Intuitive decision-making and deep level diversity in entrepreneurial ICT teams

Abstract

Intuition, which ranges in style from affective to inferential to holistic (Pretz et al. 2014) can play a central role in decision-making, as decision makers must often balance intuition with rational analytical thinking. This paper explores the influence of intuition based decision style diversity on both the emergence of transactive memory systems (TMS) and team performance in an entrepreneurial setting. The uncertainty of an entrepreneurial setting leads to a greater emphasis on intuitive thinking in the team context. However, intuition as a basis for decision-making manifests itself in a variety of forms. And not all forms of intuitive tendencies lend themselves to a team decision-making context that is cohesive, or to decisions that are strategically coherent. Given that a diversity of intuitive tendencies is likely to be found in every team setting, and as such may exert a considerable influence on the decision-making process, we develop hypotheses in order to investigate the nature of this influence. Drawing on data collected from 188 participants across 22 countries and split into 48 entrepreneurial ICT teams, our findings show strong support for the influence of intuitive decision style diversity on both team level states and team performance. The explanatory model developed has significant implications for contemporary understanding of decision-making teams, deep-level diversity, the entrepreneurial process, and TMS research.

Keywords:

• Intuition
• decision making
• diversity
• ICT entrepreneurship

1. Introduction

The development of new products and software in the entrepreneurial context is widely recognised as a team driven activity (McNamara, Dennis, et al., 2008; Mol et al., 2015; Vlaar, Fenema, et al., 2008). These richly diverse team contexts provide are responsible for significant variation in both the decision-making styles that are employed, and the decision processes that emerge (Lowry, Zhang, et al., 2010; Turban, Liang, et al., 2010). This study draws on numerous research strands to explore the impact of the intuitive decision-making styles of ICT entrepreneurs on team performance. In developing our theoretical framework we highlight conceptual and empirical shortcomings in the decision processes, team cognition, entrepreneurship, and transactive memory systems (TMS) literatures to which we direct our attention with this study.

We firstly highlight the rising interest in intuitive decision-making in contemporary scholarship and point to the neglected area of intuition based decision style diversity and its potential implications for team performance (Baldacchino, Ucbasaran, Cabantous, & Lockett, 2015; Kickul, Gundry, Barbosa, & Whitcanack, 2009). We then turn to the issue of deep-level diversity and the need to develop more nuanced and empirically grounded explanatory frameworks which consider the two-way effects of cognitive diversity in decision-making teams in general, and of intuitive decision styles in particular (Klein, Knight, Ziegert, Lim, & Saltz, 2011; Pretz et al., 2014). Such frameworks are of particular relevance to the entrepreneurship literature given the uncertain, high-risk, time pressured, and emotionally challenging nature of the entrepreneurial process and the subsequent reliance on intuition in that domain (Baron, 2008). In order to address this challenge, we designed a study which investigated the moderating effects of intuitive decision style diversity on the relationship between TMS, as a model of a decision team’s cognitive structure, and team performance. We find strong support for our contention that intuition based decision style diversity exerts a powerful influence on both the emergence of TMS and team performance in an entrepreneurial setting. We find evidence of a relationship between this diversity and team confidence in the aftermath of face-to-face (FTF) interaction, but that the basis of this confidence is fundamental to whether it ultimately has a positive or negative effect on performance. We additionally find strong evidence of a positive moderating effect for intuitive-diversity on the relationship between TMS and team performance, but only when that diversity is affective or inferential in nature.

The paper proceeds as follows. In the following sections we develop our theoretical framework and identify gaps in the literature. In section 3 we describe our research objectives and present our hypotheses. In section 4 we detail our methodology and describe our results. We discuss our findings in section 5, before offering some concluding remarks.

