Cooperativeness and competitiveness as two distinct constructs: Validating the Cooperative and Competitive Personality Scale in a social dilemma context

Abstract

The present research validated the construct and criterion validities of the Cooperative and Competitive Personality Scale (CCPS) in a social dilemma context. The results from three studies supported the notion that cooperativeness and competitiveness are two independent dimensions, challenging the traditional view that they are two ends of a single continuum. First, confirmatory factor analyses revealed that a two-factor structure fit the data significantly better than a one-factor structure. Moreover, cooperativeness and competitiveness were either not significantly correlated (Studies 1 and 3) or only moderately positively correlated (Study 2). Second, cooperativeness and competitiveness were differentially associated with Schwartz's Personal Values. These results further supported the idea that cooperativeness and competitiveness are two distinct constructs. Specifically, the individuals who were highly cooperative emphasized self-transcendent values (i.e., universalism and benevolence) more, whereas the individuals who were highly competitive emphasized self-enhancement values (i.e., power and achievement) more. Finally, the CCPS, which adheres to the trait perspective of personality, was found to be a useful supplement to more prevalent social motive measures (i.e., social value orientation) in predicting cooperative behaviors. Specifically, in Study 2, when social value orientation was controlled for, the CCPS significantly predicted cooperative behaviors in a public goods dilemma (individuals who score higher on cooperativeness scale contributed more to the public goods). In Study 3, when social value orientation was controlled for, the CCPS significantly predicted cooperative behaviors in commons dilemmas (individuals who score higher on cooperativeness scale requested fewer resources from the common resource pool). The practical implications of the CCPS in conflict resolution, as well as in recruitment and selection settings, are discussed.

Keywords:

• Cooperativeness
• Competitiveness
• CCPS
• Validity
• Social dilemma

Notes

¹In Xie et al.'s study (2006), a nine-point Likert scale was used.

²Mass mailing is a service provided by the university to generate electronic mass mailings of officially required notification information, announcements, and promotions of events and activities to all students in the university.

³The details of this experiment are not related to the aims of the current research, so we will not mention them.

⁴Note that we only included those error covariances that share similar wordings or meanings within a dimension if they were suggested by the LM test. Also, adding the error covariances did not change the estimation of the other parameters much, which suggests that the proposed model/theory fits the data well; no significant modification of the theory is needed. Practices for the following two studies are the same as for this one.

⁵Although according to Hu and Bentler (1998) index of a good fit for the SRMR should be smaller than .05, a recommendation by Schermelleh-Engel, Moosbrugger and Müller (2003) suggests that a value for the SRMR that is equal to or smaller than .10 is an acceptable fit.

⁶Group size was an experimental factor that was beyond the scope of this study; adding group size into the analysis returned similar pattern of results; group size was thus excluded from further analyses.

⁷With four error covariances in the cooperativeness dimension added, following the recommendation of the LM test.

⁸Factor mean scores of the cooperativeness and competitiveness dimensions were used here and in subsequent analyses.

⁹We also tested whether prosocials and proselfs differ in their scores for self-transcendent values and self-enhancement values, and we found that the proselfs scored significantly higher on achievement values than the prosocials. Also, when the effect of SVO was controlled for, cooperativeness was still positively correlated with achievement, benevolence, and universalism, and competitiveness was still positively correlated with achievement and power.

¹⁰With four error covariances in the cooperativeness dimension added, following the recommendation of the LM test. Note that across three studies, we added four to six error covariances to improve the model's fit. All two-factor models across all studies shared a considerable amount of error covariance (2–3). Such inconsistencies might be due to the differences in the CCPS administration in these studies. Specifically, the CCPS was administered at the end of the experiment in Study 1, and at the beginning of the experimental sessions in Studies 2 and 3.

¹¹Items of the scale were originally in Simplified Chinese. Permission to publish the full scale was obtained from the scale developer.