An experimental investigation of activity-based costing, knowledge, and judgment accuracy

This dissertation experimentally investigates the effects of activity knowledge and activity-based costing (ABC) accounting knowledge on identifying and estimating distorted product cost allocations. Accuracy in identifying and estimating distorted product cost allocations is predicted to be an interactive function of activity knowledge, ABC accounting knowledge, and the cost accounting system.; Anecdotal evidence suggests that many, if not most, firms' cost systems are distorted, primarily due to inaccurate indirect product cost allocations. Managers claim they are able to estimate the error in reported product cost allocations that are distorted. Both the identification and the estimation components of this error-estimation task are investigated. Activity and ABC accounting knowledge may be two key inputs in identifying and estimating product cost allocation distortion. Variation in the success of implementing ABC systems may be related to the interaction among activity knowledge, accounting knowledge, and the cost accounting system error. The combination of activity and accounting knowledge is predicted to increase managers' accuracy in sufficiently estimating the correct amounts of cost allocations without implementing a complex cost accounting system (i.e., many cost pools and cost drivers).; The audit and tax scholarly literatures have found relationships between judgment performance and task and audit/tax knowledge. This dissertation builds on this previous knowledge literature. First, the effect of activity knowledge and ABC accounting knowledge on judgment accuracy is extended to a managerial setting. Second, the interaction among subsets of independent variables (activity knowledge, ABC accounting knowledge, and cost accounting system) is tested.; Activity knowledge and cost accounting system error are manipulated independent variables; ABC accounting knowledge is a measured independent variable. Identification and estimation accuracy are the two dependent variables. Two experiments are conducted to investigate specification (choice of cost driver) and aggregation (choice of cost pools) subsets of the independent variables. The experimental setting allows for control of the product cost allocation to measure judgment accuracy in estimating the error in cost allocations. Aggregation and specification are based on Datar and Gupta (1994) and extends the investigation of ABC accounting and activity knowledge to finer levels.; The results reported are for two experiments with a subject pool of graduate students. The results of the experiment designed to investigate specification provide statistical support for the interactive hypotheses for both identification and estimation of product cost allocation distortion. The results of the experiment designed to investigate aggregation provide statistical support for the interactive prediction made for the estimation of distortion but not for the interactive prediction made for the identification of distortion.