2. Theoretical framework

2.1. Decision-making processes in the team context

Decision-making research at the managing team level is both broad in its scope and diverse in its theoretical influences (Klein et al., 2011; Shepherd & Rudd, 2014). The heterogeneity of conceptual approaches and disciplinary interests can in large part be assigned to the critical importance of the decision-making process to organisational outcomes in a wide variety of social settings. Adding further diversity to the literature is the inherently cross-functional nature of the demands placed on this process in organisational administration and planning. At the contextual level, the dynamism, uncertainty, and hostility of the external environment as well as the structural and political constraints of its internal counterpart create significant difficulty for management teams in their efforts to construct decision processes that are sufficiently informed and strategically coherent (Mol et al., 2015; Shepherd & Rudd, 2014). These processes must, simultaneously, be flexible enough to capitalise on the individual experience of the decision-makers and identify the most pertinent trends amongst a relentless flow of market and competitor data (Heavey & Simsek, 2014).

The development of a range of potential organisational directions on the basis of this data and the selection of those most promising among them places considerable demands on the cognitive resources of decision-making teams. Such is the intensity of these demands that organisational science has increasingly turned its attention to the intuitive (as opposed to information intensive) decision processes employed by decision-makers in plotting strategic courses. However, intuition as a basis for decision-making manifests itself in a diverse array of styles (Pretz & Totz, 2007). And it seems likely that not all forms of intuitive tendencies lend themselves to a team decision-making context that is cohesive, and as such decisions that are strategically coherent. Nevertheless it must be acknowledged that such diversity surely exists, and in existing it may exert a hitherto unexamined influence on the emergent team states that underpin the decision-making process. We will further elaborate on the importance of deep level diversity in the following section, before we turn our attention to the nature of intuition based decision-making.

2.2. Deep level diversity in decision-making teams

Team diversity can be defined as the distribution of differences amongst team members with respect to a given characteristic (Harrison & Klein, 2007). With respect to the influence of diversity on management team performance, two clear lines of thought can be distinguished in the literature. The first emphasises the value of diverse opinions and perspectives in the generation of creative and innovative decisions, and considers diversity as means of constructing a decision process or framework that is informationally or experientially enriched. The second directs its attention to internal group dynamics and places greater emphasis on the role of team cohesion and conflict in understanding the decision-making process (Klein et al., 2011). The varied experiences and information resources regarded within the first tradition as the source of innovation, are seen through this lens as a source of dysfunction arising out of contradictory assumptions and decision frames at the individual or subgroup level (Spickermann, Zimmermann, & Heiko, 2014).

In order to adequately consider the diversity problem it is necessary to further distinguish between ‘surface’ and ‘deep-level’ diversity. At the surface level, diversity can be understood in terms of more easily observable characteristics such as gender, ethnicity, or nationality (Harrison, Price, & Bell, 1998; Stahl, Maznevski, et al., 2010). In contrast, deep-level diversity refers to a less visible distribution of differences with respect to variables such as attitudes, values, tendencies, and cognitive styles (Jackson et al., 1991). The latter is of particular interest to decision theorists. Hough and ogilvie (Hough & Ogilvie, 2005) refer to cognitive style as the manner in which an individual perceives, considers, and solves problems. In this respect it plays a fundamental role in the information processing and alternative generating processes of individual decision makers. Depending upon which of the aforementioned diversity perspectives to which one adheres, diversity of cognitive styles may equally be regarded as a source of conflict and miscommunication amongst decision makers or an essential contextual variable (Fisher, Bell, Dierdorff, & Belohlav, 2012; Miller, Burke, & Glick, 1998; Shepherd & Rudd, 2014). As such it is probable that more nuanced distinctions between the types of deep-level diversity that improve or diminish decision quality are required. Such distinctions are limited however by the emphasis in the literature on surface level differences to the relative neglect of deep-level constructs (Harrison, Price, Gavin, & Florey, 2002; Klein et al., 2011; van Knippenberg & Schippers, 2007). It is to this gap that we turn our attention in this study. In the following section we will justify our selection of intuitive decision style diversity, before completing our theoretical framing with an overview of distributed cognition in the final section.

