Early-Determined Business Endgame Standings

Abstract

Teach and Patel reported that teams playing CAPSTONE arrived at their final financial standings early in the game. Based on this, they concluded that participants could feel the game was unfair and that further plays beyond the game’s early rounds might be counterproductive from a game-grade perspective, and even worse, from a learning perspective. Their study was repeated using 1,164 CAPSTONE firms in 194 competitions for 9,312 rounds of play. Their findings of early-determined first-place and last-place finishes were not confirmed based on either cumulative or round-by-round profits. The Teach and Patel research design was also applied to a larger scale game. The results found a company’s early results were low predictors of their final results which did not support the early-finish observation. Game users are cautioned that the educational benefits of any game depends on optimizing all aspects of the simulation’s teaching/learning environment and that competitive fluidity exists throughout a game’s duration.

Keywords

business games economic results learning motivation replication

It ain’t over till it’s over.

Observation by Yogi Berra, 1997 New York Mets manager

In 2007, Teach and Patel presented an article that dealt with the larger issue of how to determine whether players in a business game learned and the degree to which a firm’s profits should serve as a learning measure. Based on their experiences with CAPSTONE (CAP, Smith, n.d.), they noted that teams that took their industry’s early lead in accumulated profits after three playing rounds kept that lead for the game’s duration. Conversely, those companies that fell to the industry’s bottom after only three rounds of play ended as the industry’s losers. They concluded many players might feel the game was patently unfair because each team’s final standing was preordained well before the game’s end. From the perspectives of course management and game-based grades, they observed that it is not clear that a firm’s profits are a measure of its learning and that continuing the game to its scheduled end was not necessary as the “winners” and “losers” had been determined after a few rounds of play.

Because of the importance of Teach and Patel’s findings for the design and use of such games, this study replicates their study using CAPSTONE and an additional total enterprise-type game. A replication is important given CAPSTONE’s worldwide usage by more than 500 colleges and universities, almost 50 corporate clients, and nearly 600,000 students through fall 2010. If their observation associated with CAPSTONE is correct and can be generalized to the many other games being used for educational purposes in North America (Faria, 1998) and other parts of the world (Chang, 1997; Wolfe, 1993), a dramatic reordering of how business games are designed, used, and endorsed would have to be undertaken.

Our presentation replicates the method used in the Teach and Patel study. The study will first review the CAPSTONE replication. It then conducts the same study for a large-game replication using THE GLOBAL BUSINESS GAME (GBW; Wolfe, n.d.). Hypotheses will be tested regarding the power that a company’s early performance has on its final economic standing based on the evidence Teach and Patel produced for a firm’s cumulative and round-by-round earnings.

CAPSTONE Replication

The Teach and Patel study rank-ordered economic performance results from CAPSTONE’s use in a series of business-to-business (B2B) marketing courses. This is an appropriate application of that game as the simulation’s scenario is one where teams sell the sensors they make to original equipment manufacturers. They, in turn, use them in their own final products. Their students competed on teams of unreported sizes in 41 separate industries for eight decision rounds. The column labeled CAP in the appendix indicates the nature of the playing environment created by CAPSTONE via a number of significant game dimensions. This study’s replication data were obtained from Capsim, CAPSTONE’s publisher, for all six-firm industries who played the game SPRING 2010. Capsim estimated their game was used in integrative, program-ending capstone types of courses about 75% of the time. Their census generated results from 194 industries and 1,164 companies playing the game for eight decision rounds. This was a data set that matched Teach and Patel’s game configuration in terms of the number of firms per industry and number of decision rounds played.

Cumulative Earnings Test

Teach and Patel’s first test used cumulative earnings to determine in which round an eventual first-place or last-place firm obtained its final, sustained end performance rank. This is a valid way to score a company’s performance. As noted by Rappaport (2005), “A company’s value depends on its long term ability to generate cash to fund value-creating growth and pay dividends to its shareholders [and that] discounted cash flow [is] the standard for valuing financial assets in well-functioning capital markets” (p. 1). This approach has specifically been taken in THE EXECUTIVE GAME (Henshaw & Jackson, 1986) where the firm’s discounted earnings and cash flows were presented each quarter and round-to-date.

