Concern about forecasts of national faculty shortages and the importance of local studies

The higher education popular press has published several forecasts of faculty shortages due to the mass retirement of expansion era faculty and the forecasted low numbers of doctoral graduates. These forecasts of national shortages encouraged a local study of faculty attrition in a four-campus university system. The local study considered two factors not accounted for in national studies but important locally: whether disciplinary field of study or campus location (urban or rural) has been associated with faculty attrition in the recent past and should therefore be considered when modeling demand for new faculty. Discipline and institution location were found to have no significant effect on the likelihood of faculty to continue employment from year to year when faculty age was controlled. Although retirement rate was expected to increase, no evidence of mass retirements producing an increased demand for new faculty was found. The lack of forecasted increases in demand for faculty locally led to reconsideration of national forecast studies, especially when it was determined that the age distribution of faculty across system campuses was similar to faculty nationwide. This paper questions whether there is sufficient evidence, as presented in the popular press, to support special action to increase the supply of faculty.Presented at the Thirty-first Annual Forum of the Association for Institutional Research, San Francisco, May 26–29, 1991.     

The influence of institutional involvement in international education on United States students

This study is focused on the impact of international education upon United States students. The research question tested was: does the involvement of a US university in international education programs affect the international outlook of United States students? Seven concept areas were developed and tested on a randomly selected group of 663 United States senior students enrolled at six universities: Indiana University, the University of Pittsburgh, the University of Wisconsin at Madison, Ohio State University, the University of Illinois at Urbana, and Southern Illinois University at Carbondale. The analysis of variance tests used indicated that for the most part the involvement of a university in international education programs does make a difference in the international outlook of United States students.     

Plant cell cultures—a future source of natural products?

Plant cell cultures are a potential source of commercially important compounds, but although technical problems seem to be relatively easy to solve there are several biological hurdles still to be overcome. An analysis of the state of the art shows that a breakthrough can be expected only if research is intensified on unravelling the regulatory controls in the biosynthesis of natural products. The knowledge gained might then be used to manipulate the expression of secondary metabolism by the techniques of genetic engineering.     

The influence of deformation on barbell mechanics during the clean pull

For simplicity of biomechanical analyses, the weightlifting barbell is typically modelled as a rigid, nondeformable object. Most coaches and weightlifters, however, are aware of the elastic nature of the barbell, and its influence on the successful completion of lifting attempts. Variables such as velocity, work performed, and power output are indicators of the quality of performance during the snatch, clean, and related weightlifting pulling movements. The aim of this study was to establish whether differences exist in determining these biomechanical parameters when the centre of the barbell is analysed compared with each end of the barbell. Nine men performed three maximal-effort repetitions in the clean pull exercise at 85% of their self-reported single repetition maximum (1-RM) clean (90-155 kg) using a barbell instrumented for mechanical analysis. Results indicated that peak barbell speed was 5-30% (P < 0.05) lower for the centre of the barbell than the ends. Although differences (P < 0.05) in kinetic and potential energy were found between the centre and ends of the bar, differences between total work performed were small (< 6%; P < 0.05) and no differences were observed for average power (P > 0.05). Although approximately the same work and power occur for the centre and ends of the barbell, they manifest as different kinematics as a result of the elastic nature of the equipment. The elastic characteristics should be considered when selecting instrumentation and variables for research involving barbells. Coaches should be aware of the elasticity of barbells, including selecting appropriate viewing angles as well as understanding how deformation may affect the ends of the barbell relative to the centre.     

Reducing the grade disparities between American Indians and Euro-American students in introduction to psychology through small-group,peer-mentored,supplemental instruction

Supplemental instruction (SI) is a small-group, peer-mentored programme which is compatible with the learning preferences of American Indian students. We tested the hypothesis that SI is a compensatory strategy that reduces the differences in the grades earned in introduction to psychology by Euro-American and American Indian students. The sample consisted of 129 American Indian students and 4588 Euro-American students enrolled in introduction to psychology at a US university. As hypothesised, a multi-level model yielded a significant (p < .01) interaction between SI and ethnicity on course grade. Whereas for non-SI users, the gap between Euro-American and American Indian students was .71 grade points, for SI users, it was only .15 grade points. Strategies should be devised for increasing SI visits by students enrolled in introduction to psychology, particularly those who belong to American Indian tribes.     

Net joint moments and muscle activation in barbell squats without and with restricted anterior leg rotation

Muscle utilisation in squat exercise depends on technique. The purpose of this study was to compare net joint moments (NJMs) and muscle activation during squats without and with restricted leg dorsiflexion. Experienced men (n = 5) and women (n = 4) performed full squats at 80% one repetition maximum. 3D motion analysis, force platform and (EMG) data were collected. Restricting anterior leg rotation reduced anterior leg (P = 0.001) and posterior thigh (P < 0.001) rotations, resulting in a smaller knee flexion range of motion (P < 0.001). At maximum squat depth, ankle plantar flexor (P < 0.001) and knee extensor (P < 0.001) NJM were higher in unrestricted squats. Hip extensor NJM (P = 0.14) was not different between squat types at maximum squat depth. Vastus lateralis (P > 0.05), vastus medialis (P > 0.05) and rectus femoris (P > 0.05) EMG were not different between squat types. Unrestricted squats have higher ankle plantar flexor and knee extensor NJM than previously reported from jumping and landing. However, ankle plantar flexor and knee extensor NJM are lower in restricted squats than previous studies of jumping and landing. The high NJM in unrestricted squat exercise performed through a full range of motion suggests this squat type would be more effective to stimulate adaptations in the lower extremity musculature than restricted squats.     

Leveraging relevance cues for language modeling in speech recognition

Language modeling (LM), providing a principled mechanism to associate quantitative scores to sequences of words or tokens, has long been an interesting yet challenging problem in the field of speech and language processing. The n-gram model is still the predominant method, while a number of disparate LM methods, exploring either lexical co-occurrence or topic cues, have been developed to complement the n-gram model with some success. In this paper, we explore a novel language modeling framework built on top of the notion of relevance for speech recognition, where the relationship between a search history and the word being predicted is discovered through different granularities of semantic context for relevance modeling. Empirical experiments on a large vocabulary continuous speech recognition (LVCSR) task seem to demonstrate that the various language models deduced from our framework are very comparable to existing language models both in terms of perplexity and recognition error rate reductions.