Team Culture and Business Strategy Simulation Performance

Abstract

Many capstone strategic management courses use computer-based simulations as core pedagogical tools. Simulations are touted as assisting students in developing much-valued skills in strategy formation, implementation, and team management in the pursuit of superior strategic performance. However, despite their rich nature, little is known regarding the relationship between team-level attributes and simulation performance. This research reports the findings of a multiyear study that uncovered a clear link between specific team cultural values, as measured by the Competing Values Framework, and simulation performance. It then explores how these findings may influence the pedagogical use of simulations in the strategy classroom in areas ranging from using culture as a performance tool to diagnosis and training.

Keywords

strategy course capstone simulation culture teams performance

In a course long dominated by lectures and case discussions, computer-based simulations have increasingly become the experiential tool of choice for strategic management instruction (Tompson & Dass, 2000; Zantow, Knowlton, & Sharp, 2005). With nearly 1,000,000 users to date, leading simulations such as Capstone®, Foundation®, The Business Strategy Game®, and GLO-BUS® have grown in popularity with students and professors. The focal point of these simulations is to provide students with a hands-on experience to help them understand the key strategy goal of building a sustainable competitive advantage (cf. Thompson, Peteraf, Gamble, & Strickland, 2012, p. 3). As such, classic classroom simulation application requires that student teams develop the necessary skills to examine the competitive environment, develop a company strategy, and manage their firm’s internal capabilities (e.g., finance, marketing, production, and R&D) to implement the strategy effectively.

Overall, there are two fundamental requisites for effective implementation of simulation strategies: (a) an understanding of the simulation environment and (b) team dynamics. With regard to the simulation environment, a wide array of tools and objective measures of performance, such as balanced scorecards, financial ratios, and market share, are embedded in simulations to help students render decisions that achieve desired outcomes. However, with regard to the complexities of team dynamics (e.g., effective team formation, communication, and implementation of simulation tasks), the onus is on student teams to figure out how to achieve desired outcomes. In support of developing effective teams, a broad spectrum of literature exists that explores student team issues (Blaylock, McDaniel, Falk, Hollandsworth, & Kopf, 2009; Chen, Donahue, & Klimoski, 2004; Goltz, Hietapelto, Reinsch, & Tyrell, 2008; Jassawalla, Sashittal, & Malshe, 2009; Volkema, 2010). However, lacking in this literature is a serious discussion of ‘team culture’ implications in the strategy simulation context. This is surprising, in light of the fact that the concept of culture is discussed extensively in strategy teaching (Thompson et al., 2012), research (Sørensen, 2002; Zuckerman, 2002), and practice (Peters & Waterman, 1982). Therefore, the purpose of this study is to examine whether team-level culture is a predictor of simulation performance. We begin by reviewing relevant literature relating to individual- and team-level predictors of simulation performance. Then, using data from multiple strategic management classes, we examine linkages between team culture and simulation performance. Finally, we discuss implications for teaching and research.

Individual- and Team-Level Predictors of Simulation Performance

Research during the past six decades regarding predictors of simulation performance has yielded equivocal outcomes. Studies through the 1980s hypothesized that individual student factors, such as work experience, major, and grade point average (GPA), would predict simulation success on typical financial performance measures such as ROI (return on investment) and ROE (return on equity). However, Wolfe (1978, p. 317), in a comprehensive review of the literature, notes that the mixed findings of many of these early studies (e.g., Dill, 1961; Gray, 1972; McKenney & Dill, 1966; Potter, 1965; Vance, 1960; Vance & Gray, 1967) was possibly because of the fact that the simulation work under investigation was often conducted in teams, which likely polluted the results. Even studies using aggregated measure approaches, such as cumulative work experience or average GPA, did not capture true team-level phenomena (cf. Wolfe, Bowen, & Roberts, 1989) because strategy simulations are largely designed to be played by teams (Xu & Yang, 2010). Thus, the collective wisdom of this early research was that individual factors have limited predictive power for team simulation achievement (Gamlath, 2009). Unsurprisingly, given the results of these early studies, individual factor research has languished since the 1980s.

In contrast to the individual-focused research explored above, organizational studies have consistently found that team processes are correlated with performance (cf. Bierly, Stark, & Kessler, 2009; Carson, Tesluk, & Marrone, 2007; Chen et al., 2004; Yeatts & Hyten, 1998). Wolfe et al. (1989), in a survey of the then extant research, noted that simulation studies had primarily examined the relationships between either team-building activities or team cohesion and economic performance, with cohesion typically being represented as some measure of the degree to which team members like working together (Wellington & Faria, 1996). Collectively, the studies all discovered that team building exercises either had no effect on, or hampered, simulation economic performance. However, cohesion—an element of clan cultures (Cameron & Quinn, 2011, p. 180, see below)—has been found to affect simulation performance positively, negatively, and nonsignificantly (J. R. Anderson, 2005; Wellington & Faria, 1996; Wolfe et al., 1989). Wolfe et al’s. (1989) review of research noted that the natural emergence of cohesion within a team resulted in a positive economic impact in two studies and a nonsignificant impact in four studies. Their own findings indicated that teams that had received pregame training initially outperformed competitors that did not receive training, but the performance gap, while never overcome, was significantly lessened over time as other teams developed their own cohesiveness. Relatedly, Wellington and Faria (1996) discovered a correlation between increased performance and cohesion as the simulation progressed. Similar to Wolfe et al. (1989), Wellington and Faria (1996) found that teams that started the simulation with higher levels of cohesion ultimately outperformed teams with lower initial levels of cohesion. Conversely, J. R. Anderson’s (2005) study, using the Foundation simulation, found a relationship between increased team cohesion and decreased performance. Finally, Luxon (1992) and Neal (1997) explored competitive disposition (self-assessments regarding one’s orientation toward competitiveness)—an element of market cultures (Cameron & Quinn, 2011, p. 180, see below)—and performance. Despite controlling for simulation (MARKSTRAT2) and the measurement of competitive disposition (the Sales Performance Index), Neal (1997) and Luxon (1992) arrived at different conclusions, likely because of differences in team starting conditions. Luxon discovered a positive relationship between competitive disposition and performance, whereas Neal found a correlation between the emergence of group cohesion as the simulation progressed and economic performance, but not between competitive disposition and economic performance.

