We hypothesized ([H.sub.1]) that: the more positive the partners' experience of the project, the more likely that the project would be seen to be successful.

Exogenous constructs: Relational trust and project management competence

Trust is widely considered to be a critical factor in the success of collaborative R&D projects (Barnes et al. 2002), and it has been theorized in a number of ways in this context (e.g., Liebeskind & Oliver 1998; Davenport et al. 1999). In this study we drew on Sako's definition of trust (Sako 1992; Sako & Helper 1997; see also Krishnan et al. 2006), which is based on an expectation that partners will behave in predictable and mutually acceptable ways. According to Sako, trust is a multi-faceted construct with three components. This three-dimensional view of trust is consistent with a broad literature on the topic (see Ryan et al. 2004). Predictability is associated with stability and consistency of behaviour, and the confidence that a partner will act in the future to produce mutually beneficial outcomes.

One component is Contractual Trust, and refers to the belief that partners will adhere to agreements and keep promises. The second, Competence Trust, develops in relation to the dependability associated with the expertise, know-how, ability, and the actual performance of partners in meeting targets and obligations as distinct from their stated intention to perform (Sako 1997; Sako & Helper 1998; Das & Teng 2001). While competence-based trust focuses on immediate risk reduction, it contributes to the building of more resilient trust over time as confidence in a partner's loyalty, integrity and honesty builds (Davenport et al. 1999). Expectations of positive outcomes in longer-term partnerships can only arise from the trust associated with partners proving themselves to be equitable, fair in their dealings, having high integrity and standards of conduct, reciprocating favours, and acting in ways that go beyond the formal agreement to benefit the partnership. This is Goodwill Trust, the third component identified by Sako. This can only develop over time through repeated successful encounters and, if sufficiently strong, can lead to 'hand-shake' agreements (de Laat, 1997). For interorganizational collaborations to continue over time, trust relations have to develop their own dynamics, transcending individual relationships, to become ingrained in the routines and practices of the collaborating entity and the partners within it (Dodgson 1993).

We used the second order construct Relational Trust (which we defined, drawing on Sako & Helper 1998, as an expectation of an organizational actor that partners will behave in mutually acceptable and supportive ways during a collaboration) to reflect these three forms of trust, where a high level of relational trust is associated with 'respect for a partner's abilities, commitment based on belief in mutual benefit, and openness and honesty in objectives' (Dodgson 1993: 92), and this both reduces the perceived risks of collaboration and enhances the perceived quality of the relationship. It is commonly accepted that in R&D partnerships, which involve the commercialization of knowledge, trust not only encourages risk-taking and the sharing of proprietary information, it also mitigates the need for excessive controls and encourages looser contractual practices in what is largely an unpredictable and uncertain area (de Laat 1997; Ryan et al. 2004). In this way, trust reduces the transaction costs associated with contracting and other legal means aimed at minimizing the risks of opportunism in such partnerships.

We hypothesized ([H.sub.2]) that: the higher the level of relational trust (i.e., in terms of contractual, competence and goodwill trust) amongst the partners in the project team, the more positive will be the partners' experience of the project.

