The effect of problem-solving styles on accountants ̓materiality decisions

Vita.The primary purpose of this dissertation was to investigate possible relationships between problem-solving styles of accountants and the materiality decision-making process, Problem-solving styles were derived from Jung's theory of psychological types. They were determined using the Form G of the Myers-Briggs Type Indicator. A set of three cases containing financial data for hypothetical companies was used to ascertain accountants' materiality thresholds expressed as dollar amounts to be disclosed in financial statements, and ratings of factors considered to play important roles in the process of making materiality decisions. The design of this study emphasized an individual behavioral attribute in contrast to previous research on this accounting concept, which was predominantly results-oriented. Testing of hypotheses of the study was based on responses from a sample of accountants that were employed by three public accounting firms in Houston, Texas. Each of the fifty-six accountants in the final sample made a total of ten materiality decisions and they rated factors on the basis of importance for each of the decisions. The two principal components of the investigation were analyzed separately. Major conclusions were as follows: 1. Three models were designed to test the independence of materiality decisions. These models included problem-solving style, firm membership and professional staff level. None of these independent variables were significantly related to the amounts expressed as materiality thresholds. Further, interaction of problem-solving style with the other two attributes was not significant in any of the models tested. 2. The second aspect of the study, factor importance, directed attention to the hypotheses relating to similarity of factor ratings within and between styles. Preliminary analysis indicated that accountants with the same problem-solving style highly agreed in their evaluation of factors. Subsequent analysis on an item-by-item basis and statistical testing of the agreement exhibited in pooled orderings of factor rankings was contradictory.