an increase in implementation success, and ultimately lead to improved profit performance. 3.4.2 Sample and Data Collection Procedures Lawler, Nadler, and Mirvis (1983) indicate that “the only way to capture the change program and its effects is to assess the program before, during and after the intervention.” Moreover, longitudinal research conducted in organizations provides evidence that the effects of change efforts may appear long after the initiation of the change program (Likert 1961 ). Finally, the change program may first affect the employees and their attitude toward the organization and their 71 skills, resulting in a lag for improvements in firm’s profitability (Mirvis and Lawler 1977). Consequently, the most appropriate method of assessing the success of change is analysis of longitudinal data. Everyone involved in NPD has been invited for participation in the survey. While the first two surveys were administered at a specific time at the company headquarters, the third one was an on-line survey where participants were given a link to the survey site in an e-mail sent by a senior manager. Consistent with Ancona and Caldwell (1990), the members were asked to report their perceptions of the team’s handling of task-related group processes. In order to increase motivation to cooperate without fear of reprisals, participants were informed that their responses would remain anonymous (Huber and Power 1985). All information in this dissertation has therefore been sanitized to ensure confidentiality due to these provisions promised to the participating company and its employees. There were three rounds of data collection in about a year and a halftime frame. The NPD process at Alpha Corporation takes approximately 15 to 18 months for a really new product and three months for line extensions. As such, data collection with a minimum six month intervals enabled me to capture the entire development processes of new products with varying degrees of innovativeness. For the dependent variables of team performance and product quality, the list of projects considered by study participants were sent to a senior director of product development. This senior director, along with two aids who were 72 extensively involved in NPD efforts of the firm, provided the objective company data on these projects. In order to motivate senior managers to provide as much of the project and product performance data as possible, they were promised that the resulting publications will not be explicitly linked to their company or any product or brand name they offer. 3.4.3 Respondents and Demographics There were forty-four participants in the first, forty-six participants in the second, and only eighteen participants in the third survey administration. The company contact who sent employees the announcement e-mails for the survey administration time and location as well as the link for the web survey in the third round told us that he contacted the same 75 employees for each round. This yields a response rate of, 58.7% for the first, 61.3% forthe second, and 24.0% for the third round. The reason for a low turn out in the third administration might be attributed to the non-presence of a researcher and a dedicated administration time. Also, it can be attributed to new process implementation and survey fatigue. Overall, two surveys were eliminated: one due to missing data, and the other due to inappropriateness as the responded noted “Not directly involved/quite familiar with”. While the total number of surveys from all rounds of data collection was 107, only eight people participated in all three rounds. It is also worthwhile to note that thirty-five of the respondents participated in both the first and second administration of the survey. Table 3-1 summarizes the characteristics of the sample. 73 Table 3-1 Respondent Participation in Data Collection Rounds Also participated in the... Those participating in the... 1St Round 2nd Round 3rd Round 1St Round 44 2nd Round 35 46 3'd Round 11 9 ‘ 17 *Total number of participants in each round is on the diagonal. Although the first two rounds of data collection were done via on-site surveys where the researcher was present to answer any questions that respondents might have, company management requested that a third on-site survey administration is not done due to lack of time for majority of potential respondents to come to the survey administration location. It was therefore agreed that a web survey is to be done for the third round. The survey had questions regarding respondent’s background and experience. Overall, of the informants providing their education level, 75.4% have a college degree and 24.6% have a master’s degree or higher, indicating that most of the respondents were highly educated. Table 32 presents respondents’ tenure and experience range, average tenure and standard deviation in their current job, firm, industry and in NPD for each round of data 74 .0... 0.. .00. 0.0 .0... 0.. .0... 0.. 0.0-0.0 0.0-0.0 0.00-0.0 0.00-0.0 0000.09... 9.2 .00.. 0... .0... 0 0... .0.... 0.0. .20 0 0.... 000.20 0.00.0.0 0.01.... 0.00.20 0020.09... 2.0005 .0.... 0... .00. 0.0 .0.... N... .00. 0.. 0.8-00 0.00-0.0 0. .0.-0.. 0.00-00 0.2.0. 2.... .00. 0.0 .00. 0.0 .0.... .0 .00. 0... 0.0.0.- .0 0.00-0... 0.0.0.. 0.0.-..0 0.2.0. 00. 20.50 00.502 5. 0.502 00 0.502 20 0.502 0. 000.02 __< 0.502 00 0.502 20 0.502 a. 020202.00 0.0.20.0 020 0200.2 0020.0 Eu... .00"... .31... 0.00.. 2.. 00202090 020 0.320... 20028000 .0. 0000006 02.020000 «-0 0.20... 75 collection as well as cumulative descriptive statistics across all three rounds. Overall, the average length of the respondents’ job tenure at Alpha Corporation is 8.4 years, while professional tenure (years in industry) is 14.2 years. Moreover, on average, the respondents had 7.9 years of NPD experience, with high variability as indicated by a coefficient of variation of 1.0. As seen in the table, the respondents represent both novices and veterans in the positions, firm, industry and NPD activities. Table 3-3 shows the proportion of the functional departments represented in the surveys. 4.9% of the respondents were from the sales and manufacturing department each in the first round. These proportions almost reduced to half in the second round with a 2.4% representation of the two departments eaCh- Sales were represented by 6.7% and manufacturing by 13.3% in the last round of data collection. Respondents’ proportions from the marketing and product development department are consistent across the first two rounds, comprising approximately 32% of respondents. This figure jumps to two thirds of the respondents in the third round. Other departments primarily consist of purchasing, packaging, finance, and supply chain. The employees from these departments are represented by a total of 58.5%, 61.9%, and 13.3% in the first, second and third rounds, respectively. 76 Table 3-3 Functional Representation of the Respondents Functional Background 1St Round 2"d Round 30 Round Marketing a Product Dev. 317% 333% 667% Sales 49% 2.4% 6.7% Manufacturing 49% 24% 13.3% Other 585% 61 .9% 13.3% 3.4.4 Measures 3.4-4.1 NPD Team Performance In extant literature, innovation team performance is frequently operationalized as composed of constructs capturing effectiveness and efficiency of teams (of. Ancona and Caldwell 1992a; Henderson and Lee 1992; Leonard- Barton and Sinha 1993; Faraj and Sproull 2000; Hoegl and Gemuenden 2001). While effectiveness reflects a comparison of actual versus intended outcomes, efficiency measures on a comparison of actual versus intended inputs (Hoegl and Gemuenden 2001). Accordingly and based on the argument that effectiveness should be directly related to an innovation team’s task (Goodman, Ravlin, and Schminke 1987), NPD team effectiveness is usually operationalized as the degree is which the team meets expectations regarding the new product developed, i.e., the output of an NPD team. In this dissertation the output of an NPD team is measured by three variables: new product sales, new product’s return on investment, and new product’s quality. These measures are consistent with what has been suggested in previous literature by many scholars (of. Griffin 77 and Hauser 1996 for ROI, Clark and Wheelwright 1993 for product quality referring to the extent of the new product’s technical function). The other dimension of team performance, efficiency, is regularly measured by items such as project cost (or budget) and time-to-completion (e.g., meeting schedules for starting the full scale manufacturing), i.e., capturing the process outcomes of the NPD process. Accordingly, in this dissertation adherence to budgets and adherence to schedules are obtained for each new product project. It is worthwhile to note that using single-item measures would not reduce the predictive validity, since the performance items measured are constructs that consist of a single identifiable object different from the other. That is, although one can combine all these performance dimensions to form one composite measure of NPD team performance, no one can deny the conceptual difference between a new product’s quality and the time it takes to market this new product from ideation. Recent research suggests that since predictive validity will not be reduced for these kinds of items, single-items should be used in their operationalization in order to avoid the risk of additional items’ tapping into an undesired attribute (Bergkvist and Rossiter 2007). Self-report measures are criticized because at least some employees are unable to report their or their team’s performance accurately due to poor introspection (Locke, Latham, and Erez 1988). Therefore, in order to avoid self- evaluation bias, three executive managers were asked to compile objective company data and reported the percentage that each project was above or below 78 budget in terms of cost, whether the team was on schedule in the process, sales as well as ROI and product quality. This is in line with the extant literature where more objective ratings are suggested in performance measures such as sales as a percentage of forecasts (Clark and Fujimoto 1987). 