Sei sulla pagina 1di 20

Fire and Ice: Testing a Model on Culture and Complex Problem Solving

C. Dominik Güss 1

Journal of Cross-Cultural Psychology 42(7) 1279–1298 © The Author(s) 2011 Reprints and permission:

sagepub.com/journalsPermissions.nav DOI: 10.1177/0022022110383320 jccp.sagepub.com

DOI: 10.1177/0022022110383320 jccp.sagepub.com Abstract Extending research on complex problem solving

Abstract Extending research on complex problem solving (CPS) and dynamic decision making (DDM), this study investigated the relationship between horizontal-vertical individualism-collectivism, complex problem-solving strategies, and performance in two computer-simulated problem scenarios with different task demands. Participants were 535 students from Germany and the United States (both high on horizontal individualism), Brazil (high on horizontal collectivism), and India and the Philippines (both high on vertical collectivism). Path analyses revealed that horizontal individualism, horizontal collectivism, and vertical collectivism were related to action orientation, and action orientation and planning strategies were related to performance. Strategies were even stronger predictors of performance than the control variables computer experience and intelligence. The structural path model was confirmed for the overall sample and for each of the five national samples. Cross-cultural differences in the process model are discussed. Results showed the importance of cultural values in cognitive processing, specifically for the selection of CPS strategies.

Keywords complex problem solving, dynamic decision making, culture, values, strategy, action orientation, performance, horizontal and vertical individualism and collectivism

“All life is problem solving.”

—Sir Karl Popper

Imagine you are the commanding officer of a fire brigade. You have several trucks and helicop- ters to protect three small villages surrounded by forests from approaching fires. Fires can arise at any time and in any place, and, with strong winds, they can spread easily. Imagine you are the manager of a supermarket responsible for a coldstore full of dairy products. The coldstore tem- perature is regulated by an automated cooling device. Suddenly, this device breaks down and you have to regulate the temperature manually.

1 Psychology Department, University of North Florida, Jacksonville, Florida, USA

Corresponding Author:

C. Dominik Güss, Department of Psychology, University of North Florida, 1 UNF Drive, Jacksonville, Florida 32224, USA. Email: dguess@unf.edu.

1280

Journal of Cross-Cultural Psychology 42(7)

Hor izontal Computer .17*** Individualism Exper ience .09* .24*** .18** –.01 Ve r tical Individualism
Hor izontal
Computer
.17***
Individualism
Exper ience
.09*
.24***
.18**
–.01
Ve r tical
Individualism
.03
Action
.04
.20** *
or ientation
WINFIR E
Perfor mance
.32** *
.12**
–.12*
Planning
Hor izontal
Collectivism
–.06
.18** *
.02
.10* .11*
Ve r tical
Collectivism
Intelligence
–.10
Horizontal Computer .06 Individualism Experience –.12** –.14** Ver tical –.03 Individualism COLD- .58***
Horizontal
Computer
.06
Individualism
Experience
–.12**
–.14**
Ver tical
–.03
Individualism
COLD-
.58***
Action orientation
STORE
Performance
.06
Horizontal
Collectivism
–.13* *
–.07
.17**
Ver tical
Intelligence
Collectivism

Figure 1. Path Analysis Indicating Relationships Among Four Values, Action Orientation, Planning, Computer Experience, Intelligence, and WINFIRE Performance and COLDSTORE Performance. The figure also shows the mediating effect of action orientation on cultural values and performance and the moderating effects of computer experience and intelligence on the relationships between strategies and performance. Path loadings are standardized coefficients. Paths significant at *p < .05, **p < .01, and ***p < .001.

The WINFIRE and COLDSTORE Microworlds and Successful Strategies

We presented the fire and coldstore problems in the computer simulations WINFIRE (Gerdes, Dörner, & Pfeiffer, 1993) and COLDSTORE (Reichert & Dörner, 1988) to participants from Brazil, India, Germany, the Philippines, and the United States. The goal of this study was to test

Güss

1281

a theoretical model postulating that cultural values would predict problem-solving strategies and performance in two simulations with different demands (see Figure 1). Such simulations are called microworlds (Brehmer & Dörner, 1993) and are commonly used in the field of complex problem solving (CPS; e.g., Funke, 2001; Quesada, Kintsch, & Gomez, 2005) and dynamic decision making (DDM; e.g., Gonzalez, 2005). The advantage of micro- worlds is that they represent aspects of problems in real life, such as a situation’s complexity and dynamic development, as well as the opaqueness of some aspects, yet they offer the possibility for controlled laboratory studies (Frensch & Funke, 1995). Incidentally, during the study we received feedback from participants in the United States and Brazil, who had worked with big supermarket coldstores, that the microworld scenario mimicked exactly what employees faced in the event of an actual thermostat breakdown. The fire microworld also simulated a real-life situ- ation, such as that faced by fire fighters across the southern United States in summer 2007 or in Australia in 2008 and 2009. We chose the two simulations WINFIRE and COLDSTORE because of the different chal- lenges they pose to the problem solver. In WINFIRE, the participant has to make quick decisions under time pressure in order to extinguish and contain fires. Von Clausewitz’s (1832) two prin- ciples underlying all strategic planning (i.e., utmost concentration of sources and utmost speed) are essential in WINFIRE. Planning and action orientation were expected to lead to success. Action orientation was operationalized as the total number of decisions participants made in the simulation. In order not to favor cultures that prefer making fast decisions, we chose COLDSTORE (Dörner, 1996), which does not require the same urgency and quick response. In COLDSTORE, the participant has to monitor temperature changes to regulate and maintain an ideal of 4° Cel- sius (39° Fahrenheit). A cautious, less action-oriented strategy was expected to lead to success because more forceful changes to the control wheel lead to high temperature oscillations.

Hypothesis 1: Action orientation will be positively associated with performance in WIN- FIRE and negatively with performance in COLDSTORE.

CPS, Intelligence, and Computer Experience

Previous research on CPS has been conducted mainly in Western-industrialized countries and has often focused on individual difference variables like intelligence (Gonzalez, Thomas, & Vanyukov, 2005; Süß, Kersting, & Oberauer, 1993) and computer experience (e.g., Schaub, 2001) to explain variance in performance. Particularly performance on intelligence tests assess- ing abstract problem-solving abilities (e.g., Advanced Raven Progressive Matrices; Raven, 1977) was related to CPS performance in other versions of WINFIRE and COLDSTORE (Rigas, Car- ling, & Brehmer, 2002). Computer experience frees up working memory capacity for dealing with the actual microworld problem. Limited computer experience would require more working memory for dealing with the computer, for instance, when using the mouse. Recently, researchers have recommended exploration of not only those individual difference variables but also the process, what people are actually doing while making dynamic decisions (Brehmer, 1999), and the strategies used (Schoppek & Putz-Osterloh, 2003; Strohschneider & Güss, 1998). Thus, our focus on strategies, specifically action orientation, addresses this recommendation.

Hypothesis 2: Intelligence and computer experience will predict successful strategies (high action orientation in WINFIRE and low action orientation in COLDSTORE) and per- formance in both microworlds.