The bar charts presented in their article indicated 53.7% of the firm’s took over and retained their first-place finishes, after two rounds of play. Moreover, “over 85 percent of the firms never relinquished their lead after 3 rounds” (Teach & Patel, 2007, p. 78). Based on this statement, this replication’s hypothesis should be stated as follows, with “early in the game” defined as by or before four rounds of play or one-half of the game’s total rounds.

Hypothesis 1: The majority of firms will arrive at their industry’s ultimate first-place finish early in the game.

Figure 1 compares the bar chart results for the Teach and Patel study versus those produced by this replication effort. The bar chart shows by which round the industry’s eventual first-place company locked into its first-place finish. In the Teach and Patel study, 14.7% of the firms reached their sustained first-place ranking in Round 1, another 39% in Round 2, and another 31.7% in Round 3, with no additional firms being added in Round 4. By the end of Round 3, a total of 85.4% of firms in the Teach and Patel study reached their sustained end-game first-place finish. In contrast, it can be seen a total of only 44.6% of those in the replication group obtained their final finishes before the fourth round of play, which is about half the proportion found in the Teach and Patel study. As a result, we found a significant difference between the proportions of end-state first-place finishes shown in Figure 1 (χ² = 51.5, df = 7, p < .0001).

Figure 1.

First-place end-game arrival round by cumulative rank.

Table 1 presents game first-place end-states in a different way for the replication players. It shows that knowledge of the probable first-place finishes for firms was not known in the majority of cases until the end of the game’s fourth round. It should also be noted that by the seventh, penultimate round, 9.8% of the first-place finishes are still not known. Based on these results, the hypothesis regarding early-obtained, final high-ranked earnings is rejected when cumulative earnings are used to measure company success.

Table 1.

First-Place Finish Probabilities by Cumulative Rank-Replication Study.

	Round
Final first-place finish	1	2	3	4	5	6	7	8
Determined	14.9%	24.7%	37.1%	51.5%	63.9%	76.8%	90.2%	100.0%
Not determined	85.1%	75.3%	62.9%	48.5%	36.1%	23.2%	9.8%	0.0%

Teach and Patel also presented evidence that last-place firms arrived at their last-place finish early in the game. They found that about 46.3% of their firms arrived there by the game’s third round and stayed there for its duration. Based on their evidence, this replication’s hypothesis regarding last-place finishes must be stated as follows:

Hypothesis 2: The majority of firms will arrive at their industry’s ultimate last-place finish early in the game.

Figure 2 compares the Teach and Patel bar chart results with those associated with the replication’s players. We found a significant difference between the proportions of end-state last-place finishes shown in Figure 2 (χ² = 22.5, df = 7, p < .002). Table 2 shows that knowledge of the probable last-place finishes for the replication’s firms was not known in the majority of cases until the end of the game’s fourth round. This was also the case for all ending first-place companies. Based on these results, the hypothesis regarding early-obtained, final low-ranked earnings is rejected when cumulative earnings are used to measure company success.

Table 2.

Last-Place End-Game Arrival Period by Cumulative Rank-Replication Study.

	Round
Final last-place finish	1	2	3	4	5	6	7	8
Determined	16.0%	30.4%	43.8%	58.2%	66.5%	76.3%	86.6%	100.0%
Not determined	84.0%	69.6%	56.2%	41.8%	33.5%	23.7%	13.4%	0.0%

Figure 2.

Last-place end-game arrival period by cumulative rank.

While the Teach and Patel article emphasized end-states in their extreme forms of first and last places, in a real competition, others arrive at end-states. Table 3 indicates that for all other firms in each industry, their finishes were even more in doubt over time. The majority of those finishing third, fourth, or fifth did not know their finishes until the game’s seventh round. Thus, we see an even stronger rejection of the early-finish thesis when applied to all other firms in an industry.

Table 3.

Midrank End-Game Arrival Rounds by Cumulative Rank.

