This paper presents a study that sought to use tablet personal computers (PCs) with the InkSurvey software for improving the instruction of a probability and statistics course for engineering students at the Colorado School of Mines (CSM). The InkSurvey software was designed in the physics department at CSM for use with tablet PCs and has the benefit of supporting anonymous student-teacher interaction during instruction. Using InkSurvey, an instructor can ask a question to which students may electronically respond. The instructor can then view the anonymous responses and provide formative feedback during class. In this investigation, a pre and post attitude survey and a pre and post statistics concept inventory was administered to students in six sections of a probability and statistics course for engineers. Five instructors taught the various sections using different approaches, all of which included the tablet PCs with InkSurvey. Across sections, the students' attitudes became more negative towards statistics over the course of the semester and, based on a statistics concept inventory, there was no evidence to support that their understanding of the subject had improved. This paper explores the potential explanations for these bleak results.