2.3. Intuition based decision-making

One of the earliest characterisations of intuition advanced in scholarly work was that of a primary mode of perception operating at a subconscious level (Jung, 1971). While Jung’s framework tended to prioritise individual differences in the exercise of intuition, much of the research that followed homogenised intuition in conceptual terms (Pretz & Totz, 2007). A natural outcome of this practice was a contradictory treatment of intuition in the literature. Intuition was simultaneously recognised as an inferential heuristic process (Hill, 1987), within which decisions had become automated through practice, and a holistic one in which an array of environmental cues were integrated without conscious awareness (Hammond, 1996). In addition, a third stream of research recognised intuition in terms of a judgement based on ‘gut feeling’, or an affective confidence in an opinion to which no explicit rational support could be offered (Bastick, 1982; Epstein, 1994).

In developing a comprehensive framework for both the conceptualisation and measurement of intuition, Pretz et al. (2014) argued that affective, inferential, and holistic intuition represented distinct types of information processing within a unifying intuitive mode. A concise definition of these three types are presented below:

Affective-intuitive decisions are based on emotional reactions to decision situations, and can be understood as associative in nature, drawing on prior conditioning and emotional arousal.

Inferential-intuitive decisions are based on automated inferences, decision-making processes that were once analytical but have become intuitive with practice, and which draw on well-developed mental schemas.

Holistic-intuitive decisions are based on non-analytical process that are bottom-up, data driven, and which integrate multiple, diverse cues into immediate situational judgements.

Given the highly differentiated nature of decision formation across all three types of intuition, it seems probable that their distribution across a team has considerable implications for both the decision-making context in the first instance, and decision quality in the second. We well turn in a moment to Transactive Memory Systems (TMS) as a conceptual framing of this distributed cognition, but firstly we must briefly consider the particular importance of this intuition based diversity to entrepreneurial teams.

It is unfortunate that little attention has been paid to intuitive decision-making in the entrepreneurial context given that intuition is widely regarded as a core element of entrepreneurially driven innovation, and is often referred to as ‘the seed of any entrepreneurial action’ (Dutta & Crossan, 2005, p. 436). Intuition based decision-making is of particular relevance to the entrepreneurial setting for two key reasons. The first is that task uncertainty, such as that involved in the development of a new product or service (Baldacchino et al., 2015; Gustafsson, 2006), and the second is the typical inexperience with either a given role or environment that accompanies entrepreneurial endeavour (Baron & Ensley, 2006; Dew, Read, Sarasvathy, & Wiltbank, 2009). Indeed, the entrepreneurial setting describes at least 8 of the 10 conditions Klein (2008) refers to as features of a naturalistic decision-making environment. These include ill-defined goals and ill-structured tasks, uncertainty, ambiguity, and missing data, shifting and competing goals, dynamic and continually changing conditions, action-feedback loops, time stress, high stakes, and multiple players.

Decision-making in such circumstances, Klein argues, is not characterised by the generation of multiple alternatives and the assessment of probabilities. Rather, decision-making in the naturalistic setting proceeds on the basis of action and reaction on the basis of prior experience. Given that the wider psychology and organisation science literatures also point to the greater prevalence of intuition dependence in uncertain environments, the high-risk, time pressured, emotionally challenging nature of entrepreneurship seems a suitable setting for our investigation (Baron, 2008). We will next provide an overview of the TMS construct as a model of distributed cognition in management teams, before presenting our hypotheses.

2.4. Transactive memory systems

Our discussion of cognitive styles and decision style diversity leads us to TMS as a concept through which the cognitive structure of the decision-making team can be analysed. At its centre, TMS theory posits that team members can serve as external sources of knowledge for one another, and can therefore be relied upon to be responsible for different areas of concern or interest (Heavey & Simsek, 2014). TMS allow for an emergent understanding of who knows what in a team, or alternatively a social distribution of cognitive labour (Wegner, 1987). Teams first come to a consensus with respect to who knows what, and then over time develop communication processes that allow for collective use of individual knowledge (Mell, Van Knippenberg, & Van Ginkel, 2014). A team’s TMS can be understood in terms of three key dimensions, coordination, credibility, and specialisation, (Lewis, 2003). These are defined as follows:

Coordination refers to the extent of the effectiveness of knowledge exchange within the team.