	Round
Finish	1	2	3	4	5	6	7	8
2	1.5%	4.1%	11.3%	20.6%	37.1%	52.1%	71.6%	100.0%
3	1.5%	4.6%	7.7%	18.6%	30.9%	47.9%	66.5%	100.0%
4	1.5%	4.6%	8.2%	15.5%	25.8%	41.8%	62.9%	100.0%
5	1.5%	8.2%	12.9%	22.7%	32.0%	49.5%	69.6%	100.0%

Round-by-Round Earnings Test

The Teach and Patel study next turned its attention to a firm’s profit performance on a round-by-round basis. This is a justifiable short-term orientation. Although a long-term approach should be taken regarding evaluating a company’s financial performance, it is the current quarter’s earnings, as projected into the future, that has the greatest weight when stock annalists evaluate a company (Bandyopadhyay, Brown, & Richardson, 1995; Rappaport, 2005). From a game-playing perspective, this short-term attention is also warranted, because in a six-firm industry, a firm could be in any one of six dynamically changing performance ranks and most games last for a relatively low number of reporting periods.

Given this round-by-round orientation, Teach and Patel compared the number of times a firm was in first place versus the number of times it was in other places. Because the firm’s earnings position was no longer cumulative, it was now possible that a firm could be top-ranked in one quarter, but lower-ranked in other quarters. Thus, a firm may never be top-ranked in any given quarter, but still finish in first place on a cumulative basis. Table 4 reflects this different orientation where they noted the various ranks a firm occupied by round.

Table 4.

Frequency of Earning the Highest Per-Round Profits.

Number of times in first place	Number of firms	Number of times these firms were in last place
8	4	0
7	11	1
6	6	0
5	12	1
4	6	1
3	4	3
2	16	7
1	55	61
Never first, but were never last	97	254
Never first, and never last	35	0

Source: Teach and Patel, 2007, p. 79 (revised).

Based on this exhibit and the logic they employed, they concluded that a pattern of continued dominance of an eventual first-place finish continued through a firm’s round-by-round decisions. As noted in more detail (Teach & Patel, 2007)

There were eleven firms that placed 1st seven times, but of these eleven there was only one case in which the firm placed last during any round of play. There were 4 firms that placed 1st six times. Of the 8 opportunities in which they could have places [sic] last, only one firm placed last one time . . . Thus one would conclude that the dominance phenomenon occurred in evaluating [round by round] profits as well as when one evaluated the cumulative profits. (p. 79)

As part of this article’s replication study, Teach and Patel made available their original data. With this data, we were able to plot the rates at which firms occupied various ranks given their ultimate finish. To support the early-determined finishes hypothesis, a large number of round-by-round first-place finishes would have to occur for six or more rounds.

Figure 3 indicates for the ending Teach and Patel teams that ended as first-place finishers, they were in first place 51.2% of the time for six or more rounds. The replication group resided in their industry’s top spot a lower number of times, which indicates their finishes took more decision rounds to be determined. The difference in the distributions are significant (χ² = 26.5, df = 8, p < .001).

Figure 3.

Round-by-round ranks for final first-place firms.

Figure 4 presents similar data for those firms that ended in last place. Table 5 shows the statistics associated with each set of distributions. Using these measures, we found nonsignificant differences in their sample variances, skewness, and kurtosis. For the first-place finishers, the variance in the Teach and Patel distribution was larger, but not significantly different using an F test. The replication data are symmetric, with a skewness that ranges between −0.5 and +0.5. The Teach and Patel data are moderately skewed with a value of 0.5, but less than 1.0. Regarding their kurtosis, both can be characterized as being normally distributed based on a two-tailed test of excess kurtosis. The same observations hold for the last-place distributions. Thus, the early-determined finishes hypothesis is rejected for both the replication study finishes, and for even the Teach and Patel finishes. For the latter study’s results, their distributions should have been significantly skewed to the left, which was not the case for first- and last-place finishers.

Figure 4.

Last-place end-game arrival round-by-round rank.

Table 5.

Distribution Statistics by Finish.

	Finishes
	First place		Last place
Statistic	Replication	T/P	Replication	T/P
Sample variance	0.01	0.01	0.01	0.01
Skewness	0.32	0.67	0.17	0.80
Kurtosis	1.22	2.05	1.70	2.83

Note: T/P = Teach and Patel.

As stated earlier, other companies exist in a game. The early-finish phenomenon, if it existed, would also have an affect on their playing aspirations. Table 6 indicates that for all midranked firms, none had the majority of their finishes determined by the game’s penultimate seventh round.

