Graduation Year

2021

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Psychology

Major Professor

Michael T. Brannick, Ph.D.

Co-Major Professor

Sandra Schneider, Ph.D.

Committee Member

Michael D. Coovert, Ph.D.

Committee Member

Walter C. Borman, Ph.D.

Committee Member

Jason Beckstead, Ph.D.

Committee Member

Michael T. Braun, Ph.D.

Keywords

Choice, Diagnosis, Judgment, Medical

Abstract

Effective decision-making is critical and necessary for organizational success across a wide range of occupations, situations, and industries. However, decision making, by its nature, is not always a direct process of a single decision leading to a direct outcome. Rather, it can often become a multilevel process whereby one decision’s outcome leads to information that is used in subsequent larger or other types of decisions. The decision-making process then becomes progressively more complex and more difficult to navigate as these decisions compound within one another. Thus, decision-makers must find an appropriate way to approach such decisions. Understanding the multilevel nature of decision-making and how to optimize the final solution can have implications across a variety of areas. This dissertation aims to address those multilevel decisions in diagnostic medicine where the decision requires the assessment of multiple informational inputs. A psychometric approach was taken to look at different models pertaining to how these decisions can be made with the greatest degree of classification accuracy. Ultimately, tree-based models outperformed all other methods and were found to have the most applicability to diagnostic medicine. While some constraints related to tree-based modeling are noted, examples are shown to discuss how these models can be used to enrich current medical approaches. Possible implications for future research are examined.

Download

Share

COinS