Pupils’ Attitude Toward Technology (PATT) research began in Botswana in 1993. The research was designed to study pupils attitudes toward technology through the use of an instrument that has shown to be effective in measuring affective/behavior and cognitive attitude dimensions. The instrument used was an (English) adaptation of one created at the University of Technology Eindhoven, The Netherlands, by Drs. Jan Raat and Marc de Vries in 1984. Since that time, versions of the Dutch instrument have been used to assess the pupils' attitude toward technology in over 20 countries worldwide. One of the aims of this ex post facto study was to produce a descriptive profile of the student population based on a sample of 800 Form 5 pupils. The model employed a comparative framework controlling for differences in demographic characteristics that included GENDER, LEVEL OF TECHNOLOGICAL STUDY, and URBAN/RURAL BACKGROUND. The identification of these variables was seen as important in terms of their social significance within a changing traditional culture. Interest in comparing the findings from male and female students living in diverse rural traditional and modern urban environments, was the rationale for the investigation. Another aim was to compare responses from pupils with a technological background in school to pupils with no prior technology course work. Another aspect of the research was to encourage opportunities to share the findings in cross-cultural comparisons with research in other African countries. The Botswana instrument was modified from a study conducted in the USA (Bame, de Vries, and Dugger) and re-designed for Form 5 pupils (ages 16-21). The modified instrument was field tested during October-November, 1993, with 800 pupils in eight schools (four rural, four urban) across the nation. The instrument contains four basic parts. The first part asks pupils for a short description of what the student thinks technology is. The second part consists of 14 questions to gather demographic data about the respondents, and a survey of the technical subjects a pupil may have studied. In the third part, 58 statements were included to assess the respondent's attitude toward technology. In the fourth part, 31 items assess the pupils' concept of technology. The findings revealed that gender was a factor that affected students' attitude toward technology, as was the level of technology pupils studied in school. To a lesser extent, but still an important factor, the urban/rural backgrounds of pupils was found to combine with other variables, and thus contribute toward pupils’ attitudes and concepts of technology. The findings also showed that in general, a positive correlation was determined to exist between pupils' concept of technology and their attitudes toward it. However this relationship, while observed significant, was in magnitude, not notably strong. The contributions of PATT Research in Botswana offer both educators and program planners an instrument to assess the needs of particular pupil populations. Government and local planners need a means of monitoring formal educational efforts. As a curriculum development tool, the design offers a responsive solution to the needs of assessment and evaluation. / Ph. D.

A single-plate interferometric system was developed for the evaluation of unsteady flow generated by a passing shock in a linear transonic turbine cascade. A 15 mW Helium-Neon laser was used as the light source. An intensified charge-coupled device (CCD) was used to capture images, allowing the required 100 ns shutter times to “freeze” the passing shock. High frequency Kulite pressure transducers and miniature hot wire probes were used to capture the required known density field for the quantitative evaluation of the flowfield. Variations in the density field data as the shock passes upstream of the studied blade passage are presented, showing the capabilities of this system and detailing the unsteady flow field. The automation of the interferogram interpretation is presented. Image analysis techniques used include adaptive binarization and Hilditch line thinning. Further areas for improvement are also discussed. In addition to the interferometric study, a boundary layer transition study was also undertaken. Magnified boundary layer spark shadowgraphs were taken on both the pressure and suction surfaces of the turbine blade. Flow conditions including low (<1.0%) and high (5%) free stream turbulence values were investigated. The transition location was found to be dependent not only on the level of turbulence but also on the turbulence characteristics. Further, an investigation was made into the application of hydrodynamic stability theory to turbine blade flows. The results of this evaluation were used in the determination of appropriate length scales for tailoring free stream turbulence to more effectively induce suction surface transition. Little instability and no transition was calculated for the blade suction surface in agreement with the shadowgraph results. Free stream turbulence with frequencies near 40 kHz were found to most affect the instabilities in the boundary layer. Tailored free stream turbulence was found to produce transition on the blade suction surface. / Ph. D.