We believe that the confusion regarding the effects of team-level factors on the simulation performance relationship emanates from the fact that many studies (e.g., J. R. Anderson, 2005; Luxon, 1992; Neal, 1997) aggregated individually focused constructs and did not use true team-level measures (i.e., those explicitly designed to assess team-level phenomena). This is unsurprising since the research methods literature (cf. Arthur, Bell, & Edwards, 2007; Chan, 1998) has highlighted significant problems created by summing individual factor scores in team-level research. The lack of consensus among the results regarding simulation performance also suggests that the specific team measures employed may be part of the problem. In particular, cohesion and competitive disposition are typically measured as unidimensional constructs, which unrealistically simplifies the “team” concept (Chan, 1998; Mullen & Copper, 1994; Zaccaro, 1991). More problematically, these measures are by-products of, and reductionist approaches to, the larger phenomenon of organization culture (Sánchez & Yurrebaso, 2009).

In light of the issues associated with aggregating individual measures, we believe that using a true organization-level construct, such as culture, more accurately reflects macro-level team phenomena. Culture typically represents shared beliefs and values within groups (cf. Deal & Kennedy, 1982; Sánchez & Yurrebaso, 2009). Culture also has the advantage of being multidimensional—it is a shared collection of organizational-level values and norms and not a single factor such as “cohesion” (Deal & Kennedy, 1982, 2000; Sørensen, 2002; Zuckerman, 2002). Thus, it is not enough to be “strong” or “weak,” but rather the constellation of cultural values needs to be configured to improve firm performance (Cameron & Quinn, 2011; Deal & Kennedy, 1982; Sørensen, 2002; Zuckerman, 2002), which may partly explain the mixed results in the team simulation performance literature noted above. Strong cultures can be the reason for poor or strong outcomes and they may provide varying outcomes at different times (Cameron & Quinn, 2011; Sørensen, 2002; Zuckerman, 2002). As Zuckerman (2002) states “researchers have found that companies with the strongest cultures—where values and norms are widely shared and strongly held—tend to outperform their peers” (p. 158).

This contingent and configurational approach to culture and performance is a major theme in the literature (cf. Cameron & Quinn, 2011; Deal & Kennedy, 1982, 2000; Sørensen, 2002; Zuckerman, 2002). What is lacking is classroom-based research regarding the relationship between broader team constructs, such as culture, and their performance connections in strategy simulations. The literature insists that some culture subcomponents influence team performance in simulations, though as noted, the specific elements and the nature of their impact remains unclear, especially regarding cohesion and competitive disposition. Finally, the literature asserts that cultures, as values and beliefs systems, are likely to have different configurations that lead to competitive success in different environments (Cameron & Quinn, 2011; Schein, 2010). Because of the limited and conflicting guidance regarding specific cultural conditions that may be associated with overall team simulation performance, we approached this issue from an exploratory perspective rather than using firm hypotheses. Therefore, the following two questions guide our research:

Research Question 1: What specific team culture components are associated positively with strategy simulation performance?

Research Question 2: What team culture configurations are associated positively with strategy simulation performance?

Method

This study was conducted as part of an ongoing classroom exercise designed to help undergraduate students participating in a business strategy simulation to understand better the relationship between culture and organization performance. In each class in the study, students were directed to complete individually Cameron and Quinn’s (2011) Organizational Culture Assessment Instrument (OCAI) as it related to their own simulation teams. Individual results were then used to produce a team score on each OCAI construct. Team simulation performance was measured using both rankings (e.g., 1st place, 2nd place) and standardized performance scores. Research Question 1 was tested by regressions of team performance against team scores on each OCAI value. Research Question 2 was tested by using cluster analyses to examine the relationship between OCAI-based cultural configurations and simulation performance. The sample, instrument, and variables are described in more detail below. The analysis procedures are described in the results section.

Sample

The sample consisted of undergraduates enrolled in typical senior-level capstone business strategy courses in two public, midsized, AACSB accredited business schools in the southeastern United States. Factor analysis data to validate the study’s instrument, the OCAI, was collected during the years 2005-2008 from strategy classes taught by the first author using Capstone Business Simulation and the third author using The Business Strategy Game (BSG) as part of a regular classroom teaching exercise designed to help students understand culture in a strategic context. Data separate from the factor analysis data were collected through 2011 by the first author (who moved from the first to the second school during the study) and the second author who used the BSG. Study data were collected from all the strategy class sections taught by the first and second authors, resulting in a total sample size of 186 teams in 30 classes. The authors lectured on the importance of diversity of functional expertise (e.g., accounting, management, finance, marketing, and computer information systems) and formed teams, dependent on class size and most frequently containing four to five members per team, by the second week of the semester.

Instrument

Cameron and Quinn’s (2006, 2011) OCAI was used to measure the dimensions of team culture since it demonstrates strong reliability, displays concurrent, discriminant, and divergent validity, and is widely used in practice (Cameron, 1986; Cameron & Quinn, 2011; Kalliath, Bluedorn, & Gillespie, 1999; Quinn & Spreitzer, 1991; Wicks & St. Clair, 2007; Yeung, Brockbank, & Ulrich, 1991; Zammuto & Krakower, 1991). It also has been successfully used in business classroom instruction (Cameron & Quinn, 2011).