Credibility refers to the individual members’ beliefs about the reliability of other members’ knowledge.

Specialisation refers to the extent to which team members have distinctive and specialised knowledge domains with respect to the team’s decisions.

Effective TMS lend several advantages to team decision-making processes. Firstly they allow for specialisation which maximises the efficiency of information processing at the individual level (Heavey & Simsek, 2014). Secondly, they may facilitate both improved quality and quantity of task-related information retrieval (Mell et al., 2014). And thirdly, they can enhance information integration and elaboration (Bell, 2007). Collectively these factors contribute to superior team performance. However, TMS may also be problematic, as a poorly constructed TMS can generate team overconfidence, with a belief in the credibility of member knowledge that is unjustified. There are two main causes of this effect. Firstly, teams need a shared awareness of the distribution of team expertise, which is particularly challenging in both teams that have had limited prior socialisation and must operate in a novel task environment. Secondly, TMS can be inefficient due to inaccurate team beliefs about the location or credibility of knowledge. (Kozlowski & Bell, 2003; Peltokorpi, 2008).

As a consequence of its basis not just in the objective distribution of cognitive resources in a decision-making team, but also in team members’ subjective beliefs about the distribution of those resources, TMS serve as a construct through which we can explore both the positive and negative implications of deep-level diversity. In so doing we may also address some key gaps within the TMS literature itself. Firstly, there is a limited understanding of how it is that face-to-face (FTF) communication may contribute to either a well- or poorly-constructed TMS (O’Neill, Hancock, et al., 2015; Pridmore & Phillips-Wren, 2011; Schmidt, Montoya-Weiss, et al., 2001). Secondly, there has been little to no examination of how expert recognition may be complicated by diversity based conflict and miscommunication. Thirdly, and finally, the process through which a TMS emerges in an uncertain task context has been largely neglected in the literature (Peltokorpi, 2008).

In the following section we will advance the hypotheses we developed on the basis of the gaps we have identified pertaining to decision-making, deep-level diversity, intuition, and TMS. The theoretical framework developed is presented in Figure 1.

Figure 1. Intuition style diversity in a naturalistic decision setting.

3. Research objectives and hypotheses

Our primary objective is to investigate how intuition based decision style diversity interacts with both the emergence of TMS and team performance in an entrepreneurial setting. On the basis of the theoretical framework we have constructed, and the shortcomings identified therein, we developed the following research hypotheses.

3.1. Intuitive decision-making diversity and FTF based team confidence

Our first task is to investigate the initial effects of intuitive decision-making diversity at the team level on team confidence. As we have argued, due to the distinctive nature of the affective, inferential, and holistic decision-making styles, diversity may be manifested in important differences with respect to initial team dynamics. Considering these differences, we propose firstly that due to its emotional and non-rational quality, diversity with respect to affective-intuitive decision-making will predict lower levels of initial team confidence. Secondly, and due to its schematic and abstract quality, we expect that inferential-affective decision-making will also predict lower levels of team confidence. Thirdly, we predict that due to its bottom-up data driven quality, holistic-intuitive decision-making will predict higher levels of team confidence.

Finally, we consider the implications of these predictions for team performance. We anticipate that the despite (or perhaps because of) the diminishing effect of affective and inferential diversity on team confidence, such diversity will ultimately contribute to a stronger TMS and higher level team performance. We additionally expect that holistic-intuitive diversity will produce overconfidence in the team TMS, and will be associated with lower level team performance. Therefore out first three hypotheses are as follows:

H1a: For higher performing teams, diversity of affective-intuitive decision-making will predict lower levels of team confidence.

H1b: For both higher and lower performing teams, diversity of inferential-intuitive decision-making will predict lower levels of team confidence.

H1c: For lower performing teams, diversity of holistic-intuitive decision-making will predict higher levels of team confidence.

