University class size: Is smaller better?

This study investigates the relationship between class size and achievement at the college level. Recent meta-analyses have found a strong relationship between class size and student achievement, but few of the studies examined class size larger than 40 or university-age populations. This analysis examines a university's testing-center data-archives representing 305 sections from 24 different courses. Section sizes ranged from 13 to 1,006. A total of 16,230 test scores were analyzed. The results of this investigation reveal that at the college level, class size may be less important an influence on student achievement than some educators have thought.     

The interaction of conditioned taste aversions and schedule-induced polydipsia: Availability of alternative behaviors

Animals poisoned following exposure to saccharin subsequently avoided the schedule-induced consumption of saccharin. While this suppression was transient for subjects who had access only to the saccharin solution during the free-food presentations, recovery of schedule-induced saccharin consumption was significantly retarded for subjects who had concurrent access to saccharin and a running wheel. It has been suggested that the transient suppression of schedule-induced polydipsia by conditioned taste aversions results from the pellet-induced tendency to drink within the schedule-induced polydipsia procedure. That access to the running wheel reduces schedule-induced polydipsia in general and prolongs the suppression of schedule-induced polydipsia by taste aversions supports this view.     

Behavioral momentum in academics: Using embedded high‐p sequences to increase academic productivity

The use of high‐probability (high‐p) request sequences has enjoyed support in the applied behavioral literature as a method to increase compliance. Based on the theory of behavioral momentum, high‐probability sequences increase the rate of responding, and subsequent rate of reinforcement, within a response class. This increase in density of reinforcement results in increased responding for the response class as a whole. Early research in this area had focused mainly on compliance issues for individuals with developmental disabilities. However, more recently the utility of high‐p sequences has been examined within the context of academics. The purpose of these two experiments was to examine the use of high‐p sequences with two academic tasks—letter‐writing and mathematics problem completion. The results of these studies suggest that high‐p sequences can be used to increase academic productivity. Moreover, the addition of experimenter‐delivered reinforcers to existing high‐p sequences enhances overall behavioral persistence. Theoretical and applied implications of persistence and behavioral momentum are discussed. © 2004 Wiley Periodicals, Inc. Psychol Schs 41: 789–801, 2004.     

Trends in computer applications in science assessment

Seven computer applications to science assessment are reviewed. Conventional test administration includes record keeping, grading, and managing test banks. Multiple-choice testing involves forced selection of an answer from a menu, whereas constructed-response testing involves options for students to present their answers within a set standard deviation. Adaptive testing attempts to individualize the test to minimize the number of items and time needed to assess a student's knowledge. Figurai response testing assesses science proficiency in pictorial or graphic mode and requires the student to construct a mental image rather than selecting a response from a multiple choice menu. Simulations have been found useful for performance assessment on a large-scale basis in part because they make it possible to independently specify different aspects of a real experiment. An emerging approach to performance assessment is solution pathway analysis, which permits the analysis of the steps a student takes in solving a problem. Virtually all computer-based testing systems improve the quality and efficiency of record keeping and data analysis.