The OCAI operationalizes the Competing Values Framework (CVF)—“the dominant framework in the world for assessing organizational culture” (Cameron & Quinn, 2011, p. 35). The CVF captures values about performance at the organizational level (pp. 38-41) and demonstrates robust validity and reliability across organization and subunit levels (pp. 22-23). The OCAI’s items are constructed with a referent-shift focus, thus overcoming some of the problems of using individual factor aggregated instruments to assess group phenomena (Klein, Conn, Smith, & Sorra, 2001). It also offers both a multidimensional and configurational approach to culture with explicit organizational outcomes links.

The CVF combines two continuums regarding an organization’s beliefs and values (Quinn & Rohrbaugh, 1981, 1983). The following is a brief summary of the CVF (a more complete description can be found in Cameron & Quinn, 2011, pp. 38-51). One axis represents values of flexibility (greater discretion) versus stability (higher levels of order). The second axis represents values emphasizing either an internal focus on the organization (such as its employees or its processes) or its external relationship with the working environment.

The two axes form four organization culture types. The first, clan—flexibility and discretion coupled with internal focus and integration—stresses the importance of participation, cohesion, shared values, commitment, and high morale. Clan is referred to as the collaborate quadrant in practice. The second, adhocracy—flexibility and discretion coupled with external focus and differentiation—assumes innovation and initiative lead to success and encourages entrepreneurial, creative, and visionary behavior. It is called the create quadrant. The third, hierarchy—stability and control coupled with internal focus and integration—is characterized by a structured workplace with formal rules and policies and a focus on efficiency, timeliness, and control. It is called the control quadrant. Finally, market—stability and control coupled with external focus and differentiation—perceives the external environment as hostile with choosy consumers and a need for the organization both to be results and production oriented. It is called the compete quadrant. An organization can achieve low to high scores on each type independently of the scores on the other measures. For example, Cameron and Quinn (2011, pp. 90-91) list cultural configurations associated with specific industries where performance differences vary depending on the context.

The four CVF constructs were operationalized using the 24 questions from Cameron and Quinn’s (2011, pp. 30-32) OCAI. Following their recommendations for practice (pp. 184-185), a Likert-type scale ranging from 1 (least similar) to 5 (most similar) was added to each item to facilitate measurement in a questionnaire format, maximize independence of the response items, and follow a more conservative path of producing less differentiation between the values quadrants. As a check on the reliability and the validity of items for each of the scales, we tested the instrument via principal components analysis using a varimax rotation on a representative sample of 169 students from strategic management classes using strategy simulations. Cronbach tests revealed that the clan, market, adhocracy, and hierarchy scales displayed alphas of .84, .76, .76, and .74, respectively, well above Nunally’s (1978) .70 cutoff criteria. However, the analysis revealed that coefficients for adhocracy displayed multiple cross-loadings, demonstrating that the associated questionnaire items did not significantly discriminate from the other factors. The six adhocracy items were eliminated from the analysis. We collected additional data, for a total sample of 296 students, and conducted an additional factor analysis (see Table 1) using the remaining three constructs of clan, market, and hierarchy. The analysis revealed that the majority of the coefficients for each of the questionnaire items were well above the acceptable cutoff threshold of .40 (Hatcher & Stepanski, 1994) for each culture dimension. These three scales were retained for a confirmatory factor analysis using nested model comparisons (Kelloway, 1996, 1998).

Table 1.

Competing Values Framework Scale Factor Loadings.

Item	Clan	Market	Hierarchy
Q13	0.77	0.14	0.20
Q17	0.75	0.02	0.26
Q1	0.75	0.08	0.08
Q9	0.69	0.10	0.05
Q5	0.67	0.02	0.20
Q21	0.63	−0.13	0.45
Q15	0.04	0.77	0.11
Q19	0.10	0.75	0.07
Q11	−0.02	0.69	0.22
Q23	0.09	0.68	0.05
Q3	0.35	0.67	0.10
Q7	−0.15	0.60	0.11
Q16	−0.02	0.15	0.80
Q20	0.28	0.06	0.72
Q24	0.18	0.27	0.64
Q8	0.38	0.23	0.52
Q4	0.16	0.33	0.51
Q12	0.30	−0.04	0.42
α	.84	.76	.74
Mean	3.83	3.72	3.57

Note: N = 296. Varimax rotation.

The three-factor model was compared with alternative null (single factor) and two-factor models. Results indicated that the three-factor model (χ² = 306.52, 129 df; p < .01; goodness-of-fit index [GFI] = .90; normed fit index [NFI] = .92; relative fit index [RFI] = .91; comparative fit index [CFI] = .95) displayed superior fit with the data (see Table 2 for results of the nested model comparisons). All hypothesized item paths for their respective latent constructs in the three-factor model were significant at p < .001.

Table 2.

Confirmatory Factor Analysis Comparison of Models.

Model	χ²	df	p, Δχ²	p, Model	GFI	NFI	CFI	RFI
Null	1034.33	131	<.001	<.001	.72	.81	.84	.78
Two-factor	474.55	130	<.001	<.001	.85	.89	.92	.87
Three-factor	306.52	129	<.001	<.001	.90	.92	.95	.91

Note: df = degrees of freedom; GFI = goodness-of-fit index; NFI = normed fit index; CFI = comparative fit index; RFI = relative fit index.

Variables

The OCAI was completed by students individually in the classroom near the end or shortly after the completion of each course’s simulation (typically after 8 to 10 “decision rounds,” which included 2-3 preliminary “practice” decisions). Consistent with the time horizons of other team-focused simulation research (cf. Neal, 1997, Table 2) and other student team studies (Chen et al., 2004; Dahlin, Weingart, & Hinds, 2005), the OCAI was administered after teams had operated for approximately 2 months, thus providing adequate time for cultures to form (cf. Schein, 2010, especially p. 67, pp. 197-218). The independent culture variables hierarchy, clan, and market were then calculated as the average score for each team on each measure. Possible score ranges for each measure were 6 to 30.