Table 6.

Midrank End-Game Arrival Round-by-Round Rank.

	Round
Place	1	2	3	4	5	6	7	8
2	18.0%	19.0%	21.0%	29.2%	32.0%	36.9%	47.2%	100.0%
3	16.9%	15.9%	19.0%	19.0%	25.1%	33.8%	36.9%	100.0%
4	17.4%	15.4%	22.6%	21.5%	27.2%	37.4%	37.4%	100.0%
5	20.5%	21.5%	19.0%	31.3%	31.3%	35.4%	45.1%	100.0%

Large-Game Replication

The Teach and Patel article encouraged replication efforts to investigate the early-finish phenomenon they believed they had experienced. Although not a direct replication effort, Bernard and de Souza (2009) later investigated early dominance within game play for two different online games played synchronously (SIMCO, 2005, 2006, 2007, 2008; SIND, 2002, 2006) for 7 to 8 rounds. Their study, however, defined dominance differently and used the firm’s stock price rather than its earnings as the company’s performance measure. Moreover, the simulations were not total enterprise games. Still, their findings supported the observation that early dominance is attained by firms playing SIND and SIMCO.

An additional replication effort was mounted by Biggs and Fritzsche (2010) in response to Teach and Patel’s call for a test of their results. Their study used data from 19 teams playing the large-scale online and synchronous BUSINESS POLICY GAME (Cotter & Fritzsche, 2005). They were unable to reproduce the results of early first- and last-place finishes for firms that started off very well or poorly in their competitions. The study, however, covered 20 rounds of play while also using an 8-round cutoff to emulate the 8 rounds associated with the Teach and Patel effort. Because of this, the Biggs and Fritzsche study did not exactly replicate the length of game play involved, nor the results of an 8-round game.

This current study used THE GLOBAL BUSINESS GAME, WORLD EDITION (GBW, n.d.) as its large-scale game. The game is very complex, is played online in a synchronous manner, and can require up to 326 decisions per round. In contrast, a maximum of 75 decisions per round is required by CAPSTONE. The appendix compares CAPSTONE’s characteristics to the GBW’s. Only those recent GBW competitions conforming to six companies per industry playing eight decision rounds with profits being the indicator of financial success were collected. This game configuration generated 240 round-by-round results.

Cumulative Earnings Tests

The results produced by the large game are presented in Figure 5. In the GBW’s case, 40.0% of the final first-place finishes occurred in the first round with an additional 20.0% occurring in the following period. Overall, the majority of its final first-place finishes happened by the time the game had been played for three periods. These results support the Teach/Patel early finish thesis.

Figure 5.

First-place end-game arrival periods by cumulative rank.

The results presented in Figure 6 indicate the early last-place finish phenomenon was not associated with the large-scale game. Although each company arrived at its final resting place in different rounds, on a cumulative earnings basis, it took at least six rounds, or the game’s last half before their final last-place standings were achieved (χ² = 16.1, df = 14, p > .001).

Figure 6.

Last-place end-game arrival periods by cumulative rank.

Based on these two examinations, we find mixed results regarding early-finish determinations when using cumulative profits as a measure of company success. The hypothesis cannot be rejected for first-place finishes because 80.0% of these finishes were obtained by the game’s third round. The hypothesis regarding final last-place finishes is rejected. In that case, only 20.0% of the finishes were determined by the game’s third round with all the other finishes determined during or following the game’s sixth round.

Round-By-Round Earnings Tests

The bar charts presented in Figures 7 to 8 indicate the frequency and rounds in which sustained first- and last-place finishes were obtained. Figure 8 shows that none of the first-place finishes were determined until 40.0% of the games had reached their seventh round. The remaining finishes had to wait until the game’s last round for first place to be determined. Figure 8 indicates 20.0% of the last-place finishes were decided in the competition’s first round. The remaining finishes were not determined until the game’s fifth and last round.

Figure 7.

First-place end-game arrival round-by-round rank.

Figure 8.

Last-place end-game arrival round-by-round rank.

Based on this visual sighting, the early-finish hypotheses must be rejected based on round-by-round results. Because of the small and empty cell sizes associated with the GBW’s data, no chi-square tests could be employed.