Rapid changes in the drafting profession have left gaps in skills and resources which hinder the ability of community college drafting instructors to provide students with the advanced skills needed for employment. To minimize these problems and to facilitate the integration of CAD into drafting programs, local and state authorities must plan to allocate adequate resources to each program. They must also provide avenues for instructors to become proficient with CAD and to assess the instructor's perceptions about their needs. The purpose of this study was to identify areas of perceived need of community college drafting instructors in Virginia. A survey was mailed to 36 community college drafting instructors. For 14 categories of CAD skills, instructors were asked to indicate their (1) technical skill, (2) need for technical training, (3) level of CAD integration, (4) integration skill, and (5) need for integration training. The instructors were also asked to identify resource needs. The findings suggest that a high level of instruction exists in CAD categories related to traditional drafting tasks. Fewer instructors have ventured into the CAD categories related to design. Many of the instructors who have attempted to teach these categories feel uncertain about their technical skills and skills in integrating advanced CAD categories into their drafting curriculum. From the instructors' responses, it is concluded that a need for resources exists to help instructors to use and integrate CAD categories related to design and customization. / Ph. D.

Intensive fish production in a recirculating aquaculture system facility is a complex bioengineering operation involving a sensitive balance among physiological, water quality, and management components of the overall system. Warm and nutrient-rich wastewater discharged from controlled-environment fish production facilities is a loss of heat energy and nutrients in addition to being potentially harmful to the environment. The operators of such systems need sophisticated management tools if the operation is to be both commercially successful and environmentally friendly. Effluent heat and nutrients can be recovered using hydroponics in a greenhouse attached to the recirculating aquaculture system facility. A computer model was developed to simulate system performance and to help determine design parameters for an integrated fish production-hydroponics system. The aquaculture component of the model predicts (a) fish growth-dependent feeding, (b) diurnal metabolic waste production/accumulation in the fish culture water, and (c) quality, quantity and frequency of wastewater discharge. The hydroponics component computes optimum greenhouse size and models the performance of vegetable plants in terms of nutrient-uptake, water use, and growth. SUCROS and TOMGRO, plant growth models with modifications for water use and nutrient uptake, were used to simulate lettuce and tomato performance, respectively. To validate the plant models, experiments were conducted in a greenhouse utilizing aquacultural wastewater as the hydroponic solution to produce lettuce and tomatoes. Plant growth, water quality (nutrient-uptake), water use, and environmental conditions were monitored. Lettuce and tomato growth was accompanied with significant reductions in nitrogen and phosphorus levels of the wastewater. Water use by plants strongly depended on solar radiation and plant growth stage. At harvest, nine-week-old lettuce weighed 160 g/plant (average) at a density of 40 plants/m². Tomato yielded 2.4 kg/m² after 17 weeks. However, the tomato fruits did not reach maturity during this time. After 20 weeks, the tomato yield was 3.1 kg/m² and some fruits showed maturity. The use of the model as a management tool for making decisions on optimum greenhouse area for a given recirculating aquaculture system size is demonstrated. The effect of fish stocking density and greenhouse heat loss factor on the optimum greenhouse size are also demonstrated. For an optimum greenhouse size, water use and nutrient-recovery from the effluent by lettuce and tomato plants are quantified. / Ph. D.