We chose two dependent variables, both of which were calculated using each class’s final simulation-generated performance measures: score and rank. For Capstone, score was the default weighted sum of annual profit, cumulative profit, stock price, return on equity, return on sales, return on assets, and asset turnover, resulting in simulation-generated whole number values ranging from 0 to 100. For The Business Strategy Game, score was the default combination of earnings per share, return on equity, stock price, credit rating, and firm image rating, resulting in simulation-generated whole number values ranging from 0 to 110. Since slightly different score ranges existed between the two simulations, we standardized the data collected from the use of each simulation before combining the data for use in our analyses. Rank was calculated by converting performance scores to simple rankings (e.g., 1 for first place, 2 for second place).

Results

Descriptive statistics and intercorrelations for the variables (clan, hierarchy, and market cultures; performance score and rank; instructor; school; and number of teams per class) are presented in Table 3. The descriptive statistics indicate that the sample varied in performance and along the cultural and control variables examined.

Table 3.

Correlations and Descriptive Statistics (N = 186).

	Mean	SD	1	2	3	4	5	6	7
1. Clan culture	22.184	3.214
2. Market culture	22.928	3.148	.157
3. Hierarchy culture	20.495	2.378	.504	.284
4. Instructor	0.516	0.501	−.166	−.020	−.275
5. School	0.774	0.419	.271	−106	−273	−.523
6. Number of teams	6.366	1.048	.175	−.068	.178	−.650	.189
7. Performance score	0.000	1.000	.050	.594	.200	.001	.208	−.027
8. Performance rank	3.575	1.868	−.011	−.657	−.084	−.209	.098	.292	−.714

Note: All correlations greater than .16 are significant at p < .05. All correlations greater than .20 are significant at p < .01. All correlations greater than .24 are significant at p < .001.

Research Question 1, concerning the relationship between culture and simulation performance, was analyzed using ordinary least squares (OLS) regression. The dependent variables for performance (score and rank) were regressed on market, clan, and hierarchy cultures in two separate analyses. Analyses also were conducted to control for number of teams per class, instructor (and thus, simulation), and school.

The OLS results for Research Question 1 are presented in Table 4. The models’ adjusted R² were .38 (p < .001) and .49 (p < .001) for the performance score and performance rank measures, respectively. Results from both analyses revealed that the market culture was a significant predictor of simulation performance while clan and hierarchy were not. Specifically, teams with the highest level of market culture displayed the highest level of overall performance (rank, β = −0.39, p < .001; score, β = 0.18, p < .001) suggesting that teams who develop a highly competitive results-orientated culture perform the best. The negative coefficient in the rank analysis is the result of coding high performers with the rank of one. The rank-based regression also revealed that number of teams was significant (β = 0.36, p < .01) whereas the score-based regression indicated that both instructor (β = 0.43, p < .05) and school (β = 0.60, p < .001) were significant. We tested for the presence of multicollinearity using tolerance and variance inflation factor (VIF) diagnostics and determined that multicollinearity did not affect the betas, as all tolerances were above the critical value of .10 and all VIFs were well below the critical value of 10.

Table 4.

Ordinary Least Squares Regression Results.

	Model 1 (Performance Score)		Model 2 (Performance Score)		Model 3 (Performance Rank)		Model 4 (Performance Rank)
Variables	β	Standard Error	β	Standard Error	β	Standard Error	β	Standard Error
Instructor	0.34	0.22			−0.02	0.41
School	0.69***	0.21			0.19	0.38
Number of teams	0.03	0.09			0.50**	0.17
Instructor			0.43*	0.18			−0.09	0.31
School			0.60***	0.17			0.47	0.29
Number of teams			0.11	0.08			0.36**	0.13
Clan culture			−0.04†	0.02			0.01	0.04
Market culture			0.18***	0.02			−0.39***	0.03
Hierarchy culture			0.03	0.03			0.02	0.05
F	3.86*		19.97***		5.81***		31.01***
R²	.06		.40		.09		.51
Adjusted R²	.04		.38		.07		.49
df	182		179		182		179
N	186		186		186		186

†

p < .10. *p < .05. **p < .01. ***p < .001.

Cluster analysis was used to identify team culture configurations for Research Question 2. After preprocessing the data using the ACECLUS procedure in SAS, we used Ward’s minimum variance method for clustering the culture dimensions of market, clan, and hierarchy. To identify the optimal cluster solutions, we used the following decision criteria in conjunction with visual inspection of the tree plots: (a) the clusters explain at least 65% of the overall variance, (b) with an additional cluster increasing the overall fit by less than 5%, (c) with a local peak in the Cubic Clustering Criterion, (d) with a local peak in the pseudo F statistic combined, (e) with a small value of the pseudo t2 statistic and a larger pseudo t2 statistic for the next cluster fusion. When discrepancies existed across these rules, we relied on visual inspection of the tree plots and prioritized our use of each rule in the order in which they are listed above. These decision criteria are consistent with those used in prior configurational research (e.g., Fiegenbaum & Thomas, 1990; Ketchen, Thomas, & Snow, 1993; Marlin, Ritchie, & Geiger, 2009) and with clustering stopping rules recommended by the SAS Manual (SAS Institute, 2000) and by SAS Technical Report A-108 (SAS Institute, 1983).

Once our cluster analysis was complete, we ran pairwise means comparisons on the identified clusters, each representing a different team culture configuration to determine whether the clusters differed on the underlying team culture dimensions and thus whether our analysis produced different team culture configurations. Next, least-squares means (i.e., means adjusted for our three controls: number of teams, instructor, and school) comparisons were performed to test for and identify specific performance differences between the identified configurations.