Another way to indicate that the early-finish phenomenon did not exist for this study’s large-scale game was to indicate in what round the majority of companies had their finishes determined. Table 7 indicates that 60.0% of the firms had to wait until the game’s last round to learn of the company’s finish. Table 8 shows that the majority of last-place finishes did not occur until most competitions had reached their fifth round.

Table 7.

First-Place Finish Probabilities by Round Rank.

	Round
Final first-place finish	1	2	3	4	5	6	7	8
Determined	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	40.0%	100.0%
Not determined	100.0%	100.0%	100.0%	100.0%	100.0%	100.0%	60.0%	100.0%

Table 8.

Last-Place Finish Probabilities by Round Rank.

	Round
Final last-place finish	1	2	3	4	5	6	7	8
Determined	20.0%	20.0%	20.0%	20.0%	60.0%	60.0%	60.0%	100.0%
Not determined	80.0%	80.0%	80.0%	80.0%	40.0%	40.0%	40.0%	100.0%

Discussion

Most of the emphasis in this article, and that by Teach and Patel, has been placed on the companies that were either first or last in their industries. This emphasis unfortunately overlooks the journey taken by each company over its life course. It is this journey, rather than the firm’s end-state, that is the focus of the experiential approach to business education. As noted in this replication effort, the paths taken by firms in business games are irregular. More importantly, the journey itself is littered with problems and opportunities. Each of these is a teaching/learning point to be utilized by the team and the instructor. Figure 9 presents the paths taken by the average CAPSTONE firm by finish. The firms that finished in first-place to fourth-place ranks in the last round, on average started between the third and fourth rank in the first round of play. In this case, we found no statistically significant differences in the starting ranks of these firms. Those who ultimately finished in their industry’s first through fourth earnings ranks were in a statistical dead heat (p < .01) at the beginning of the second round. At one time, the average fifth-place firm rallied in the fifth round and almost tied the earning performance of those who ended in fourth place. The average round ranks shown here actually smoothed out what a company experiences. Figure 10 shows the range of actual results for the CAPSTONE firms that finished in third place. While the average path was smooth, the range of actual round results varied widely with the round results varying the greatest in the last three playing periods.

Figure 9.

CAPSTONE average finish paths.

Figure 10.

Performance ranges for third-place finishers.

Regarding the debate on the use of earnings or profits to measure a team’s success in a business game, it has been stated the only legitimate end for a firm’s activities should be to maximize profits (Friedman, 1970) and that asking or demanding anything else is counterproductive for the firm’s participants, its owners, and society in general. It could be that emphasizing other criteria, such as forecasting accuracy, or scoring high on benchmark measures or a company scorecard, leads to lower profits while also generating lower learning levels.

With deference to Friedman and others, the reality is that for a variety of reasons, criteria other than profit are used in the “real world” for evaluating the performance of firms for a variety of reasons, some of which follow. First, in many instances, cash flow is more important than profit. Firms can make a profit, but still lack sufficient cash and therefore go bankrupt. Second, in start-up firms, many money-losing rounds occur; so if profit is used as the criteria, such firms have failed. Third, things such as growth rates and efficiencies measures are often used as evaluation components for the performance of firms. The ability to forecast accurately can be quite important for growth whether with the market or by gaining market share. The efficient use of resources is important not only for profitability reasons, but for societal reasons as well. Fourth, in the “real world,” we pay people big money to turn companies around, but in the early stages, we may not see significant changes in profits. Fifth, if profits are the be all and end all, why is so much press given to stock price, which is influenced not only by profits, but other factors as well. It would appear that one could make an argument that if only one criterion is to be used, for a publicly held company, it should be stock price. Sixth, many “real world” firms have adopted the Balanced Scorecard (Kaplan & Norton, 1996), which includes criteria other than profits. The fact that many of these items vary across industries also indicates that instructors need to tailor the performance criteria to the nature of the industry in which the students are operating. It is also the case that in using simulations in educational institutions, the emphasis is on learning through active involvement, which includes analysis to learn about relationships. If we only focus on profits, we may fail to have students undertake the analysis to see the why and how of profits. If the Balanced Scorecard is being used in the “real world,” are we not doing students a disservice if we fail to show them how it is used for evaluating the performance of a firm? Thus, while profits should be a measure used to evaluate profit-oriented firms, they should not be the only measure in simulation environments any more than it is in the “real world.”