1282

Journal of Cross-Cultural Psychology 42(7)

Culture,Values, and CPS

Culture can be understood as implicit and explicit knowledge shared by a specific group of people and transmitted from generation to generation (Smith, Bond, & Kagitcibasi, 2006). Val- ues are an aspect of cultural knowledge relevant to this study because they can be defined as abstract, transsituational goals (e.g., Hofstede, 2001) that may act as guiding principles for the selection of specific strategies in complex and novel problem situations. Our main argument is that value orientations can influence the selection of specific CPS and DDM strategies in a problem situation (see Ajzen and Fishbein’s [1980] theory of reasoned action). Following Berry’s (2004) eco-cultural model, these strategies are learned in specific cultural environments. Strategies regarded as adaptive and likely to lead to success would prob- ably be applied to novel problem situations. How those strategies are acquired can be explained by action and sociocultural theories, which stress the role of social interactions (Vygotsky, 1978). Such social interactions occur in a cultural context and offer similar opportunities for learning, which Lave (1991, p. 67) called “situated social practice.” Thus, cross-cultural differences can emerge and strategies that might be adaptive in one cultural environment might not be adaptive in another. The general model, however, postulating that values influence strategy selection and that strategies influence per- formance, would apply across cultures.

Hypothesis 3: The structural path model will be invariant across the five national samples.

Horizontal and Vertical Individualism and Collectivism and Strategy Selection

This study extends existing research on individualism-collectivism (I-C) and power distance (Hofstede, 2001), also called conservatism and hierarchy (Schwartz, 1994). For a long time, individualism and collectivism were considered two poles of the same dimension (e.g., Fischer et al., 2009; Hofstede, 2001). Thus, a person with high individualistic values could not have high collectivistic values at the same time. It is possible, however, that individualism and collectivism are two independent dimensions and people can have high or low scores on both dimensions at the same time (Triandis, Chen, & Chan, 1998). Therefore, I-C together with V-H have been com- bined and assessed as four relatively independent value preferences in previous studies with vari- ous cultural and ethnic groups to further refine the broad I-C dimension (e.g., Triandis & Gelfand, 1998): Horizontal individualism (HI), vertical individualism (VI), horizontal collectivism (HC), and vertical collectivism (VC). HI favors equality and focuses on the self and unique self-iden- tity; VI accepts inequality and focuses on the self and competition; HC favors equality and focuses on the group and caring for the group; and VC accepts inequalities and focuses on the group and self-sacrifice for the group (Li & Reb, 2009; Singelis, Triandis, Bhawuk, & Gelfand, 1995; Triandis et al., 1998). Validity of the measurement and measurement equivalence have been investigated and shown in various cultural samples (e.g., Chiou, 2001). To this point, most studies have investigated strategies referring to the I-C dimension alone. Individualistic values, for example, were related to more planning and to a more action-oriented problem-solving strategy (Mann et al., 1998; Ohbuchi, Fukushima, & Tedeschi, 1999). Collec- tivistic values were related to more caution and less action orientation (Güss, 2004; Ohbuchi et al., 1999). The positive relationship between action orientation and individualism has been differentiated in Komarraju, Dollinger, and Lovell’s (2008) study, which investigated how HI, VI, HC, and VC predicted five conflict management styles (dominating, integrating, avoiding, obliging, and compromising). Both horizontal and vertical individualism were related

Güss

1283

to a dominating style. The dominating style was characterized by low concern for others, high concern for self, forcefulness, and action. HC was related to an obliging style. An obliging style was a nonconfrontational style with high concern for others, low concern for self, and cautious action. VC was related to an avoidant style of conflict management, simply not dealing with a conflict, which was characterized by low concern for others and low concern for self (see also Kim-Jo, Benet-Martínez, & Ozer, 2010). It would seem adaptive for individualists, especially horizontal individualists, to focus more on planning and to follow a forceful action-oriented strategy because individual goals and inde- pendent self would dominate with less need to take social factors and others into consideration. Especially horizontal individualists would perceive themselves as being in control over deci- sion alternatives and possible actions in social situations. As Nurmi (2008) showed, extent of planning was related to internal control and estimation of the likelihood that goals can be reached in the future. If the self-concept in collectivistic cultures is predominantly interdepen- dent (Markus & Kitayama, 1991), extending beyond the person and including others, then an individual’s planning will try to consider their perspectives as well, making predictions harder and leading to less control and increased uncertainty (Knoblich & Scott, 2003). A horizontal collectivist with high concern for the views of others would have, then, a tendency to give in and to proceed more cautiously. This cautiousness would be even more pronounced for a verti- cal collectivist, who would try to avoid dealing with the problem and place group needs over self needs. Empirical support comes from Brockner et al. (2001), who showed that a horizontal compared to a vertical orientation was related to a higher desire for action in decision making.

Hypothesis 4: Individualism, specifically HI, will be positively associated with planning and action orientation. Collectivism, specifically VC, will be negatively associated with planning and action orientation.

Selection of Countries and Expected Differences in Values

To test the hypotheses summarized in the postulated model in Figure 1, it was necessary to select participants from several countries—choosing a number large enough to enable a broader frame of comparison, but not including so many countries that completion of the study would be near impossible. First, the participants had to be from industrialized countries because they needed to have some familiarity with computer use to work on the simulated microworlds. Second, the five countries were selected for their potential to vary in the cultural value dimensions relevant to this study. We tried to follow Smith, Peterson, and Schwartz’s (2002) advice:

Thus, there is a dearth of studies that adequately test systematic, theoretically grounded relations between cultural values and behaviors across a sample of cultures that is suffi- ciently broad as to include the major sources of global variation within modern societies. (p. 190)

Although no existing study has addressed HI, VI, HC, and VC in the selected five countries simultaneously, previous studies have shown that the United States (rank 91 in Hofstede, 2001) and Germany (67) have higher individualistic values, Brazil (38) and the Philippines (32) have high collectivist values (Hofstede, 2001), and India (48) has both individualistic and collectivist values (Sinha & Tripathi, 1994). The Philippines (94) and India (77) have high power distance, Brazil (69) has medium, and the United States (40) and Germany (35) have low power distance values (Hofstede, 2001). Hofstede’s individualism correlated positively with affective autonomy (.85) in Schwartz’s (1994) student samples. Hofstede’s power distance correlated positively with

1284

Journal of Cross-Cultural Psychology 42(7)

conservatism (.70)—emphasizing security, conformity, and tradition—in Schwartz’s study. Hofstede’s characterization of the five countries was confirmed in Schwartz’s studies. Additionally, one study compared HI, VI, HC, and VC in U.S. and Filipino student samples and found that U.S. participants had, as expected, significantly higher scores on VI and HI and significantly lower scores on VC compared to Filipino participants (Cukur, de Guzman, & Carlo, 2004). No significant differences were found on the HC dimension.

Method

Participants

Participants were 535 students from two different educational institutions in each of the follow- ing countries: Brazil (n = 97), Germany (n = 104), India (n = 97), Philippines (n = 104), and United States (n = 133). Students were from schools of arts and sciences, social sciences, and business. Samples were comparable according to course or major and gender. Females made up 63% of all participants. The age range was 18 to 49, with an average age of 22.1 years (SD = 4.44). None of the participants had taken part in other CPS experiments prior to this study. To test for possible within-country variability, we compared the performance in WINFIRE and COLD- STORE between the participants from two universities in every country. None of the differences were significant (ps > .11).

