Three factors, potential (P), motivation (M), and development (D) constitute the frame of reference proposed in this research for leadership development: P x M x D. The three factors are presented here in an interactive (multiplicative), rather than an additive relationship because of our prior assumption that if one of the factors is absent (or has zero value), the product of the multiplication will be zero, namely no leadership. This assumption has yet to be examined empirically, although it has been raised in the past on the basis of common sense. On the other hand, it may be argued that while one or two of the components (almost certainly potential and motivation) are essential for leadership, the third (development) is possible and additive, but not essential. This, then, is the aim of the present research: to examine the nature of the relationship between the three components. At the same time, the research assumption is that all three components are required in the leadership process and that the absence of one of them will invalidate the equation, in other words, will not produce effective leadership. 15. SUBJECT TERMS Leadership, potential to lead, motivation to lead, leadership development, self-efficacy, locus of control, anxiety, attachment, optimism, personality SECURITY CLASSIFICATION OF 19. LIMITATION OF ABSTRACT 20. NUMBER OF PAGES

Taking a constructivist, collaborative experiential learning approach to education and training of global managers, we designed an on-line, 4-week virtual multicultural team project and tested its effect on the development of management students' cultural intelligence, global identity, and local identity. The total sample of 1221 graduate management students, assigned to 312 virtual multicultural teams, consisted of four cohorts, each participating in one 4-week project; one project was conducted every year between 2008 and 2011. All projects were designed in the same way, according to principles of collaborative experiential learning, and offered a psychologically safe learning environment that enabled trust building. Data on cultural intelligence, global identity, and local identity were collected by way of web-based questionnaires at the beginning and at the end of the project, as well as 6 months later. Team trust was assessed in the middle of the project. Hierarchical linear modeling analyses revealed that cultural intelligence and global identity, but not local identity, significantly increased over time and that this effect lasted for 6 months after the project had ended. Trust as a team level factor moderated the project's effect on team members' cultural intelligence and global identity, with significant effects under moderate to high rather than low levels of trust.

This study returns to the question that occupied "trait approach" scholars in the early days of leadership research: identification of the major capacities required for leadership. The conceptual and methodological progress that has been made in psychology since the trait approach enables us to formulate models and deal with variables that did not exist in the early days of leadership research. We argue that three types of psychological capacities are essential for leadership: (a) self confidence, expressed and measured by three variables-internal locus of control, low level of trait anxiety, and self-efficacy; (b) proactive orientation, expressed by optimism; and (c) capacities required for prosocial relationships, expressed by secure attachment styles.A series of questionnaires was administered to 402 soldiers from Infantry and Armored corps who were nearing the end of 3 months' basic training. A sociometric questionnaire examining the peers' and commanders' evaluations of the soldiers' leadership capacities was used to evaluate each soldier and to classify the soldiers dichotomously as leaders and nonleaders.The findings reveal significant differences between leaders and nonleaders in all the variables that were defined as psychological capacities to lead. Leaders have more internal locus of control, a lower level of anxiety, higher self-efficacy, and more optimism, and they rank higher in the measure of secure attachment style. Anxiety, locus of control, and attachment style were found to be significant in the regression equation, but trait anxiety was found to be the most discriminant variable. The implications of the findings in the light of relevant psychological models are discussed.