Cluster analysis, analysis of variance (ANOVA), and pairwise means testing results are presented in Table 5. Our cluster analysis for Research Question 2 produced four team culture configurations with cluster sizes ranging from 37 (in Configuration 4) to 63 (in Configuration 1). The results of ANOVA and pairwise tests indicate significant between-configuration differences (p < .05) along all three team culture dimensions suggesting that the cluster analysis produced distinct configurations. More specifically, Configuration 1 ranked high on market culture and was highest on hierarchy culture, yet was second highest on clan culture; Configuration 2 ranked highest on market culture, however, hierarchy and clan culture were low; Configuration 3 ranked the highest on clan culture, yet hierarchy culture and market culture were low; Configuration 4 displayed low hierarchy culture scores, and was also lowest on both market culture and clan culture. We also found that culture Configurations 1 and 2 outperformed (p < .05) Configurations 3 and 4, which are discussed in more detail in the next section.

Table 5.

Cluster Analysis Results.

	Cluster Means				Mean Comparisons*
	1	2	3	4
Clan culture	22.99	21.77	24.45	19.02	3 > 1 > 2 > 4
Market culture	24.73	24.90	21.01	19.26	2, 1 > 3 > 4
Hierarchy culture	22.88	19.37	19.07	19.35	1 > 2, 4, 3
Performance score^a	0.34	0.40	−0.52	−0.55	2, 1 > 3, 4
Performance rank^a,b	2.91	2.92	4.37	4.73	1, 2 > 3, 4
N	63	48	38	37	186

Least square means controlling for instructor, school, and number of teams.

Lower values of rank indicate higher levels of performance.

p < .05.

Discussion

This study explored linkages between team culture and simulation performance in undergraduate capstone strategy courses. We found that market values and certain cultural configurations affect a team’s simulation performance. In the following paragraphs, we explain the role of culture as a performance diagnostic in simulation-based environments.

Culture and Team Simulation Performance

This study’s primary finding is that team culture accounts for a large portion of variance in simulation performance. Specifically, after accounting for the control variables, the culture dimensions in this analysis explained 34% and 42% of the simulation score and rank outcomes, respectively. Interestingly, our findings indicate that the highly competitive results orientation of the market culture (Cameron & Quinn, 2011) is associated with positive financial outcomes in popular strategy simulations. This lends credence to the idea that simulations are achieving their stated pedagogical goals of encouraging competitively focused behavior.

Our results are also consistent with organizational research that asserts that “strong culture equals high performance” (e.g., Deal & Kennedy, 1982, 2000; Peters & Waterman, 1982; Sørensen, 2002; Zuckerman, 2002) as the highest performing teams in our study did indeed display strong cultures. However, while cultural strength is a necessary condition for performance, it is not sufficient. Our results indicate that it is important to consider the types of culture and alignment with outcomes in order to gain maximum performance benefits.

Our findings also provide specific insights regarding linkages between the multidimensional nature of culture and simulation performance. High performing teams (i.e., members of Configurations 1 and 2) either had the absolute strongest market culture or simultaneously displayed strong to moderate scores on all three cultural dimensions. This reinforces the notion that high-performing cultures are composed of varying levels of well-configured, multidimensional values. Furthermore, the relationship between culture and performance depends on the environmental context. In the case of this study, cultures that included market orientations were rewarded with high levels of simulation performance. In contrast, teams that lacked specific culture configurations performed poorly. For example, the highest absolute clan scores (e.g., teams in Configuration 3) were not sufficient to achieve strong performance. Indeed, despite the presence of strong values, our findings suggest that culture must contain key specific performance supporting aspects (in our case, strong market values present in Configurations 1 and 2) and culture’s other dimensions must be aligned to support, or not interfere with, success (as with Configuration 2). Finally, firms in Configuration 4, with low scores on all dimensions emerged as classic “weak culture” organizations with all their attendant performance problems. Thus, our results provide instructors with multiple opportunities to explore culture, which is often an abstraction for many students (Bisel, Messersmith, & Keyton, 2010; Colakoglu & Littlefield, 2011). Instructors are also equipped with relevant examples that illustrate how cultural configurations can either aid or inhibit performance.

Our findings also shed new light on prior equivocal results related to cohesion (clan), competitive disposition (market), and simulation performance (J. R. Anderson, 2005; Luxon, 1992; Neal, 1997; Wellington & Faria, 1996; Wolfe et al., 1989). Cameron and Quinn (2011, pp. 64-65) argue that organizations often progress through a cultural configuration “life cycle” over time from initial adhocracy dominance to clan, hierarchy, and finally market dominance. From this perspective, it is conceivable that early simulation studies that linked clan prominence with high performance actually consisted of teams that possessed high market value levels that were unmeasured. Conversely, research that found cohesion performance problems possibly studied teams “stuck” in the early life cycle stage. These teams may not have possessed, and the studies did not measure, the strong market value orientation of the later stage (Cameron & Quinn, 2011, p. 92). In sum, earlier confusion regarding antecedents of simulation performance may have been clarified through the use of the multidimensional CVF to provide possible evidence where teams were at in the cultural development life cycle.

Our results also provide evidence for links between cultural clarity and performance. In the regression analysis involving the performance score, the highest performing configuration displayed one primary value—market—whereas the second highest displayed a configuration of high market, clan, and hierarchy values (the results were inverted in the rank regression). While both configurations displayed significant performance outcomes, we speculate that this may be evidence that some teams had not completed their developmental cycle, or were transitioning from one configuration to another, and thus incurred attentional distractions, such as continuing to focus on clan and hierarchy issues. This may have prevented them from achieving maximal performance vis-à-vis the market-focused cluster, even while it aided performance compared with the other two configurations. Conversely, the alternative explanation from the performance rank regression results, but less well-aligned with CVF theory, is that the highest performing teams were able to devote excess/slack attentional resources to maintaining strong cultures on all three dimensions as they progressed through the cycle.

