Determinants of confidence and their effect on epistemic judgments: Implications for auditor calibration bias

Auditor decisions regarding the causes of accounting misstatements can have an audit effectiveness and efficiency. Specifically, overconfidence in one's decision can lead to an ineffective audit, whereas underconfidence in one's decision can lead to an inefficient audit. This dissertation explored the implications of providing various types of information cues to decision-makers regarding an Analytical Procedure task and investigated the relationship between different types of evidence cues (confirming, disconfirming, redundant or non-redundant) and the reduction in calibration bias. Information was collected using a laboratory experiment, from 45 accounting students participants. Research questions were analyzed using a 2 x 2 x 2 between-subject and within-subject analysis of covariance (ANCOVA). ^ Results indicated that presenting subjects with information cues dissimilar to the choice they made is an effective intervention in reducing the common overconfidence found in decision-making. In addition, other information characteristics, specifically non-redundant information can help in reducing a decision-maker's overconfidence/calibration bias for difficulty (compared to easy) decision-tasks. ^

Social service provision, efficiency labor and economic development

This dissertation provides a theory of the effects and determinants of an economy's level of social services. The dissertation focuses on how the provision of social services will affect the effort decisions of workers, which will ultimately determine the economy's level of output. A worker decides on how much effort to contribute in relation to the level of social services he/she receives. The higher the level of social services received, the lower the cost—disutility—from providing effort will be. The government provides public infrastructure and social services (i.e. health services) in accordance with the economy's endowment of effort. In doing so, the government takes the aggregate effort endowment as given. Since, with higher individual work effort the higher the economy's total level of effort, failure by workers to coordinate effort levels will result in possible instances of low effort, low social services and low output; and, other instances of high effort, high social services and high output. Therefore, this dissertation predicts that in the context of social services, coordination failures in effort levels can lead to development traps. ^