Instruments

Language. The instruction sheets for the simulations and surveys were originally developed in English, then translated from English into German and Brazilian Portuguese using the translation- backtranslation method (Brislin, 1986) by multilingual and local researchers. As Indian and Filipino participants were bilingual and the mode of instruction was mostly English, the ques- tionnaires were administered to them in English. Cultural values. HI, VI, HC, and VC were assessed with a 9-point Likert-type scale measure (Singelis et al., 1995) consisting of eight items for each of the four dimensions—for example, “I prefer to be direct and forthright when discussing with people” (HI); “When another person does better than I do, I get tense and aroused” (VI); “To me, pleasure is spending time with others” (HC); “I usually sacrifice my self-interest for the benefit of the work group” (VC). In our sample, Cronbach alpha coefficients were .61 for HI (Brazil: .53, Germany: .60, India:

.71, Philippines: .51, United States: .62), .68 for VI (Brazil: .58, Germany: .59, India: .63, Philip- pines: .62, United States: .79), .71 for HC (Brazil: .55, Germany: .63, India: .81, Philippines: .59, United States: .73), and .74 for VC (Brazil: .62, Germany: .58, India: .73, Philippines: .46, United States: .63) (see Table 1). The reliabilities were not ideal but were similar to those of Singelis et al. (1995) and other studies (e.g., Kemmelmeier et al., 2003). We conducted several confirmatory factor analyses to ensure cross-cultural applicability of the instrument (see van de Vijver & Leung, 1997) receiving the highest goodness-of-fit indices for the four factor solution: χ 2 = 1717.48 (df = 458), p < .001, χ 2 /df = 3.75, GFI = .815, Std RMR = .086, RMSEA = .077, NNFI = .79. The fit indices are similar to those of Singelis et al., although they do not indicate an ade- quate fit with the data. 1 Manipulation check. One-way between-groups ANOVAs were conducted to test for cultural differences in the four value dimensions (see Table 2). The five countries differed significantly in all four values in the expected directions (ps < .001). Comparing grand mean centered data, taking into consideration culture-specific answering tendencies, Germany, United States, and Brazil showed highest HI; Germany, United States, and India highest VI; Brazil, India, and

Intell

.85

Compexp

.30***

Table 1. Correlations and Reliabilities of Values, Strategies, Performance, and Possible Moderators in WINFIRE (F) and COLDSTORE (Co)

–.27***

–.23***

PerfCo

.81

.62***

–.23***

–.22***

ActCo

.88

–.16***

–.26***

.23***

PerfF

.12*

.80

.23***

–.17***

–.20***

.24***

.17***

PlanF

.38***

.28***

–.20***

.25***

.15**

ActF

–.10*

.78

–.18***

.27***

.23***

–.37***

–.21***

–.10*

–.12*

VC

.74

.40***

–.12**

.15**

–.10*

.12*

–.09*

HC

–.07

.71

–.05

.32***

–.15**

–.11*

VI

.68

.07

.08

–.02

.08

.03

.01

.16***

.15***

.16***

.13**

.11*

.11*

–.12*

HI

.07

–.02

.61

–.01

ActCo PerfCo (rev.)

Intelligence

Compexp

Overall

PlanF PerfF

VC
ActF

VI
HC

HI

1285

P>B,U,I,G and B,U>G I,P,U >B,G I,B,P >U,G I,P>U>B,G B,G,U,P>I and B>P G>I,U,P>B B>G,I,U,P and G>P I,P>U,B,G

U,G,P,B>I and U,G>B G,P,U,B>I and G>U,B G>P,I,B and U>I,B

U,G,B>P>I G>P,U,B,I and P,U>I

B,G,U,I>P and B>I

Post hoc

I,P,B>U>G

I,P,B>U,G

.12

.22

.08

.04

.10

.12

.12

.29

.08

.16

.08

.16

.09

η p 2

.43

.21

.11

4, 474

4, 468

4, 416

4, 510

4, 510

4, 510

4, 462

4, 452

4, 510

4, 510

4, 504

4, 475

4, 503

4, 501

4, 501

4, 501

df

16.86***

43.21***

33.99***

31.07***

10.18***

15.74***

14.90***

21.40***

12.90***

4.75***

10.65***

93.90***

11.99***

23.91***

15.06***

11.63***

F

593.52 (261.30) 667.12 (222.00) 395.80 (199.73) 727.86 (226.14) 675.32 (203.45) 507.13 (287.51)

United States

69.58 (22.98)

55.58 (13.65)

116.81 (61.34)

35.55 (4.85)

2.58 (0.64)

3.64 (3.09)

5.55 (1.13)

6.76 (1.05)

5.34 (1.42)

7.01 (0.95)

22.22 (5.72)

.87

.59

–.61

–.81

84.59 (14.30)

51.16 (10.74)

106.35 (63.85)

Philippines

35.73 (4.36)

1.84 (0.70)

4.08 (3.18)

6.81 (0.90)

7.15 (0.93)

5.75 (1.24)

7.45 (0.88)

19.83 (2.73)

–1.02

.03

.41

.71

Table 2. Means and Standard Deviations of All Variables for the Five Cultures and Overall

87.42 (10.94)

61.83 (47.62)

33.49 (5.22)

1.46 (0.61)

48.98 (8.63)

1.74 (1.60)

6.93 (0.95)

7.32 (1.06)

5.83 (1.17)

6.68 (1.15)

22.03 (2.59)

India

–.76

.16

.33

.83

63.69 (23.12)

56.60 (14.56)

107.12 (40.57)

38.99 (3.51)

2.55 (0.50)

5.05 (3.41)

4.56 (1.03)

6.46 (0.95)

4.83 (1.29)

6.41 (1.04)

22.77 (3.64)

Germany

–.97

.90

–.66

.91

82.45 (15.51)

84.05 (51.64)

34.87 (5.35)

2.32 (0.76)

48.63 (5.71)

3.38 (2.72)

4.96 (1.27)

7.18 (0.96)

4.30 (1.28)

7.02 (0.95)

23.79 (5.03)

Brazil

–1.44

–.89

1.43

1.13

Note. Standard deviations are provided in parentheses.

77.41 (20.29)

52.32 (11.68)

96.61 (57.18)

35.69 (5.00)

3.60 (3.07)

5.75 (1.41)

6.96 (1.04)

5.23 (1.40)

6.94 (1.04)

22.11 (4.44)

2.18 (.77)

Overall

.77

.80

–.43

–.93

Vertical Individualism (VI) Horizontal Collectivism (HC) Vertical Collectivism (VC)

Horizontal Individualism (HI)

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

Computer experience

Performance (revers.)

HC grand mean cent

VC grand mean cent

grand mean cent. grand mean cent.