Shifts in weed populations to herbicide-resistant biotypes are occurring more frequently. In two adjacent field studies from 1991 through 1994, crop rotations and herbicide programs affected control and densities of common lambsquarters, common ragweed, smooth pigweed, redroot pigweed, jimsonweed, goosegrass, stinkgrass, large crabgrass, smooth crabgrass, fall panicum, and yellow nutsedge. Generally, the continuous use of the same herbicide or herbicides with similar selectivities resulted in proliferation of tolerant weed species. Corn, tomato, and soybean yields were affected mostly by crop rotations, rainfall, and weed control. Herbicide rotations or combinations must include herbicides that are efficacious on the target weed species to preclude weed shifts. Approximately 5 million smooth pigweed plants in Painter, VA were treated with imazethapyr or nicosulfuron from 1992 to 1994, and no ALS-inhibitor-resistant plants were identified. Smooth pigweed in Marion, MD and Oak Hall, VA, and livid amaranth in Warren County, NJ were reportedly escaping control from imazaquin or imazethapyr. In greenhouse studies, control of smooth pigweed from Marion and Oak Hall was 3 to 18% by 560 or 1120 g ai/ha imazaquin. Control of smooth pigweed from Painter was 81% by 70 g ai/ha imazethapyr. Control of livid amaranth from New Jersey was 8 to 15% by 560 g/ha imazethapyr. Field, greenhouse, and laboratory studies were conducted on Marion [resistant (R) biotype] and Painter [susceptible (S) biotype] smooth pigweed to characterize herbicide resistance in the R biotype. The R biotype was resistant at high levels to imazaquin and imazethapyr, and was cross-resistant at low levels to rmsulfuron and chlorimuron in the greenhouse. Both biotypes were equally susceptible to ASC-67040, CGA-152005, flumiclorac, halosulfuron, lactofen, metribuzin, nicosulfuron, pendimethalin, primisulfuron, pyrithiobac, and thifensulfuron in field or greenhouse studies. ALS enzyme assays confirmed target site-based resistance to imazaquin, imazethapyr, and rimsulfuron but not to chlorimuron in the R biotype. Metabolism of ¹⁴C-chlorimuron was more rapid in the R than in the S biotype which may explain the low level of whole plant resistance to chlorimuron in the R biotype. The occurrence of herbicide resistance and patterns of cross-resistance in weeds cannot be predicted. / Ph. D.

Seismicity in the central United States has typically been defined by the New Madrid Seismic Zone. Recent discoveries of paleoliquefaction evidence outside that Zone suggest prehistoric earthquake magnitudes in the Wabash Valley far exceed historical events. This geotechnical engineering study of paleoliquefaction sites in southern Indiana and along the Wabash River in Illinois has provided an opportunity to estimate magnitudes and attenuation characteristics associated with strong prehistoric Wabash Valley earthquakes. In-situ soil strength parameters have been measured at 22 sites over an area in the Wabash River drainage approximately 250 km N-S, and 180 km E-W. These parameters have been used in liquefaction susceptibility analyses to estimate moment magnitude (M) and surficial ground motion amplitudes associated with each of four separate paleo-earthquakes. In addition, site response studies based on a semi-theoretical model of eastern North America bedrock motions have been used to compare results of the engineering study with seismological predictions of ground motions associated with large earthquakes in the region. The site response study has been used to develop an attenuation relationship for eastern North America soil sites that mirrors attenuation relationships for the western United States. Minimum magnitude estimates are based on both distribution of paleoliquefaction evidence and on geotechnical estimates of stresses required to induce the evidence observed. A seismic energy intensity approach is then used to help improve the magnitude and surface acceleration estimates. The approach allows revised magnitude estimates by comparing peak accelerations required to induce the liquefaction effects observed to the peak surface motions predicted by the seismological model. The results of this study suggest the Wabash Valley paleoseismic events had magnitudes of M6.9, M7.1, M7.3 and M7.8. Uncertainty exists in the analysis, but these estimates are considered reliable to within 0.25 to 0.5 magnitude unit. The geotechnical estimates of peak meizoseismal accelerations and regional attenuation characteristics were also found to be consistent with seismological predictions. The estimates of seismic parameters based on this study greatly exceed those of all historical events, suggesting the Wabash Valley seismic hazard may be underestimated by local building codes, and that seismic design requirements should be reassessed. / Ph. D.

