The thinking preferences of 487 students at the University if North Carolina at Charlotte were evaluated with the Hermann Brain Dominance Instrument (HBDI) 1 at the beginning of the fall 1995 term. These beginning engineering, computer science, and engineering technology students were grouped in teams of four to seven students for projects in their ENGR 1201 Introduction to Engineering Practice and Principles or EGET 3071 Professional Development in Engineering Technology courses. As much as possible the teams were multidisciplinary. The engineering teams included computer science, electrical, mechanical, civil, and general (undeclared) majors; the engineering technology teams included electrical, mechanical, civil, and manufacturing engineering technology majors. Half of the teams in each of the two courses were selected with consideration of the HBDI profiles of the students in an attempt to form heterogeneous or so-called "whole-brain" teams; the other half of the teams had random distributions of thinking preferences. At the end of the term, the team projects (and the team learning process) were evaluated by faculty teams in order to test the hypothesis that heterogeneous teams as a whole will have better problem-solving outcomes even when the homogeneous student teams have been taught about thinking preferences and their implications for group dynamics and communications. This is an important issue: can the cost and labor involved in using the HBDI (especially at the freshman level) be justified by its contribution to improved team outcomes? This paper reports the preliminary findings of the first phase of a longitudinal study at UNC Charlotte examining the relationships between the make-up of thinking style profiles of teams and the outcomes they produce. The current study attempts to validate earlier studies of the thinking preferences of engineering students conducted at the University of Toledo 2. The study will add detailed quantitative and qualitative assessment data over a five year projected course. This paper reports initial anecdotal findings; early statistical assessment will be available at the ASEE '96 Annual Conference.

No abstract