Action orientation

Action orientation

Dependent Variables

Control variables

Performance

COLDSTORE

Intelligence

Planning

WINFIRE

Values

HI

VI

Age

1286

Güss

1287

Germany highest HC; and Philippines and India highest VC. It is surprising, though, that Brazil also had high HI scores, which might be related to our student samples. WINFIRE. In the WINFIRE microworld (Gerdes et al., 1993), the participant assumes the role of a firefighting commander. Participants can choose between nine command options, for exam- ple, “goal” to send a unit to a desired location or “extinguish” to command units to extinguish a fire. The duration of the WINFIRE simulation consisted of 111 cycles, each lasting 6 seconds, after which the area of burning forest was updated. WINFIRE lasted 11 minutes. Performance in WINFIRE was measured by the percentage of protected area during each of its 111 cycles. In this study, the Cronbach alpha value for protected forest in cycles 42, 54, 90, and 111 was .80 for the overall sample (.74, .79, .80, .81, and .80 for Brazil, Germany, India, Philippines, and United States, respectively). Overall, participants saved 52.33% (SD = 11.69) of the forest at the end of the game. Without any intervention or commands from the firefighting commander, 45.18% of the forest would have been saved. Two strategies were identified from participants’ saved computer data files. Action orienta- tion, the total number of units moved during the entire simulation, indicated problem solving for the various fires. The correlation of actions in cycles 1 to 55 and cycles 56 to 111 was .68 overall (Brazil: .66, Germany: .44, India: .55, Philippines: .76, United States: .63; ps < .001). Cronbach alpha coefficients were .78 overall (Brazil: .74, Germany: .59, India: .68, Philippines: .83, United States: .76). Planning, the number of strategic distributions of units operationalized as the total number of patrol, search, clear, and goal commands given before the first fire started in cycle 19, thus referred to any action taken at the beginning of the simulation versus simply waiting. COLDSTORE. In COLDSTORE (Reichert & Dörner, 1988, following an idea by McKinnon & Wearing, 1985), the goal is to switch the control wheel to keep the temperature stable at an opti- mal temperature in order to keep products from freezing or spoiling. The duration of the experi- ment is 100 cycles, each cycle lasting 8 seconds, after which the temperature is updated according to the position of the control. COLDSTORE lasted 13.3 minutes. Performance in the COLDSTORE simulation was measured by the total deviations from the goal temperature summed for each cycle. The fewer deviations from the goal temperature, the more successful the participant was. In this study, the Cronbach alpha value for the overall per- formance variable in cycles 1 to 33, 34 to 66, and 67 to 100 was .84 for the overall sample (.65, .88, .71, .69, and .85 for Brazil, Germany, India, Philippines, and United States, respectively). In COLDSTORE, action orientation was operationalized as total number of control wheel adjust- ments. The correlation of total control wheel adjustments in cycles 1 to 33 and 66 to 100 was .64 overall (Brazil: .46, Germany: .64, India: .40, Philippines: .44, United States: .72; ps < .001); Cronbach alpha coefficients was .88 overall (Brazil: .81, Germany: .90, India: .73, Philippines:

.77, United States: .90). It was not possible to operationalize planning adequately from saved computer protocol data. Possible demographic covariates. We assessed several demographic variables such as gender, age, and years of computer experience. Nonverbal intelligence. Intelligence was assessed with the Test of Nonverbal Intelligence (TONI- 3; Brown, Sherbenou, & Johnsen, 1997). This test was selected because it is relatively culture- free, language-free, and it assesses nonverbal aspects of intelligence related to problem solving such as analogous reasoning and deduction with 45 abstract-figural items. The test has been vali- dated in various ethnic groups and with other intelligence tests. Retest reliabilities ranged from .79 to .95, and Cronbach alpha values were in the .90s (Brown et al., 1997). In our samples, the Cron- bach alpha values were lower but satisfactory with an overall alpha of .85 (.91, .71, .87, .76, and .79 for Brazil, Germany, India, Philippines, and the United States, respectively).

1288

Journal of Cross-Cultural Psychology 42(7)

Procedure

The author collected data in the five countries over a 3-year period with an onsite team of researchers in each country. Participants’ rights were protected, and applicable human research guidelines were followed. Demographic and intelligence-test data were collected in group ses- sions (2 hours), and the WINFIRE and COLDSTORE microworlds were administered in indi- vidual sessions (2 hours). Participants received either course credit or were paid for their participation. In the group meeting, participants filled out the consent form, the demographic questionnaire, the two value surveys, and the TONI-3. In the individual meeting, each participant worked on WINFIRE and COLDSTORE and answered surveys regarding the two simulations. Written instructions for each simulation were provided and 3-minute test games were played before the actual simulation started to familiarize participants with the microworlds. The experi- ment ended with an interview about participants’ impressions and experiences regarding the microworlds. All the decisions participants made during the simulations were automatically saved to computer files. Data were not complete for every participant due to computer problems, electrical power failures, and missing participant data on some survey items.

Overview of Analyses

Initially, descriptive statistics, including means, standard deviations, correlations, 2 and reliabili- ties, were analyzed and cross-cultural differences were investigated. Then, I tested each model for WINFIRE and COLDSTORE using path analysis as a variant of structural equation modeling (SEM) in Mplus (Muthén & Muthén, 1998-2006). The goal was to test the associations outlined in the structural models in Figure 1. The compatibility of the specified models and the observed data were evaluated for the WINFIRE and COLDSTORE models separately as strategy and performance variables differed between them. The maximum likelihood method was used to estimate path coefficients in the models. Standardized coefficients were presented in the WIN- FIRE and COLDSTORE models to facilitate comparison. 3 Missing values due to incomplete data were incorporated in Mplus using maximum likelihood method assuming missing data is random (see also Enders, 2001). However, two cases with missing values on all variables in COLDSTORE and one case with missing values on all variables in WINFIRE were not included in the analyses. After analyzing the pan-cultural path model, path analyses with multigroup comparisons were conducted to detect a possible moderation effect of culture. To evaluate the path models, several fit indices were used. Using criteria guidelines on the interpretation of the indices (e.g., Hu & Bentler, 1995; Schumaker & Lomax, 1996), values greater than .90 on the Comparative Fit Index (CFI) and the Tucker-Lewis Index (TLI) indicated good fit. Values of less than .05 for the RMSEA indicated good fit, while values above .10 indicated poor fit (McDonald & Ho, 2002). Standardized root mean square residual (SRMR) should ideally be below .10 (Hu & Bentler, 1999). Constrained and unconstrained models were then compared applying the chi-square dif- ference test and pairwise comparisons were conducted to investigate which paths differed between cultural samples.

Results Pan-Cultural Path Analyses and Relationships Among Variables

Hypotheses 1, 2, and 4 specified the relationships between the variables in the model. The overall pan-cultural unconstrained model for WINFIRE presented in Figure 1 demonstrated adequate fit

Güss

1289

indices: χ 2 (4) = 10.32, p = .04; CFI = .968, TLI = .833, RMSEA = .054, and SRMR = .016. The proportion of the variability in action orientation that is explained by specific values (HI, VI, HC, VC), computer experience, and intelligence is 12% (R 2 = .116, p < .001). The proportion of vari- ability in planning that is explained by specific values (HI, VI, HC, VC), computer experience, and intelligence is 9% (R 2 = .091, p < .001). Moreover, the proportion of the variability in WIN- FIRE performance that is accounted for by action orientation, planning, computer experience, and intelligence is 12% (R 2 = .123, p < .001). According to Kline (2005), “standardized path coefficients with absolute values less than .10 indicate a ‘small’ effect; values around .30 a ‘typical’ or ‘medium’ effect; and ‘large’ effects may be indicated by coefficients with absolute values ≥ .50” (p. 122). Hypothesis 1 was confirmed in WINFIRE, showing small to medium effects for the positive significant paths from action orien- tation and planning to performance. According to Hypothesis 2, intelligence and computer expe- rience would predict successful strategies (high action orientation and planning in WINFIRE) and performance. Path coefficients showed that both predicted strategies, but only intelligence was directly related to performance. Hypothesis 4 stated that individualism, specifically HI, would be positively associated with planning and action orientation, and collectivism, specifi- cally VC, would be negatively associated with planning and action orientation. Significant stan- dardized path coefficients partially supported this hypothesis. HI predicted planning and action orientation but not VC. HC was negatively related to action orientation. The overall unconstrained model for COLDSTORE presented in Figure 1 showed very good fit indices: χ 2 (4) = 4.10, p = .39; CFI = 1.000, TLI = .999, RMSEA = .007, and SRMR = .010. The proportion of the variability in action orientation that is explained by specific values (HI, VI, HC, VC), computer experience, and intelligence is 12% (R 2 = .124, p < .001). The proportion of the variability in COLDSTORE performance that is accounted for by action orientation, com- puter experience, and intelligence is 41% (R 2 =.409, p < .001). Hypothesis 1 predicted a negative relationship between action orientation and performance in COLDSTORE. The path coefficient supported this hypothesis, showing a large effect. The higher the action orientation, the more deviations from the target temperature—the lower the perfor- mance. Hypothesis 2 was mostly supported. Paths from intelligence and computer experience to action orientation and to performance were significant with the exception of the path from intel- ligence to performance. Hypothesis 4 was partially confirmed. Only VC was related to action orientation; none of the other values were.