3.4.4.2 Antecedents of NPD Team Performance Where possible and appropriate, existing measures were used for the variables. Measurement items for the antecedent variables are mostly adapted from Cooper (1993) and Cooper and Kleinschmidt (1986) and all items were anchored from O=very poor to 10=excellent. For the dependent variable of new product success company officials will be asked to provide objective results. These results will be sales of the new product, return on assets, profit margin and return on investment relative to its stated objective (as a true %). Consequently, this study differs from previous studies in marketing where these items have previously been used as subjective measures of new product performance (cf. Moorman 1995). As such, common method variance was alleviated by obtaining objective performance figures. In this dissertation, proficiencies of task execution in NPD process phases are formulated as formative as opposed to reflective constructs. This is deemed appropriated as the underlying items measure the proficiency of executing different activities, which collectively form the total set of activities in an NPD phase. Moreover, these items are conceptually different enough that they do not necessarily have a high correlation. For example, items used for measuring proficiency in the discovery phase inquire about opportunity identification, first 79 rung-J. : decision to go ahead with the project, initial market appraisal, market research, and decision to move into a full-fledged development project. As can be seen, the items used are comprehensive of the tasks an NPD team would perform in the front-end of each NPD project and therefore cover the domain content when all of them are used. Deleting any one of them would transform the construct to something else because a formatively measured latent construct is determined by its indicators (Diamantopoulos and Winklhofer 2001). In conclusion, proficiencies in discovery phase are formed by five items, development phase by five items and commercialization phase by two items. Means, standard deviations, correlations between all variables are shown in Table 3-4. As can be seen, the correlations between the antecedent variables are all below 0.8, which would have indicated high collinearity (Kennedy 1985). Reliabilities for NPD task phase execution proficiency constructs are not calculated because formative measures are not subject to reliability. This is because they are formed by comprehensive coverage of the construct domain as opposed to predictions of correlation between items in that space (Bagozzi 1994) 3.4.4.3 Moderators Perceived NPD Team Cross-Functionality. To measure cross functionality of NPD teams, respondents were presented with a single item and were asked how the project was handled. This is consistent with the operationalizations of the variable in the literature (cf. Griffin 1997a). The respondents were given a binary response option: as a multidisciplinary, 80 multifunctional team, with identified members from various functions — product development, marketing, manufacturing, purchasing, etc. — as active members of the team; or as a functional effort: each function doing its part of the project, but not really as a team. Perceived NPD Team Leader Existence and Accountability. A dichotomous “Yes/No” type item was used to measure this variable, which posed the respondent “Was there a project leader, clearly identified, who led and was accountable for the project from the early stages through to launch?” Inherent Level of New Product lnnovativeness. Similar to other studies that investigated product innovativeness, (of. Song and Montoya-Weiss 1998; Atuahene-Gima and Evangelista 2000), we operationalized the inherent level of product innovativeness (lLI) as a dichotomous variable where lLl=1 when the product is a really new product, and lLl=0 when the product is an incremental modification or extension of an existing product. 3.5 ANALYSIS AND RESULTS 3.5-1 Seemingly Unrelated Regression (SURE) In this dissertation, Hypotheses 1 through 4 were tested using seemingly unrelated regression with identical regressors (SURE). SURE entails a series of endogenous variables that are considered as a group due to close conceptual relationships. These endogenous variables are therefore expected to be significantly correlated. Consequently, the unobserved disturbance (or error) terms cannot be assumed to be uncorrelated, and a series of OLS regressions 81 Eva-z move. 3X... 00.. 00.0 00.. 00. 00. 00. 00. 00. 2000300 0.002000 20 00.0 2.0 00. 00. 00. 00. 00. 200s. 00.. E00. E00. E00. .200. 500. :00. 5.00. 0.0020 8000.00.02.58 .0 00.. 500. 00. E.- 00. 00.- 00.- 0.0020 202520.00 .0 00.. 00. 00.- 00.- .2.- 0..- 00020 20.6005 .0 00.. $00. $00. :00. E00. 0020.020< 0500200 .0 00.. E00. E00. 500. 02.0.0200. .0080 .0 00.. 500. .500. 5000 8000.0 .0 00.. E00. 50 .0 00.. 00000 .0 0 0. 0 0 0 0 0 0 00_00_.0> 20000.00 02250.50 020 00:00.05 02.020000 0.0 0302. 82 will not yield efficient estimates (Pindyck and Rubinfeld 1998). As seen in Table 3-4, bivariate correlations between the outcome variables are all highly significant (p<0.01, with the only exception between ROI and schedule adherence, which is significant at p<0.05). Utilization of SURE is therefore appropriate in such a case as this, as it is a regression technique that solves a set of equations simultaneously and allows for error covariances among the equations (Zellner 1962; Greene 2003). In other words, consistent estimation of the coefficients do not require the regressors (Xh’s) and error terms (ug’s) to be uncorrelated when h¢g (Wooldridge 2002). Table 3-5 presents the SURE results obtained by using SAS software. All SURE regressions are significant at p<0.1 (except where ‘Sales’ is the dependent variable, which is significant at 10%) and the smallest adjusted R- squared is 0.21. 0.70 is the highest adjusted R-squared when the predictors are regressed against budget adherence. Hypothesis 1a posits a significant positive relationship between NPD process phase proficiencies and outcomes. As can be seen, discovery phase activities are positively related to new product sales and schedule adherence, but the coefficients are not significant. Discovery phase is negatively related to ROI, product quality and budget adherence, but these relationships are not significant. Hypothesis 1b expected that development phase activities have a positive relationship with the outcomes, and the results provide support for this hypothesis as demonstrated by positive coefficients. Moreover, the development 83 mmmtm>hm>05§ “2.50.01 ‘0 Exad 2.09005: H ‘1: o 5.030.238... .0003 800:- u 01:. 0 5.008.000.5086 u .00-0 0 BX...- . 0.x...- .220. 84 00. 0... 00. 00. 00. 00.00000 00.00.02 5!. 5.vn SVn EVQ o fiva 82005290 2200050.”. .0: 0.0 A0 0. 00.0 0: 00.0 .0.. 00.0 .0.. 00.0 00. 020..-“. 00. 00. 00.- 00. :00. 2 -. Sh . 0-0 .. 2000000005500 2. 00.- 00. .00.- 00. Q . 0..: .. 0-0 - 020252000 00. 00. 00. E00 . 00. ..= . 50 . 0-0 - 20.6005 00.- 00. 00.- 00.- 00.- Q - 0..: - 0-0 0.00.- .50 0.- 00.- E00.- 00.- :_ . 2000000005500 0..- 00. 00. .00. 00. 2 . 020500.000 00.- .00.- .0..- 0.00.- 0..- 0..: - 20.6005 500. 00. 00. 00.- 00.- 50 - 8000.00.02.58 0 0.- 00.- 00.- 00.- 00. 5.. . 0202.80.00 :00.- 00. 00. 00. 00.- 050 . 20.6005 :00. 00.- 00.- 00.- 00.- -..-o . 8000.00.02.58 0.00.- 20 0.- 00.- .00.- 00. 0-0 - 020502000 300. :00. 00. 00. 00.- 00-0 . 20.6005 00. 00.- 00.- 500.- 00.- 2000000005500 :00. E0 0. 00. :0 0. 00. 205020.00 00. 00.- 00.- 00.- 00. 20.6005 .20< 0300200 .200. 00000 30000 .0000... .02 00.00 00.00..0> .9200... 0030...; 0:20:38 000002 8000.002 0230 0-0 0300 phase activities significantly improve ROI (p<0.05), budget adherence (p<0.01) and schedule adherence (p<0.01). Hypotheses 1c argued that commercialization execution proficiency enhances new product outcomes. There is only directional support for commercialization’s effect on schedule adherence since the coefficient is positive, but insignificant. Moreover, sales, ROI, product quality and budget adherence are all hindered by proficient execution of commercialization phase tasks. However, only ROI is significantly reduced as commercialization improves. Hypothesis 2a posits that the interaction between task execution proficiency in cross-functionality positively affects outcomes. All of the coefficients between this interaction variable and NPD outcomes are positive except for sales, thereby providing directional support for this hypothesis- Moreover, the coefficients between this interaction is significant for budget adherence (p<0.05) and schedule adherence (p<0.01). Hypothesis 2b posits a positive interaction effect between development phase and cross functional team use. Unfortunately, only this interaction improves only sales while hindering all other outcomes. Moreover, the negative coefficients are significant for ROI (p<0.10), budget adherence (p<0.01), and schedule adherence (p<0.05). The third hypothesis in the second set, H2c, expected a significant positive effect of the interaction between commercialization phase task proficiency and cross- functionality on outcomes. While insignificant, coefficients between this interaction and sales, ROI, product quality, and budget adherence are all 85 negative, full support is found as demonstrated by the significant (p<0.01) and positive coefficient for schedule adherence. Hypothesis 3a argues that existence of an accountable team leader interacts with discovery phase task execution proficiency to improve outcomes. Although none of them are statistically significant, the results provide directional support for ROI, product quality and budget