This paper examines and critiques several learning style inventories (LSI) in an effort to find the optimal research instrument for a proposed empirical study of both the quality of courseware and the impact of courseware characteristics on variations of learning effectiveness due to learner characteristics. Five LSIs are examined: (1) Dunn LSI; (2) Entwistle's styles of learning; (3) Hunt's conceptual levels; (4) Kolb's LSI; and (5) two preference scales for students, the Grasha-Riechmann Student Learning Style Scales and Owens and Straton's Learning Preference Style for Students. The strengths and shortcomings of each model or instrument are detailed, reasons for choosing the Kolb LSI as the study instrument are given, and a brief outline of the proposed empirical study is provided. The text is supplemented by one figure and three tables. (15 references, 13 in English and 2 in German) (Author/EW)