Path Analysis for Each Country

In a next step, the models for the two microworlds were tested in all cultural groups separately (see Figures 2 and 3). Overall data fit the models reasonably well, indicating that the theoretical models hold in different cultural contexts (see Table 3). In WINFIRE, the German and Filipino TLI and RMSEA indicated inadequate fit with the data, though. Modification indices (5.05 and 6.48, respectively) showed that freeing the correlation between HI and performance in the Ger- man sample and freeing the correlation between planning and performance in the Filipino sam- ple would improve the model fits. In COLDSTORE, all but the German model demonstrated excellent fit. One explanation is the correlation between computer experience and performance, as the modification index of 5.22 indicates. Freeing correlations and parameters to increase model fit—if not theoretically justified—is, however, a widely criticized procedure that we did not pursue (see Kline, 2005; MacCallum, Roznowski, & Necowitz, 1992).

1290

Journal of Cross-Cultural Psychology 42(7)

1290 Journal of Cross-Cultural Psychology 42(7) HI .07 Comp_exp .23* –.04 .19 † –.03 .07 .28**
HI .07 Comp_exp .23* –.04 .19 † –.03 .07 .28** .20 † .18 † –.15
HI .07
Comp_exp
.23*
–.04
.19 †
–.03
.07
.28**
.20 †
.18 †
–.15
.20 †
.04
–.04
.09
.3 0* *
.14
.15 †
.14
.12
.34***
–.06
.26*
.13
.09
.06
.04
.04
–.20*
VI –.07 .06
.07
Actio n
10
.12
.20 †
or ientation
.27* *
–.19 †
.31* *
.20*
.1 4
–.18 †
.36***
.21*
WINFIRE
.25*
.08
Perfor mance
.04
.38***
–.35***
.38***
–.11
–.03
.05 .18 † .0 7
Planning
HC
–.16
–.05 .1 2
.15 .07 –.08 –.03
.04
.0 2
.24***
.04
–.14
–.06
.0 9 †
.13
.04
–.08
.0 8
.09
–.21*
.10
.18 †
.01
.1 2
–.01
–.03
.23*
–.12
.21*
.19*
–.03
–.00
Intelligence
VC
Figure 2. Path Analysis Showing Relationships Among Cultural Values, Strategies, Computer Experience,
Intelligence, and Performance in WINFIRE

Regular typeface = Brazil, italics = Germany, underline = India, boldface = Philippines, and regular typeface again = United States. All coefficients are standardized estimates. p .10. *p .05. **p .01. ***p .001.

HI .26** Comp_exp .09 .05 .18 .00 –.02 .02 .00 –.18 † –.01 –.14* .00
HI
.26**
Comp_exp
.09
.05
.18
.00
–.02
.02
.00
–.18 †
–.01
–.14*
.00
–.17
.01
VI
–.12 .09 –.07
.21*
–.01
.02
Action
.44*** .59***
.33*** .40***
.61***
COLDSTORE
orientation
Performance
.04
.06
–.02
.01
–.22**
–.09
HC
.03
–.07
.14
–.12*
.15
.07
.04
–.17*
.01
.06
–.06
–.05
.07
–.02
Intelligence
VC

Figure 3. Path Analysis Showing Relationships Among Cultural Values, Strategies, Computer Experience, Intelligence, and Performance in COLDSTORE

Regular typeface = Brazil, italics = Germany, underline = India, boldface = Philippines, and regular typeface again = United States. All coefficients are standardized estimates. p .10. *p .05. **p .01. ***p .001.

Güss

1291

Table 3. Fit Indexes for the Proposed Unconstrained WINFIRE and COLDSTORE Models in Each Country

Country

χ 2

df

p

CFI

TLI

RMSEA

SRMR

R 2 Accted. for in Performance

R 2 Accted. for in Activity

R 2 Accted. for in Planning

WINFIRE

Brazil

2.16

4

.71

1.00

1.45

0.00

0.02

0.07

0.07

0.15

Germany

7.87

4

.10

0.87

0.31

0.09

0.04

0.13

0.13

0.06

India

3.39

4

.50

1.00

1.11

0.00

0.02

0.13

0.28

0.07

Philippines

9.18

4

.06

0.90

0.48

0.11

0.04

0.23

0.13

0.14

United States

5.42

4

.24

0.92

0.60

0.05

0.02

0.09

0.05

0.13

COLD-STORE

Brazil

3.60

4

.46

1.00

1.06

0.00

0.03

0.22

0.13

Germany

11.37

4

.03

0.84

0.49

0.13

0.05

0.39

0.05

India

1.38

4

.85

1.00

1.90

0.00

0.02

0.16

0.06

Philippines

2.90

4

.57

1.00

1.14

0.00

0.03

0.26

0.07

United States

0.15

4

.99

1.00

1.24

0.00

0.00

0.42

0.08

Multiple Group Analysis to Test Culture as Potential Moderator

According to Hypothesis 3, the structural path model will be invariant across the five national samples. Path analyses with multigroup comparisons were conducted to investigate the extent culture moderated the relations specified in the models. Constrained and unconstrained models were compared. First, we constrained all paths, but not the means and variances of the dependent variables (i.e., strategies and performance), as we expected (and found) cultural differences. In the unconstrained models, all parameters were free to vary by culture. The fit indices for the unconstrained model in WINFIRE were χ 2 (32) = 46.62, p = .05; CFI = .90, TLI = .68, RMSEA = .07, and SRMR = .04. The fit indices for the constrained model (all direct effects equality constrained) in WINFIRE were χ 2 (84) = 124.04, p = .003; CFI = .74, TLI = .67, RMSEA = .07, and SRMR = .08. The fit indices for the unconstrained model in COLD- STORE were χ 2 (28) = 29.66, p = .38; CFI = .99, TLI = .98, RMSEA = .02, and SRMR = .04. The fit indices for the constrained model in COLDSTORE were χ 2 (56) = 67.99, p = .13; CFI = .92, TLI = .91, RMSEA = .05, and SRMR = .06. Comparing chi-square results of constrained and unconstrained models (Byrne, 2006), results showed significant differences in WINFIRE, χ 2 D (52) = 77.42, p = .01, and not in COLDSTORE, χ 2 D (28) = 38.33, p = .09, indicating that over- all there was a difference in the WINFIRE models among the cultural groups. We then investigated which paths differed significantly between countries and focus only on the paths from values to strategies and from strategies to performance. Pairwise comparisons of the 17 paths in WINFIRE were conducted. 4 The paths from action orientation and planning strat- egies to performance did not differ significantly between any of the five cultural samples in WINFIRE. Analyzing the paths from the four values to strategies in WINFIRE, only the path from HC to action orientation differed. HC was related negatively to action orientation in the Indian sample. The path in the Indian sample differed significantly from the ones in the Filipino (pairwise z test –2.08), U.S. (pairwise z test –2.70), and Brazilian samples (pairwise z test 1.99). Paths from VI to planning differed significantly between India and the United States (pairwise z test –2.83) and between Philippines and the United States (pairwise z test –2.83). Whereas VI was positively related to action orientation in the U.S. sample, it was negatively related to action orientation in the Indian and Filipino samples.