adherence. Moreover, this interaction hinders sales (p>0.10) and schedule adherence (p<0.01). There is also only directional support for Hypothesis 3b, as demonstrated by positive coefficients between the interaction of development phase task execution and team leader accountability and sales. For the remaining four outcome variables: ROI, product quality, budget and schedule adherence, the coefficients are negative and insignificant. Thirdly, contrary to anticipations in hypothesis 3c, the interaction between commercialization and team leader accountability hinder sales and ROI, but improves product quality and budget adherence and thus provide directional support only. The coefficient between this interaction and schedule adherence is in the expected direction and significant at 1% providing full support for the hypothesis. Fourth set of hypotheses posits a positive effect of the interactions between the task execution proficiencies in discovery, development and commercialization phases and inherent level of product innovativeness on outcomes. No support is found for the effect of the interaction between discovery and team leader accountability on outcomes as demonstrated by negative coefficients. The interaction between development execution and team leader 86 accountability marginally improve ROI (p<0.10). There is also directional support for this interaction’s effect on sales, product quality and budget adherence. Finally, the interaction of commercialization and product innovativeness is negatively related to all outcomes. This interaction significantly reduces ROI (p<0.01), budget adherence (p<0.01) and schedule adherence (p<0.05). Table 36 presents the summary findings for the hypotheses tested via SURE. 3.5-2 One-Way Multivariate Analysis of Variance (MANOVA) Hypotheses 5 a, b, and c posited that the proficiency of discovery, development, and commercialization phase task executions improve over time. Before conducting the appropriate statistical significance test, first, means and standard deviations for task execution proficiency ratings for each phase are calculated and tabulated. As seen in Table 3-7, by the end of the third round, proficiencies in all three phases are higher than where they were in the first round of data collection, which suggests that proficiencies might have improves over time. Next, these mean ratings are plotted across rounds to get a pictorial representation of the progress. Figure 3-2 shows that discovery phase task proficiency has monotonically increased from round 1 to round 3. On the other hand, development phase proficiency was almost constant in rounds 1 and 2, but has increased in the final round. Finally, although the value of commercialization task phase execution proficiency has dropped in round 2, it increased dramatically in round 3, resulting in the final round mean being higher than in round 1. 87 Table 3-6 Summary of Hypotheses Testing Hypotheses Expected Result Conclusion Dependent Variable: Sales + + Directional support H1a: Discovery Proficiency H1 b: Development Proficiency + + Directional support H1c: Commercialization Proficiency + - No support H2a: Discovery Pro. * C-F + - No support H2b: Development Pro. * C-F + + Directional support H2c: Commercialization Pro. * C-F + - No support H3a: Discovery Pro. * TLA + - No support H3b: Development Pro. * TLA + + Directional support H3c: Commercialization Pro. * TLA + - No support H4a: Discovery Pro. * ILI + - No support H4b: Development Pro. * ILI + + Directional support H4c: Commercialization Pro. * lLl + - No support Dependent Variable: ROI + - No support H1a: Discovery Proficiency H1b: Development Proficiency + +** Full support H1c: Commercialization Proficiency + -*** No support H2a: Discovery Pro. * C-F + + Directional support H2b: Development Pro. * C-F + -* No support H2c: Commercialization Pro. * C-F + - No support H3a: DiScovery Pro. * TLA + + Directional support H3b: Development Pro. * TLA + - No support H30: Commercialization Pro. * TLA + - No support H4a: Discovery Pro. * ILI + -** No support H4b: Development Pro. * ILI + +* Full support H4c: Commercialization Pro. * lLl + -*** No support 88 Table 3-6 (cont’d). Hypotheses Expected Result Conclusion Dependent Variable: Product Quality H1a: Discovery Proficiency + - No support H1b: Development Proficiency + + Directional support Htc: Commercialization Proficiency + - No support H2a: Discovery Pro. * C-F" + + Directional support H2b: Development Pro. * C-F + - No support H2c: Commercialization Pro. * C-F + - No support H3a: Discovery Pro. * TLAb + + Directional support H3b: Development Pro. * TLA + - No support H3c: Commercialization Pro. * TLA + + Directional support H4a: Discovery Pro. * ILI° + -* No support H4b: Development Pro. * ILI + + Directional support H4c: Commercialization Pro. * ILI + - No support Dependent Variable: Budget Adherence H1a: Discovery Proficiency + - No support H1 b: Development Proficiency + -*** No support H1c: Commercialization Proficiency + Directional support H2a: Discovery Pro. * C-F + +** Full support H2b: Development Pro. * C-F + +*** Full support H2c: Commercialization Pro. * C-F + - No support H3a: Discovery Pro. * TLA + + Directional support H3b: Development Pro. * TLA + - No support H30: Commercialization Pro. * TLA + - No support H4a: Discovery Pro. * ILI + +** Full support H4b: Development Pro. * lLl + Directional support H4c: Commercialization Pro. * ILI + +*** Full support 89 Table 3-6 (cont’d). Hypotheses Expected Result Conclusion Dependent Var.: Schedule Adherence H1a: Discovery Proficiency + - No support H1 b: Development Proficiency + *** No support H1c: Commercialization Proficiency + Directional support H2a: Discovery Pro. * C-F + +*** Full support H2b: Development Pro. * C-F + -** No support H2c: Commercialization Pro. * C-F + -*** No support H3a: Discovery Pro. * TLA + +** Full support H3b: Development Pro. * TLA + Directional support H3c: Commercialization Pro. * TLA + -*** No support H4a: Discovery Pro. * ILI + - No support H4b: Development Pro. * ILI + + Directional support H4c: Commercialization Pro. * ILI + ** No support *p<0. 10; “p<0.05; ***p<0.01 a C- =Cross—Funclionality b TLA: Team Leader Accountability 0 ILI: Inherent Level of Product lnnovativeness Table 3-7 Mean Ratings of NPD Phase Execution Proficiencies across Rounds Discovery Development Commercialization Phase Phase Phase Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Round 1 5.97 1.39 6.38 1.43 5.92 1.28 Round 2 6.14 1.47 6.39 1.53 5.75 0.99 Round 3 6.34 1.21 6.97 1.17 6.41 1.15 Overall 6.10 1.39 6.47 1.44 5.93 1.16 Figure 3.2 Mean Proficiency of Task Execution Rating for each NPD Phase 7.00 4 6.75 - 6.50 ~ 6.38 l- ---------- " 6.25 ~ 6.00 - 5.97 5.92 5.75 ~ 5.75 5.50 0 , Round 1 Round 2 Round 3 {—0— Discovery Phase —I - Development Phase - -A- - Commercialization Phase Since there are three dependent variables (mean task execution proficiency in each phase) and one factor (round of data collection) one-way multivariate analysis of variance (MANOVA) procedures were used to test Hypotheses 5 statistically. It is worthwhile to note that if there were no missing data, MANOVA results would have been identical to those obtained by three ANOVAs where a separate NPD phase task proficiency is the dependent variable in each. MANOVA procedure excludes data on all dependent variables for a case, if a score is missing on any one dependent variable (Green and Salkind 2005). In 91 other words, if the value for discover phase task proficiency is missing for case, say 45, then, case 45 is excluded from analysis even though development and/or commercialization proficiency values are present in the data set. For MANOVA to be valid, the variance-covariance matrices must be equal for the groups (Hair et al. 1998). As seen in Table 3—8, which presents results of Box’s test, the covariance matrices are not statistically different (p=0.21). Table 3-8 Box’s Test for NPD Phase Execution Proficiencies across Rounds Box's M 16.47 F-value 1 .29 df1 12 df2 10210.88 p-value 0.21 The software used for this analysis was SPSS. Table 3-9 presents MANOVA results. Table 3-9 MANOVA Results for NPD Phase Execution Proficiencies across 190 .45 0.03 97 .18 0.05 Hotelling's Trace 0.06 0.97 Roy's Largest Root 0.05 1.68 Rounds Hypothesis Error Partial Eta Effect Value F df df Sig. Squared Intercept Pillai's Trace 0.97 908.55 3 96 .00 0.97 Wilks' Lambda 0.03 908.55 3 96 .01 0.97 Hotelling's Trace 28.39 908.55 3 96 .02 0.97 Roy's Largest Root 28.39 908.55 3 96 .03 0.97 Round Pillai's Trace 0.06 0.98 6 194 .44 0.03 Wilks' Lambda 0.94 0.97 6 192 .44 0.03 6 3 92 In conclusion, although mean ratings have either increased or decreased for individual items, the discovery, development and commercialization phase activity execution proficiency have not changed over the past two years, after three rounds of data collection. 