Instructors will want to carefully plan how they present these insights. If culture is introduced early in a course, teams might be advised simply to treat its diagnosis as one aspect of relative competitive strength or weakness to be used in strategy analysis and formation. Teams may also be encouraged to diagnose and monitor their own cultures as well as their competitors’ culture as part of a key component of simulation success. If culture is introduced after simulation completion, it presents an opportunity, regardless of the results, to explore its impact in the context of the larger strategy–performance–competitive advantage framework. We believe that decisions regarding how to present and analyze culture will be especially important for instructors who favor teaching strategy using a structural perspective.

Culture Diagnoses and Training in Simulation Environments

Tough questions arise from our results for those teaching culture from a strategy process perspective, namely how to engage in assessment, training, and intervention? There is also a need to ensure that we are obtaining a reliable measure of our phenomena—although our findings suggest a strong link between specific cultural configurations and simulation performance, they are not absolute predictors of success. Furthermore, culture as a collection of shared values emerges over time and varies in terms of intensity and integration (Rousseau, 1990; Schein, 2010).

Thus, premature assessment of culture, particularly in the time span of a simulation, might result in spurious conclusions and ineffective actions either by the instructor or the team. Conversely, an educator might conclude that it is best to identify each team’s culture early in the course and immediately invoke unique interventions aimed at correcting any obvious issues, or the instructor may choose to provide consistent cultural training to all teams prior to the onset of simulation play to improve overall performance. However, the desirability and effectiveness of such potential interventions needs to be considered in relation to the relatively short course time frames available for teams to progress from cultural “forming” to “performing.” Furthermore, while Cameron and Quinn (2011) indicate a natural time-based serial cultural developmental model in organizations, we have little guidance regarding whether classroom-based simulation teams can make “quantum leaps” from founding conditions to market orientations without having to progress through intermediate configurations, or whether the competing values systems can develop in parallel. Therefore, instructors need to weigh the critical time constraints needed for active interventions to manipulate culture, and the chances of their success, against the possibility of allowing culture to develop naturally within teams. Finally, instructors’ decisions have to be made against the background of which approach is likely to support better their overall learning and assessment objectives.

Finally, many teachers using simulations are already engaged in some form of team intervention—should they engage in more? For example, to maximize team member diversity and ensure fairness, instructors often undertake the team assignment process themselves. Others allow students to form teams with the advice that teams should aim for maximum member functional diversity. Thus, it is plausible that an instrument could be designed and administered to students prior to forming teams to assess their affinity for various CVF measures and/or to provide data for subsequent cultural training. This approach may be particularly applicable for process-focused strategy instructors interested in improving learning regarding team effectiveness. Thus, composing teams based on a number of diagnostic assessments may accelerate the forming to performing process and lead to attainment more quickly of an effective team culture if this is indeed a critical learning objective.

However, attempts by teams to engage in development and change activities aimed at achieving market-dominated cultures (if early reliable assessments can be made) may not be the wisest use of their energies. If team training and development activities are to be undertaken, they could arguably be more valuable in the near term if aimed at directly improving hierarchical values, such as the conduct of meetings and effectiveness of communications (Baker, 2010) before focusing on the development of market values. Furthermore, culture may not be the most immediately important factor, such as in crisis situations where simulation teams make serious misjudgments. Culture is only one aspect to be considered by a team in plotting a strategy for competing in a simulation game. In summary, the decision about how to interpret this research and a team’s management of its own culture is but one piece of information to be considered by students in plotting and executing strategy.

Limitations and Future Research

One limitation of this study is that it was undertaken at two midsized public universities in the southeastern United States. It would be interesting to test our research questions in other milieu, such as private colleges and larger institutions where student environments differ. Furthermore, this study centered on a single culture framework (Cameron & Quinn, 2011), albeit the one most widely used in practice. Other frameworks, such as Hofstede’s (1980), might afford additional insights.

Second, this study examined simulation performance and did not assess student learning. Since simulation providers are increasingly attempting to connect their products to outcomes assessment, we need additional studies that demonstrate strong simulation performance equals “learning.” Our results indicate that upward of 50% of simulation performance is tied to team culture; therefore, it is worthy to ask, what are the student learning outcomes? The connection between learning and simulation performance remains woefully underexplored (P. H. Anderson & Lawton, 2009; Xu & Yang, 2010).

As noted earlier, Cameron and Quinn (2011, pp. 64-65) assert that organizations often follow a predictable cultural life cycle development path. Our analysis presents tentative evidence that successful teams managed to travel further down this path by developing dominant market cultures while clan- and hierarchy-focused teams did not progress as far. Consequently, earlier research that used only cohesion measures likely focused on the initial stages of culture development and change and thus produced mixed findings through their exclusion of later emerging cultural values. Likewise, our results and CVF theory may indicate that configurations that demonstrate cultural clarity, where one value is stronger than the other two, will achieve simulation success. This suggests the following three propositions:

Proposition 1: Teams will typically follow a progression from clan, to hierarchy, to market cultures over the duration of a simulation.

Proposition 2: Teams progressing the fastest through the stages—from clan, to hierarchy, to market cultures—will demonstrate higher levels of simulation performance.

Proposition 3: Teams with culture configurations dominated by one culture type will demonstrate higher levels of simulation performance than teams with multiple cultural values of similar strength.

If these propositions are verified, instructors will have more incentive to revisit the questions regarding team training and development while remembering the caveats previously discussed.

Finally, a fruitful research direction might explore whether adhocracy values affect the simulation environment. The CVF measures are unique (Cameron & Quinn, 2011), and culture’s impact is contingent on the environment (Sørensen, 2002; Zuckerman, 2002). Simulations’ rules, processes, and relatively predictable environments give teams scant need to adopt innovative values. Thus, our lack of an adhocracy factor likely is more the nature of simulation environments rather than instrumentation issues. However, theory suggests that simulations that introduce significant exogenous shocks—such as unexpected tariff wars—could witness the emergence of adhocracy values. This suggests a fourth future research proposition:

Proposition 4: Simulation environments that introduce significant exogenous shocks into game play will see the emergence of distinct OCAI adhocracy values.

