Essential competencies of exceptional professional software engineers

Abstract

This dissertation presents a differential study of exceptional and non-exceptional professional software engineers in the work environment. The first phase of the study reports an in-depth review of 20 engineers. The study reports biographical data, Myers-Briggs Type Indicator test results, and Critical Incident Interview data for 10 exceptional and 10 non-exceptional subjects. Phase 1 concludes with a description of 38 essential competencies of software engineers. Phase 2 of this study surveys 129 engineers. Phase 2 reports biographical data for the sample and concludes that the only simple demographic predictor of performance is years of experience in software. This variable is able to correctly classify 63% of the cases studied. Phase 2 also has the participants complete a Q-Sort of the 38 competencies identified in Phase 1. Nine of these competencies are differentially related to engineer performance. A10 variable Canonical Discriminant Function is derived which is capable of correctly classifying 81% of the cases studied. This function consists of three biographical variables and seven competencies. The competencies related to Personal Attributes and Interpersonal Skills are identified as the most significant factors contributing to performance differences.