3.6 CONCLUSIONS The results revealed that cross-functionality interacts with discovery to enhance both budget and schedule adherence, but its interaction with commercialization facilitates only schedule adherence. Moreover, the interactions of cross-functional team use and NPD phase task proficiencies do not improve market-based performance figures of sales, ROI, and product quality. It has been almost half a century since McCarthy (1959) argued that executives should not expect to have great NPD success just because they initiate formation of committees or new product development departments in their firms and that organizing is not an important factor in NPD unless the entire firm is interested and willing to take chances to succeed in NPD. He noted that NPD can only succeed in an environment where teamwork is present for those accountable in the development of new products as well as a united effort on the part of all supporting departments of a firm. When organizations go through change, members frequently interpret the meaning of the change in relation to the functions they belong to (Dutton and Jackson 1987). As such, turf barriers as well as interpretive barriers arise (Hutt 1995), causing stress in cross-functional relationships, thereby making the 93 institutionalization of change a non-accomplishable task. This may be a viable explanation for the insignificant interaction effect of cross-functional team use and development and commercialization task proficiencies. Although NPD teams have gone through several projects, NPD phase task proficiencies have not improved, contrary to expectations. Slater and Narver (1995) argue that despite the superiority of the potential for efficiency of new processes, it may be rejected by organization members. This may be why a significant difference in task execution proficiency is not observed in any of the three NPD phases. Another explanation to insignificant improvement in task executions can be explained by use of newly acquired skills. Learning can result in behavior change and subsequently performance increases on a continuum from direct application to indirect use of the experience (Slater and Narver 1995). But there is also a third type of use; a more effective, less action-oriented in nature, which is results in increased job satisfaction and at the same time fosters dissonance with the change in place should not be ignored. It may very well be the case that proficiency in task execution would not improve unless team members reach a threshold level of satisfaction with the new processes employed. This is also supported by Mirvis and Lawler’s (1977) arguments that a change program may first affect the employees and their attitude toward the organization and their skills, resulting in a lag for improvements in firm’s profitability. Therefore, future studies may capture NPD team satisfaction as an intervening variable and if possible, continue monitoring change for a longer period. However, at least in terms mean rating, increase is observed in development and commercialization phases as opposed to almost no change in discovery phase. 95 CHAPTER 4 CONTRIBUTIONS AND FUTURE RESEARCH DIRECTIONS 4.1 INTRODUCTION The primary objective of this dissertation was to advance the . understanding of NPD team use in organizations. To achieve this objective, two separate studies related NPD team use in organizations were conducted. While the first study provided a meta-analytic review of the extant literature, the second took a more in depth look into how NPD team use and other organizational structures such as a team leader’s existence in all NPD phases can contribute to the success of the implementation of a growth strategy with new product introductions. As noted earlier, numerous extant studies present apparent contradictory results on the effects of certain antecedents on NPD team performance. Many insights are gleaned from the meta-analytic review, including the gaps in the current body of knowledge as well as what the accumulated knowledge reveals on the effects of several antecedents of NPD team performance. In the second study, the differential effects of proficient NPD task execution on market-based and process-based success measures were identified. Moreover, the interaction of these tasks with type of NPD process used, team leader existence and accountability, and degree of inherent innovativeness of a new product are related to outcomes. Finally, the changes in the proficiency of NPD phase task execution over time is observed in an effort to 96 shed some light whether learning by doing pans out in successive executions of tasks. The remainder of this chapter is organized as follows: first, theoretical contributions of the two studies are delineated. Then, managerial contributions are discussed. Finally, limitations of the study followed by future research suggestions are presented. 4.2 STUDY #1: ANTECEDENTS OF NPD TEAM PERFORMANCE: A META- ANALYTIC REVIEW 4.2-1 Theoretical Contributions This study contributes to theory in several ways. First, it extends recent meta-analysis on organizational teams (cf. Stewart 2006), by examining NPD teams only and offers specific insights. As noted previously, NPD teams differ from many types of organizational teams such as top management teams, or even from other project teams such as an enterprise resource planning software implementation team. To this end, extant literature is thoroughly searched and a theoretical framework for investigating NPD team performance is constructed. Borrowing from extant literature reviews, antecedent variables were grouped into meaningful sets to provide direction for future research. One result obtained while constructing the framework was the necessity to decompose NPD team performance into dimensions such as effectiveness, efficiency, and attitudinal outcomes. Next, correlation effect sizes of antecedents on NPD team performance are examined. Since NPD team performance is also treated as a multidimensional construct, differing effect sizes of antecedents on three types of 97 performance are observed. One insight gleaned from the meta-analysis is the low number of studies’ examination of the antecedents of overall NPD team performance. This shows the high rigor of the literature on the topic as different dimension of performance might be influenced differently by same antecedents or not influenced at all. Investigating the effects of antecedents on separate NPD team performance dimensions reflected itself in the granularity of the results obtained. For example, it was revealed that the effect size that goal clarity and job satisfaction predict NPD team overall performance, effectiveness and efficiency are consistently large across performance type. On the other hand, some factors are found to demonstrate differing effect sizes on various performance types. One such antecedent is outcome-based rewards. This variable affects overall and process-based performances much strongly than it affects market-based performance. The situation is reversed for internal communication, which has a medium effect size on market-based performance as opposed to a small effect size on process-based performance. Another interesting result was the low effect of an important variable, physical distance, on all performance types examined. However, this does not imply to discontinue using this variable in future studies as its effect might be strongly mediated by other factors. One such factor is found to be cohesion. The subsequent multivariate analysis revealed such a significant relationship. 98 4.2.2 Managerial Contributions The major managerial contribution of the meta—analytic review is finding that the any managerial level might have differential effects on different types of performance. For example, senior managers vying to improve the success of the new products their firms introduce should not rely on outcome-based rewards for an NPD team, but rather facilitate internal communication. This might be done through increasing the reach and range of the information technologies available to the team members. In terms of team leader-related variables, team leader competence has a larger effect on overall performance as opposed to a small effect size of team leader power. This suggests that extra caution should be spent to assign the most appropriate team leader for a particular NPD project. For example, an NPD team leader, with a lot of internal NPD experience might not be a good candidate for a new project that entails significant involvement of external suppliers. One interesting result is low effect size of formalized process use for overall team performance while the effect size is medium for team effectiveness and efficiency, respectively. Although there is literature hinting that formalized processes with teams does not significantly influence overall NPD team performance, scholars such as Voss (1985) also find that for new software development formal project teams do not improve other success measures such as team effectiveness. This finding suggests that implementing a formalized NPD process with the use of teams may not be in and of themselves sufficient to reap the expected benefits. The medium effect sizes found for team 99 effectiveness and efficiency may be the result of other elements present in those teams. One such factor could be the existence of a dedicated NPD team leader in all NPD phases. 4.2.3 Future Research Directions There are a limited number of studies in the literature that investigate the effects of team psychosocial traits on both team internal processes and NPD team performance. Consequently, future researchers are encouraged to include variables such as norms and beliefs in their studies. I» This study concluded that physical distance reduces cohesion, which ultimately hinders performance. However, with the advent of information technologies, the negative effect of physical distance could be reduced or reversed. Studies investigation the effect of information technology on this relationship would be helpful for firms that are increasingly employing vitual teams. 