Building on hypothesis H1c, we anticipate that where team confidence is predicated on holistic-intuitive diversity, it will also be associated with stronger TMS Credibility due to the team member confidence in the data driven quality of knowledge exchanged. We therefore advance hypothesis H2:

H2: Where team potency is positively associated with holistic-intuitive diversity, it will predict stronger TMS credibility.

3.2. Intuitive decision-making diversity and transactive memory systems

Our second set of hypotheses consider the moderating effect of intuitive decision-style diversity on team TMS and team performance. We firstly predict that as teams address problems over time, due to the insights generated by reference to sophisticated mental schemata (inferential) and associations with prior experience (affective), TMS that are based on affective- and inferential-intuitive diversity will be associated with a higher level of team performance. We further suggest that due to inaccurate assumptions and overconfidence generated by initial FTF interaction, TMS that are based on holistic-intuitive diversity will be associated with a lower level of team performance. We therefore advance the following hypotheses:

H3a: Amongst higher performing teams, diversity of affective- and inferential-intuitive decision-making will predict a stronger TMS

H3b: Amongst lower performing teams, diversity of holistic-intuitive decision-making will predict a stronger TMS

Having presented our study’s hypotheses, we will now describe the research setting, measurement methods, and data analysis arising out of our investigation.

4. Methodology

4.1. Research setting

The study is set in the context of a European wide ICT entrepreneurship initiative which supports aspiring young tech entrepreneurs interested in co-founding new international information, communication and technology (ICT) start-ups through an initial, week long start-up event wherein team members meet and interact, followed by a 12 week virtual incubation phase. Participants were drawn from 22 different countries, and formed multi-national teams, consisting of at least 3 nationalities in each team. The programme offered an ideal setting for our investigation for five reasons. Firstly, the week-long FTF interaction of team members would afford us the scope to analyse how decision-making effectiveness in later phases was impacted by initial interactions (Kennedy, Vozdolska, et al., 2010). Secondly, the programme would allow us to investigate our hypotheses in the context of virtual teams, a literature especially concerned with TMS (Griffith & Neale, 2001; Peltokorpi, 2008) and wherein there have been recent calls from leading scholars for a more nuanced understanding of how diversity interacts with innovation and creativity (Gilson, Maynard, Jones Young, Vartiainen, & Hakonen, 2015). Fourthly, the programme provided us with an entrepreneurial setting for our investigation that allowed us to address a number of the aforementioned gaps in the literature. Finally, the ICT focus of the programme allowed for a contribution to understanding of decision-making in the context of software development teams that were not only virtual, but were also agile in nature. To date, decision-making research in agile teams has had a rational and post-cognitive emphasis, which neglects the underlying diversity of teams and decision-making in a fluid and natural setting (Drury, Conboy, & Power, 2012; Moe, Aurum, & Dybå, 2012).

4.2. Sample

The sample consisted of 48 remote working entrepreneurial ICT teams (188 team members). Of the study’s participants 129 were male and 59 were female, and the average age was 25.3 (SD = 3.34). In total 22 nationalities were represented in the sample. Participants were issued with questionnaires at three points in the programme: (1) one week prior to the scrum event measuring the four types of intuitive decision styles; (2) 3 days after the end of the scrum event measuring team confidence and (3) at the end of the 12 week programme measuring the teams’ transactive memory systems. All questionnaires had response scales which ranged from 1 = Completely Disagree to 7 = Completely Agree, with some items reverse scored.

4.3. Measures

Intuition: To measure the intuition based decision styles of participants we adapted the types of intuition scale (TIntS) developed by Pretz et al. (2014). The affective-intuitive decision style was captured with 6 items, such as ‘I often make decisions based on my gut feelings, even when the decision is contrary to objective information.’ The inferential-intuitive construct was measured with a further 6-item scale with items such as ‘Familiar problems can often be solved intuitively’, or ‘My intuitions are based on my experience.’ Finally the holistic-intuitive dimension was measured using a 3-item scale (sample item: ‘When tackling a new project, I concentrate on big ideas rather than the details’). Cronbach’s alphas for the scales were .74, .75, and .67 respectively. These findings indicate that the scales are reliable and internally consistent.

