Expectations,observations, and violations: Comparing their effects on course ratings

Knowledgeable observers generally agree that students can give reasonably unbiased assessments of their classes, as long as proper instruments and procedures are used. These same observers agree, however, that more research is needed to understand the subtle dynamics which affect evaluations. Students' expectations (obtained on the first day) and the violation of these expectations are two possible factors. Their effects on students' final interest and future intentions are compared with students' actual reports of the class given at the end of the semester (n=196). Regression analysis using a simultaneous entry of expectations, expectancy violations, and final reports shows that students' reports at the end of the course account for almost all of the explained variance in the course rating items.     

Controlled choice assignments: A new and more effective approach to school desegregation

The authors argue that conventional mandatory and voluntary school desegregation techniques have failed to provide minority students with maximum feasible relief from invidious discrimination because they do not redress the root causes of segregation. They demonstrate that residential-based student assignment plans as well as plans that give students limited access to a few educationally enriched magnet schools are not asymmetrical remedies to wrongful segregation and thus work to perpetuate the racial and social inequities that they purport to correct. The authors offer an alternative approach to school desegregation—controlled choice—that operates independently from residential housing patterns and which explicitly seeks to maximize personal choice within a racially unitary, equitable, and educationally enhanced system of public schooling. Controlled choice was originally pioneered in Cambridge, Mass. during the early 1980s and has recently been adopted in Little Rock, Ark., and Fall River, Mass. The policy is also being actively reviewed by several of the nation's largest school districts including Seattle, Memphis, and Boston.     

Long-range course planning by college students

A group of 633 college students at a large midwestern public university was asked fifteen questions pertaining to personal long-term course planning. The results indicated that only 48% of the students sketched out a tentative listing of at least two courses per term for one or more subsequent semesters when first planning courses for the current semester. Long-range planning was positively correlated with the number of years students had already spent in college and with students' perception of the following factors: course planning utility, degree of certainty about pursuing one's current first choice of career, and degree of commitment in using college coursework as a means of achieving a career goal (all p < 0.001). Long-range planning was negatively correlated with the anticipation of a possible change in academic interests. Such planning was not significantly correlated with students' entrance examination aptitude scores or grade point averages. Thus, students' long-range planning of courses was associated more strongly with career commitment than with recent academic performance or aptitude.     

A new look at learning-disabled and normal-achieving children's word-pair memorization ability

An 11-parameter Markov model of stages of learning, which was developed by Brainerd et al., will be presented. The focus will be on parameter interpretation in terms of long-term memory (LTM) processes. Next the model will be used to examine the contribution of these various LTM processes to Schooltype and Age differences in a word-pair memorization experiment. Subjects were 8 and 11-year-old normal-achieving (NA) children from regular elementary schools and 8 and 11-year-old children from special schools for learning disabled (LD) children. One of the main conclusions is that the Schooltype × Age interaction which was demonstrated for traditional performance measures, can be explained by LD children's little developmental progress in storage processes and deviations from normal development in aspects of the acquisition of retrieval algorithms. No deficiences were, however, detected in the LD children's rerention ability early in the task, and in the so-called heuristic retrieval operations.     

Masculinity and femininity in twin children: genetic and environmental factors

Genetic and environmental origins of individual differences in masculine and feminine personality attributes were investigated in a sample of 38 monozygotic and 32 dizygotic twin pairs (total N = 140) during pre- and early adolescence. Self-report measures of both masculine and feminine characteristics were obtained for each child using 2 standardized instruments: the Children's Personality Attributes Questionnaire (CPAQ) and the Adolescent Self-Perception Inventory (ASPI). Multivariate biometrical analyses revealed significant genetic influences in all measures, accounting for 20%-48% of the observed variation in each. Environmental influences, which explained the remaining 52%-80% of variance in masculinity and femininity, were apparently specific to each individual and not shared by members of the same twin pair. The results underscore the importance of considering both genetic and environmental factors in gender-role development, particularly in studies of family resemblance.