4.3 STUDY #2: A LONGITUDINAL STUDY OF ORGANIZATIONAL CHANGE BY IMPLEMENTATION OF NEW PRODUCT-DRIVEN GROWTH STRATEGY: PROFICIENCY OF PERFORMING NPD PROCESS AND THE MODERATING EFFECT OF CROSS-FUNCTIONAL TEAM USE 4.3-1 Theoretical Contributions The major and unique contribution of this study is to provide evidence on the effect of organizational factors and processes on NPD outcomes in a firm that is going through organizational change towards innovation. While examining these relationships, it also provides a deeper look into the variables that influence marketing strategy implementation. Specifically, interactions between structural variables such as use of the type of process, cross-functionality and the 100 existence of a team leader and behavioral factors such as phase task execution skills are examined. For example, while cross-functionality does not directly enhance performance, it interacts with discovery to enhance both budget and schedule adherence, and it interacts with commercialization to facilitate schedule adherence. As such, phase task proficiency interacts with cross-functionality to enhance process-based outcomes. Another contribution of this study was to find evidence that development task proficiency improves ROI, budget and schedule adherence. As these were the only significant direct effects of NPD phase execution proficiencies, the results suggest that well performing NPD tasks do not influence outcomes, but would act as a necessary condition for the effect of other factors to be revealed. This is demonstrated by many significant interaction effects found in this study. Finally, the results revealed that, contrary to expectations, NPD phase task proficiencies do not improve over time, at least for the observation period of this study. This suggests that significant learning has not occurred in NPD teams across projects. 4.3-2 Managerial Contributions This study will especially be useful for managers that are going through change in their innovation strategies and processes as it is the context of data collection. A major managerial implication of the study is that task execution proficiencies during NPD process phases have divergent effects on NPD team performance outcomes, which include both market-based and process-based 101 outcomes. For example, development activities significantly improve ROI, budget and schedule adherence, while commercialization hinders ROI. The highly significant effect on schedule adherence found for the interaction of commercialization task proficiency and team leader existence and accountability suggests that managers should ensure that team leader stays with the project throughout the whole process. However, as team leader accountability hinders schedule adherence in discovery, senior managers cultivate an environment where every team member is accountable in earlier NPD phases. 4.3.3 Future Research Directions One interesting finding of this study was to find no significant predictor of new product sales. In this study, a maximum of two year sales figures of new products were captured. New products that have been in the market for a longer time might have developed more customer base due to word of mouth reputation or ongoing advertising. Future studies therefore might consider product market duration as a moderating variable. Finally, this study found evidence that NPD phase task proficiencies do not improve over time. Schein (1990) argues that unlearning must be promoted to overcome resistance to learn new things. Future research should therefore investigate the moderating role of unlearning of old habits in marketing implementation success. 102 APPENDIX 103 CODING SHEET FOR STUDIES Author name(s) Publication status: published, unpublished, or working paper Year of publication Type of paper: empirical, conceptual, or analytic (i.e., mathematical proof such as a game theoretical paper like Cohen, Eliashberg, and Ho 1996) Sample size (usable sample size) Journal type (first tier or second tier) where: First tier: Journal of Marketing, Journal of Marketing Research, Journal of Product Innovation Management, Management Science, Organization Science, Academy of Management Journal, Administrative Science Quarterly, Journal of the Academy of Marketing Science, lEEE Transactions on Engineering Management, Information Systems Research, MIS Quarterly Second tier: Research Technology Management, PDMA Proceedings, Research Policy, R&D Management, Technovation, Industrial Marketing Management, Journal of Business and Industrial Marketing, Journal of Management, European Journal of Innovation Management, Decision Sciences, Creativity and Innovation Management, etc. Research Context Organization size: small, large, or both where small = number of employees < 500 or annual sales < $1 billion Team type: cross-functional, other, or both Team proximity: co-located, partially dispersed, or virtual Innovation type: product, process, or other Type of product developed: goods, services, and software Product innovativeness degree: incremental, radical, or both Data collection countries: United States, other, or both Technological innovation degree: high tech, low tech, or both Industry type: manufacturing, non-manufacturing, or both - “Software developers’ are assumed to be manufacturers. - ‘Telecommunications’ is services such as phone services from Verizon, etc., i.e., non-manufacturing except in Hauptman and Hirji (1996) since the authors explicitly note that in their sample, firms in the telecommunications industry are manufacturers. - ‘Capital goods’ is assumed to be products provided by financial firms, i.e. non-manufacturing. Measurement and Methods . Sampling method: random or other, e.g., deliberate . Method used for data collection: survey, case study (e.g., in-depth interviews, company records, etc.), or both . Performance/outcome measurement item plurality (e.g., team performance): single, multiple items, or both . Performance/outcome measure subjectivity: subjective, objective, or both 104 . Data temporality: longitudinal or cross-sectional . Informants: manager/senior managers, team leader, team members, or more than one type . Method(s) used for analysis: univariate only, multivariate only, or both univariate and multivariate where - univariate: frequency, means, cross tabs, etc. and - multivariate: regression, ANOVA, MANOVA, factor analysis, path analysis, SEM, correlations, etc. . Data type: primary, secondary, or both . Correlations . Reliabilities 105 REFERENCES *Akgun, Ali E. and Gary S. Lynn (2002a), “Antecedents and Consequences of Team Stability on New Product Development Performance,” Journal of Engineering and Technology Management, 19 (3-4), 263-286. *Akgun, Ali E. and Gary S. Lynn (2002b), “New Product Development Team Improvisation and Speed-to-Market: An Extended Model,” European Journal of Innovation Management, 5(3), 1 17-129. *Akgun, Ali E., Gary S. Lynn, and John C. Byrne (2006), “Antecedents and Consequences of Unlearning in New Product Development Teams,” Journal of Product Innovation Management, 23 (1), 73-88. *Akgun, Ali E., Gary S. Lynn, and Richard Reilly (2002), “Multidimensionality of Leaning in New Product Development Teams,” European Journal of Innovation Management, 5 (2), 57-72. *Akgun, Ali E., Gary S. Lynn, and Cengiz Yilmaz (2006), “Learning Process in New Product Development Teams and Effects on Product Success: A Socio- Cognitive Perspective,” Industrial Marketing Management, 35 (2), 210-224. *Akgun, Ali B, John C. Byrne, Halit Keskin, and Gary S. Lynn (2006), “T ransactive Memory System in New Product Development Teams,” IEEE Transactions on Engineering Management, 53 (1), 95-111. Alchian, Armen Albert (1959), “Cost and Output,” in Moses Abramovitz et al. (eds.). The Allocation of Economic Resources: Essays in Honor of B. F. Haley, Stanford University Press: Stanford, CA. Alderfer, Clayton P. (1977), “Organization Development,” Annual Review of Psychology, 28, 197-223. *Ancona, Deborah G. and David F. Caldwell (1992a), “Bridging the Boundary: External Activity and Performance in Organizational Teams,” Administrative Science Quarterly, 37 (4), 634-665. *Ancona, Deborah Gladstein and David F. Caldwell (1992b), “Demography and Design: Predictors of New Product Team Performance,” Organization Science, 3 (3), 321 -341. *Ancona, Deborah Gladstein and David F. Caldwell (1990), “Improving The Performance of New Product Teams,” Research Technology Management, 33(2), 25-29. 106 *Ang, Soon and Sandra A. Slaughter (2001), “Work Outcomes and Job Design for Contract versus Permanent Information Systems Professionals on Software Development Teams,” MIS Quarterly, 25 (3), 321-350. Arrow, Kenneth J. (1962), “The Economic Implications of Learning by Doing,” The Review of Economic Studies, 29 (3), 155-173. *Atuahene-Gima, Kwaku and Felicitas Evangelista (2000), “Cross-Functional Influence in New Product Development: An Exploratory Study of Marketing and R&D Perspectives,” Management Science, 46 (10), 1269-1284. Bagozzi, Richard P. (1994), “Structural Equation Models in Marketing Research: Basic Principles,” in Principles of Marketing Research, Richard P. Bagozzi, (ed.), Blackwell: Cambridge, MA, 317-384. *Barczak, Gloria and Edward F. McDonough III (2006), “Exploring the Relationship between Trust, Project Commitment and Performance in New Product Development Teams,” Working Paper. *Barczak, Gloria and Edward F. McDonough Ill (2003), “Understanding the Impact of Contexts on Factors Leading to Global Team Performance,” Working Paper. Barczak, Gloria (1995), “New Product Strategy, Structure, Process and Performance in the Telecommunications Industry,” Journal of Product Innovation Management, 12 (3), 224-234. *Barczak, Gloria and David Wilemon (1989), “Leadership Differences in New Product Development Teams,” Journal of Product Innovation Management, 6 (4), 259-267. Beer, Michael and Anna Elise Walton (1987), “Organization Change and Development,” Annual Review of Psychology, 38, 339-367. Bentler, Peter M. (2004), EOS 6 Structual Equations Program Manual, Multivariate Software Inc.: Encino, CA: Bergkvist, Lars and John R. Rossiter (2007), “The Predictive Validity of Multiple- Item versus Single-Item Measures of the Same Constructs,” Journal of Marketing Research, 44 (2), 175-185. Berry, Lilly M. and John P. Houston (1993), Psychology at Work: An Introduction to Industrial and Organizational Psychology, Brown and Benchmark: Madison, WI. 107 Bollen, Kenneth A. (1989), Structural Equations with Latent Variables, John Wiley and Sons: New York, NY. Bonoma, Thomas V. (1984), “Making Your Marketing Strategy Work,” Harvard Business Review, 62 (2), 69-76. Brown, Shona L. and Kathleen M. Eisenhardt (1995), “Product Development: Past Research, Present Findings, and Future Directions,” Academy of