If this proposition is supported by research, instructors will then have additional opportunities to adjust their simulations as a pedagogical tool for helping students gain further insights about the strategic role of culture in organizations.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Anderson

J. R.

(2005). The relationship between student perceptions of team dynamics and simulation game outcomes: An individual level analysis. Journal of Education for Business, 81(2), 85-90.

Anderson

P. H.

Lawton

(2009). Business simulations and cognitive learning: Developments, desires, and future directions. Simulation & Gaming, 40, 193-216.

Arthur

Jr. Bell

S. T.

Edwards

B. D.

(2007). A longitudinal examination of the comparative criterion-related validity of additive and referent-shift consensus operationalizations of team efficacy. Organizational Research Methods, 10, 35-58.

Baker

D. F.

(2010). Enhancing group decision making: An exercise to reduce shared information bias. Journal of Management Education, 34, 249-279.

Bierly

P. E.

III Stark

E. M.

Kessler

E. H.

(2009). The moderating effects of virtuality on the antecedents and outcome of NPD team trust. Journal of Product Innovation Management, 26, 551-565.

Bisel

R. S.

Messersmith

A. S.

Keyton

(2010). Understanding culture and communication through a gyroscope metaphor. Journal of Management Education, 34, 342-366.

Blaylock

B. K.

McDaniel

J. L.

Falk

C. F.

Hollandsworth

Kopf

J. M.

(2009). A borrowed approach for a more effective business education. Journal of Management Education, 23, 577-595.

Cameron

K. S.

(1986). Effectiveness as paradox: Consensus and conflict in conceptions of organizational effectiveness. Management Science, 32, 539-553.

Cameron

K. S.

Quinn

R. E.

(2006). Diagnosing and changing organizational culture: Based on the competing values framework (Rev. ed.). San Francisco, CA: Jossey-Bass.

10.

Cameron

K. S.

Quinn

R. E.

(2011). Diagnosing and changing organizational culture: Based on the competing values framework (3rd ed.). San Francisco, CA: Jossey-Bass.

11.

Carson

J. B.

Tesluk

P. E.

Marrone

J. A.

(2007). Shared leadership in teams: An investigation of antecedent conditions and performance. Academy of Management Journal, 50, 1217-1234.

12.

Chan

(1998). Functional relations among constructs in the same content domain at different levels of analysis. Journal of Applied Psychology, 83, 234-246.

13.

Chen

Donahue

L. M.

Klimoski

R. J.

(2004). Training undergraduates to work in organizational teams. Academy of Management Learning & Education, 3, 27-40.

14.

Colakoglu

Littlefield

(2011). Teaching organizational culture using a projective technique: Collage construction. Journal of Management Education, 35, 564-585.

15.

Dahlin

K. B.

Weingart

L. R.

Hinds

P. J.

(2005). Team diversity and information use. Academy of Management Journal, 48, 1107-1123.

16.

Deal

T. E.

Kennedy

A. A.

(1982). Corporate cultures: The rites and rituals of corporate life. Reading, MA: Addison-Wesley.

17.

Deal

T. E.

Kennedy

A. A.

(2000). The new corporate cultures: Revitalizing the workplace after downsizing, mergers, and reengineering. Cambridge, MA: Perseus.

18.

Dill

W. R.

(1961). The educational effects of management games. In Dill

W. R.

Jackson

J. R.

Sweeney

J. W.

(Eds.), Proceedings of the conference on business games as teaching devices (pp. 61-68). New Orleans, LA: Tulane University.

19.

Fiegenbaum

Thomas

(1990). Strategic groups and performance: The U.S. insurance industry, 1974-84. Strategic Management Journal, 11, 197-215.

20.

Gamlath

S. L.

(2009). Field testing two simulation games: Do winners win consistently? On the Horizon, 17, 388-396.

21.

Goltz

S. M.

Hietapelto

A. B.

Reinsch

R. W.

Tyrell

S. K.

(2008). Teaching teamwork and problem solving concurrently. Journal of Management Education, 32, 541-562.

22.

Gray

C. F.

(1972, April). Performance as a criterion variable in measuring business gaming success: An experiment with a multiple objective performance model. Paper presented at the Southeastern American Institute of Decision Sciences Conference, Memphis, TN.

23.

Hatcher

Stepanski

E. J.

(1994). A step-by-step approach to using the SAS system for univariate and multivariate statistics. Cary, NC: SAS Institute.

24.

Hofstede

G. H.

(1980). Culture’s consequences. Thousand Oaks, CA: Sage.

25.

Jassawalla

Sashittal

Malshe

(2009). Students’ perceptions of social loafing: Its antecedents and consequences in undergraduate business classroom teams. Academy of Management Learning & Education, 8, 42-54.

26.

Kalliath

T. J.

Bluedorn

A. C.

Gillespie

D. F.

(1999). A confirmatory factor analysis of the competing values instrument. Educational and Psychological Measurement, 59, 143-158.

27.

Kelloway

E. K.

(1996). Using LISREL for structural equation modeling: A researcher’s guide. Thousand Oaks, CA.: Sage.

28.

Kelloway

E. K.

(1998). Using LISREL for structural equation modeling. Thousand Oaks, CA: Sage.

29.

Ketchen

D. J.

Thomas

J. B.

Snow

C. C.

(1993). Organizational configurations and performance: A comparison of theoretical approaches. Academy of Management Journal, 36, 1278-1313.

30.