Transactive Memory System: To measure the teams’ transactive memory systems we adopted the TMS specialisation, credibility, and coordination sub-scales developed by Lewis (2003). The specialisation scale was measured by 3 items such as ‘Each team member has specialised knowledge of some aspect of our project.’ The credibility dimension was measured by another 3-item scale with items such as ‘I was confident relying on the information that other team members brought to the discussion.’ Finally the coordination scale was composed of 3 items such as ‘Our team had very few misunderstandings about what to do.’ Reliability analysis of the sub-scales produced alphas of .87, .73, and .83. The overall scale generated an alpha of .89. As such all scales were reliable and had strong internal consistency.

Team Confidence: To operationalize team confidence we adapted Guzzo, Yost, Campbell, and Shea’s (1993) group potency scale. This scale has been widely used and repeatedly validated in a variety of research settings (de Jong, de Ruyter, & Wetzels, 2005). We constructed a three item scale composed of items such as ‘Our team feels it can solve any problem it encounters’. Reliability analysis generated a Cronbach’s alpha of .93, indicating very robust internal consistency for our scale.

Team Performance: After a period of 12 weeks, teams were independently assessed by a panel composed of incubator directors, seed fund investors, and innovation/new venture researchers. Assessment criteria were drawn from categories of interest typically employed by incubator directors and venture capitalists for new ventures at the nascent stage of development, and were revised and approved by the expert panels. Given the critical necessity of attracting seed funding or acceleration support for ICT new ventures this was deemed the most appropriate measure of team performance.

4.4. Treatment of the data

Correlation coefficients for the variables of interest are provided in Table 1. As our level of analysis was that of the team, individual level answers were aggregated to the group level taking the mean for TMS and Potency scales, and variance for the TIntS measures in line with our hypotheses. In order to test the appropriateness of this aggregation we computed the intraclass correlation coefficients (ICC 1) for all measures. A high level of reliability was found across all averages, ranging from .67 (holistic-intuitive) to .87 (TMS specialisation). The values computed justified our aggregation of measures to the team level.

Table 1. Variable correlation coefficients.

	1	2	3	4	5	6
1 Affective
2 Inferential	.08 (–.14)
3 Holistic Big-Picture	−.29 (–.10)	.03 (.34)
4 Team Confidence	−.46** (.–13)	−.10 (.21)	.28 (.38***)
5 TMS Coordination	.55** (.08)	.52** (.02)	−.07 (.53***)	−.10 (.02)
6 TMS Credibility	.14 (–.14)	.54** (–.12)	−.20 (.39**)	.26 (.39***)	.63*** (.62***)
7 TMS Specialisation	.33 (–.02)	.53** (.14)	−.05 (.59***)	−.07 (.11)	.70*** (.85***)	.65*** (.67***)

N = 47 (18/29);

^* p < .1;

^** p < .05;

^*** p < .01.

4.5. Data analysis and results

H1a: For higher performing teams, diversity of affective-intuitive decision-making will predict lower levels of team confidence.

The correlation coefficient for higher performing teams was -.46. In order to test the robustness of this relationship we conducted a regression analysis on affective-intuitive diversity and our potency variable. The model explained 16.5% of the team potency variance amongst higher performing teams (β = -.46, p < .05). As such we can accept our first hypothesis.

H1b: For both higher and lower performing teams, diversity of inferential-intuitive decision-making will predict lower levels of team confidence.

Correlation coefficients for both higher and lower performing teams were -.10 (p > .1) and .21 (p > .1) respectively. Since both correlations were non-significant, we reject our hypothesis that diversity of inferential-intuitive decision-making would predict lower levels of team confidence.

H1c: For lower performing teams, diversity of holistic-intuitive decision-making will predict higher levels of team confidence.

Our third prediction with respect to team confidence was that amongst lower performing teams, diversity of holistic-intuitive decision-making would be positively associated with team confidence. The correlation coefficient for the relationship between the two variables was .38 (p < .01). We further tested our prediction with a regression analysis of the two variables. Our regression model explained 11.2% (β = .38, p < .05) of the team potency variance amongst the lower performing teams. As such we accept hypothesis H1c.