Management Review, 20(2), 343-378. Brown, Steven P. and Robert A. Peterson (1993), “Antecedents and Consequences of Salesperson Job Satisfaction: Meta-Analysis and Assessment of Causal Effects,” Journal of Marketing Research, 30 (1), 63-77. Calantone, Roger J. and Robert G. Cooper (1981), “New Product Scenarios: Prospects for Success,” Journal of Marketing, 45 (2), 48-60. Calantone, Roger J. and C. Anthony di Benedetto (1988), “An Integrative Model of the New Product Development Process: An Empirical Validation,” Journal of Product Innovation Management, 5 (3), 201 -215. Calantone, Roger J., C. Anthony di Benedetto, and Richard Divine (1993), “Organizational, Technical, and Marketing Antecedents for Successful New Product Development,” R&D Management, 23(4), 337-351. Calantone, Roger J., Jeffrey B. Schmidt and X. Michael Song (1996), “Controllable Factors of New Product Success: A Cross-National Comparison,” Marketing Science, 15 (4), 341 -358. Campbell, John P. and Marvin D. Dunnette (1968), “Effectiveness of T-Group Experiences in Managerial Training and Development,” Psychology Bulletin, 70, 73-104. ”Carbonell, Pilar and Ana I. Rodriguez (2006), “Designing Teams for Speedy Product Development: The Moderating Effect of Technological Complexity,” Journal of Business Research, 59 (2), 225-232. ”Carmel, Erran (1995), “Cycle Time in Packaged Software Firms,” Journal of Product Innovation Management, 12(2), 110-133. Cespedes, Frank V. (1991). Organizing and Implementing the Marketing Effort: Text and Cases, Addison-Wesley Publishing Company: Reading, MA. Chakravarthy, Balaji S. and Peter Lorange (1991), Managing the Strategy Process, Prentice Hall: Englewood Cliffs, NJ. 108 *Chen, Ming-Huei (2006), “Understanding the Benefits and Detriments of Conflict on Team Creativity Process,” Creativity and Innovation Management, 15 (1 ), 105-116. *Chen, Jiyao, Richard R. Reilly, and Gary 8. Lynn (2005), “Team Empowerment, Uncertainty, and Performance in New Product Development,” PDMA Research Forum Proceedings, 77-92. Child, John (2005), Organization: Contemporary Principles and Practice, Blackwell Publishing: Malden, MA. Clark, Kim B. and Steven C, Wheelwright (1993), Managing New Product and Process Development, The Free Press: New York, NY. Clark, Kim B. and Fujimoto, Takahiro (1987), “Overlapping Problem Solving in Product Development,” Working Paper, Harvard University, Graduate School of Business Administration, Paper No. 87-048. Clark, Kim B. and Fujimoto,'Takahiro (1991), Product Development Performance: Strategy. Organization, and Management in the World Auto Industry, Harvard Business School Press: Boston, MA. *Cohen, Morris A., Jehoshua Eliashberg, and Teck H. Ho (1997), “An Anatomy of a Decision- -Support System for Developing and Launching Line Extensions,” Journal of Marketing Research, 34 (1), 117- 129. *Cohen, Morris A., Jehoshua Eliashberg, and Teck H. Ho (2000), “An Analysis of Several New Product Performance Metrics,” Manufacturing and Service Operations Management, 2(4), 337-349. *Cohen, Morris A., Jehoshua Eliashberg, and Teck H. Ho (1996), “New Product Development: The Performance and Time-to—Market Tradeoff,” Management Science, 42 (2), 173-186. Cohen, Susan G. and Diane E. Bailey (1997), “What Makes Teams Work: Group Effectiveness Research from the Shop Floor to the Executive Suite,” Journal of Management, 23(3), 239-290- Cohen, Jacob W. (1977), Statistical Power Analysis for the Behavioral Sciences, revised edition, Academic Press: New York, NY. Cohen, JacobW. (1986), Statistical Power AnaIySIs for the Behavioral Sciences, 2"d edition, Academic Press: New York, NY. Cooper, Robert G. (1994), “Third Generation New Product Processes,” Journal of Product Innovation Management, 11 (1), 3-14. 109 Cooper, Robert G. (1993), Winning at New Products: Accelerating the Process from Idea to Launch, 2nd ed., Addison-Wesley: Reading, MA. Cooper, Robert G. (1992), “The NewProd System: The Industry Experience,” Journal of Product Innovation Management, 9 (3), 1 13-127. Cooper, Robert G. (1979), “The Dimensions of Industrial New Product Success and Failure,” Journal of Marketing, 43 (Summer), 93-103. Cooper, Robert G. and Elko J. Kleinschmidt (1986), “An Investigation into the New Product Process: Steps, Deficiencies, and Impact,” Journal of Product Innovation Management, 3 (2), 71 -85. ”Cordero, Rene, George F. Farris, and Nancy DiTomaso (1998), “Technical Professionals in Cross-functional Teams: Their Quality of Work Life,” Journal of Product Innovation Management, 15 (6), 550-663. Dalton, Melville (1965), “Managing the Managers”, Human Organization, 14, 4- 10. Day, George S. (1994), “The Capabilities of Market-Driven Organizations,” Journal of Marketing, 58 (4), 37-52. “Denison, Daniel R., Stuart L. Hart, and Joel A. Kahn (1996), “From Chimneys to Cross-Functional Teams: Developing and Validating a Diagnostic Model,” Academy of Management Journal, 39 (4), 1005-1023. Devine, Dennis J. (2002), “A Review and Integration of Classification Systems Relevant to Teams in Organizations,” Group Dynamics: Theory, Research and Practice, 6 (40), 291 -310. Diamantopoulos, Adamantios and Heidi M. Winklhofer (2001), “Index Construction with Formative Indicators: An Alternative to Scale Development,” Journal of Marketing Research, 38 (2), 269-277. Dougherty, Deborah and Trudy Heller (1994), “The Illegitimacy of Successful Product Innovation in Established Firms,” Organization Science, 5 (2), 200- 218. Druskat, Vanessa Urch and D. Christopher Kayes (2000), “Learning versus Performance in Short-Term Project Teams,” Small Group Research, 31 (3), 328-353. Dutton, Jane E. and Susan E. Jackson (1987), “Categorizing Strategic Issues: Links to Organizational Action,” Academy of Management Review, 12 (1 ), 76- 90. 110 Dwyer, Larry and Robert Mellor (1991), “Organizational Environment, New Product Process Activities, and Project Outcomes,” Journal of Product Innovation Management, 8 (1 ), 39-48. Ettlie, John E., William P. Bridges, and Robert D. O’Keefe (1984), “Organization Strategy and Structural Differences for Radical versus Incremental Innovation,” Management Science, 30(6), 682-695. *Faraj, Samer and V. Sambamurthy (2006), “Leadership of Information Systems Development Projects,” IEEE Transactions on Engineering Management, 53 (2), 238-249. *Faraj, Samer and Lee Sproull (2000), “Coordinating Expertise in Software Development Teams,” Management Science, 46 (12), 1554-1568. *Faraj, Samer, Patricia J. Guinan, and V. Sambamurthy (2003), “Control Strategies and Performance of Software Development Teams: The Mediating Role of Experienced Ambiguity,” Working Paper. *Fedor, Donald B., Soumen Ghosh, Steven D. Caldwell, Todd J. Maurer, and Vinod R. Singhal (2003), “T he Effects of Knowledge Management on Team Members' Ratings of Project Success and Impact,” Decision Sciences, 34 (3), 513-539. Friedlander, Frank and L. Dave Brown (1974), “Organization Development,” Annual Review of Psychology, 25, 313-341 . Garcia, Rosanna and Roger J. Calantone (2002), “A Critical Look at Technological Innovation Typology and lnnovativeness Terminology: A Literature Review,” Journal of Product Innovation Management, 19(2), 110- 132. Gladstein, Deborah L. (1984), “Groups in Context: A Model of Task Group Effectiveness,” Administrative Science Quarterly, 29 (4), 499-517. Glass, Gene V., Barry McGaw, and Mary Lee Smith (1981), Meta-Analysis in Social Research, Sage Publications: Beverly Hills, CA. *Gobeli, David H., Harod F. Koenig, and Iris Bechinger (1998), “Managing Conflict in Software Development Teams: A Multilevel Analysis,” Journal of Product Innovation Management, 15 (5), 423-435. Goodman, P. S., E. C. Ravlin and M. Schminke (1987), “Understanding Groups in Organizations,” Research in Organizational Behavior, 9, 121 -173. 111 Granovetter, Mark S. (1973), “The Strength of Weak Ties,” American Journal of Sociology, 78 (6), 1360-1680. Green, Samuel B. and Neil J. Salkind (2005), Using SPSS for Windows and Macintosh: Analyzing and Understanding Data, 4th ed., Prentice-Hall: Upper Saddle River, NJ. Greene, William H. (2003), Econometric Analysis, 5th ed., Prentice Hall: Upper Saddle River, NJ. “Griffin, Abbie (1997a), “The Effect of Project and Process Characteristics on Product Development Cycle Time,” Journal of Marketing Research, 34 (1 ), 24-35. Griffin, Abbie (1997b), “PDMA Research on New Product Development Practices: Updating Trends and Benchmarking Practices,” Journal of Product Innovation Management, 14 (6), 429-458. Griffin, Abbie and John R. Hauser (1996), “Integrating Mechanisms for Marketing and R&D,” Journal of Product Innovation Management, 13(3), 191-215. Griffin, Abbie and John R. Hauser (1993), “The Voice of the Customer,” Marketing Science, 12 (1), 1-27. Griffin, Abbie and John R. Hauser (1992), “Patterns of Communication among Marketing, Engineering, and Manufacturing-A Comparison between Two New Product Teams,” Management Science, 38(3), 360-373. *Guinan, Patricia J., Jay G. Cooprider, and Samer Faraj (1998), ”Enabling Software Development Team Performance during Requirements Definition: A Behavioral versus Technical Approach,” Information Systems Research, 9 (2), 101-125. Gupta, A. K. and David L. Wilemon (1990), “Accelerating the Development of Technology-Based New Products,” California Management Review, 32(2), 24-53. Gupta, A. K., S. P. Raj, and David Wilemon (1986), “A Model for Studying R&D- Market Interface in the Product Innovation Process,” Journal of Marketing, 50 (2), 7-17. Hackman, J. Richard (1987), The Design of Work Teams, in J. W. Lorsch (ed.), Handbook of Organizational Behavior, Prentice-Hall: Englewood Cliffs, NJ. 112 Hair, Joseph F. Jr., Rolph E. Anderson, Ronald L. Tatham, and William C. Black (1998), Multivariate Data Analysis, 5th ed., Prentice-Hall: Upper Saddle River, NJ. *Hauptman, Oscar and Karim K. Hirji (1996), “The influence of process concurrency on project outcomes in product development: An empirical study of cross-functional teams,” lEEE Transactions on Engineering Management, 43(2), 153-164. Henard, David H. and David M. Szymanski (2001), “Why Some New Products Are More Successful Than Others,” Journal of Marketing Research, 38 (3), 363-375. Henderson, John C. and Soonchul Lee (1992), “Managing I/S Design Teams: A Control Theories Perspective,” Management Science, 38(6), 757-777. Henke, John W., A. Richard Krachenberg, and Thomas F. Lyons (1993), “Cross- Functional Teams: Good Concept, Poor Implementation,” Journal of Product Innovation Management, 10 (3), 216-229. “Hirst, Giles and Leon Mann (2004), “A Model of R&D Leadership and Team Communication: The Relationship with Project Performance,” R&D Management, 34 (2), 147-160. Hitchcock Darcy and Marsha Willard (1995), Why Teams Can Fall and What to Do About It: Essential Tools for Anyone Implementing Self-Directed Work Teams, Irwin: Chicago, IL. *Hoegl, Martin and K. Praveen Parboteeah (2006), “Team Reflexivity in Innovative Projects,” R&D Management, 36(2), 113-125. “Hoegl, Martin and Luigi Proserpio (2004), “Team Member Proximity and Teamwork in Innovative Projects,” Research Policy, 33 (8), 1153-1165. *Hoegl, Martin and Hans Georg Gemuenden (2001), “Teamwork Quality and the Success of Innovative Projects: A Theoretical Concept and Empirical Evidence,” Organization Science, 12(4), 435-449. “Hoegl, Martin, Ernst Holger, and Luigi Proserpio (2007), “How Teamwork Matters More as Team Member Dispersion Increases,” Journal of Product Innovation Management, 24 (2), 156-165. *Hoegl, Martin, K. Praveen Parboteeah, and Charles L. Munson (2003), “T eam- LeveI Antecedents of Individuals' Knowledge Networks,” Decision Sciences, 34 (4), 741 -770. 113 *Hoegl, Martin, K. Praveen Parboteeah, and Hans Georg Gemuenden (2003), “When Teamwork Really Matters: Task lnnovativeness as a Moderator of the Teamwork-Performance Relationship in Software Development Projects,” Journal of Engineering and Technology Management, 20 (4), 281 -302. Huber, George P. and D. J. Power (1985), “Retrospective Reports of Strategic Level Managers: Guidelines for Increasing Their Accuracy,” Strategic Management Journal, 6 (2), 171 -180. Hume, David ([1739] 1977), A Treatise on Human Nature, Dutton: New York, NY. Hunter, John E. and Frank L. Schmidt (2004), Methods of Meta-Analysis: Correcting Error and Bias in Research Findings, 2"d ed., Sage Publications: Thousand Oaks, CA. Hutt, Michael D. (1995), “Cross-Functional Working Relationships in Marketing,” Journal of the Academy of Marketing Science, 23 (4), 351 -357. Ilgen, Daniel R., John R. Hollenbeck, Michael Johnson, and Dustin Jundt (2005), “Team in Organizations: From lnput-Process-Output Models to IMOI Models,” Annual Review of Psychology. 56, 517-43. Jacobson, R. and David A. Aaker (1987), “T he Strategic Role of Product Quality,” Journal of Marketing, 51 (4), 131-144. Kanter, Rosabeth M. (1983), The Change Masters: Innovation and Entrepreneurship in the American Corporation, Simon and Schuster: New York, NY. *KeIIer, Robert T. (2001), “Cross-functional Project Groups in Research and New Product Development: Diversity, Communications, Job Stress, and Outcomes,” Academy of Management Journal, 44 (3), 547-555. *KeIIer, Robert T. (1986), “Predictors of the Performance of Project Groups in R&D Organizations,” Academy of Management Journal, 29 (4), 715-726. *KeIIer, Robert T., Scott D. Julian, and Ben L. Kedia (1996), “A Multinational Study of Work Climate, Job Satisfaction, and the Productivity of R&D Teams,” lEEE Transactions on Engineering Management, 43 (1 ), 48-55. Kennedy, P. (1985), A Guide to Econometrics, MIT Press: Cambridge, MA. King, Nigel and Neil Anderson (2002), Managing Innovation and Change: A Critical Guide for Organizations, 2nd ed., Thomson: London, UK. 114 Kirca, Ahmet H., Satish Jayachandran, and William O. Bearden (2005), “Market Orientation: A Meta-Analytic Review and Assessment of Its Antecedents and Impact on Performance,” Journal of Marketing, 69 (2), 24-41. *Kratzer, Jan, Roger Th. A. J. Leenders, and Jo M. L. van Engelen (2004), “Stimulating the Potential: Creative Performance and Communication in Innovation Teams,” Creativity and Innovation Management, 13 (1 ), 63-71. Larson, Erik and David Gobeli ( 1988), “Organizing for Product Development Projects,” Journal of Product Innovation Management, 5 (3), 180-190. Lawler, Edward E. III, David A. Nadler and Philip H. Mirvis (1983), “Organizational Change and the Conduct of Assessment Research,” in Assessing Organizational Change: A Guide to Methods, Measures, and Practices, Stanley E, Seashore, Edward E. Lawler III, Philip H. Mirvis and Cortlandt Cammann (eds.), John Wiley & Sons: New York, NY. *Lee, Chuan and Wen-Jung Chen (2007), “Cross-Functionality and Charged Behavior of the New Product Development Teams in Taiwan’s Information Technology Industries,” forthcoming article in Technovation. *Leenders, Roger Th. A J., Jo M. L. van Engelen, and Jan Kratzer (2007), “Systematic Design Methods and the Creative Performance of New Product Teams: Do They Contradict or Complement Each Other?,” Journal of Product Innovation Management, 24 (2), 166-179. *Leenders, Roger Th. A. J., Jo M. L. van Engelen, and Jan Kratzer (2003), “Virtuality, Communication, and New Product Team Creativity: A Social Network Perspective,“ Journal of Engineering and Technology Management, 20, 69-92. Leonard-Barton, Dorothy and Deepak K. Sinha (1993), “Developer-User Interaction and User Satisfaction in Internal Technology Transfer,” Academy of Management Journal, 36 (5), 1 125-1 139. Leonard-Barton, Dorothy (1992), “Core Capabilities and Core Rigidities: A Paradox in Managing New Product Development,” Strategic Management Journal, 13 (Summer), 111-125. Levinthal, Daniel (1997), Three Faces of Organizational Learning: Wisdom, Inertia, and Discover, in Technological Innovation: Oversights and Foresights. Raghu Garud, Praveen Rattha Nayyar, and Zur.Baruch Shapira (eds), Cambridge University Press: New York, NY. 115 Lewin, K, R. Lippit, and R. K. White (1939), “Patterns of Aggressive Behavior in Experimentally Created ‘Social Climates’,” Journal of Social Psychology, 10, 171 -199. Likert, Resnais (1961), New Patterns of Management, McGraw-Hill: New York, NY. Lipsey, Mark W. and David B. Wilson (2001), Practical Meta-Analysis, Sage Publications: Thousand Oaks, CA. Locke, E. A., G. P. Latham, and M. Erez (1988), “T he Determinants of Goal Commitment,” Academy of Management Review, 13, 23-39. *LoveIace, Kay, Debra L. Shapiro, and Laurie R. Weingart (2001), “Maximizing Cross-Functional New Product Teams’ lnnovativeness and Constraint Adherence: A Conflict Communications Perspective,” Academy of Management Journal, 44(4), 779-793. *Lynn, Gary 8., Ali E. Akgun, and Halit Keskin (2003), “Accelerated Learning in New Product Development Teams,” European Journal of Innovation Management, 6 (4), 201-212. *Lynn, Gary 8., Richard R. Reilly, and Ali E. Akgun (2000), “Knowledge Management in New Product Teams: Practices and Outcomes,” IEEE Transactions on Engineering Management, 47 (2), 221 -231. *Lynn, Gary S.,Riichard B. Skov, and Kate D. Abel, (1999), “Practices that Support Team Learning and Their Impact on Speed to Market and New Product Success,” Journal of Product Innovation Management, 16 0, 439- 454. *Madhavan, Ravindranath and Rajiv Grover (1998), “From Embedded Knowledge to Embodied Knowledge: New Product Development as Knowledge Management,” Journal of Marketing, 62 (4), 1-12. Mankin, D., Susan G. Cohen, and T. K. Bikson (1996), Teams and Technology: Fulfilling the Promise of the New Organization. Harvard Business School Press: Boston, MA. March, James G. and Herbert A. Simon (1958), Organizations, Wiley: New York, NY. Marks, Michelle A., John E. Mathieu, and Stephen J. Zaccaro (2001), “A Temporally Based Framework and Taxonomy of Team Processes,” Academy of Management Review, 26 (3), 356-376. 116 McCarthy, E. J. (1959), “Organization for New-Product Development?,” The Journal of Business, 32 (2), 128-132. McDonough, Edward F. III (2000), “Investigation of Factors Contributing to the Success of Cross-Functional Teams,” Journal of Product Innovation Management, 17 (3), 221-235. *McDonough, Edward F. Ill (1993), “Faster New Product Development: Investigating the Effects of Technology and Characteristics of the Project Leader and Team,” Journal of Product Innovation Management, 10 (3), 241 - 250. *McDonough, Edward F. III and Gloria Barczak (1992), “The Effects of Cognitive Problem-Solving Orientation and Technological Familiarity on Faster New Product Development,” Journal of Product Innovation Management, 9 (1 ), 44- 52. *McDonough, Edward F. III and Gloria Barczak (1991), “Speeding Up New Product Development: The Effects of Leadership Style and Source of Technology,” Journal of Product Innovation Management, 8 (3), 203-211. McDonough, Edward F. III and Richard Leifer (1986), “Effective Control of New Product Projects: The Interaction of Organization Culture and Project Leadership,” Journal of Product Innovation Management, 3 (3), 149-157. *McDonough, Edward F. III, Kenneth B. Kahn, and Abbie Griffin (1999), “Managing Communication in Global Product Development Teams,“ IEEE Transactions on Engineering Management, 46 (4), 375-386. McGrath, Joseph E. (1984), Groups: Interaction and Performance, Prentice-Hall: Englewood Cliffs, NJ. Meyers, Patricia W. and David Wilemon (1989), “Learning in New Technology Development Teams,” Journal of Product Innovation Management, 6 (2), 79- 88. Mirvis, Philip H. and Edward E. Lawler III (1977), “Measuring the Financial Impact of Employee Attitudes,” Journal of Applied Psychology, 62 (1 ), 1-8. *Moenaert, Rudy K., Filip Caeldries, Annouk Lievens, and Elke Wauters (2000), “Communication Flows in International Product Innovation Teams,” Journal of Product Innovation Management, 17 (5), 360-366. *Moenaert, Rudy K., William E. Souder, Arnoud De