Klein

K. J.

Conn

A. B.

Smith

D. B.

Sorra

J. S.

(2001). Is everyone in agreement? An exploration of within-group agreement in employee perceptions of the work environment. Journal of Applied Pyschology, 86(2), 3-16.

31.

Luxon

(1992). Competition, performance and learning: Implications for business games and simulations. Unpublished master’s thesis. University of Canterbury, New Zealand.

32.

Marlin

Ritchie

W. J.

Geiger

S. W.

(2009). Strategic group membership and nonprofit organization performance. Nonprofit Management and Leadership, 20, 23-39.

33.

McKenney

J. L.

Dill

W. R.

(1966). Influences on learning in simulation games. American Behavioral Scientist, 10(2), 28-32.

34.

Mullen

Copper

(1994). The relations between group cohesiveness and performance: An integration. Psychological Bulletin, 115, 210-227.

35.

Neal

D. J.

(1997). Group competitiveness and cohesion in a business simulation. Simulation & Gaming, 28, 460-476.

36.

Nunnally

J. C.

(1978). Psychometric theory (2nd ed.). New York, NY: McGraw-Hill.

37.

Peters

T. J.

Waterman

R. H.

Jr. (1982). In search of excellence: Lessons from America’s best-run companies. New York, NY: Harper & Row.

38.

Potter

G. B.

(1965). An exploratory study of psychological factors in business simulation games. Unpublished master’s thesis, University of Illinois at Urbana-Champagne.

39.

Quinn

R. E.

Rohrbaugh

(1981). A competing values approach to organizational effectiveness. Public Productivity Review, 5, 122-140.

40.

Quinn

R. E.

Rohrbaugh

(1983). A spatial model of effectiveness criteria: Towards a competing values approach to organizational analysis. Management Science, 29, 363-377.

41.

Quinn

R. E.

Spreitzer

G. M.

(1991). The psychometrics of the competing values culture instrument and an analysis of the impact of the organizational culture on quality of life. In Woodman

Pasmore

(Eds.), Research in organizational change and development (Vol. 5, pp. 115-142). Greenwich, CT: JAI Press.

42.

Rousseau

D. M.

(1990). Assessing organizational culture: The case for multiple methods. In Schneider

(Ed.), Organizational climate and culture (pp. 153-192). San Francisco, CA: Jossey-Bass.

43.

Sánchez

J. C.

Yurrebaso

(2009). Group cohesion: Relationships with work team culture. Psicothema, 21(1), 97-104.

44.

SAS Institute. (1983). SAS technical report A-108: Cubic clustering criterion. Cary, NC: Author.

45.

SAS Institute. (2000). SAS/STAT user’s guide (8th ed.). Cary, NC: Author.

46.

Schein

E. H.

(2010). Organizational culture and leadership (4th ed.). San Francisco, CA: Jossey-Bass.

47.

Sørensen

J. B.

(2002). The strength of corporate culture and the reliability of firm performance. Administrative Science Quarterly, 47, 70-91.

48.

Thompson

A. A.

Peteraf

M. A.

Gamble

J. E.

Strickland

A. J.

(2012). Crafting and executing strategy: The quest for competitive advantage: Concepts and cases (18th ed.). New York, NY: McGraw-Hill Irwin.

49.

Tompson

G. H.

Dass

(2000). Improving students’ self-efficacy in strategic management: The relative impact of cases and simulations. Simulation & Gaming, 31, 22-41.

50.

Vance

S. C.

(1960). Management decision simulation. New York, NY: McGraw-Hill.

51.

Vance

S. C.

Gray

C. F.

(1967). Use of a performance evaluation model for research in business gaming. Academy of Management Journal, 10, 27-37.

52.

Volkema

R. J.

(2010). Designing effective projects: Decision options for maximizing learning and project success. Journal of Management Education, 34, 527-550.

53.

Wellington

W. J.

Faria

A. J.

(1996). Team cohesion, player attitude, and performance expectations in simulation. Simulation & Gaming, 27, 23-40.

54.

Wicks

A. M.

St. Clair

(2007). Competing values in healthcare: Balancing the (un)balanced scorecard. Journal of Healthcare Management, 52, 309-323.

55.

Wolfe

(1978). Correlations between academic achievement, aptitude, and business game performance. Exploring Experiential Learning: Simulations and Experiential Exercises, 5, 316-324.

56.

Wolfe

Bowen

D. D.

Roberts

C. R.

(1989). Team-building effects on company performance: A business game-based study. Simulation & Gaming, 20, 388-408.

57.

Yang

(2010). Student learning in business simulations: An empirical investigation. Journal of Education for Business, 85, 223-228.

58.

Yeatts

D. E.

Hyten

(1998). High-performing self-managed work teams: A comparison of theory to practice. Thousand Oaks, CA: Sage.

59.

Yeung

A. K. O.

Brockbank

Ulrich

(1991). Organizational culture and human resources practices: An empirical assessment. In Woodman

R. W.

Pasmore

W. A.

(Eds.), Research in organizational change and development (Vol. 5, pp. 59-81). Greenwich, CT: JAI Press.

60.

Zaccaro

S. J.

(1991). Nonequivalent associations between forms of cohesiveness and group related outcomes: Evidence for multidimensionality. Journal of Social Psychology, 131, 387-399.

61.

Zammuto

R. F.

Krakower

J. Y.

(1991). Quantitative and qualitative studies of organizational culture. In Woodman

R. W.

Pasmore

W. A.

(Eds.), Research in organizational change and development (Vol. 5, pp. 83-114). Greenwich, CT: JAI Press.

62.

Zantow

Knowlton

D. S.

Sharp

D. C.

(2005). More than fun and games: Reconsidering the virtues of strategic management simulations. Academy of Management Learning & Education, 4, 451-458.

63.

Zuckerman

(2002). Strong corporate cultures and firm performance: Are there tradeoffs? Academy of Management Executive, 16, 158-160.