H2: Where team potency is positively associated with holistic-intuitive diversity, it will predict stronger TMS credibility.

As hypothesis H1c was confirmed, we examined the relationship between team confidence and TMS credibility amongst the lower performing teams in our sample. The coefficient for the relationship was .39 (p < .05). Regression analysis of the variables explained 11.2% of the variance (β = .35, p ≤ .05). We therefore find support for hypothesis H2. A summary is provided in Figure 2.

H3a: Amongst higher performing teams, diversity of affective- and inferential-intuitive decision-making will predict a stronger TMS

H3b: Amongst lower performing teams, diversity of holistic-intuitive decision-making will predict a stronger TMS

Figure 2. Findings and effects for H1a, H1b, H1c, and H2.

We will now present our analysis of the relationship between all three intuitive decision-making styles and the emergent TMS for the teams in our study. Turning first hypothesis H3a, correlation coefficients for affective-intuitive diversity and the three TMS elements were .55 (coordination), .14 (credibility), and .33 (specialisation). These findings indicated that a predictive relationship between affective-intuitive diversity and TMS strength did indeed exist for high performing teams, but acting primarily through the coordination dimension. For inferential-intuitive diversity the coefficients were .52 (coordination), .54 (credibility), and .53 (specialisation), suggesting a significant relationship with all TMS elements for high performing teams. For hypothesis H3b, the coefficients for holistic intuitive diversity and the three TMS dimensions were .53 (coordination), .39 (credibility), and .59 (specialisation), lending support to our hypothesis. We conducted hierarchical regression modelling to further test these relationships. The results of this analysis are presented in Table 2.

Table 2. Hierarchical regression analysis pertaining to hypotheses 3a and 3b.

(Step) Variable Entered	High Performing TMS	Low Performing TMS
(1) Inferential-Intuitive	.60***
Contribution to R²	.32*** (8.82)
(2) Inferential- x Affective-Intuitive	.70*
Overall R² for H3a	.42* (3.82)
(3) Holistic-Intuitive		.58***
Overall R² for H3b		.34*** (13.61)

Standardised coefficients (β) are reported, F-statistics are reported in parentheses.

^* p < .1;

^** p < .05;

^*** p < .01.

Analysis revealed strong support for the predicted interaction between inferential- and affective-intuitive diversity and the TMS strength of high performing teams. The model accounted for 42% of TMS strength variance, with inferential and affective decision styles accounting for 32% and 10% of the variance respectively. We also found strong evidence in support of H3b, with the holistic-intuitive decision style explaining 34% of TMS strength variance amongst lower performing teams. Based on these findings we can conclude that team variance across all three forms of intuitive decision styles investigated moderated the relationship between TMS strength and team performance in our study sample. We will discuss these findings in the following section. A summary is provided in Figure 3.

Figure 3. Findings and effects for H3a and H3b.

5. Discussion

Integrating theory and research on team decision-making, deep-level diversity, intuitive decision styles and transactive memory systems, we established as our research objective an investigation of how intuition based decision style diversity interacts with both the emergence of TMS and team performance in an entrepreneurial setting. Following from our review of the literature, we identified a number gaps on the basis of which we developed and tested 7 hypotheses in our study. Finding support for 6 of these hypotheses, our study has a number of significant implications for the theoretical framework developed.

Firstly, we provide an original and nuanced account of the two-way effects of intuitive decision-making diversity at the deep-level in the team context (Gilson et al., 2015; Klein et al., 2011; Pretz et al., 2014). We found that the distribution of intuitive tendencies had a range of effects that were both complex and ultimately significant in their moderation of the TMS-performance relationship. A key contribution of the study is the insight provided into the manner in which deep-level diversity influences FTF interaction in a manner which has considerable implications for later outcomes (O’Neill et al., 2015; Pridmore & Phillips-Wren, 2011; Schmidt et al., 2001). Diversity of holistic-intuitive decision-making predicted high levels of team confidence in the immediate aftermath of the FTF phase, but only amongst lower performing teams. In contrast affective-intuitive diversity, as a consequence of its less rational and data driven nature, negatively impacted upon initial team confidence but ultimately was associated with higher level team performance. We indeed find that not all combinations of intuitive tendencies lend themselves to a team decision-making context that is cohesive and as such decisions that are strategically coherent, but equally – some do. These findings contribute to, and provide some scope for bridging the gap between, contending schools of thought with respect to the influence of the deep-level diversity on team states by illustrating the manner in which diversity may act in both directions (Spickermann et al., 2014).

