Gender differences in levels of transfer and conceptual and procedural knowledge in engineering statistics students

This study investigated gender differences in introductory statistical knowledge for a sample of engineering undergraduate students with advanced math training (two semesters of college calculus). The participants were enrolled in a required three-hour introductory engineering statistics course at Arizona State University during the fall 1999 semester. Two schemes were used to classify assessment items. One classification was based upon level of transfer, with near and far categories identified. The second method of classification distinguished between conceptual and procedural knowledge.; No gender differences were found in overall statistics knowledge or in the pattern of performance on the items based upon the use of either classification scheme. Performance differences were found for the entire sample of students on both the level of transfer and type of knowledge tests. Students obtained higher scores on the near transfer and procedural knowledge scales.; Transfer was defined as the ability to apply knowledge to situations that differed from those encountered during training. Near transfer situations closely resembled those encountered during training, while far transfer situations differed to a greater extent. At least two factors may account for the small difference in performance on the two transfer scales. One explanation is that only a narrow range of the far transfer continuum was investigated in this study. Another explanation is that students received instruction that included only a limited number of different item formats for each concept or procedure.; Procedural knowledge was defined in this study as the application of the appropriate algorithm, rule or procedure. Most of the procedural items required computation of a numerical answer and/or application of a hypothesis testing procedure. Conceptual knowledge was defined as the ability to make connections between statistical ideas.; Students found procedural items moderately easier than conceptual items. Possible explanations for this finding are that conceptual items may have required more domain-specific memory, may not have relied as heavily upon the students' prior knowledge base in mathematics, or that a large part of the instructional emphasis was placed upon application of procedural routines to solving short “story” problems.; Classifying assessment items according to level of transfer and type of knowledge allowed the two categorization systems to be interrelated. Assessment items that require far transfer also require conceptual knowledge. Specifically, procedural items at a far level of transfer comprise one type of conceptual knowledge. While far transfer implies conceptual knowledge, conceptual knowledge does not necessarily imply far transfer.