Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




This research was designed to provide empirical evidence concerning the effects disclosing different levels of segment data has on the decision making performance of sophisticated users (Chartered Financial Analysts) of published corporate financial statements. The primary surrogate for influence on decision making was the existence of statistically different net earnings projections. To accomplish the research objective CFAs were supplied with financial statements of a hypothetical company disclosing varying levels (six) of segment disclosures and were asked to predict corporate earnings for 1981 and to indicate the range in which they were 95% confident true corporate earnings would fall. Five hypotheses were stated and the data obtained from the CFAs were analyzed with various statistical methods to determine whether the segment disclosures influenced: (1) the CFAs' predictions of corporate earnings, (2) variability in earnings predictions (communicative ability among CFAs), and CFAs' confidence in their predictions of corporate earnings. Level of segment data disclosed did significantly affect CFAs' average predictions of corporate earnings. Tests indicated, however, that once revenue and profitability data were furnished no further changes in predictions occurred as additional segment data was furnished. Segment earnings variability and level of segment disclosure did not interact to affect CFAs' predictions. There was a significant change in communicative ability of the disclosures when groups with no segment data were compared to the other five groups. There was no significant differences when the no segment data group was excluded. This implies that once basic segment revenue data is presented disclosure needs are met. Similar results were obtained regardless of whether the analysis was conducted when segment earnings variability was small, large, or both combined. CFAs' confidence in their earnings predictions was not enhanced as a result of being provided increased levels of segment data. Their range of confidence was generally much smaller when segment earnings variability was small. The findings of the study do not provide conclusive evidence regarding the need and usefulness of the required segment disclosures studied. The findings suggests that the presentation of basic revenue and profitability data meets the segment disclosure needs of the sophisticated financial statement user.