Meyer, and Dirk Deschoolmeester (1994), “R&D-Marketing Integration Mechanisms, 117 Communication Flows, and Innovation Success,” Journal of Product Innovation Management, 11 (1), 31-45. “Montoya-Weiss, Mitzi M., Anne P. Massey, and Michael Song (2001), “Getting It Together: Temporal Coordination and Conflict Management in Global Virtual Teams,” Academy of Management Journal, 44 (6), 1251-1262. Montoya-Weiss, Mitzi and Roger J. Calantone (1994), "Determinants of New Product Performance: A Review and Meta-Analysis," Journal of Product Innovation Management, 11 (5), 397-417. Moorman, Christine (1995), “Organizational Market Information Processes: Cultural Antecedents and New Product Outcomes,” Journal of Marketing Research, 32 (3), 318-335. Noble, Charles H. and Michael P. Mokwa (1999), “Implementing Marketing Strategies: Developing and Testing a Mmanagerial Theory,” Journal of Marketing, 63(4), 57-73. Olson, Eric M., Orville C. Walker, and Robert W. Ruekert (1995), “Organizing for Effective New Product Development: The Moderating Role of Product lnnovativeness,” Journal of Marketing, 59 (1 ), 48-62. Page, Albert L. (1993), “Assessing New Product Development Practices and Performance: Establishing Crucial Norms,” Journal of Product Innovation Management, 10(4), 273-290. “Pelled, Lisa H. and Paul S. Adler (1994), “Antecedents of Intergroup Conflict in Multifunctional Product Development Teams: A Conceptual Model,” IEEE- Transactions on Engineering Management, 41 (1 ), 21-28. *Pelled, Lisa Hope, Kathleen M. Eisenhardt and Katherine R. Xin (1999), “Exploring the Black Box: An Analysis of Work Group Diversity, Conflict, and Performance,” Administrative Science Quarterly, 44 (1), 1-28. Pindyck, Robert S. and Daniel L. Rubinfeld (1998), Econometric Models and Economic Forecasts, 4th ed., McGraw-Hill Irwin: Boston, MA. *Pinto, Mary Beth and Jeffrey K. Pinto (1990), “Project Team Communication and Cross-Functional Cooperation in New Program Development,” Journal of Product Innovation Management, 7 (3), 200-212. *Pinto, Mary Beth, Jeffrey K. Pinto, and John E. Prescott (1993), “Antecedents and Consequences of Project Team Cross-Functional Cooperation,” Management Science, 39 (10), 1281 -1 297. 118 Prahalad, C. K. and Richard Bettis (1986), “The Dominant Logic: A New Linkage Between Diversity and Performance,” Strategic Management Journal, 7 (6), 485-501. Prahalad. C. K. and Gary Hamel (1990), “The Core Competence of the Corporation,” Harvard Business Review, 68 (3), 79-92. *Qiu, Tianjiao, William Oualls, Jonathan Bohlmann, and Deborah E. Rupp (2006), “The Effect of lnteractional Fairness on the Performance of Cross- Functional Product Development Teams: A Multilevel Mediated Model,” Working Paper. Rafii, Farshad (1995), “How Important is Physical Co-Location to Product Development Success,” Business Horizons, 38 (Jan-Feb), 78-84. Rosenau, Milton C., Abbie Griffin, George A. Castellion and Ned F. Anscheutz (1996), PDMA Handbook on New Product Development, John Wiley and Sons: New York, NY. Rosenthal, Robert (1995), “Writing Meta-Analytic Reviews,” Psychological Bulletin, 1 18 (September), 183-192. Rothwell, R., C. Freeman, A. Horlsey, V. T. P. Jervis, A. B. Roberson and J. Townsend (1974), “SAPPHO Updated-Project SAPPHO Phase II,” Research Policy, 3, 258-291. Rubenstein, David and Richard W. Woodman (1984), “Spiderman and the Burma Raiders: Collateral Organization Theory in Action,” Journal of Applied Behavioral Sciences, 20 (1 ), 1-21. *Sarin, Shikhar and Vijay Mahajan (2001), “The Effect of Reward Structures on the Performance of Cross-Functional Product Development Teams,” Journal of Marketing, 65 (2), 35-53. *Sarin, Shikhar and Christopher McDermott (2003), “The Effect of Team Leader Characteristics on Learning, Knowledge Application, and Performance of Cross-Functional New Product Development Teams,” Decision Sciences, 34 (4), 707-739. Schein, Edgar H. (1990), “Organizational Culture,” American Psychologist, 45 (2), 109-119. *Sethi, Rajesh (2000a), “New Product Quality and Product Development Teams,” Journal of Marketing, 64 (2), 1-14. 119 *Sethi, Rajesh (2000b), “Superordinate Identity in Cross-Functional Product Development Teams: Its Antecedents and Effect on New Product Performance,” Journal of the Academy of Marketing Science, 28 (3), 330-344. *Sethi, Rajesh and Carolyn Y. Nicholson (2001), “Structural and Contextual ‘ Correlates of Charged Behavior in Product Development Teams,” Journal of Product Innovation Management, 18 (3), 154-168. *Sethi, Rajesh, Somendra Pant, and Anju Sethi (2003), “Web-based Product Development Systems Integration and New Product Outcomes: A Conceptual Framework,” Journal of Product Innovation Management, 20 (1 ), 37-56. “Sethi, Rajesh, Daniel C. Smith, and C. When Park (2001), “Cross-Functional Product Development Teams, Creativity, and the lnnovativeness of New Consumer Products,” Journal of Marketing Research, 38 (1 ), 73-85. “Shim, Duksup and Muhsin Lee (2001), “Upward Influence Styles of R&D Project Leaders,” IEEE Transactions on Engineering Management, 48 (4), 394-413. Sinkula, James M. (1994), “Marketing Information Processing and Organizational Learning,” Journal of Marketing, 58 (1 ), 35-45. Slater, Stanley F. and John C. Narver (1995), “Market Orientation and the Learning Organization,” Journal of Marketing, 59 (3), 63-74. Song, X. Michael and Mark E. Parry (1994), “R&D-Marketing Interface in Japanese High-Technology Firms: Hypotheses and Empirical Evidence,” Journal of the Academy of Marketing Science, 21 (2), 125-133. Song, X. Michael and Mark E. Parry (1996), “The Dimensions of Industrial New Product Success and Failure in State Enterprises in the People’s Republic of China,” Journal of Product Innovation Management, 11 (2), 105-118. Song, X. Michael, and Mark Parry (1997), “A Cross-National Comparative Study of New Product Development Processes: Japan and the United States,” Journal of Marketing, 61 (2), 1 -18. Song, X. Michael and Mitzi M. Montoya-Weiss (1998), “Critical Development Activities for Really New versus Incremental Products,” Journal of Product Innovation Management, 15(2), 124-135. Souder, William E. and Rudy K. Moenaert (1992), “An Information Uncertainty Model for Integrating Marketing and R&D Personnel in New Product Development Projects,” Journal of Management Studies, 29 (4), 485-512. 120 Souder, William E. (1987), Managing New Product Innovations, Lexington Books: Lexington, MA. Stewart, Greg L. and Murray R. Barrick (2000), “Team Structure and Performance: Assessing the Mediating Role of Intrateam Process and the Moderating Role of Task Type,” Academy of Management Journal, 43 (2), 135-148. Stewart, Greg L. (2006), “A Meta-Analytic Review of Relationships between Team Design Features and Team Performance,” Journal of Management, 32 (1 ), 29-54. *Stock, Ruth Maria (2006), “Interorganizational Teams as Boundary Spanners between Supplier and Customer Companies,” Journal of the Academy of Marketing Science, 34 (4), 588-599. Sundstrom, Eric, Kenneth P. De Meuse, and David Futrell (1990), “Work Teams: Applications and Effectiveness,” American Psychologist, 45(2), 120-133. ”Susman, Gerald l. and Judith M. Ray (1999), “Test of a Model of Organizational Contributors to Product Development Team Effectiveness,” Journal of Engineering and Technology Management, 16 (3), 223-245. Tatikonda, Mohan V. and Mitzi M. Montoya-Weiss (2001), “Integrating Operations and Marketing Perspectives of Product Innovation: The Influence of Organizational Process Factors and Capabilities on Development Performance,” Management Science, 47 (1 ), 151 -172. “Thamhain, Hans J. (2003), “Managing Innovative R&D Teams,” R&D Management, 33 (3), 297-311. *Troy, Lisa 0., David M. Szymanski, and P. Rajan Varadarajan (2001), “Generating New Product Ideas: An Initial Investigation of the Role of Market Information and Organizational Characteristics,” Academy of Marketing Science Journal, 29 (1 ), 89-101. Thompson, James D. (1967), Organizations in Action, McGraw-Hill, New York, NY. Tushman, Michael L. and R. Katz (1980), “External Communication and Project Performance: An Investigation into the Role of Gatekeepers,” Management Science, 26 (11), 1071-1085. Ulrich, Karl T. and S. D. Eppinger (1995), Product Design and Development, McGraw-Hill: New York, NY. 121 Urban, Glen L. and John R. Hauser (1993), Design and Marketing of New Products, 2nd ed., Prentice Hall: Englewood Cliffs, NJ. *Valle, Sandra and Lucia Avella (2003), “Cross-Functionality and Leadership of the New Product Development Teams,” European Journal of Innovation Management, 6 (1 ), 32-47. *Van den Bulte, Christophe and Rudy K. Moenaert (1998), “The Effects of R&D Team Co-Iocation on Communication Patterns among R&D, Marketing, and Manufacturing,” Management Science, 44 (1 1 ), S1 -S18. Van de Ven, Andrew H. and Marshall Scott Poole (1995), “Explaining Development and Change in Organizations,” Academy of Management Review, 20 (3), 510-540. Van de Ven, Andrew H., A. L. Delbecq and R. Koenig, Jr. (1976), “Determinants of Coordination Modes within Organizations,” American Sociological Review, 41, 322-338. *Voss Christopher A. (1985), “Determinants of Success in the Development of Applications Software,” Journal of Product Innovation Management, 2 (2), 122-129. Wolf, Frederic M. (1986), Meta-Analysis: Quantitative Methods for Research Synthesis, Sage Publications, Beverly Hills, CA. Wooldridge, Jeffrey M. (2002), Econometric Analysis of Cross Section and Panel Data, MIT Press: Cambridge, MA. Zellner, Arnold (1962), “An Efficient Method of Estimating Seemingly Unrelated Regressions and Tests for Aggregation Bias,” Journal of the American Statistical Association, 57, 348-368. Zirger, Billie Jo and Modesto A. Maidique (1990), “A Model of New Product Development: An Empirical Test,” Management Science, 36 (7), 867-883. “References with asterisk are included in the meta -analytic re view. 122 lllllll‘lljlflljlllllljljfllIjllljjl