Gabrielsen, Eva Susanne Brogger2014-11-142014-11-142014-09https://hdl.handle.net/11299/167864The United World College Red Cross Nordic (RCN) is a unique high school: 200 students representing roughly ninety countries pursue the two-year International Baccalaureate degree in an environment characterized by diversity, multicultural learning, intergroup cooperation, and peace education. The boarding school’s mission is to “make education a force to unite people, nations and cultures for peace and a sustainable future,” but to date, no formal assessments have evaluated the effectiveness of the multicultural peace education program on students’ decreased outgroup bias and prejudice. This longitudinal field study sought to evaluate RCN’s multicultural peace education program by assessing changes in students’ intergroup bias: negative outgroup emotions, desire for social distance, and generalized ethnocentrism.Three established social-psychological models of intergroup contact, the contact hypothesis, the intergroup contact model, and the common ingroup identity model, were compared and contrasted with data-driven linear mixed-effects built on the models’ contact conditions to determine whether a theoretical or data-driven model was the best fit for the current sample. The contact hypothesis specifies five contact conditions: equal status, common goals, intergroup cooperation, institutional support, and acquaintance potential. The intergroup contact model uses the five contact hypothesis conditions as a foundation, and adds that intergroup salience is important in intergroup contact situations. The common ingroup identity model specifies a common ingroup identified with by all participants in intergroup contact as the facilitator for success.A balanced panel design was established to survey students four times over their two years at RCN: within the first week of the first year, within the last week of the first year, within the first week of the second year, and within the last week of the second year. 272 students representing 96 countries participated in at least one of the four survey collections. Twenty-eight students, representing either peaceful or conflicted countries, were interviewed at the beginning and at the end of their RCN education. The third, and cross-sectional, sample upon which the analyses herein rest was of 256 United World College alumni, representing 77 countries and ten UWCs.By Time 4, students reported spending less time each week on their extra-academic commitments, but reported no change in the hours spent with friends. Perceptions of the presence of facilitating contact conditions and students’ self-reported intergroup anxiety decreased from Time 1 to Time 2, increased from Time 2 to Time 3, and decreased again from Time 3 to Time 4, the total change over time was significant.Students reported non-significant decreases in negative outgroup emotions, desire for social distance, and generalized ethnocentrism. Data-driven intergroup contact models were the consistent best fit for the data, for all three measures of intergroup bias. The particular covariates differed for the three: for negative outgroup emotions the covariates included equal status, common goals, and intergroup salience; for need for social distance the covariates included equal status, intergroup cooperation, and intergroup salience; for generalized ethnocentrism the covariates included intergroup cooperation and intergroup salience. For negative outgroup emotions and desire for social distance, adding intergroup salience as a predictor to the common ingroup identity model’s LME improves the models’ fit above and beyond all the best-fitting models. Cross-sectional alumni data corroborated findings from the longitudinal data on current students. The 256 alumni surveyed differed significantly from Time 4 students on almost all measures. Alumni reported significantly higher levels of religiosity, liberal political orientation, trust, empathy, and the eight contact conditions, and significantly lower levels of generalized ethnocentrism, intergroup anxiety, closedmindedness and social dominance orientation. Older alumni did not significantly differ from younger alumni, but men differed from women. Because of the cross-sectional nature of the data, LME models were inappropriate and linear regressions were used to establish whether the best-fitting LME models held in the alumni sample; they did not.Content analyses of 25 complete pre- and post-interviews showed different trends for the students from conflicted countries compared to students from peaceful countries. Themes that arose include increased knowledge of conflicts around the world, critical thinking and awareness of own self, ability and willingness to trust and show empathy toward others, conflict management skills, and finally, growth. Additionally, themes that emerged spontaneously across the interviews included consistency in thinking and an increased importance assigned values such as health, happiness, and freedom. Using the Linguistic Inquiry Word Count software, the following were found to differ significantly between Time 1 and Time 4: English proficiency, talking about family, the use of negative emotion words, talking about feelings, and insight.Several suggestions are presented to explain the lack of changes in intergroup bias. Because intergroup salience was found to be one of the most significant contact conditions, recommendations are provided for how RCN can continue to capitalize on representativeness among its students. The importance of RCN identity, too, is explored. Because pure contact is not enough, RCN must continue its work of facilitating contact conditions to encourage and increase the instances of successful intergroup contact.enContact hypothesisIntergroup contactLongitudinal dataModel comparisonProgram evaluationUnited World CollegePsychologyThesis or Dissertation