In defense of the Teach and Patel effort, it should be recognized that a test of the Spearman rank-order correlation coefficients found that each round’s firm rank ordering, either measured by cumulative or round-by-round earnings, was correlated with the ending-round firm rank ordering at the 0.05 or better level of statistical significance. We argue, however, that the magnitude of the correlation coefficient was not sufficient, or strong enough, to cause players to lose interest or slack off in their competition as contended by Teach and Patel. As can be noted in Table 9 by the round-by-round profits, the rank-order correlation coefficient does not reach 0.70 until Round 6. A correlation coefficient of 0.70 implies that only 48.6% of the variance in the ending round rank ordering is explained whereas the majority of the variance, 51.4%, remains unexplained. Even in Round 7, 32.6% of the variation in the rank ordering of the ending round remains unexplained.

Table 9.

Mean Ranked Correlation Coefficients.

	Round
Statistic	1	2	3	4	5	6	7	8
Correlations by cumulative profits
Correlation (rho)	0.15	0.40	0.53	0.68	0.74	0.88	0.97	1.00
Variance explained	2.3%	15.9%	27.6%	45.7%	54.7%	78.1%	93.7%	100.0%
Unexplained variance	97.7%	84.1%	72.4%	54.3%	45.3%	21.9%	6.3%	0.0%
Correlations by round profit
Correlation (rho)	0.05	0.28	0.35	0.49	0.54	0.70	0.82	1.00
Variance explained	0.2%	7.9%	12.4%	23.6%	29.1%	48.6%	67.4%	100.0%
Unexplained variance	99.8%	92.1%	87.6%	76.4%	70.9%	51.4%	32.6%	0.0%

Conclusion

Given the results obtained by Teach and Patel, versus those obtained by a larger census of CAPSTONE companies, an investigation should be conducted as to why their results, which they contended may have had demoralizing consequences, were obtained. Our replication effort used the same simulation, but with different participants and different instructors in different institutional settings. Therefore, on average, the different results might be attributed to differences between the nature of the original study’s players and instructors and those in the replication. This examination could adopt the observations by Burns, Gentry, and Wolfe (1990) regarding the causal elements of a business gaming environment that determine the experience’s outcomes reproduced here. In our study, the concepts covered in the experiential were controlled because the same game was used. Still, there may be differences in team sizes, instructor coaching, and where the course itself was sequenced within the school’s curriculum.

It is the areas of the educator considerations, student attributes, and the conduct and nature of the game that are perhaps the major areas of concern. The game could have been either too difficult or too easy for its players. Players may not have been held accountable for their actions both within their teams and within the course itself. For educators themselves, they may or may not have used additional resources to enhance the experience or practiced a “sink or swim” attitude toward team coaching. All these elements have an effect on the different results individual instructors can obtain from the same business game.

It is hoped this study demonstrates the value of replication efforts. The yeoman-like practice of replication is essential if a science or body of knowledge is going to make progress. While replications are de rigueur in the fields of physics and medicine, especially pharmaceuticals, they are rare in the social sciences. This rarity is especially true in the field of business education. This is unfortunate as the ability to replicate another’s findings assures that the original results were valid and reliable and possibly be generalized to other situations. It is believed the lack of replication studies, despite their lack of glamour, goes to the heart of why so much of what has been published in the business gaming field has not led to a cumulative science (Wolfe & Crookall, 1998).

Those using business games have often used the firm’s cumulative and/or round-by-round profits as a measure of the amount of learning obtained by its players. We agree with Teach and Patel’s assertion that a firm’s profit performance may not indicate organizational or individual learning. It could be reasoned that an amount of organizational learning occurred. This would have happened because each company’s players either learned what the game rewarded, or were been better at applying what they had previously known to the dilemmas presented by the game’s model. In addition, because a firm’s results come from a group effort, nothing is directly known about the contributions of each player or the degree to which the group’s dynamics allowed optimal participation from the team’s players. On the other hand, the use of profits as an indicator of a firm’s success by instructors is warranted. Profits are the “bottom line” and this ethic must be recognized, although many other penultimate measures of an organization’s success could be added to the evaluation mix.