1292

Journal of Cross-Cultural Psychology 42(7)

Paths from VC to planning differed significantly between Philippines and Germany (pairwise z test –1.99) and between Philippines and the United States (pairwise z test –2.57). Whereas VC was positively related to planning in the Filipino sample, it was negatively related to planning in the German and U.S. samples.

Discussion

The goal of this study was to test a theoretical model postulating that values influence strategies and strategies influence CPS performance in two different situations including the control vari- ables of computer experience and intelligence. Four specific predictions were made:

The first hypothesis about a positive relationship of action orientation and performance in WINFIRE and a negative relationship of action orientation and performance in COLDSTORE was confirmed (i.e., many deviations from target temperature), indicating that participants adjust their strategy according to the different task-demands. The second hypothesis about intelligence and computer experience predicting successful strate- gies (high action orientation in WINFIRE and low action orientation in COLDSTORE) and perfor- mance in both microworlds was mostly confirmed. Consistent with previous studies, computer experience (Schaub, 2001) and intelligence (Gonzalez et al., 2005) were influential and predicted strategies in both microworlds in the expected directions. Both variables predicted more action ori- entation in WINFIRE and less action orientation in COLDSTORE. Intelligence, however, did not significantly predict performance in COLDSTORE and computer experience did not significantly predict performance in WINFIRE. This is partly due to the strategies. Strategies predicted perfor- mance better than computer experience and intelligence, showing the relevance of studying strate- gies in CPS as Schoppek and Putz-Osterloh (2003) suggested. Our results support Brehmer’s (1999) claim that researchers in the field of decision making should focus on what people actually do. The third hypothesis about structural invariance of the path model across the five national samples was mostly confirmed. Both WINFIRE and COLDSTORE models manifested a notable fit with the data for the overall pan-cultural model and for the models of each country (with the exception of the German models and the Filipino WINFIRE model). The inadequate fit for the German sample could be related to the finding that German and U.S. participants perceived both WINFIRE and COLDSTORE as less realistic than participants in Brazil, India, and the Philip- pines. Realism of the microworlds was assessed with one Likert-type scale question. Perceived realism, however, did not negatively affect performance. On the contrary, German and U.S. par- ticipants had the highest performance in both microworlds. Another reason for the not ideal fit of both models in the German sample could be related to methodological problems. The German sample had, compared to the other national samples, the lowest reliability of action orientation and the lowest Cronbach alpha in nonverbal intelligence and (together with the Filipino sample) the lowest reliabilities in the four values (see also Soh & Leong, 2002, for a critique of reliability and validity of the value instrument). Chi-square difference tests showed significant differences between constrained and uncon- strained models only in WINFIRE. Pairwise comparisons of the structural paths in WINFIRE were conducted to investigate which paths differed between cultural samples. No differences among cultures were found for the paths from the two strategies to performance. Cross-cultural differences for the paths from the four values to the two strategies were found for the path from HC to action orientation and for the paths from VI and VC to planning. An interesting finding refers to VI and VC relating differently to action/planning across indi- vidualistic and collectivistic samples. To summarize, whereas VI was positively related to action orientation in the U.S. sample, it was negatively related to action orientation in the Indian and Filipino samples. Whereas VC was positively related to planning in the Filipino sample, it was

Güss

1293

negatively related to planning in the German and U.S. samples. One possible explanation for this result is that IC has different meanings and implications for concrete behaviors such as action and planning in collectivistic versus individualistic settings (see also Fischer et al., 2009). The first part of the fourth hypothesis about a positive association of action orientation and planning for HI and about a negative association of action orientation and planning for VC was confirmed. Results showed that HI was positively related to action orientation only in WINFIRE. Contrary to our prediction, VC was not negatively but positively related to action orientation in COLDSTORE. Why did participants with high VC not proceed cautiously in COLDSTORE? Indian and Filipino participants had significantly higher VC scores and significantly more actions in COLDSTORE compared to U.S. and German participants. Results might be due to specific task characteristics. Whereas WINFIRE is transparent, COLDSTORE is not. Participants ini- tially do not know and cannot see why the temperature is oscillating. Participants with high VC values may have felt more comfortable dealing with non-transparency, leading them to be more active in COLDSTORE. In this case, VC could be highly related to uncertainty avoidance values. The high uncertainty avoidance values for Germany (65 in Hofstede, 2001; 5.4 in House et al., 2004) and medium ones for the United States (46, 4.2), compared to low values for India (40, 4.0) and the Philippines (44, 3.7), could support this explanation. Thus, future research could investigate the role of uncertainty avoidance values in CPS strategies and performance.

Limitations

Although the theoretical models for WINFIRE and COLDSTORE were mostly confirmed, the results offer interesting insights and leave some questions. First, not all values were related to strategies and the percentage of variance in performance explained by the four cultural values was relatively small. Therefore, it is very likely that other cultural variables could also influence strategies and performance. One possible variable, uncertainty avoidance, was discussed previ- ously. Uncertainty avoidance is related to risk perception and dealing with unpredictable events and could therefore influence CPS strategy selection. Second, cultural differences in other vari- ables, such as performance orientation or assertiveness—as assessed in the Globe study—could also be related to performance in CPS tasks. Germany and the United States ranked high on these two value dimensions compared to Brazil, India, and the Philippines (House et al., 2004). Ideally, future research would control for those and other possible cultural variables of interest. The relationships between values and strategies differed between the two microworlds. VC, for example, was not related to action orientation in WINFIRE and was positively related to action orientation in COLDSTORE. Thus, values do not always trigger the same strategies. The results showed that participants, regardless of their value orientations, tried to adjust their behav- ior to the specific task demands of the two microworlds.

Conclusion

CPS is a key ability for adapting to changes in the material and social environment (Güss, Tua- son, & Gerhard, 2010). The environmental and social demands from which problem-solving strategies develop differ between cultures (Berry, 2004; Strohschneider, 2001). When people are confronted with novel problems, they rely on culturally developed and learned strategies to solve them (Güss & Wiley, 2007). The present study addressed a specific aspect of culture, namely cultural values, that could be related to particular CPS strategies. Results of this study are consistent with other research that shows the influence of cultural values on cognitive strategies. Tsagaris (2007), for example, investigated the relationship of the four value orientations (HI, VI, HC, and VC) and 13 thinking styles based on Sternberg’s theory

1294

Journal of Cross-Cultural Psychology 42(7)

of mental self-government. The four values predicted thinking styles better than several demo- graphic variables, and thus, the data show how closely cultural values are connected to prefer- ences in cognitive processing. Notwithstanding potential limitations related to sample and method, the present study makes important advances in understanding cultural influences on CPS performance. The two models tested in this study provide support for a novel theoretical claim that abstract cultural values related to perception of self and others affect CPS strategies, in turn affecting performance in the two dynamic microworlds WINFIRE and COLDSTORE. As the variance accounted for by par- ticular values was not too high, further research is needed to investigate other cultural variables, such as uncertainty avoidance, performance orientation, or social axioms (Bond et al., 2004) that could influence CPS strategies and performance. Future research could focus also on the process by which participants adjust their strategies in the microworlds as their strategic approach might change during the course of the microworlds. The findings of this research are theoretically relevant. They challenge a still dominant cogni- tive psychological paradigm that assumes individuals’ thinking and problem solving are context- and culture-free. On the contrary, culture influences even more basic cognitive processes such as perception (e.g., Segall, Campbell, & Herskovits, 1966), higher cognitive processes such as cat- egorization and reasoning (Medin & Atran, 2004; Nisbett, Peng, Choi, & Norenzayan, 2001), and problem solving (Güss & Wiley, 2007) as it occurs in a specific sociocultural context. Results of this study showed the impact of cultural values on precise strategies that optimally affect per- formance or functioning.

