2 resultados para factor analytic model
em University of Connecticut - USA
Resumo:
It is important to check the fundamental assumption of most popular Item Response Theory models, unidimensionality. However, it is hard for educational and psychological tests to be strictly unidimensional. The tests studied in this paper are from a standardized high-stake testing program. They feature potential multidimensionality by presenting various item types and item sets. Confirmatory factor analyses with one-factor and bifactor models, and based on both linear structural equation modeling approach and nonlinear IRT approach were conducted. The competing models were compared and the implications of the bifactor model for checking essential unidimensionality were discussed.
Resumo:
Truth and Reconciliation Commissions (TRC) have emerged in the last few decades as a mechanism for a state to overcome widespread, grave, human rights violations. There are numerous approaches to a TRC all with an ultimate goal: that formerly warring factions, perpetrators, witnesses, and victims can move forward as a united people. I propose that the provision of amnesty is critical to the success of a TRC. I hypothesize that the form of amnesty chosen (i.e. blanket v. conditional amnesty) determines the revelation of truth and realization of justice, which in turn dictates whether a TRC can achieve reconciliation. To test this hypothesis, I use two case studies: South Africa, which has utilized conditional amnesty, and Sierra Leone which has employed blanket amnesty. I create a model for measuring reconciliation. I can then look at the implications of both types of amnesty and assess which, in the end, is more effective. My overarching conclusion is that the provision of conditional amnesty is more effective than blanket amnesty in achieving reconciliation. Ultimately, I hope that this conclusion can be generalized to other TRCs.