Future studies should be conducted that measure the true amount of before/after learning that has been achieved by each game’s player and the role that different outcome expectations have on the amount of learning that takes place. Other lines of inquiry might look at the impact of the number of firms in an industry and/or the number of rounds of game play. It may be that these factors influence how easy or difficult it is to achieve and maintain a first-place finish or to become last-place finishers. Finally, to complicate the analysis even further, studies should be conducted that look at interval economic performances rather than ordinal performances, as the later can greatly mask what is happening between the industry’s firms (Biggs, 1978).

Footnotes

Appendix

Business Game Teaching/Learning Elements

	CAP	GBW
Number of products	One product for five market segments. Can expand to eight models.	Two, but tailored by quality levels for the country markets served.
Product type(s)	Electronic sensor	25” and 27” television sets
Home country	United States	Can be the United States, Mexico, Germany, Spain, Japan, or Thailand
Active subsidiaries	None	One in the Home Country plus the addition of up to five more.
Factories	One	Six
Factory operations	Two shifts	Two shifts plus overtime
Factory maintenance	No	Yes—By factory and by three types of equipment in each factory
Factory options	No	Build new, expand, contract, decommission, liquidate, or transfer all to other operating units. Can also sell off capacity and subcontract as strategic alliances
Quality control	Ten options regarding quality initiatives	Two simultaneous types of programs.
Research and development	Product performance, size, and mean time between a product’s failure	Product development resulting in patentable features with slight process benefits
Factory workers	Assigned by the simulation	Experienced and inexperienced with different salaries and training needs scheduled by shift and product
Factory foreman	No	Yes—By coverage required in each country
Robotics	No	Two types plus attending technicians
Raw materials	Automatically ordered by the simulation	Six—Advance ordering of two major groups with three quality levels each
Capacity options	Increase or decrease factory size and the factory’s labor to robotics ratio	Eighteen possibilities—Assembly line capacity and two types of robots within each factory
Funds transfers	No	Yes
Sales promotion	By medium	Budgeted by country markets and products
Fluctuating exchange rates	No	Yes
Sales offices	Automatically hired by the simulation	Yes—Options to start-up and shut-down multiple sales offices in each country
Sales force	Part of sales budget by model	Country assignments, quits, hiring, firing, training budget, base salaries, and commission rates
Employee training	Factory workers	Factory workers, sales representatives, and robotic technicians
Distribution centers	No, but part of sales budget by model	Yes
Wholesalers	No	Two types
Bonds	Yes—With no mandatory repayments with bond calls	Yes—With mandatory quarterly payments with bond calls
Stocks	Yes—With dividends and Treasury Stock purchases	Yes—With dividends and Treasury Stock purchases
Short-term loans	1-year loan	90-day loan
Short-term investments	No	Yes
Minimum decisions	50	24
Maximum decisions	75	326
Research reports	No	Twelve covering actual unit sales, one-quarter sales forecasts, advertising budgets, quality levels, and sales force compensations by products and countries
Companies per industry	2-6	3-9
Computer-generated events	None	Twelve Critical Incidents
Strategic alliances	No	Yes—Patent licenses, capacity sales/purchases, and subcontracting
Computer role	Online server	Online server
Decision support materials	Pro forma Income Statements and Pro forma Cash Flow Report and auxiliary spreadsheets	Built-in Pro forma Income Statements and Pro forma Cash Flow Report and auxiliary spreadsheets for production scheduling, raw materials purchases, and currency translations

Note: CAP = CAPSTONE (http://www.capsim.com).

GBW = THE GLOBAL BUSINESS GAME, WORLD EDITION (J. Wolfe, n.d., http://www.onlinegbg.com).

Acknowledgements

We would like to express our appreciation to Dan Smith for providing us with the CAPSTONE results used in this study. We also acknowledge the value of the comments made by Dick Teach on earlier drafts of this article.

Authors’ Note

One of this article’s authors was the creator of one of the games used in this study. That author was not associated with the other game used in this replication.

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Joseph Wolfe, PhD, New York University’s Stern School of Business, is the author of a number of Internet-based top-management games. He is also an Association for Business Simulation and Experiential Learning (ABSEL) Fellow and one of its past-presidents.