Acknowledgment

This research would not have been possible without the support of friends and colleagues abroad. I would like to thank especially M. Cristina Ferreira, Nadia, Cilio Ziviani, Miguel Cal, Anizaura, Rodrigo Cer- queira, Fabricia Mac-Cord, Livia de Melo, Daniela N. Oliveira, Anizaura L. Rodrigues de Souze, and Camila Sagae, in Brazil; Krishna Prasaad Sreedhar, S. Raju Samuelpa, Ajay Kesavan, T. K. Velayudham, and Ibrahim Syed, in India; President Bautista, Father Paul, Geraldine Sanil, Lorainne Matulag, Sally Maximo, Llyod Orduna, Oliver Pangan, and Peter Tuason, in the Philippines; Jochen Kaiser, Doreen Druschel, Lena Michael, Elizabeth Rebholz, Susanne Ress, Raphaela Schneider, Lena Wesseler, and Steffi Zurmühlen, in Germany; Shona Bailey, Emma Glencross, Tina Hicks, Chris Parker, Zara Peisker, Elizabeth Sanz, Vanessa Teixeira, Brian Wiley, Jennifer Williams, Ma. Teresa Tuason, and F. Dan Richard, in the United States; and the many students who participated. Especially, I wish to thank Dietrich Dörner for providing the two microworlds and Cameron McIntosh for providing feedback on statistical and path analytical procedure.

Declaration of Conflicting Interests

The author declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Financial Disclosure/Funding

The author disclosed receipt of the following financial support for the research and/or authorship of this article. The research reported herein has been supported by the National Science Foundation (Grant Nrs. 0218203 and 0349997) and through a research fellowship of the Alexander von Humboldt Foundation.

Notes

1. The Singelis et al. (1995) measure was cross-validated with a decision scenario instrument also assessing HI, HC, VI, and VC (Triandis et al., 1998). Cross-validation correlations indicated accept- able construct validity. I did not include Triandis et al.’s instrument in further analyses because it yields nominal data and psychometric properties are not accessible in the literature or in this study.

Güss

1295

2. Correlation tables for each country were omitted due to space limitations but are available upon request.

3. Maximum likelihood parameter estimates and modification indices for analyses of WINFIRE and COLDSTORE path models across five countries were omitted due to space limitations but are avail- able upon request.

4. A table with all pairwise z tests of path coefficients among all five countries is available upon request.

References

Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall. Berry, J. W. (2004). An ecocultural perspective on the development of competence. In R. J. Sternberg & E. L. Grigorenko (Eds.), Culture and competence (pp. 3-22). Washington, DC: American Psychological Association. Bond, M. H., Leung, K., Au, A., Tong, K.- K., Reimel de Carrasquel, S., Murakami, F., et al. (2004). Culture- level dimensions of social axioms and their correlates across 41 cultures. Journal of Cross-Cultural Psychology, 35, 548-570. Brehmer, B. (1999). Reasonable decision making in complex environments. In P. Juslin & H. Montgomery (Eds.), Judgment and decision making. Neo-Brunswikian and process-tracing approaches (pp. 9-21). Mahwah, NJ: Lawrence Erlbaum. Brehmer, B., & Dörner, D. (1993). Experiments with computer-simulated microworlds: Escaping both the narrow straits of the laboratory and the deep blue sea of the field study. Computers in Human Behavior, 9, 171-184. Brislin, R. W. (1986). The wording and translation of research instruments. In W. J. Lonner & J. W. Berry (Eds.), Field methods in cross-cultural psychology (pp. 137-164). Beverly Hills, CA: Sage. Brockner, J., Ackerman, G., Greenberg, J., Gelfand, M. J., Francesco, A. M., Chen, Z. X., et al. (2001). Culture and procedural justice. The influence of power distance on reactions to voice. Journal of Experi- mental Social Psychology, 37, 300-315. Brown, L., Sherbenou, R. J., & Johnsen, S. K. (1997). Test of Nonverbal Intelligence TONI-3 (3rd ed.). Austin, TX: Pro-Ed Incorporated. Byrne, B. M. (2006). Structural equation modeling with EQS: Basic concepts, applications, and program- ming (2nd ed.). Mahwah, NJ: Lawrence Erlbaum Associates Publishers. Chiou, J.- S. (2001). Horizontal and vertical individualism and collectivism among college students in the United States, Taiwan, and Argentina. Journal of Social Psychology, 141, 667-678. Cukur, C. S., de Guzman, M. R. T., & Carlo, G. (2004). Religiosity, values, and horizontal and vertical individualism-collectivism: A study of Turkey, the United States, and the Philippines. The Journal of Social Psychology, 144, 613-634. Dörner, D. (1996). The logic of failure. New York: Holt. Enders, C. K. (2001). A primer on maximum likelihood algorithms available for use with missing data. Structural Equation Modeling, 8, 128-141. Fischer, R., Ferreira, M. C., Assmar, E., Redford, P., Harb, C., Glazer, S., et al. (2009). Individualism- collectivism as descriptive norms: Development of a subjective norm approach to culture measurement. Journal of Cross-Cultural Psychology, 40, 187-213. Frensch, P., & Funke, J. (Eds.). (1995). Complex problem solving: The European perspective. Hillsdale, NJ:

Lawrence Erlbaum Associates Inc. Funke, J. (2001). Dynamic systems as tools for analysing human judgment. Thinking and Reasoning, 7,

69-89.

Gerdes, J., Dörner, D., & Pfeiffer E. (1993). Interaktive Computersimulation “Winfire” [The interactive computer simulation “Winfire”]. Bamberg, Germany: Otto-Friedrich-Universität Bamberg, Lehrstuhl Psychologie II.

1296

Journal of Cross-Cultural Psychology 42(7)

Gonzalez, C. (2005). Decision support for real-time dynamic decision making tasks. Organizational Behav- ior and Human Decision Processes, 96, 142-154. Gonzalez, C., Thomas, R. P., & Vanyukov, P. (2005). The relationship between cognitive ability and dynamic decision making. Intelligence, 33, 273-286. Güss, C. D. (2004, January). Decision making in individualistic and collectivistic cultures. Online Readings in Psychology and Culture (Unit 4, Chapter 3). International Association for Cross-Cultural Psychology:

Available at: http://orpc.iaccp.org/index.php?option=com_content&view=article&id=30%3Aguss43&

catid=22%3Achapter&Itemid=4

Güss, C. D., Tuason, M. T., & Gerhard, C. (2010). Cross-national comparisons of complex problem-solving strategies in two microworlds. Cognitive Science, 34, 489-520. Güss, C. D., & Wiley, B. (2007). Metacognition of problem-solving strategies in Brazil, India, and the United States. Journal of Cognition and Culture, 7, 1-25. Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviors, institutions, and organizations across nations (2nd ed.). Thousand Oaks, CA: SAGE. House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., & Gupta, V. (Eds.) (2004). Culture, leadership, and organizations. Thousand Oaks, CA: SAGE. Hu, L., & Bentler, P. M. (1995). Evaluating model fit. In R. H. Hoyle (Ed.), Structural equation modeling:

Concepts, issues, and applications (pp. 76-99). Thousand Oaks, CA: SAGE. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for the fit indexes in covariance structure analysis: Conven - tional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55. Kemmelmeier, M., Burnstein, E., Krumov, K., Genkova, P., Kanagawa, C., Hirshberg, M., Erb, H.- P., et al. (2003). Individualism, collectivism, authoritarianism in seven societies. Journal of Cross-Cultural Psychology, 34, 304-322. Kim-Jo, T., Benet-Martínez, V., & Ozer, D. J. (2010). Culture and interpersonal conflict resolution styles:

Role of acculturation. Journal of Cross-Cultural Psychology, 41(2), 264-269. Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York: The Guilford Press. Knoblich, G., & Scott, J. J. (2003). Action coordination in groups and individuals: Learning anticipatory control. Journal of Experimental Psychology: Learning, Memory, and Cognition, 29, 1006-1016. Komarraju, M., Dollinger, S. J., & Lovell, J. L. (2008). Individualism-collectivism in horizontal and verti- cal directions as predictors of conflict management styles. International Journal of Conflict Manage- ment, 19, 20-35. Lave, J. (1991). Situating learning in communities of practice. In L. B. Resnick, J. M. Levine, & S. D. Teasley (Eds.), Perspectives on socially shared cognition (pp. 63-82). Washington, DC: American Psy- chological Association. Li, A., & Reb, J. (2009). A cross-nations, cross-cultures, and cross-conditions analysis on the equivalence of the Balanced Inventory of Desirable Responding. Journal of Cross-Cultural Psychology, 40(2), 214-233. MacCallum, R. C., Roznowski, M., & Necowitz, L. B. (1992). Model modification in covariance structure analysis: The problem of capitalization on chance. Psychological Bulletin, 111, 490-504. Mann, L., Radford, M., Burnett, P., Ford, S., Bond, M., Leung, K., et al. (1998). Cross-cultural differences in self-reported decision making style and confidence. International Journal of Psychology, 33, 325-335. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and moti- vation. Psychological Review, 98, 224-253. McDonald, R. P., & Ho, M. (2002). Principles and practice in reporting structural equation analysis. Psy- chological Methods, 7, 64-82. McKinnon, A. J., & Wearing, A. J. (1985). Systems analysis and dynamic decision making. Acta Psycho- logica, 58, 159-172. Medin, D. L., & Atran, S. (2004). The native mind: Biological categorization and reasoning in development and across cultures. Psychological Review, 111, 960-983.

Güss

1297

Muthén, L. K., & Muthén, B. O. (1998-2006). Mplus user’s guide (4th ed.). Los Angeles: Muthén & Muthén. Nisbett, R. E., Peng, K., Choi, I., & Norenzayan, A. (2001). Culture and systems of thought: Holistic versus analytic cognition. Psychological Review, 108, 291-310. Nurmi, J.- E. (2008). Planning, motivation, and evaluation in orientation to the future: A latent structure analysis. Scandinavian Journal of Psychology, 30, 64-71. Ohbuchi, K.-I., Fukushima, O., & Tedeschi, J. T. (1999). Cultural values in conflict management: Goal orientation, goal attainment, and tactical decision. Journal of Cross-Cultural Psychology, 30, 51-71. Quesada, J., Kintsch, W., & Gomez, E. (2005). Complex problem-solving: A field in search of a definition? Theoretical Issues in Ergonomics Science. Special Issue: Cognitive Research with Microworlds, 6, 5-33. Raven, J. (1977). Advanced raven progressive matrices. Oxford, UK: Oxford Psychologists Press. Reichert, U., & Dörner, D. (1988). Heurismen beim Umgang mit einem “einfachen” dynamischen System [Heurisms in handling a “simple” dynamic system]. Sprache & Kognition, 7, 12-24. Rigas, G., Carling, E., & Brehmer, B. (2002). Reliability and validity of performance measures in micro- worlds. Intelligence, 30, 463-480. Schaub, H. (2001). Persönlichkeit und Problemlösen [Personality and problem solving]. Weinheim, Ger- many: Beltz, Psychologie Verlags Union. Schoppek, W., & Putz-Osterloh, W. (2003). Individuelle Unterschiede und die Bearbeitung komplexer Probleme [Individual differences and complex problem solving]. Zeitschrift für Differentielle und Diag- nostische Psychologie, 24, 163-173. Schumaker, R. E., & Lomax, R. G. (1996). A beginner’s guide to structural equation modeling. Hillsdale, NJ: Lawrence Erlbaum Associates. Schwartz, S. H. (1994). Beyond individualism/collectivism: New cultural dimensions of values. In U. Kim, H. C. Triandis, Ç. Kagitçibasi, S.- C. Choi, & G. Yoon (Eds.), Individualism and collectivism. (pp. 85-119). Thousand Oaks, CA: Sage. Segall, M. H., Campbell, D. T., & Herskovits, M. J. (1966). The influence of culture on visual perception. Indianapolis, IN: Bobbs-Merrill. Singelis, T. M., Triandis, H. C., Bhawuk, D. S., & Gelfand, M. J. (1995). Horizontal and vertical dimensions of individualism and collectivism: A theoretical and measurement refinement. Cross-Cultural Research, 29, 240-275. Sinha, D., & Tripathi, R. C. (1994). Individualism in a collectivist culture: A case of coexistence of oppo- sites. In U. Kim, H. C. Triandis, Ç. Kagitçibasi, S.- C. Choi, & G. Yoon (Eds.), Individualism and col- lectivism. Theory, method, and applications (pp. 123-136). Thousand Oaks, CA: Sage. Smith, P., Bond, M. H., & Kagitcibasi, C. (2006). Understanding social psychology across cultures: Living and working in a changing world. London: Sage. Smith, P. P., Peterson, M. F., & Schwartz, S. H. (2002). Cultural values, sources of guidance, and their rel - evance to managerial behavior. A 47-nation study. Journal of Cross-Cultural Psychology, 33, 188-208. Soh, S., & Leung, F. T. L. (2002). Validity of vertical and horizontal individualism and collectivism in Singapore: Relationships with values and interests. Journal of Cross-Cultural Psychology, 33, 3-15. Strohschneider, S. (2001). Kultur – Denken – Strategie: Eine indische Suite [Culture – thinking – strategy:

An Indian suite]. Bern, Germany: Hans Huber. Strohschneider, S., & Güss, D. (1998). Planning and problem solving: Differences between Brazilian and German students. Journal of Cross-Cultural Psychology, 29, 695-716. Süß, H.- M., Kersting, M., & Oberauer, K. (1993). Zur Vorhersage von Steuerungsleistungen an computer- simulierten Systemen durch Wissen und Intelligenz [Predicting control performance in computer-simulated systems by means of knowledge and intelligence]. Zeitschrift für Differentielle und Diagnostische Psychologie, 14, 189-203. Triandis, H. C., Chen, X. P., & Chan, D. K.- S. (1998). Scenarios for the measurement of collectivism and individualism. Journal of Cross-Cultural Psychology, 29, 275-289.

1298

Journal of Cross-Cultural Psychology 42(7)

Triandis, H. C., & Gelfand, M. J. (1998). Converging measurement of horizontal and vertical individualism and collectivism. Journal of Personality Social Psychology, 74, 118-128. Tsagaris, G. S. (2007). The relationships between thinking style preferences, cultural orientations and aca- demic achievement. Dissertation Abstracts International Section A: Humanities and Social Sciences, 67(9-A), 3290. van de Vijver, F. J. R., & Leung, K. (1997). Methods and data analysis for cross-cultural research. Thou- sand Oaks, CA: SAGE. Von Clausewitz, C. (1832). Vom Kriege [On war]. Berlin, Germany: Dümmlers Verlag. Vygotsky, L.S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press.