A preliminary in vitro digestion study was conducted to investigate the effect of roughage level in the diet of steers producing fecal waste, drying method and NaOH treatment, on in vitro dry matter digestibility of cattle fecal waste. All three factors significantly (P<.01) affected in vitro dry matter digestibility of cattle fecal wastes. In vivo digestion trials were conducted to study the apparent digestibility of untreated and NaOH treated fecal waste from steers fed rations containing different roughage levels. In each of two trials, three yearling steers were fed a 50% roughage ration and three were fed a 10% roughage ration. High- and low-roughage rations contained 13.4 and 13.6% crude protein, dry basis. Apparent digestibility of dry matter was 67% for the steers fed the high-roughage and 78% for those fed the low-roughage ration. One-half of the feces from steers fed each ration was treated by adding 3% NaOH (w/w), wet basis. All feces were then dried at 120 C for 24 hours. Dried, untreated high-roughage (UHR) fecal waste, treated high-roughage (THR) fecal waste, untreated low-roughage (ULR) fecal waste, and treated low-roughage (TLR) fecal waste contained 18.0, 14.9, 20.7 and 17.0% crude protein and 20.5, 17.8, 15.7 and 12.4% crude fiber, dry basis, respectively. A digestion trial was conducted with 30 wether lambs fed the following rations: (1) 100% basal, (2) 75% basal and 25% UHR fecal waste, (3) 75% basal and 25% THR fecal waste, (4) 75% basal and 25% ULR fecal waste, and (5) 75% basal and 25% TLR fecal waste. The basal contained 30% roughage and analyzed 11.7% crude protein, dry basis. Apparent digestibility of proximate and Van Soest components in the basal ration was generally decreased by substitution of all kinds of fecal waste, with the exception of fiber components in the ration containing NaOH treated low-roughage waste. TDN and metabolizable energy content of all the rations containing dried cattle fecal waste were lower (P<.01) than for the basal ration. Apparent digestion coefficients for cattle fecal wastes, calculated by difference, were: 24.3, 52.9, 52.9 and 67.0 for dry matter; 32.1, 20.8, 49.1 and 29.1 for crude protein; and 3.2, 43.0, 32.6 and 60.1 for crude fiber in UHR, THR, ULR and TLR waste, respectively. In general, apparent digestion coefficients for ULR waste were much higher than those for UHR waste. Untreated high-roughage waste appears to be of limited protein and energy value for refeeding to ruminants. However, untreated low-roughage waste appears to have considerable refeeding value, being comparable to good quality orchardgrass hay with respect to digestible protein and energy content. NaOH treatment of high- and low-roughage cattle fecal wastes resulted in a significant increase in apparent digestibility of most components studied. The increases in apparent digestibility of fiber components of the NaOH treated wastes were sizable, especially for the high-roughage waste. Addition of the high level of NaOH used in this study may not be economically feasible. However, addition of lower levels, as have been used in treatment of other low quality roughages, may result in sufficient improvement in nutritive value to be economically feasible. / Doctor of Philosophy

Three digestion trials were conducted with six yearling steers to study apparent digestibility of fecal waste from steers fed a ground, high roughage finishing ration. In the first trial steers were fed a basal ration containing approximately 50% roughage. The ration contained 14.3% crude protein and 22.8% crude fiber, dry basis. Apparent digestibility was 65% for dry matter and 66% for crude protein. The feces collected during this trial were dried in a forced air oven at 120 C and ground. The chemical composition of the feces was 13.2% crude protein, 31.4% crude fiber, 2.8% ether extract, 5.4% ash, 38.8% NFE, 70.9% cell walls and 44.8% ADF, dry basis. For trials 2 and 3 a switch-back design was used. In each trial one-half of the steers were fed the basal and the remainder were fed a ration in which dried cattle feces collected during trial 1 were substituted for 20% of the basal ration. / Master of Science