To operationalize the second order construct of Relational Trust, which is reflected by three first order constructs (Contractual Trust, Competence Trust, Goodwill Trust), we drew on the work of Cummings and Bromiley (1996) because they had a validated trust scale and their research was focused on developing a conceptualization and measure of trust in relation to how it affects transaction governance (Bigley & Pearce 1998). We also drew on Ryan et al. (2004) because they had specifically focused on R&D collaborations and had used Sako's validated trust scale. Both studies assisted us in devising a measure that could be used in the context of project level R&D partnering (Kadefors 2004). We used seven of Ryan et al.'s items from their non-validated trust scale (marked 'R' in Appendix 1), and added an additional six from Cummings and Bromiley's Organizational Trust Inventory (marked 'C&B' in Appendix 1) to create our three trust scales. We started with four items for Contractual Trust (two from Ryan et al. and the other two from Cummings and Bromiley), four for Competence Trust (three from Ryan et al. and one from Cummings and Bromiley), and five for Goodwill Trust (two from Ryan et al. and the rest from Cummings and Bromiley), as shown in Appendix 1. In some cases scale items were modified slightly to make them more suitable for our study context, and for about one-half of them we reverse worded the item (requiring that we subsequently reverse-coded the responses). Following conventional scale design practice aimed at mitigating the possibility of a response set effect (e.g., Parker & Rea 1997), we reverse worded items in order to encourage respondents to read each of the trust scale questions more carefully. We did not assume that a reverse worded item was a measure of distrust, which requires different constructs and theorizing (see Bigley & Pearce 1998; Lewicki et al. 1998). Nor do our scales constitute a continuum from trust to distrust as we were identifying degrees of trust (i.e., from low to high levels) in a working relationship.

Our preliminary qualitative research had also suggested that a relationship focus, as indicated by the building and maintenance of trust among the partners, is necessary but not sufficient to create a positive collaboration experience. Collaborative projects are established specifically to achieve particular goals, and for this a task focus--and its associated project management capabilities--among the project team members is essential. The importance of effective project management systems and procedures for the success of projects, and especially for projects with high levels of uncertainty and risk, has long been emphasized. Empirical studies of cross-sector collaborative research have confirmed the central importance of project management (e.g., Barnes et al. 2002, 2006). Effective project management involves two distinct processes. In the first, project planning, objectives are clearly specified, timelines are defined, and tasks and resources are allocated among project team members. A clear specification of project objectives has been associated with cooperative relationship success (e.g., Barnes et al. 2002; Mora-Valentin et al. 2004), and effective planning can ensure that partners' expectations are made explicit and a community of common interests can be built and sustained. In the second process, project control, progress is monitored and corrective action is taken if necessary. This is usually achieved through the institution of formal systems and procedures as well as through the management actions of a project manager or leader, whose role is often seen to be critical for project success (e.g., Dodgson 1993; Barnes et al. 2002). The more effective this process is, the higher the likelihood that project objectives will be achieved and the more satisfying will be the experience of the collaboration for the participants.

We hypothesized that (H3): the higher the level of project management competence in the project team, the more positive will be the partners' experience of the project.

The construct of Project Management Competence (which we defined as the capability to effectively plan and manage a project, through the application of knowledge, skills, tools and techniques, so that it delivers on its objectives within constraints of time and resources) we believe has not been measured before, and we operationalized it by drawing on the effective project management factors identified by Barnes et al. (2002), adding to them three other competencies that are widely accepted in the project management literature (e.g., Project Management Institute 1996). An exploratory factor analysis of the initial 11 items revealed that they loaded on two principal components, corresponding to Project Planning Capability (6 items) and Project Control Capability (5 items). So, in the model Project Management Competence was conceived as a second order construct reflected by these two first order constructs which in turn measured the capabilities of the two processes that constitute the practice of project management (Appendix 1).

RESEARCH METHODS

Development of the survey instrument

A questionnaire was designed to test the theoretical model, and a first step was to operationalize the constructs in the model. This was achieved by, wherever possible, using and/or modifying existing scales (e.g., that for Partner Collaboration Experience), but for a number of the constructs no such scales existed (e.g., Project Management Competence) and so these were developed specifically for the survey. After the addition of questions to collect supplementary data (e.g., on the respondents' backgrounds), an initial draft was pilot tested with 6 experienced CRC project managers. This resulted in modifications to some of the questions and to the instructions. The final questionnaire consisted of 19 questions, most of which could be answered in terms of a 6-point Likert-type scale, ranging from 'completely agree' to 'completely disagree'. In completing the questionnaire, respondents were asked to focus on the CRC project that they had most recently completed (or were soon about to complete) as project leader in the last two years.

Print Get the Mag Weekly Updates