These findings also provide a response to a number of important gaps in the TMS literature. We firstly address the limited understanding of how it is that FTF communication may contribute to either a well- or poorly-constructed TMS (Lowry, Schuetzler, et al., 2015; Peltokorpi, 2008). Overconfidence based on holistic-intuitive diversity predicted stronger TMS credibility but only amongst lower performing teams. As such the findings demonstrate that conflict or initially low confidence levels arising out intuitive decision style diversity should not necessarily be seen as a cause for alarm. What matters is the basis of the team’s confidence or lack thereof. We therefore suggest that both decision support practitioners and project team managers would do well to both explore and adjust to the types of intuitive diversity present in project teams. For example, the detrimental effect of affective-intuitive diversity on team confidence did not lead to team failure in this study, but it seems plausible that in other circumstances this could easily be the case. An awareness of this diversity could allow for the both the early identification of the causes of team conflict and the preparation of interventions for such instances. Similarly, our findings explanatory account for the manner in which expert recognition may be complicated by diversity based conflict and miscommunication could also be of assistance to both practioners and researchers in this domain (Peltokorpi, 2008).

Thirdly, our findings respond to calls for an overdue examination of how intuitive decision-making is manifested in entrepreneurial outcomes (Baldacchino et al., 2015) Our study finds strong support for the influence of all three forms of intuition based diversity on entrepreneurial outcomes. More importantly, the model developed offers significant evidence of the positive influence of a TMS on entrepreneurial team performance when that TMS is based on affective- and inferential-intuitive decision styles. We suggest two possible explanations for this effect, both of which merit further testing. Our first proposal is that diversity of affective- and inferential-intuitive decision styles are more likely to generate insights that mobilise prior knowledge and prior learning, and that this is of significant value in the context of high pressure and time constrained entrepreneurial environments. Our second proposal is that the irrational quality of affective-intuitive thinking and the abstract quality of inferential-intuitive thinking lends itself to scepticism that constrains the natural exuberance of early stage entrepreneurial teams. This in turn may prevent the emergence of poorly constructed TMS. Both these possibilities provide scope for further research.

Fourthly, and finally, our findings offer a deep-level analysis of decision-making in both virtual and agile type teams that provides a counter-balance to the more rational and post-cognitive analytical frameworks which have traditionally been more prevalent in those fields. In so doing this study adds to the literature a framework that emphasises the cognitive dynamics that underpin the fluid and fast-moving decision environment that characterises the agile process in particular (Drury et al., 2012). In this respect we introduce a fresh argument for greater consideration of underlying team diversity, attention to which may yield significant benefits for actors in a variety of roles who are concerned with the software development process (Moe et al., 2012).

6. Conclusion

Scholars across the varied disciplines of entrepreneurship, information systems, small group research, and strategic decision-making stand to uncover much new ground through further exploration of the influence of intuition on both team level outcomes and the performance of new product and new venture teams. Our study offers some beginning to this analysis but is also constrained by a number of limitations. Testing of our models with a larger sample size, or in organisations with either greater maturity or environmental complexity would serve to both confirm our findings and generate a new range of theoretical insights. We nevertheless contend that the findings of this study provide sufficient evidence to conclude that intuition style diversity offers exciting new avenues for research to scholars across all of the aforementioned fields. In the age of big data, information systems that are more fine-tuned to both individual and team level decision-styles are ever more essential to the refinement of decision-making processes.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was supported by the Horizon 2020 EU-XCEL Project [644801].