It is now recognized that reproductive hormones such as estrogen influence not only classical targets (eg. reproductive tissues), but may also act on non-classical target sites such as the immune system. A better understanding of the effects of estrogen on the immune system is of paramount importance since: (i) increasing numbers of women around the world take estrogen-containing oral contraceptives, some times for most of their reproductive life; (ii) estrogen is often prescribed as a replacement therapy to postmenopausal women; and (iii) a large number of pesticides, insecticides, and phytoestrogens (plant-derived estrogens) have been found to have hormone disrupting effects as evidenced by altered development of reproductive and immune functions in wild species. The precise effects of estrogen on the normal immune system are not well known. The overall objective of this work has been to better understand the role of sex hormones on the B cell function of normal mice. It is hoped that this will lead to an improved understanding of the pathogenesis and treatment of immune-related disorders such as autoimmune diseases and cancer. These studies were accomplished by parenteral administration of estrogen to nonautoimmune C57BL/6 mice, a widely used strain in immunology. In these mice, we found that treatment with estrogen, but not 5a-dihydrotestosterone. induced the expression of a wide variety of IgG and IgM autoantibodies and heteroantibodies that are associated with autoimmune and infectious diseases. These include antibodies to cardiolipin and other membrane phospholipids, dsDNA and acetone-killed Brucella abortus strain RB51. Importantly, the expression of anti-dsDNA and anti-cardiolipin antibodies was sustained for several months after the removal of the exogenous source of estrogen. This indicates that the immunomodulatory effect of estrogen is long-lasting. These antibodies have a marginal degree of crossreactivity with other antigens and belong mainly to IgG2b subisotype. These findings were confirmed at the cellular level, where we have shown that estrogen-treated mice have increased numbers of plasma cells in the spleen, and that these plasma cells actively secrete IgM and IgG immunoglobulins as assessed by ELISPOT. Further, higher immunoglobulin yield per cell was evident in estrogen-treated than in placebo-treated controls in the spleen and bone marrow. Interestingly, we found that splenic lymphocytes had an increase in antibody-forming cells for all specificities tested. Active antibody-forming cells from bone marrow preferentially recognized autoantigens, cardiolipin and dsDNA. Functional analysis on the viability of the splenic lymphocytes showed that in vivo exposure to estrogen resulted in: (a) increase in the proportion of cells dying by apoptosis, and (b) an increased proportion of lymphocytes that were actively proliferating as assessed by cell cycle analysis. Culturing of B cells in the absence of any deliberate stimulus showed the B cells from estrogen-treated mice underwent active proliferation and resisted death by apoptosis more compared to controls. We also found that despite the autoproliferative character of splenic B cells, they were able to respond adequately to stimulation with anti-CD40 antibodies, IL-4 and lipopolysaccharides. B cells from mice treated with estrogen had a marked reduction in their susceptibility to apoptosis when cultured in the presence of such stimuli. Together these studies indicate that normal mice exposed to estrogen may express a variety of autoantibodies, show signs of B cell hyperactivity, have defects in susceptibility of B cells to apoptosis as well as the ability to proliferate in the absence of stimulation. It is hoped that these studies would enhance our understanding of the immunomodulatory role of estrogen in health and in a wide range of disorders such as autoimmune and cancer disorders. / Ph. D.

