We present statistical evidence that when students voluntarily write reflective "journal entry" essays on assigned reading, their performance on multiple-choice quizzes on the reading is improved compared to students who do not complete such essays. A Smirnov non-parametric test and a pair-wise comparison analysis were used to establish the validity of the hypotheses.

No abstract

Traditional grading of student work has several inherent problems including a granularity in grades which is too fine, passive student involvement, and a questionable connection between the grade and the learning objectives. We have developed and refined a quality based assessment process that addresses some of these problems. The quality assessment process has three, closely integrated aspects: 1) a three level assessment outcome (Meets Expectations, Exceeds Expectations, Needs Improvement), 2) the use of checklists to define expectations, and 3) the active involvement of students. The three levels of assessment are based on a three level customer needs model developed by Kano. In the context of the Kano model, faculty are viewed as customers and student work is the product. Three part checklists consistent with this model are used; these checklists facilitate the assessment of technical correctness, other course standards, student readiness, as well as the level and quality of the effort. The students are expected to use the checklists to determine how to Meet or Exceed Expectations on the assignment.

A general, sufficient test for proving a local minimum to be the global minimum is presented. The test uses a special lower bounding function and methods for constructing this special bounding function for univariable or multivariable unconstrained signomial design problems are given. Two simple design problems illustrate the use of the constructions and test.

No abstract