Thirty-eight farm sales occurring over a five year period in two adjacent Coastal Plain counties and 84 farm sales in a Piedmont county were studied in an effort to identify and analyze factors affecting sale price. The effect of soil productivity and landscape characteristics were of particular interest in this study. The effect of soil productivity on farm sale values was found to be significant in both study areas but was somewhat Overshadowed by other considerations. Factors affecting sale price in the Coastal Plain study area were: (1) inflation, (2) location, (3) proportion of farm in cropland, (4) soil productivity, and (5) value of improvements. Eighty percent of the variation in sale price per hectare could be accounted for by a multiple regression equation which included variables representing each of these factors. Factors affecting sale price in the Piedmont study area were: (1) value of improvements, (2) proportion of farm in cropland, (3) inflation, (4) soil productivity, and (5) size of flue-cured tobacco allotment. A multiple regression equation including variables representing each of the factors accounted for slightly less than 50% of the variation in sale price per hectare. The effect of tobacco allotment on sale price was found to be considerably diminished from that found by previous studies. Crop yield ratings were more highly correlated with sale price per hectare than were SCS land capability classes among the Piedmont farm sales. The opposite was true for Coastal Plain sales. Classes I, II, and III lands were found to be positively correlated with sale price per hectare while Classes IV, VI, and VII were negatively correlated with sale price per hectare in the case of Coastal Plain farm sales. None of the three land capability classes were significantly correlated with sale price per hectare in the case of the Piedmont farm sales. A weighted average of all land classes occurring on each farm was found to be significantly related to sale price. Crop yield ratings for forages, soybeans, and small grains were found to be significantly related to sale price in the case of the Piedmont farms. Yield ratings for small grains were also found to be significantly related to sale price in the case of the Coastal Plain soils. Indices of soil productivity were found to be more highly correlated with sale price among the Coastal Plain farm sales than was the case among the Piedmont sales. The soils and topography of the Coastal Plain are more conducive to intensive crop production. In both study areas, a closer relationship was found between soil productivity and sale price in those soil associations which were well suited to intensive cropping. Farm sale prices rose rapidly in both study areas during the five year period studied. Farm sale values increased at an average annual rate of 47% in the Coastal Plain sales. In both study areas, about 2% of all farms were involved in bona fide sales within a given year. / Ph. D.

A historical overview of the principalship is the story of a position in constant flux, responding continuously to a changing society and the demands that those changes bring to bear on the educational system. If the 1960s and early 1970s can be described as periods of uncertainty for the principalship, the late 1970s and early 1980s can be recognized as the period when the principalship came into its own, as it gained recognition as a key position in the determination of effective schools. Research studies conducted during the 1960s emphasized the need for the principals to assume a leadership role, with an emphasis on management of personnel resources for performance, morale, and productivity. Studies during the 1970s and early 1980s revealed insights into the nature of the principalship, and focused on those characteristics exhibited by effective principals. Faced by ever present societal demands for accountability, efficiency, and effectiveness at a time when the economy was on the downswing, coupled with constraints being placed upon their management prerogatives, principals were faced with many challenges during the 1970s and 1980s. During that period, research findings revealed that the nature of the principalship was reminiscent of a rapidly changing society. Time was a scarce commodity for principals who lived a fast paced, complex, and highly personal role, characterized by brevity of tasks, continuous face-to-face encounters, which often required on-the-spot decision making. Effective principals were described as diverse, proactive, people-oriented, risk takers, who were, for the most part, non-beleaguered and able to deal with the highly energized and demanding job. Another common characteristic of principals was their major stressor--dealing with teacher performance and evaluation, particularly when interpersonal conflicts and forced resignation were eminent. The second greatest stressor was perceived to be a threat to the physical status and/or job security of principals. In order to deal with their highly demanding and stressful jobs, principals were observed employing various coping mechanisms that enabled them to maintain control in often ambiguous roles. / Ed. D.

根據民國88至90年教育部統計處出版的資料，本文首先利用資料包絡分析法評估個別國立大學之技術效率；然後，以Tobit截斷迴歸模型探討可能造成國立大學技術效率差異之因素。技術效率評估結果顯示，在不同投入產出組合下，國立大學整體的技術效率平均值介於115.30﹪與122.17﹪之間，並且，規模效率平均值大於1，代表：台灣國立大學在產出上仍存在改善空間；且造成其不效率之原因，主要歸咎於純技術無效率及規模無效率。Tobit截斷迴歸實證結果顯示，大學部學生比例與技術效率之關係為負向，競爭程度的提高可提昇國立大學的經營效率，師範與科技類大學的技術效率低於其他大學。 關鍵詞：技術效率、資料包絡分析法、Tobit截斷迴歸方法 JEL Classification：L83、M11、M21

The empirical literature on mergers, market power and collusion in differentiated markets has mainly focused on methods relying on output and/or panel data. In contrast to this literature we suggest a novel approach that allows for the detection of collusive behavior among a group of firms making use of information on the spatial structure of horizontally differentiated products. By estimating best response functions using a spatial econometrics approach, we focus on differences in the strategic interaction in pricing between different groups of firms as well as on differences in price levels. We apply our method to the market for ski lift tickets using a unique data set on ticket prices and detailed resort-specific characteristics covering all ski resorts in Austria. (authors' abstract) / Series: Department of Economics Working Paper Series