A methodology for progressive failure analysis is presented that is applicable to thin composite structures undergoing large deformations. The method is based on a nonlinear shell theory that accounts for geometric and material nonlinearities. An isoparametric, displacement-based finite element formulation is used to discretize the nonlinear shell equations and an iterative solution is obtained via the Newton-Raphson method. Phenomenological failure criteria are used to predict the onset of damage in the material. Damage is treated using nonlinear constitutive relationships where the material properties are degraded to simulate loss of load-carrying capability. Unlike traditional ply discount models where properties are degraded in an entire ply, the damage models are evaluated at several material points within each ply during the element integration. This allows damage to be localized within an element or ply, and reduces finite element mesh sensitivity since localized damage can develop even in a relatively coarse mesh. An experimental program was conducted to validate the progressive failure analysis. Six foot diameter graphite/epoxy frames—representative of ring stiffeners for rotorcraft fuselage—were tested in quasi-static crush tests that simulated crash loading of the rotorcraft. Excellent agreement was obtained between the analysis and experiments: the analysis successfully predicted the failure load, the magnitude of unloading at failure, and the residual stiffness of the frames after failure for several different frame configurations. Several parameters affecting the progressive failure analysis are investigated including nonlinear solution procedures, material failure data, damage models, and finite element mesh effects. / Ph. D.

The objective of this exploratory study was threefold. First, it was to identify the political risk events that drive the level of equity involvement of multinational enterprises (MNEs) in hotel projects in the Caribbean region. Second, it was to learn how these forces influence specific financial parameters involved in the funding of these projects. The final dimension of the study’s objective was to learn to what extent the stage of tourism development had a moderating effect on these parameters. It was hypothesized that the occurrence of political risk events influence the willingness of multinational hotel companies to participate in hotel projects, given a particular stage of the tourism development. The definition of political risk events is based upon Friedmann and Kim’s (1988) list. Butler’s (1980) Concept of Tourist Area Cycle of Evolution was used as the basis for consideration of the influence of political risk events. Key investment parameters, on which managerial decisions were to be made such as level of investment, mode of investment and risk premium requirements were examined. Methodology for the study was a modified Delphi technique conducted in three rounds of panel considerations. Panelists were senior executives, drawn from the Caribbean Hotel Association’s Corporate Forum and from international hotel investment and consulting firms. These executives were asked indicate what they perceived the level of risk associated with each variables to be. In addition, they were asked to rank 12 Caribbean countries in terms of the level of tourism development. During the third round their tasks were to rate the variables again and to indicate, based on rating scales provided, how they would invest in three designated countries: Haiti, the British Virgin Islands and Puerto Rico. Analysis of the results of the panel considerations indicated that when political risk was perceived to be high, investment levels were low and the associated risk premiums were high. The non-equity mode was preferred for all countries. Results of the study should be useful to the decision makers in the hospitality industry, and to policy makers and scholars of international tourism development. / Ph. D.

Concurrent Engineering has brought much attention in recent years to engineering design and its impact on issues such as costs, cycle-time, quality, and other life-cycle processes. Coupled with global markets and the rapid rate of technology advancements, the need for improved methods and supporting tools for engineering design is significant. To date, advances for engineering design have predominantly focused on tasks that are well into the latter stages of product development. Advances for early design (where over 60% of lifecycle costs are committed) still remain largely investigational, specialized, and rarely consider the requirements for functional abstraction and detail necessary in a concurrently engineered development process. In general, methodologies have taken either a top-down or a bottom-up approach to design, and as such, have virtually guaranteed the continued separation of abstraction and detail during conceptualization. This research proposes a methodology based on the blending of top-down and bottom-up approaches. Toward this end, a framework for integrated conceptual design is presented. Three mechanisms are central to the framework definition, including: 1) a functional modeling environment supporting concept model building, knowledge capture, and reuse, 2) a components knowledge-base supporting configuration, and 3) an integrated design domain accessible from the functional modeling environment interconnecting tools, analysis routines, and data sources necessary for design synthesis, analysis, and evaluation. Based upon an object-oriented paradigm and semantic reasoning, the framework for functional modeling and configuration has been designed in detail. A Windows-based user interface has been prototyped and enables designers to both visualize and compose conceptual models using a building-block approach. An example design problem in the domain of power conversion systems is provided and demonstrates the methodology for technologically sophisticated products where conceptualization crosses electrical, mechanical, and software domains (mechatronics). A use case model and object class diagrams describe and document the framework function and architecture. / Ph. D.