Contact: jwolfe8125@aol.com.

William D. Biggs, PhD, The Pennsylvania State University, is professor emeritus, School of Global Business, Arcadia University. He is a fellow and past-president of ABSEL, and has coauthored a textbook for human resources, a portfolio management simulation, and articles related to simulations, experiential learning, and management. In 2011, he conducted simulation workshops in Poland under a Fulbright Senior Specialists grant.

Contact: biggs@arcadia.edu.

Steven C. Gold, PhD, is a professor of economics and finance in the Saunders College of Business at the Rochester Institute of Technology. He is a fellow and past-president of ABSEL, and has authored four computerized simulation games with such notable publishers as McGraw-Hill and Macmillan. His most recent simulation is designed to teach economics, called BEAT THE MARKET, published with Gold Simulations.

Contact: stevengold@saunders.rit.edu.

References

Bandyopadhyay

S. P.

Brown

L. D.

Richardson

G. D.

(1995). Analysts’ use of earnings forecasts in predicting stock returns: Forecast horizon effects. International Journal of Forecasting, 11, 429-445.

Bernard

R. R. S.

de Souza

M. P.

(2009). Dominance in online business game competitions. In Developments in business simulation and experiential learning (Vol. 36, pp. 287-290). Reprinted in the Bernie Keys Library, 10th ed. Available from http://ABSEL.org

Biggs

W. D.

(1978). A comparison of ranking and relational grading procedures in a general management simulation. Simulation & Gaming, 9, 185-200.

Biggs

W. D.

Fritzsche

D. J.

(2010). Using accumulated profits to assess performance in simulations. In Developments in business simulation and experiential learning (Vol. 37, pp. 183-189). Reprinted in the Bernie Keys Library, 10th ed. Available from http://ABSEL.org

Burns

A. C.

Gentry

J. W.

Wolfe

(1990). A cornucopia of considerations in evaluating the effectiveness of experiential pedagogies. In Gentry

J. W.

(Ed.), The guide to experiential learning (pp. 253-278). New York, NY: Nichols.

Chang

(1997). The use of business gaming in Hong Kong academic institutions. In Developments in business simulation and experiential learning (Vol. 24, pp. 218-219). Reprinted in the Bernie Keys Library, 10th ed. Available from http://ABSEL.org

Cotter

Fritzsche

D. J.

(2005). Business policy game: An international strategy simulation. Available from www.bpg.bpgsim.biz/game/bpg.html

Faria

A. J.

(1998). Business simulation games: Current usage levels—An update. Simulation & Gaming, 29, 295-308.

Friedman

(1970, September 13). The social responsibility of business is to increase its profits. The New York Times Magazine, pp. 122-126.

10.

Henshaw

R. C.

Jackson

J. R.

(1986). The executive game. Homewood, IL: Richard D. Irwin.

11.

Kaplan

R. S.

Norton

D. P.

(1996). Balanced scorecard: Translating strategy into action. Boston, MA: Harvard Business School Press.

12.

Rappaport

(2005). The economics of short-term performance obsession. Financial Analysts Journal, 61, 65-79.

13.

SIMCO. (2005, 2006, 2007, 2008). Retailing simulation. Florianópolis, Brazil: Bernard Sistemas. Available from www.bernard.com.br

14.

SIND. (2002, 2006). Manufacturing simulation. Florianópolis, Brazil: Bernard Sistemas. Available from www.bernard.com.br

15.

Smith

(n.d.). CAPSTONE. Available from http://www.capsim.com

16.

Teach

Patel

(2007). Assessing participant learning in a business simulation. In Developments in business simulation and experiential learning (Vol. 34, pp. 76-84). Reprinted in the Bernie Keys Library, 10th ed. Available from http://ABSEL.org

17.

Wolfe

(1993). A history of business games in English-speaking and post-socialist countries: The origination and diffusion of a management education and development technology. Simulation & Gaming, 24, 446-463.

18.

Wolfe

(n.d.). The GLOBAL BUSINESS GAME, WORLD EDITION. Available from http://www.onlinegbg.com

19.

Wolfe

Crookall

(1998). Developing a scientific knowledge of simulation/gaming. Simulation & Gaming, 29, 7-19.